Astronomers’ Observing Guides
For other titles published in this series, go to www.springer.com/series/5338
Brian Cudnik
Lunar Meteoroid Impacts and How to Observe Them with 116 Illustrations
Brian Cudnik Houston, TX USA
[email protected] Series Editor Dr. Mike Inglis, BSc, MSc, Ph.D. Fellow of the Royal Astronomical Society Suffolk County Community College New York, USA
[email protected]
ISBN 978-1-4419-0323-5 e-ISBN 978-1-4419-0324-2 DOI 10.1007/978-1-4419-0324-2 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2009930463 © Springer Science+Business Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Foreword
Foreword
The genesis of modern searches for observable meteoritic phenomena on the Moon is the paper by Lincoln La Paz in Popular Astronomy magazine in 1938. In it he argued that the absence of observed flashes of meteoritic impacts on the Moon might be interpreted to mean that these bodies are destroyed as luminous meteors in an extremely rarefied lunar atmosphere. The paper suggested the possibility of systematic searches for such possible lunar meteors. With these concepts in mind, I was surprised to note a transient moving bright speck on the Moon on July 10, 1941. It appeared to behave very much as a lunar meteor would – except that the poorly estimated duration would lead to a strongly hyperbolic heliocentric velocity. Thus, the idea of systematic searches for both possible lunar meteors and meteoritic impact flashes was born. It was appreciated that much time might need to be expended to achieve any positive results. Systematic searches were carried out by others and myself chiefly in the years 1945–1965 and became a regular program at the newly founded Association of Lunar and Planetary Observers, or ALPO. It was obvious that such searches were best attempted with the lunar background as faint as possible. Thus, one opportunity was on the earthlit regions soon before or after New Moon. In concept the best chance then was with a crescent as close to New Moon as possible, but in practice there were complications: the need for a dark sky free of dawn or twilight lighting and the requirement that the Moon be not too near the observer’s horizon. A second kind of opportunity came during lunar eclipses, with a preference for the darker eclipses. It was appreciated that a single observation of an apparent meteoritic impact (stationary flash) or apparent lunar meteor (moving bright spot) was inconclusive. It was necessary to have a second observation of the same feature at the same time and at the same position on the Moon. The ALPO observers did indeed report many flashes and many moving bright specks. However, there was never the essential duplicate confirming observation. In truth it proved very difficult to get independent and widely separated amateur observers to carry out simultaneous observations of the whole earthshine or a selected specific portion thereof. (An excellent time for an observer in Illinois in terms of dark sky and Moon’s altitude may be useless for his cooperating colleague in Colorado or Georgia). Of course, we now know that the lunar atmosphere is far too tenuous to permit lunar meteors to exist. We also know that the very great majority of meteoritic impact flashes will be too faint and too brief in duration to be detected by telescopic visual observers. It remained for Brian Cudnik and others to achieve the first confirmed observations of impact flashes during the Leonid epoch in 1999. It is easy, of course, to dismiss the old unconfirmed observations as nothing but illusions. However, some of the observers, including myself, were confident of having seen something, regardless of ideas about its interpretation. Very few of the v
Foreword
objects recorded can be terrestrial meteors coming from the direction of the Moon. Perhaps there is a very slight chance that the observers witnessed some unknown or poorly studied phenomenon.
vi
Walter H. Haas
Contents
Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
xi
Chapter 1 Widespread Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Brief History of Impacts in the Early Solar System . . . . . . . . . . . . The Impact that Built the Moon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Brief Look at Some Other Significant Impacts . . . . . . . . . . . . . . . Are Impacts Still Happening Today? . . . . . . . . . . . . . . . . . . . . . . . . . .
3 3 6 8 15
Chapter 2 Lunar Impact Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Maria vs. Highland Craters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Young vs. Old Craters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appearance vs. Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How to Recognize Different Types of Features . . . . . . . . . . . . . . . . .
17 17 18 21 24
Chapter 3 Remarkable Collisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Historic Impacts of Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The AD 1178 Lunar Impact Event . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Taurid Complex Objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The “Lunar Flare” Event of 1953 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27 27 28 29 30
Chapter 4 Possible Impact Phenomena (PIPs) . . . . . . . . . . . . . . . . . . . . . . . . . . Probable Appearance and Classification of PIPs . . . . . . . . . . . . . . . . 1955–2008: Additional Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A catalogy of Coordinated and/or Shower-Related Events as Documented by ALPO and IOTA . . . . . . . . . . . . . . . . . . . . The Perseid Meteor Shower: 12–14 August 2002 . . . . . . . . . . . . . . . .
35 35 43
Chapter 5 Beyond the Moon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Target Jupiter: The Shoemaker-Levy 9 Multiple Impact Event . . . . Crater Types and Morphologies Throughout the Solar System . . . Other Historic and Probable Impact Sightings on Other Worlds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
71 72 74
Contents
Part I The Moon and Meteoroids
48 59
92 95
vii
Part II An Observer's Guide to Lunar Meteor Impacts, Past and Present Chapter 6 Guide to Observing Impact Features on the Moon . . . . . . . . . . . . . Maria vs. Highland Cratering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Young vs. Old Craters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appearance vs. Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . How to Recognized Different Types of Features . . . . . . . . . . . . . . . A “Top 100” List of significant Impact Structures to Observe . . . . A Link to Astronomical League Observing Clubs Related to the Moon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contents
Chapter 7 Impacts Today . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lunar Transient Phenomena and Lunar Meteoroid Impacts . . . . . Some Possible Causes of LTP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Narrowing Down the Causes of LTP to Three: Electrostatic Levitation, Tidal/Thermal Stresses, and Meteoroid Impacts . . . . .
viii
Chapter 8 Lunar Impact Observation Programs . . . . . . . . . . . . . . . . . . . . . . . . Past Campaigns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Current Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Space Missions that Impacted/Will Impact the Moon . . . . . . . . . . Mission Statement of the ALPO Lunar Meteoritic Impact Search Program: A Vision of Lunar Impact Research . . . . . . . . . . .
99 101 102 104 106 109
110 111 111 112 114
121 121 122 123
125
Chapter 9 Observing Impacts as They Happen (with Contributions by Many Members of the International Occultation Timing Association) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Geometric Considerations and Preparations Needed for Making Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Techniques for Visual Observations . . . . . . . . . . . . . . . . . . . . . . . . . Techniques for Video Observations . . . . . . . . . . . . . . . . . . . . . . . . . . Twelve Examples of Observer Equipment Setup and Use . . . . . . . Some Examples of Products and Resources . . . . . . . . . . . . . . . . . . . Putting it All Together: A Lunar Meteor Observing Plan . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
128 129 130 133 142 150 152
Chapter 10 Finding Collisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Automated Impact Detection Software . . . . . . . . . . . . . . . . . . . . . . . The Use of Registax for Automated Lunar Meteor Detection . . . . Increasing the Probability of Detection with LunarScan . . . . . . . .
153 153 154 158
Chapter 11 Spurious Flash or True Impact Event? . . . . . . . . . . . . . . . . . . . . . . . How to Identify True Lunar Meteoritic Impact Events . . . . . . . . . . GLR Dark Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Validation of Lunar Flashes: A Network of Observers for Simultaneous Patrols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification of the Flash Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . Single Observer Validation of Lunar Meteor Impacts . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
165 165 165
127
168 171 175 178
Chapter 12 Professional and Amateur Collaboration . . . . . . . . . . . . . . . . . . . . Introduction: Pro-Am Collaborations . . . . . . . . . . . . . . . . . . . . . . . . Examples of Professional Research in Lunar Meteor Impacts . . . . Advances in Amateur Lunar Meteor Observations . . . . . . . . . . . . .
181 181 182 189
Appendix A References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Appendix C Impact Candidates Observed by ALPO/LMIS Likely to be Cosmic Ray Hits or Other Spurious Phenomena . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1999: Lunar Geminid Impact Candidates . . . . . . . . . . . . . . . . . . . . May 2000: Lunar Eta Aquarid Impact Candidates . . . . . . . . . . . . . The Leonid Meteor Storms of 2001 . . . . . . . . . . . . . . . . . . . . . . . . . The Perseid Meteor Shower: 12–14 August 2002 . . . . . . . . . . . . . .
209 209 209 210 211 211
Appendix D A Simple Method for Timing Videotaped Occultations (and Lunar Meteor Impact Flashes) . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . System Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . System Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Timing the Occultation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
215 215 215 215 217 219
Contents
Appendix B Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Appendix E Equipment Checklist and Vendors . . . . . . . . . . . . . . . . . . . . . . . . . . 221 Appendix F Details of Shortwave Time Signals for Astronomical Timings . . . 225 Appendix G Stellar Resources for Comparison and Calibration . . . . . . . . . . . . 229 Finding Limiting Magnitudes for Visual and Video Camera Observation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 Standard Stars for Extinction Correction and Flux Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Appendix H Impact Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
ix
Introduction
Introduction
On November 18, 1999, the first scientifically confirmed lunar meteoritic impacts were recorded in the form of pinpoint flashes that resulted from the collision of the Moon with debris within the Leonid meteoroid stream. (The locations of the first seven of these appear in Fig. 1; an image of the “A” impact appears in Fig. 2). These fragments, traveling at 71 km/s (44 mi./s) impacted the Moon with such force that the optical flashes of these explosions were visible from nearly a quarter of a million miles away. Similar events were observed during the 2001 Leonid display, with two events each confirmed independently by at least two observers. In all of these cases, the impacts had the appearance of stars, ranging in brightness from third to eighth magnitude, and each appearing for less than 1/20th of a second. These events resulted in the rebirth of the lunar meteoritic impact observing section of the Association of Lunar and Planetary Observers and have revived interest in lunar impact phenomena and lunar transient phenomena (LTP) (also called transient lunar phenomena). LTP is a term that refers to lunar change. The Moon was once thought to be a completely dead world geologically speaking, with no observable change. However, there are many astronomers, some of whom are very experienced observers, who have no doubt that they observe change on the Moon in the form of fogs and hazes; localized color changes; flashes of various brightnesses, sizes, frequencies, and colors; and other dramatic visible manifestations. The problems with these observations are that they cannot be easily reproduced and also often lack a second,
Fig. 1. The historic moon map showing the locations of the first six Lunar Leonid Meteors of 1999 (Courtesy of David Dunham and IOTA)
xi
Introduction xii
Fig. 2. The flash of Impact A, observed visually by Brian Cudnik, as imaged by Dr. David Dunham who was videotaping at the time. Note the edge of the faintly Earthlit limb of the Moon (Courtesy of David Dunham)
independent source to confirm them. Yet, it is impossible to simply dismiss these observations. To add validity to these kinds of sightings, groups of people within several astronomical organizations have made the effort to coordinate the observers and standardize the methods in the hopes of putting together a catalog of scientifically confirmed LTP events. This author is interested in the varieties of LTP attributed to one particular cause – meteoritic impacts. This one cause, depending on local circumstances, may produce visible signatures in many forms, the most common being the single, very brief, point-like flash on the unilluminated (night) side of the Moon (point flashes are not so easily seen on the day side of the Moon due to the bright lunar surface). More rarely, clouds of debris kicked up by an impact event may be observed on the dayside of the Moon. Many reports of LTP events resembling meteoritic impacts are documented from at least the 1100s. One of the most famous of these reports includes the possible sighting of an impact event by a group of monks in June 1178. Another, much more recent event is the 1953 Stuart “Lunar Flare” (more details on these and other notable impacts are provided in Chapter 3). Proof of the Lunar Flare’s occurrence
Introduction
was thought to have been found early in 2003 by researchers studying Clementine imagery for signs of a fresh impact crater in the location of the event. However, this was later refuted by careful measurement of the crater’s position versus that of the flare, as well as the appearance of the same “fresh crater” in a 1919 print of the same region of the Moon – 34 years prior to the observation of the Stuart flare. Documentation of many events like the ones mentioned above appears in NASA’s Technical Report R-277. An estimated 10–12% of these LTP events are probably manifestations of meteoroid impacts. Such events were defined by their appearance: star-like, brief, pinpoint flash, appearing on the night portion of the Moon, and are presented in the catalog in Chap. 4. Also considered for inclusion in the catalog are reports of well-defined dust clouds, possibly from impacts on the sunlit surface of the Moon. Meteoritic impact events are scientifically interesting for many reasons. According to modern geologic theory, almost all of craters now seen on the Moon were formed by the impact of debris leftover from the formation of the solar system during its early history (between 3.9 and 4.5 billion years ago). The rate of crater formation decreased rapidly near the end of that time, and then slowly leveled off as the leftover material continued to crash into the Moon. Since debris still occupy interplanetary space in the form of dust, meteoroids, and asteroids, impacts on the Moon and other solar system bodies are still happening. The thick atmosphere of the Earth protects our planet from most impacts, but the Moon has no atmosphere (except for an extremely tenuous cloak of sodium), so any object on a collision course with the Moon impacts our satellite directly. The smallest objects merely kick up small dust clouds, well below the resolution of ground-based telescopes, but larger objects occasionally do collide with the Moon, generating a very brief, but visible flash of light detectable by ground-based instrumentation (if the event occurs on the night side of the Moon as seen from the Earth). These happen about as often as a fireball (or bolide) in Earth’s atmosphere, but due to the randomness of the events (except during times of annual meteor showers) it is very difficult to obtain an observation of quality, let alone a scientifically confirmed one. Even if an impact was unambiguously identified, it is far from certain the origin of the impacting object, since only a dimensionless point flash is seen. It is up to dedicated organizations, such as the Lunar Meteoritic Impact Search section of the American organization of the Association of Lunar and Planetary Observers, the Unione Astrofili Italiani Lunar Section, the Geologic Lunar Research Lunar Impact Section (also in Italy), the British Astronomical Association, and other such groups, to set up and maintain systematic programs to regularly monitor the Moon for such events. One example of the opportunistic nature for learning important space physics and chemistry happened in the 1990s when the impacts of pieces of comet Shoemaker-Levy with Jupiter were observed by the Galileo probe’s photopolarimeter radiometer. Scientists were able to get lots of information from the impact, including light curve profiles, energy of the impactor, duration of the impact flash, and how large the plume grew. Resolution in both space and time of the visible flash provided much needed information about the impact dynamics and provided real reference frames for the mathematical modeling of these high-speed collisions in space. Improved physical modeling of the impact dynamics can provide estimates of the size and mass of meteoroids from both meteor streams and sporadics. We will be able to then get an estimate of the numbers of objects with at least several orbiting in the vicinity of the Earth. If a lunar meteoroid impact is bright enough,
xiii
Introduction xiv
low-resolution spectra can be obtained, providing composition information of the lunar surface. Such an attempt was made when Lunar Prospector was intentionally crashed into the Moon’s surface to search for water ice in shadowed craters at the lunar poles. Although the Prospector’s impact flash was not seen, a great deal of interest was stirred up in the professional community by the potential information that this encounter would generate. Finding large reservoirs of water ice would dramatically aid human exploration of space. Even though the Prospector was a large object, it was traveling at a very slow speed compared to most interplanetary impacts. Natural meteoroid collisions with the Moon can produce a far greater energy release due to their higher impact velocities and by doing this they can serve as a spectral probe for lunar water or other constituents. The European Space Agency’s SMART-1 spacecraft crashed into the dark portion of the Earth-facing lunar hemisphere on September 3, 2006. A very brief impact flash was visible in the infrared with the 3.6-m CFH Telescope in Hawaii. More information on the results of the SMART-1 impact is presented in Chap. 8. The spacecraft LCROSS is scheduled to be driven into the lunar polar region in mid-2009, out of view of Earth-based observers, except for any plume or cloud kicked up by the collision that rises high enough into the sunlight to be visible through ground-based telescopes. The purpose of this book is to assemble relevant material concerning the lunar meteoritic impact phenomena into one central reference and provide the amateur astronomer with resources to enable effective participation in an ongoing campaign to learn more about this specific form of LTP. In addition, this book provides a brief overview of cratering in general and how craters of different forms happen on different planets and relates this to the currently accepted history of the solar system. We also look at the “crater main sequence” on the Moon and discuss how this relates to the size of the impactor and the age of the impact crater and demonstrate how amateur astronomers can identify craters of various shapes, sizes, and ages. This book draws material from a large number of sources and includes procedures, directions, and advice to carry out a successful search for lunar meteoritic phenomena. These sources include experienced lunar meteoritic and asteroidal occultation video observers from the International Occultation Timing Association (IOTA) and the Association of Lunar and Planetary Observers (ALPO), the professional literature, and books of various levels detailing the dynamics of impact phenomena and crater morphologies. Relevant web sites are included as well. It takes a coordinated effort by many individuals to make a successful lunar meteoritic impact-monitoring program. At least two independent observers have to document an impact event in order to validate it. To get good statistics on the number of events per unit time, a much larger number of confirmed events are needed. So far, as of September 5, 2008, 129 confirmed meteoroid impacts on the Moon have been documented by the meteoroid environment personnel at the NASA Marshall Space Flight Center, and these provide an excellent start to look at the impacts from a statistical point of view. I hope that this book, along with the efforts of interested astronomical organizations, will motivate observers to add to the success of NASA, IOTA, ALPO, and others by committing to a long-term effort to monitor the Moon for these events. We need help, not only during the best annual showers of the year, but also during the lesser showers and the week or so per month when the Moon is favorably placed to intercept sporadic interplanetary
Introduction
debris. One desired outcome is the significant reduction of isolated observations – the occasional observer that reports seeing something happen on the Moon but has no one to confirm that observer’s candidate event. The report is tantalizing, but unfortunately of little more value than that. Of more value is the single observer that has taken steps to minimize spurious events and increase the quality of one’s observations by following guidelines presented here and elsewhere.
xv
Part I
The Moon and Meteoroids
Widespread Evidence
Widespread Evidence
Chapter 1
A Brief History of Impacts in the Early Solar System Before we discuss the impacts of lunar meteoroid and the ways to observe them, we will look at the impacts of meteoroid from a historical perspective. More specifically, we consider the impact evidence in the form of meteoroid and asteroid scars or craters that dot the surfaces of most worlds with solid surfaces. While investigating the formation of Solar System and history, we see evidences of how it came into existence: from the Sun itself to the most distant comet, we see signs that give us clues to the story of the beginning of the Solar System (Figure 1.1 shows an artist's impression as an impact that may have occurred 65 million years ago). We see obvious patterns in the Solar System including the fact that all eight major planets orbit close to a common plane and in the same direction. Most of the planets rotate in the same direction as they orbit, and the planes of their equators are fairly close to the planes of their orbits. Their satellites orbit in the same plane as the equator of their host planet. We also see patterns in what makes up various worlds at various distances from the Sun. Rocky and metallic material dominate the composition of the planets and asteroids in the inner Solar System (to about 5 AU or 5 times the Earth-Sun Distance). Ice and rocky material dominate beyond 5 AU. Most of the lighter elements (hydrogen and helium) exist beyond the asteroid belt. We also see that all the airless places (with a few very interesting exceptions) have heavily cratered surfaces, which provides another clue into the early history of the Solar System. Observations of various objects outside the Solar System also support our current understanding of how the Solar System evolved. There are other planetary systems and dust disks around other stars. The dust disks and protoplanetary disks show various stages of development that are in line with our theories. Computer simulation results also confirm our theories. The following summary of our assumption of the formation of the Solar System covers a time period ranging from 4.8 billion to 3.9 billion years ago. The Sun was thought to have started nuclear fusion at its center (converting hydrogen to helium, thereby releasing lots of energy, the process that makes a star a star by definition) about 4.6 billion years ago. According to the most widely accepted theory, the Solar System started out as a huge cloud of gas and dust, called the Solar Nebula (the process is summarized in Fig. 1.2). The Solar Nebula was made up of mostly hydrogen, with some helium and
3
Widespread Evidence
Fig. 1.1. Artist’s depiction of an asteroid impacting Earth
a few percentages of heavier elements. These elements were mostly silicates, aluminum, iron, calcium, oxygen, carbon, and nitrogen. A shockwave from a nearby supernova shook up the cloud and started a slow contraction. This cloud also began to rotate, very slowly at first but gaining speed as the nebula shrunk. The inward pull of gravity was balanced with the outward centrifugal force of the spinning nebula, and an accretion disk was formed. At the center of the disk, materials that would eventually form the Sun was collected. The temperature of the inner portion rose dramatically, due to the change of energy from gravitational to heat. Farther than about 1.5 AU from the protosun in the accretion disk, temperatures dropped dramatically. Grains within the nebula became coated with various ices of water, carbon dioxide, ammonia, and methane. The temperature began to decrease in the inner Solar System as tiny particles started to coalesce (stick together); metallic and rocky clumps began to form. These accumulated to form larger particles, with the largest beginning to exert a sufficient gravitational force to attract more materials and grow more rapidly. They collided repeatedly to form planetesimals of the size of present-day asteroids. The largest of these planetesimals became planetary embryos which continued to grow. At least five worlds, and possibly seven, from the size of the Moon to the size of the Earth, arose from this process, with remaining planetesimals providing the raw material for the countless impact events that characterize this period of heavy bombardment. In the meantime, the outer planets developed along with their major satellites. Weaker solar gravity resulted in larger planetary embryos, with the two largest (soon to become Jupiter and Saturn) sweeping up most of the debris and smaller planetary embryos. The remaining two (Uranus and Neptune) did not have as 4
1. Rotating Cloud of Gas and Dust
Widespread Evidence
Protosun
2. Inside the cloud, material collapses to form an accretion disk, with the proto-sun at center
Disk-Volatiles
Planetesimals-Refractories 3. Planetesimals form and coalesc; these sweep up the material around then. The Jovian planets form from volatiles and the asteroids and Kuiper Belt objects are the left overs
4. Present-Day Solar System
Fig. 1.2. Diagrams illustrating the process of formation of the solar system
much gravity to accumulate material (and the amount of material might have been less than what Jupiter and Saturn had to work with), so they remained smaller. The four planets of the outer Solar System swept the remaining material and eventually pushed much of it beyond Neptune’s orbit where it resides to this day as the Kuiper Belt. In the center of the Solar System, the Sun’s nuclear furnace ignites, sending shock waves throughout the Solar System (as is evident from the shock heating in the cometary samples returned by the recent Stardust mission) and dispersing the Solar Nebula. The inner planets are heated and melted: from the outside through large number of collisions, and within by radioactive heating. This leads to completely molten worlds, where almost all the metal sinks to the center and the lighter material floats to the top. This process is called differentiation and is considered to have occurred in all the planets and major satellites The Era of Heavy Bombardment, evidence of which exists to this day in the heavily cratered surfaces of the Moon and many other worlds, lasted from about 4.6 to 4.0 billion years ago. There is also evidence of a cataclysmic bombardment phase of the Earth and Moon centered around 3.9 billion years ago. Subsequently, the impact flux, or the number of impacts on a given planet in a given time frame, dropped to low levels which continue to this day (Figs. 1.3 and 1.4). 5
Widespread Evidence
Fig. 1.3. Plot of impact flux (rate of impact per unit time) versus time (the “4.5” means 4.5 Billion years ago; courtesy of the Lunar and Planetary Laboratory, the lunar cataclysm page: http://www.lpl.arizona.edu/SIC/impact_cratering/lunar_cataclysm/Lunar_Cataclysm_Page)
The Impact that Built the Moon Shortly after the outermost part of the Earth cooled, a Mars-sized body was on a collision course with the Earth. This collision, at a shallow angle, resulted in reforming the Earth to the one we know today, and forming the Moon. However, this theory has been proposed relatively recently and subsequently affirmed to be valid. Three theories have dominated our attempts to explain the origin of Moon. These theories are summarized below: •• The Moon was spun off Earth early in their history-the fission hypothesis. The major drawback of this theory includes the physical impossibility of the event as well as the distinct chemical differences between the Earth and the Moon. •• The Moon formed alongside the Earth, like the major satellites of the outer Solar System. This theory is also inconsistent with what we actually observe in terms of composition differences and orbital dynamics •• The Moon was captured by the Earth. Physically, this is very difficult, and even if it were to occur, the present-day orbit of the Moon is too circular (it would have been much more elliptical). A fourth theory, an advanced one, involves the impact of a Mars-sized object. According to this theory, the Mars-sized object obliquely impacted the Earth, stripping off several percentages of the Earth’s mass. This mass formed a ring around the Earth which ultimately accreted to form the Moon. The material consisted of parts of the mantles of Earth and the impactor, which is supported by data from Apollo. At first, the Moon formed very close to the Earth, but tidal interactions between the Earth and the Moon resulted in the Moon slowly migrating outward to its present distance. 6
Planetestimals
More Planetestimals
Impact/Radioactive melting/differentiation Mars-sized and cooling, collisions object fragmentation, cratering
Material thrown-out clumps together to form the Moon
Widespread Evidence
How the Earth-Moon system was built
Collision
Earth-Moon system
Fig. 1.4. Diagrams showing the process of formation of the Moon
High-powered computer simulations of this collision were run repeatedly, with the starting conditions changed slightly during each run. After a number of iterations, one simulation produced a Moon that closely matched the real Moon. This confirmed the proposed theory. In addition, it was thought that Mercury underwent a similar catastrophe, but in this case the outermost layer was stripped off without forming a satellite, leaving a planet high in metal content. The Moon’s orbital shape was thought to have evolved considerably in only a few hundred years; the Moon started much closer to the Earth, orbiting it in only a few days. Intense bombardment from meteoroids and asteroids around 4.45 billion years ago partially melted the crust, forming the lunar magma ocean. This ocean solidified and crystallized 4.3 billion years ago, but asteroids and meteoroids continued to impact the lunar surface. About 3.8 billion years ago, the last of the large, frequent asteroidal impacts occurred, resulting in impact basins up to 1,000 km (600 mi) wide on the Moon (and possibly on Mercury, Mars, Callisto, and other worlds). Later, the impact frequency dropped to very low levels, which continues to this day. 7
Widespread Evidence
Lunar volcanism was active for up to one billion years after its formation. Molten rock flowed onto the surface of the Moon through cracks in the crust, spreading out and filling low regions in impact basins. This lava cooled quickly and formed fine-crystalled basalts and basaltic lava plains. Volcanic activity ceased significantly around 3 billion years ago, with low-level activity persisting until 1 billion years ago, after which the Moon became completely dead, in geological terms. As the Moon orbited the Earth, the gravitational interaction between the two resulted in tidal effects (the bulge of the Earth’s oceans), much greater than what is experienced today. The friction between the solid Earth and the tidal bulge acted as a brake to slow down the Earth’s rotation (the day lengthened to 18 hours by 900 million years ago, and 22 hours by 370 million years ago). At the same time, the bulge tugged on to the Moon, speeding up its orbit, causing it to spiral outward until it reached its present distance which averages around 384,000 km (239,000 mi). This outward spiraling continued into the distant future until, some 2 billion years. Thus, the Earth and the Moon are tidally locked to each other, like the present-day Pluto-Charon system. Earth has the same hemisphere facing toward the Moon, and vice-versa, as both rotate and the Moon orbits once every 42 days.
A Brief Look at Some Other Significant Impacts Many impact craters appear on each of the terrestrial planets and their satellites, and most of the satellites of the outer planets. Larger asteroids imaged by radar and spacecraft show craters in copious amounts. Even the Earth’s surface is home to at least 172 craters1 preserved to varying extents. In the vast majority of cases, a crater is formed when one object collides with a planetary surface, removing material from that surface in the process and forming a (generally) bowl-shaped depression in the surface. I describe how this works in a little more detail in the next section. Impact signatures in the form of craters and basins are widespread: from Mercury to Triton, craters can be found on almost any Solar System object with a solid surface. Although it is most often seen as bowl-shaped depressions of many sizes, these features can also show up as tectonic-type features associated with larger, more complex craters and impact basins. These can appear as wall terraces, multi-ring mountain scarps, and radial fracture (fault) systems. It is thought that the vast majority of impact craters were formed during the era of heavy bombardment, when the planetary system was filled with debris left over from its formation. As the planets formed, grew in size, and solidified, they swept up the debris, forming craters as the debris impact the planet surfaces. Eventually, most of this debris was swept up by the planets, thrown into the Sun, confined to the asteroid belt or Kuiper belt, captured as satellites, or ejected from the Solar System. These images (Figs. 1.5–1.16) provide examples of the widespread evidence. These images were obtained from the NASA Photojournal database of images. The PIA number is the identifier for each image. The description, as it appeared with each image, as well as some additional information in a few cases is provided. Chapter 5 discusses in detail the craters on various worlds throughout the Solar 8
Widespread Evidence Fig. 1.5. PIA02941: “Mercury’s south pole”. Image credit NASA/JPL/Northwestern University
Fig. 1.6. PIA00479: “Venus - Complex Crater 'Dickinson' in NE Atalanta Region”. Image credit NASA/JPL
9
Widespread Evidence
System and gives a brief overview of how the morphologies of craters change under different conditions (for instance, different worlds, different surface types, different gravity fields, the presence and absence of an atmosphere, and the size and physical nature of the impacting object).
Fig. 1.7. PIA09305: “On Mars fractures amidst small impact craters”. Image credit NASA/JPL/ASU
10
Widespread Evidence Fig. 1.8. “Comet Tempel 1”, NASA/JPL-Caltech/UMD (PIA02128); image credit NASA/JPL-Caltech/UMD
Fig. 1.9. PIA00078: “Gaspra, Deimos, and Phobos Comparison”. Image credit NASA/JPL
11
Widespread Evidence
Fig. 1.10. PIA01515: “Bright Ray Craters in Ganymede's Northern Hemisphere”. Image credit NASA/JPL
Fig. 1.11. PIA01648: “Impact Craters on Icy Callisto; Doh crater and Asgard”. Image credit NASA/JPL/ASU
12
Widespread Evidence Fig. 1.12. PIA09019: “Rhea Craters in Relief”. Image credit NASA/JPL/Space Science Institute
Fig. 1.13. PIA06064: “The Face of Phoebe”. Image credit NASA/JPL/Space Science Institute
13
Widespread Evidence
Fig. 1.14. PIA02217: “Miranda Image” credit NASA/JPL
Fig. 1.15. PIA00039: “Titania - Highest Resolution” Voyager Picture; image credit NASA/JPL
14
Widespread Evidence Fig. 1.16. PIA02208: “Triton”. Image credit NASA/JPL
Are Impacts Still Happening Today? In a word, yes. Not only are there numerous examples of impact craters on solid bodies in the Solar System (again, pointing to a large number of impact events that have occurred mostly in the distant past), but also the Earth’s atmosphere daily shows evidence that impacts are occurring even now. This evidence is in the form of meteors, or “shooting stars” that grace the night skies of Earth (and other planets with atmospheres). Most of these are rather faint, but a few of them can become quite bright, rivaling Venus or even the Full Moon in brightness for a very brief time. The smallest (i.e. “rather faint” and telescopic) meteors are caused by flecks of dust colliding with and burning up in the upper atmosphere, while the larger fireballs are the result of pebbles and stones entering the atmosphere and burning up. It is possible that these pebbles and stones (depending on the velocity) would produce, under the right conditions, a visible impact signature, in the form of a pinpoint flash, if it collided with the dark part of the Moon while being observed with a modest sized Earth-based telescope. To determine the frequency of occurrence, we can make a rough estimate of the daily impact frequency (over the entire lunar surface), within a factor of 10 or so, by finding out how many fireballs occur each day in the Earth’s atmosphere and divide that number by 14 (since the smaller Moon, having 1/14 the cross-sectional area of the Earth, has a smaller target size proportionately; only target size is considered in this simple illustration, not the differences in mass). Observational evidence of impacts occurring in our day and age are mounting. For example, the Meteoroid Environment Office at the Marshall Space Flight Center has documented 129 impact events on the Moon from November 2005 to September 2008. We not only have these documented events, along with the lunar 15
Widespread Evidence 16
meteoroid impacts which were first confirmed with the Leonid storms of 1999, but also evidence of very recent impacts on Mars and Jupiter. The famous ShoemakerLevy 9 multiple impact event of July 1994 (see chapter 5, “Beyond the Moon”, for more information about this particular event) was the first collision ever witnessed by a wide audience on another world. A meteor was even observed by Voyager 1 at Jupiter in 1979. In 2006, a paper was published detailing 20 new craters on Mars observed by the Mars Global Surveyor spacecraft between 1999 and 2003. Most of these impacts appear as dark splotches on the surface, with a few appearing as brighter spots. They range in size from a few meters to several hundreds of meters. Some of them look like comets, while others look as though someone took a handful of dark sand and threw it on the surface, making the streaky, comet-like image on the Martian surface. Judging by the appearance of the impact marks on the surface, it appears that in many cases, the impactor broke apart into many pieces before striking the surface. A detailed list of these Martian impacts is presented in Chapter 5. Also, Chapters 3 (Notable Collisions) and 4 (A Catalog of PIPs: Possible Impact Phenomena) have evidence that impacts are occurring right up to the present day.
Chapter 2
The Moon is an airless body, devoid of the atmosphere that Earth has to protect it from the impacts of meteoroids. In the case of the Earth, the atmosphere shields the ground from all but the larger (and much rarer) meteoroid collisions. We see the collision between the meteoroid and the atmosphere as a “falling” or “shooting” star, sometimes leaving a brief luminous trail of ions in the atmosphere. On the Moon, where there is no air, any meteoroids collide directly with the lunar surface, without being slowed by air. Over the age of the Solar System, meteoroids and asteroids of all sizes impacted the Moon and other celestial bodies, producing the pockmarks that we know as craters. It was not until the middle of the twentieth century that craters on the Moon were found to be a result of the impact. Also, it was not until the advent of spacecraft exploration that craters were found on other worlds, such as Mars and Mercury. I will give a brief overview of lunar impact features (craters) in the following sections. I look at how the number of craters changes with region (Maria versus Highland Craters), how the appearance of craters on the Moon change with age (Young versus Old Craters), how the appearance of the crater changes as it gets bigger (Appearance versus Size), and how one can recognize different types of features.
Lunar Impact Features
Lunar Impact Features
Maria vs. Highland Craters A casual look at the first quarter or waxing gibbous Moon through a low-power telescope reveals two tones of gray: the darker gray regions seem smooth and largely devoid of craters while the brighter regions are rough and heavily cratered. In fact, many areas of the highlands appear saturated with craters of all sizes. The reasons behind the dichotomy are asteroid impacts and past volcanic activity of the Moon. During the tail end of the era of heavy bombardment, the Moon was impacted by asteroid-sized objects, whose craters were later filled with molten rock. The molten rock appeared during a time when the Moon underwent a period of high volcanic activity, where molten rock seeped through cracks in the surface and filled the low-lying basin areas. Since this happened after most of the big impacts of the heavy bombardment, we observe the lack of craters in the Maria versus the brighter, older highlands. Impacts after Maria solidified are few and far between. Figure 2.1, courtesy Don Pearce of Houston, Texas, was taken on 27 July 2008 and shows the Moon two days after the Last Quarter. Major impact features, as well as
17
Lunar Impact Features
Fig. 2.1. Waning crescent moon showing differences in cratering density between maria (darker gray) and the highlands (gray-white). Image taken on 27 July 2008 through a 6-inch refractor with a Nikon D50 camera by Don Pearce of Houston, TX
the Earth-facing Maria (mostly Oceanus Procellarum) are seen in this image. Looking closely at this image, the Maria appears as smooth region with few craters and darker shades of gray, while the highlands show up brighter and rougher, with many impact craters. It appears that the main difference between Maria and highland cratering is the number of craters per unit area. The appearance of a crater of a given size on the Maria (example: Copernicus) is largely similar to the one on the highlands (example: Tycho, but this one is much younger than Copernicus). One can get an estimate of how old an airless surface is by counting the number of craters per unit area, which gives the crater density, usually expressed as the number of craters per million square kilometers. The older the surface, the more the craters, hence, higher the density of craters per unit area (the technique of counting craters is introduced briefly in the next section and discussed in detail in Chapter 6). Not only does counting craters help us understand the geologic evolution of a surface, but it also enables the identification of surface and subsurface processes in that world.
Young vs. Old Craters For individual craters, due to exposure to radiation and micrometeoroids, appearance changes over time. Fresh material is usually brighter and bluer; “weathered” material darker and redder. An example of a fresh crater is shown arrowed in 18
Lunar Impact Features
Fig. 3.2, next chapter. Professional planetary geologists sometimes use a parameter called “Optical Maturity” or OMAT to gauge the age of a feature or part of feature.2 One method used to determine the OMAT parameter of a feature is with multispectral images, images sensitive to the iron oxide and titanium oxide content of the lunar surface, as well as other spectral bandpasses. This uses an approach similar to color studies of the Moon by LTP observers: by comparing the intensity of a feature in one wavelength with the intensity in another, a ratio is determined, which leads to the OMAT parameter. OMAT changes across the face of a large feature, and the way this change occurs determines the age of the feature. For example, young craters have high OMAT values near the rim, but the values drop off steeply away from the rim. Older craters have very low OMAT values near the rim, and the change of value is flat, moving away from the rim. The ejecta patterns show up best at full Moon: younger craters have ejecta blankets and rays that have a high contrast with the surroundings, but older craters have blankets that blend well with the background, making them harder to see. An example of a young crater is Tycho, about 100-million-years old, with an OMAT profile as described above. An example of a mature (older) crater is Copernicus, 810-million-years old. Other examples of older craters include Eudoxus, Aristillus, and Lichtenberg. Age estimates for some of the craters in the “Top 100” List of Impact Features to observe are provided along with a few other physical parameters. In addition to changes in brightness and contrast of ejecta blankets with age, the outline and sharpness of the central crater changes over time as well. Besides the crater counting mentioned above, the age of the crater can also be determined by context: if the surface is heavily cratered and the crater itself appears to have been disturbed by others, it is true that the crater of interest is older. On the other hand, if the crater stands almost alone in a flat Maria plane, or if it appears well-defined and undisturbed amidst a background of more degraded craters, then this is probably a younger crater. These lines of reasoning come from the geologic principle known as the “law of superposition” where the younger features appear to be “on top” of an older feature, since the younger feature will likely be “placed” at a later time in history. A good example of a mix of craters of various ages is visible in the image of the Clavius crater (Fig. 2.2): the large crater Clavius shows a flat floor as a result of flooding by molten rock; but many small craters are found on the floor. Also, notice the craters on the rim of the large crater (at least four, including one with a central peak are evident) and compare the sharpness of the large Clavius crater with that of the superimposed central peak crater along the top (north) edge. Look closely at the image to find craters of various sizes and degrees of sharpness, including some which show advanced stages of being erased or obliterated by younger craters. Crater counting provides a method of determining the approximate age of a region. On the Moon, the greater the number of craters in a given region, the older the surface: the Maria has fewer craters and hence, is younger while the highlands have more craters and are older. Counting craters gives an excellent estimate of age, and the estimate can be rather accurate as long as some guidelines are followed. Craters of volcanic origin, such as the volcanic pits along Hyginus Rille, do not get counted. Secondary craters, sometime hard to distinguish against the background mix of pre-existing craters, do not get counted. Counts are made in areas that formed at the same time in geologic history, and over areas of similar sizes.
19
Lunar Impact Features
Fig. 2.2. Clavius and its environs. Image courtesy of Klaus Brasch
After carefully counting the craters, an observer can compile a sequence of relative ages, from youngest to oldest. The ages can be confirmed with samples from these regions, which can yield absolute ages in millions and billions of years. Apollo samples provide starting points for the absolute calibration of crater counts, but ideally we would want samples from both the absolute oldest and youngest parts of the Moon’s surface. All the lunar Maria and large basins are represented by samples which have made it possible to assign estimates of absolute ages to these features. Craters on an airless surface are modified or destroyed by a handful of processes. These processes are considered to have been active at one point in the Moon’s history or are still active today and include later impacts (great and small, including a constant “shower” of micrometeoroid impacts) or flooding by lava flows. The rate of obliteration of lunar craters is greatest for the smallest craters and increases with increasing crater size. Impacts break up bedrock, produce regolith and redistribute surface material and are the most important processes that affect the surface. Apollo seismographs recorded the seismic signatures of 70–150 impacts per year in the 100 g to 1,000 kg range. More common are the micrometeoroid bombardment which includes erosion, ionization, vaporization, and lateral transport of material over short distances. The erosion rate of the lunar surface is about 1 mm per 1 million years. The seismic shocks of impacts introduce vibrations that cause a down slope movement of material on craters and other topographic features. One particular Apollo image shows a boulder that had rolled down a hill, leaving a trail on the hillside. To summarize, young craters show bright ejecta rays, sharp rims, prominent ejecta blankets, secondary craters, and a fresh, bright appearance overall. Aging craters, on the contrary, show a darker, more degraded appearance: the rays disappear, secondary craters become subdued and disappear, the rough-textured 20
ejecta blanket takes on a smoother texture, the rim sharpness decreases, and any terraces are modified by radial channels. Simple craters are partially filled by ejecta from later impacts as they age, and their profiles change over time. Young simple craters have round, bowl-shaped profiles with raised rims while older simple craters have flat floors and rounded rims. Complex craters also get shallower over time and get filled by lava or impact ejecta. Sometimes the central peak is partially covered, at other times it is completely buried. Not only can crater counts provide a chronology for a region but also the changes in crater morphology can give an indication of the age of a particular feature.
Craters on the Moon (and on Moon samples) range in size from 0.1 µm to 1,600 km (4.0 × 10–9 in. to 1,000 mi), with those <1 cm called “microcraters” or “zap pits”. We do not see any microcraters on Earth because of the atmosphere shielding the surface. Craters with diameters larger than 300 km (186 mi) are called basins. Earth does not show any basins because geologic and weathering activity erases them over time. Figures 2.2 and 2.3 show areas on the Moon that display a wide range of craters of different sizes. Many of the characteristics that are described in
Lunar Impact Features
Appearance vs. Size
Fig. 2.3. The region of Sabine & Ritter, showing craters of various types and sizes. Maria and highland terrain can be seen in the image, which is courtesy of Richard Hill and Jim Loudon Observatory.
21
Lunar Impact Features
the following sections are visible in these images. I encourage the reader to look at these images as the following sections are read. Factors that influence the appearance and size of craters on the Moon include the impactor velocity, density, and size. Lunar regolith also plays a role in the appearance of the crater. Those less than a few meters in size show up as shallow depressions without raised rims. An impactor of average density traveling with an average velocity of around 20 km/s (12 mi/s) will produce a crater between 10 and 20 times the diameter of the impactor. The simplest craters, larger than tens of meters but less than about 15 km (9 mi) in diameter, are bowl-shaped, with a smooth, slightly concave floors; sharp, well-defined rims; and steep inner walls; no central peak; and a small debris field with blocks up to 1 m across. Craters larger than 20 km (12 mi) in diameter become more complex: those ranging in diameter from 20 to 175 km (12–105 mi) tend to have central uplift features that take the form of a central mountain peak or group of peaks. They usually have relatively flat floors and more complex walls, with some slumping (collapsing or caving in) of material forming terraces on the inner wall of the crater. Larger craters (those bigger than about 175 km or 105 mi across) tend to have complex, ring-shaped central peak clusters. Larger still, those with diameters greater than 300 km (190 mi), are no longer called craters, but are called impact basins, of which there are about 40 on the Moon. There are many examples of each type of crater on the Moon. An excellent example of the simple bowl-shaped crater is the Moltke Crater (Fig. 2.4). This crater is about 7 km (4 mi) in diameter and shows a modest debris or ejecta field.
Fig. 2.4. Moltke coutersy of NASA/Apollo 10 photograph AS10-29-4324
22
Lunar Impact Features
Craters of a given size range show common features: as size changes, the features change, to an extent that scientists have generated a crater main sequence, similar to the stellar main sequence. As one goes from smaller crater sizes to larger sizes, there appears to be some form of continuity of development, like the stars encountered as one moves up the stellar main sequence. Crater morphologies depend primarily on the energy of the impact (just as stellar spectral type depends primarily on the star’s mass) and the crater’s state of preservation. Other factors that determine crater morphology include the gravitational strength of the target, the presence or lack of atmosphere, and the target surface (and subsurface) composition. Simple craters (<15 km or < 10 mi) show a circular outline, a bright inner wall, and a parabolic shadow reflecting a bowl-shaped profile. One example is the 13 km (8 mi) wide feature Mösting A, which is up to 2.7 km (1.7 mi) deep. Another is Bessel Crater (Fig. 2.5) which is 2 km (1.2 mi) deep. Larger craters (15–50 km/9–31 mi) are more complex with slumping material. The outline is polygonal in shape instead of circular, and surrounding a floor littered with mounds of material that slid down walls. In some cases, the walls are terraced and the floors flat, with one or more central peaks. The rims are raised and the surrounding terrain shows a radial pattern of rays with secondary impacts up to several kilometers across. Triesnecker is an example of a crater in this size range; it measures 26 km (16 mi) across and is up to 2.7 km (1.7 mi) deep. Euler is another such example (Fig. 2.6). Craters larger than 150 km (90 mi) diameter have rings of peaks instead of the central peak(s). Many of these are classified as basins (large craters with distinct central rings), especially if they are 200–300 km (125–190 mi) across or more. Ringed basins such as the Orientale basin display concentric mountain basins and depressions. Rings can extend to 500 km (310 mi) from the center and mountains that form the rings can peak as high as 500 m. Basins are thought to have formed from the impact of asteroid-sized bodies or comets and their impacts cover most of the lunar surface with their debris and ejecta. There are some 43 basins catalogued on the Moon that have diameters greater than 220 km (140 mi). There are
Fig. 2.5. Bessel crater; courtesy of NASA/Part of Apollo 15 Panoramic photograph AS15-9328.
23
Lunar Impact Features
Fig. 2.6. Euler crater, courtesy of NASA / Part of Apollo 17 Metric photograph AS17-2923.
about equal numbers of basins on the near side and far side; those on the near side are generally flooded with lava basalt to form the familiar Maria.
How to Recognize Different Types of Features One of the best times to observe craters is between 3 days after New to 2 days prior to Full, and again from 2 days after Full to 3 days prior to New. This will vary with ecliptic angle, elongation, and so forth, but the thing in common is the low sun angle which casts shadows that make identifying the physical character of a feature much easier; it adds 3-D relief to the scene. The full phase (or within one day of Full) is best for observing ray structures and bright spots on the lunar surface. 24
Lunar Impact Features Fig. 2.7. The region of the moon from Copernicus to Eratosthenes, showing craters of various types and sizes. Maria and highland terrain can be seen in the image, which is courtesy of Richard Hill and Jim Loudon Observatory.
A reliable approach to recognizing different types of features is by studying the morphology of the feature (Craters of various sizes and morphologies are depicted in Figure 2.7). Features to look for include the presence or absence of a central peak or complex of peaks. The appearance of this central part of the crater formation will help you get an idea of the size range of the crater itself. Also look at the structure of the walls of the crater. Is the structure intact or is it broken, and if broken, by how much? Are there secondary craters on the floor of the main crater or among the ejecta blankets of the main crater? What does the ejecta blanket look like: is it well-developed or quite faded? How does its appearance change with sun angle? How do all of the impact features change with changing sunlight? Fig. 2.8 provides another useful example with which you can have practice at identifying different types of features. Where does the crater occur, in the Maria or highland regions? Does it appear to have disturbed its neighboring craters or does it stand out seemingly untouched. Does it interact with any other craters in the region? The “top 100” list of lunar impact structures to observe (provided in Chapter 6), along with the list for the Astronomical League’s two Lunar Clubs, will allow picking and identifying various types of craters. It is also worth noting that you will get more out of your observing session if you draw the features you are interested in and include your responses to the above questions. Any good book on observing and drawing the Moon will be helpful if you are starting from scratch. By drawing the features and making notes, you will have a permanent record of what you saw, and it will enable you to take your notes and make an estimate of the age of the feature, and compare the estimate with published values. Some of the craters in the “top 100” list have published ages that you can use to see how well you estimated their ages on your own. 25
Lunar Impact Features
Fig. 2.8. Image PIA02321: “Single Still Image Full Resolution”; image courtesy of NASA/JPL/Cassini Imaging Team/University of Arizona). Use this image (and Figure 2-7) to try your hand at recognizing different kinds of features. NASA/JPL/Space Science Institute
26
Chapter 3
Remarkable Collisions
In the previous chapters, we reviewed the abundant observational evidence that impact events were widespread early in the history of the Solar System. We also discussed the effects age has on impact features. In this chapter and in Chapters 4 and 7, we will see that impacts continue (at a much lower frequency) to the present era and are observable even today. This chapter highlights some of the notable impact events of the last millennium. Besides the many noted LTP’s that resemble the collision between a small meteoroid and the Moon (i.e. pinpoint flashes of light on the lunar surface, clouds, and obscurations that are very localized, Chapter 7), a number of notable collisions of meteoroids and various targets occurred during the twentieth century. These events include the 1908 Tunguska Event, the 1953 Stuart Lunar Flare event, the impact events recorded by the Apollo Seismic Lunar Experiments, and the Shoemaker-Levy 9/Jupiter event (artist's visualization appears as Fig 3.1). In each of these cases, scientists have had an opportunity to study impact events under different conditions. In the 1908 case, the event was studied only after the impact, and in both the 1908 and 1953 cases, there is uncertainty as to the true cause of each event. In the case of the Apollo experiments, there was little doubt as to the origin of the seismic signals (either moonquakes or meteoroid impacts). And with the Shoemaker-Levy 9 event, scientists had over 16 months to prepare for the actual impact events they knew were coming. As a result, unprecedented data and images were collected depicting the collision of comet fragments with a gas giant planet. In addition to the events observed on the Earth, Moon, Mars, and Jupiter during the twentieth century, evidence of even greater collisions from the past is easily found on both the Earth and Moon. While the Moon's evidence is obvious to anyone with a small telescope, it takes a bit more searching to find the evidence on Earth. This is due to the presence of a thick atmosphere on Earth, along with the surface and subsurface processes that tend to smooth things over, namely the craters left by the impacting objects. The Moon, lacking such processes, shows nearly all of its craters almost unaltered from their initial formation. The Earth hides most of its craters, or at least has erased most of the evidence, but these craters are available for investigation for those willing to search. In contrast to the Earth’s lack of fresh craters, thousands of craters are visible on the lunar surface, even with low magnification.
Remarkable Collisions
Historic Impacts of Interest
27
Remarkable Collisions
Fig. 3.1. An artist’s rendition of the Shoemaker-Levy 9 comet heading toward Jupiter
One notable exception to the hard-to-find craters on Earth is the Barringer Meteor Crater, located 35 miles east of Flagstaff, Arizona, USA. Of all the known craters on Earth, this one is best preserved, one of the youngest, and looks most like its better-known cousins on the Moon. It was thought to have been caused by a meteoroid of some 50 m (160 ft) across about 50,000 years ago. The crater itself is 1.6 km (nearly a mile) wide, 175 m (570 ft) deep and its rim rises 45 m (150 ft) above the surrounding plains of Northern Arizona.
The AD 1178 Lunar Impact Event Once thought to be a significant impact event (and still is, by most), which created the 22 km (14 mi) crater Bruno, and believed to be only 800-years old, the 1178 June 18 event is considered by some to be a coincidence sighting of an atmospheric meteor superimposed on the Moon’s cusp. What follows is a description of the event by the English monk, Gervase of Canterbury. “[On the evening of June 18, 1178] after sunset when the Moon had first become visible a marvelous phenomenon was witnessed by some five or more men...Now there was a bright new Moon...its horns were tilted toward the east; and suddenly the upper horn split in two. From the midpoint of the division a flaming torch sprang up, spewing out, over a considerable distance, fire, hot coals, and sparks. Meanwhile the body of the Moon which was below writhed, as it were, in anxiety...the Moon throbbed like a wounded snake. Afterwards it resumed its proper state. This phenomenon was repeated 28
There is speculation as to the true nature of this event, for which no other eyewitness reports are known. The event was thought to be the impact of a large meteoroid or asteroid on the Moon's surface, which produced the crater now known as Bruno. If such an event occurred on the Moon, there would have been several million tons of debris escaping the Moon’s gravitational pull, much of this spiraling toward Earth under the influence of Earth's gravity. Spectacular meteor storms, up to 50,000 bright meteors per hour, would have followed and would have been witnessed by the Earth’s population over the subsequent several weeks. With an interesting exception, we have no reports of such spectacular meteor storms. Reports of the event of “Countless stars flew west” on 11 October 1178 by Korean astronomers seem to paint a picture of the expected aftermath of an asteroid impact on the Moon. If this is true, it would seem that the June 1178 event is likely a genuine impact event. As far as the superimposed terrestrial meteor is concerned, the probability of the monks being in the right place at the right time to witness a meteor in the exact location (superimposed on the Moon’s limb) is very, very small, but it is not quite negligible, which could explain why they were the only ones to report this event. You are invited to think about this, consider the evidence, and make a conclusion. The Earth was a target of an impact 730 years later. On the morning of 30 June 1908, a huge explosion near the Tunguska River downed trees within a 20 km (12 mi) radius of the explosion. This event actually occurred in the atmosphere, about 5–10 km (3–6 mi) up, with no meteoroids recovered from the ground. The current, most widely accepted explanation is that a piece of comet entered the atmosphere and exploded under the stress of entry. Based on eyewitness accounts of the fireball’s entry, an orbit was calculated, which roughly matched that of Comet Encke. Thus, it is possible that the Tunguska meteoroid was a piece of the comet that had somehow broken off (perhaps the result of another impact) and was eventually intercepted by the Earth. Even to this day, this explanation is not 100% certain.
Remarkable Collisions
a dozen times or more, the flame assuming various twisting shapes at random...Then after these transformations the Moon from horn to horn...took on a blackish appearance. The present writer was given this report by men who saw it with their own eyes, and are prepared to stake their honour on an oath that they have made no addition or falsification in the above narrative.”3
The Taurid Complex Objects There is some evidence that suggests that the 1178 event may be the source of the extremely slow Corvid meteor shower, observed only in June of 1937. Evidence also points to the 1178 event as being the source of some of the impact-related seismic events observed by the ALSEP (Apollo Lunar Surface Experiment Program). Until a sample of the Bruno crater floor is collected and analyzed, we may never know for sure whether the event witnessed by the Monks in June 1178 was an actual meteor impact on the Moon or something much closer to home. Almost 800 years after the 1178, the ALSEP provided objective evidence of moonquakes and meteoroid impacts. A cluster of 43 recorded impacts from 22 through 26 June 1975 seems to point toward the Corvid meteor shower mentioned above. During the 29
nearly 8 year campaign to record seismic events on the Moon, the ALSEP logged 1743 meteor impact events with 95 being considered major events (potentially observable from the Earth). A man-made impact of a spent rocket stage was used as a reference event against which the other events were gauged. It is possible that the Apollo seismic observations of 17 to 27 June 1975, the 1908 Tunguska event, the Corvid meteor shower of 1937, the 1178 Bruno event, and lesser events have a common ancestor: the Taurid Complex Objects Stream. The “objects that make up the Taurid Complex are survivors of a long history of impacts and collisions that continuously reduce large objects to small ones.”4 This stream consists of objects, from millimeter size up to kilometer size, that appear to come from a common ancestor or set of ancestors. This stream can be considered as a loosely organized meteor stream, yet the closest to an annual stream are the June Taurid meteors. Several additional items may belong to this complex, such as the asteroids 1978 SB and 2201 Oljata, the Farmington meteorite, the Kagarlyk chondrite (which fell on the same day as the Tunguska event, but is not associated with the event itself), the Taurid meteor stream, and others. In addition to watching Earth’s skies during this time of year for increased meteor activity, a useful way to sample this complex is by monitoring the Moon in late June of every year that it is favorably placed in search of impact flashes that may come from these objects. If an unusually high number of these events occur on a regular basis, it is likely due to the members of the complex.
Remarkable Collisions
The “Lunar Flare” Event of 1953
30
The image in Figure 3.2 shows what looks like a very young crater, complete with fresh, bright ejecta (arrowed). This appears to be an ordinary crater with one exception: it lies in the location of an event that was recorded by Dr. Leon H. Stuart, a respected radiologist in Tulsa, Oklahoma, on film on 15 November 1953. This event came to be known as the "Lunar Flare" or "Stuart's Event" and was also observed visually by Stuart himself. Dr. Stuart used his home-built 8-in. f/8 Newtonian telescope to obtain the photographic image of the lunar flare. The flare was determined to last at least 8–10 s and was situated about 15 km (9 mi) southeast of the lunar feature Pallas (Fig. 3.2). This event was once thought (and still is by many) to be the signature of an explosion produced by the collision of a small asteroid with the Moon. This would make Stuart the only individual known to have visually observed and photographed a large meteoroid impact (Figs. 3.3 and 3.4). Nearly 50 years later, Dr. Bonnie J. Buratti of the Jet Propulsion Laboratory reported a 1½-kilometer-wide crater, thought to be the resultant crater from the 1953 event. This was found in data from the orbiter Clementine, which mapped the entire Moon in 1994. This crater matched the position of the flare as seen in the image. The crater appears fresh, bright, and bluish, which are the characteristics of very young craters. Over time, due to the bombardment of cosmic rays, solar radiation, and micrometeoroids, the ejecta degrades from fresh, bright, and bluish, to old, dark, and reddish. This is a process known as space weathering and is considered to take place very slowly, over millions of years. Scientists can estimate the approximate age of a crater by measuring its color and brightness.
Remarkable Collisions
Fig. 3.2. The feature initially thought to be the result of the “1953 Lunar Flare” event. (Photo courtesy JPL, Dr. Bonnie Buratti)
Fig. 3.3. The “1953 Lunar Flare Event” as imaged by Dr. Stuart in November 1953. Courtesy of Lenny Abbey
31
Remarkable Collisions
Fig. 3.4. Close-up of the region containing the Lunar Flare event. Courtesy of Lenny Abbey.
32
Not only did the blueness and brightness of this fresh crater match with the expectations, but the size and shape were also similar to what would be expected by the impact of an object of a particular size estimated based on the brightness of the impact flash. The estimated size was about 20 m (65 ft), producing a crater of one to two kilometers (0.62–1.24 miles) across. In comparison, the crater found in the Clementine images was 1.5 km (0.93 mi.) across. The impact was also calculated to have released energy of about 0.5 megatons (35 times more powerful than the Hiroshima atomic bomb dropped in the final days of World War II). Events of this magnitude are estimated to occur on the lunar surface once every half-century. No likely candidates were found in photographs taken from the Lunar Orbiter spacecraft in 1967. These findings, shortly after being published in the refereed journal Icarus, were contested by Dr. John E. Westfall of the Association of Lunar and Planetary Observers. He had found that the bright feature seen by Clementine also appears in a series of telescopic plates taken decades before Stuart snapped his photo. Dr. Westfall noted that the feature was "pretty obvious" in photographs taken with the Mount Wilson 100-in. (254-cm) Hooker telescope in 1919, on plates taken in 1937 with the 36-in. (91-cm) refractor at Lick Observatory, and in others obtained with Catalina Observatory's 61-in. (155-cm) reflector in 1966. Additional pieces of evidence against this feature being the resultant crater of the 1953 Lunar flare surfaced, including the fact that such young, bright features should be a lot more common on the lunar surface that is what is actually seen. If 20-m-wide objects impact the Moon’s surface once every 50 years or so (which is often enough to make Stuart's sighting statistically plausible), then the lunar landscape should sparkle with tens of thousands of bright, fresh-looking craters. Impact specialist Dr. Alan W. Harris of the Space Science Institute noted (and was also stated by Dr. Jay Melosh, another impact expert) that 8 s is an impossibly long time for a fireball to last which results in such a small crater. Finally, the position of the crater is offset by a full 1°, or 30 km (19 mi), from the center of the flare, based on careful measurements of Mr. Stuart’s image by Sky & Telescope editors Dennis di Cicco and Gary Seronik, and by Westfall himself.
Remarkable Collisions
Based on the available evidence, it seems more and more likely that the 1953 Lunar flare was not an impact of a meteor, but some other phenomenon. One possibility was that the flare was a "point meteor," headed directly at Stuart as he observed and photographed the event, but that is ruled out by the flare's duration, as well as the very small probability that such an event would occur at such a location during such a short exposure (and stay circular for that long a period of time). The perfectly round spot seems to rule against spurious effects related to flaws in the optical system or image processing or nature of the plate used (i.e. a stray reflection or emulsion defect). The best explanation seems to be that the Lunar Flare was another form of LTP, caused by outgassing resulting from the thermal expansion of the surface, releasing gas rapidly. The location near the terminator, as well as event duration argues in favor of this explanation. Alternatively, another explanation, which may be related to the outgassing, is the triboelectric effect, where the gas becomes electrified as a result of contact with the solar wind. This electrified gas then produces an electrostatic discharge which is visible from the Earth. Unfortunately, with the original plate of this event lost forever, it is quite likely that the true nature of the “Lunar Flare” may never be conclusively found. A sample of impact events on three worlds was described above and illustrates the fact that such events are still happening today. While the SL9 impact, and events as large as Tunguska and the lunar event of 1178 AD are quite rare, potentially more common are events such as the Leonid impacts of 1974 (seismic), 1999, and 2001 (both optical), and large bolides that are regularly seen to burn bright in the Earth’s atmosphere. It is possible that a fairly large object may be discovered either by an observatory before a lunar collision or during a collision by a dedicated lunar meteoritic impact observing program. Such a program would have a tremendous potential to reveal scientifically useful information concerning the population of small- to medium-sized objects in interplanetary space in the Earth-Moon region. If a lunar monitoring program is coupled to a systematic atmospheric monitoring program for the same set of annual streams, valuable information about each stream can be obtained. A wide range of sizes of interplanetary objects, from dust specks to large boulders, could be surveyed. In addition, information on stream structure, size distribution, and variations in a parameter known in meteoritics as the population index can be obtained. Also, the physical dynamics of impact events and collisions under a wide variety of conditions could be studied. In addition to all this, knowledge of impact frequencies could provide clues into the dynamical evolution of the interplanetary environment in the present and recent past. Finally, such a comprehensive program could establish just how common these impact flash events really are, and what actual threat they pose to human settlement operations on the Moon and beyond.
33
Chapter 3
Remarkable Collisions
In the previous chapters, we reviewed the abundant observational evidence that impact events were widespread early in the history of the Solar System. We also discussed the effects age has on impact features. In this chapter and in Chapters 4 and 7, we will see that impacts continue (at a much lower frequency) to the present era and are observable even today. This chapter highlights some of the notable impact events of the last millennium. Besides the many noted LTP’s that resemble the collision between a small meteoroid and the Moon (i.e. pinpoint flashes of light on the lunar surface, clouds, and obscurations that are very localized, Chapter 7), a number of notable collisions of meteoroids and various targets occurred during the twentieth century. These events include the 1908 Tunguska Event, the 1953 Stuart Lunar Flare event, the impact events recorded by the Apollo Seismic Lunar Experiments, and the Shoemaker-Levy 9/Jupiter event (artist's visualization appears as Fig 3.1). In each of these cases, scientists have had an opportunity to study impact events under different conditions. In the 1908 case, the event was studied only after the impact, and in both the 1908 and 1953 cases, there is uncertainty as to the true cause of each event. In the case of the Apollo experiments, there was little doubt as to the origin of the seismic signals (either moonquakes or meteoroid impacts). And with the Shoemaker-Levy 9 event, scientists had over 16 months to prepare for the actual impact events they knew were coming. As a result, unprecedented data and images were collected depicting the collision of comet fragments with a gas giant planet. In addition to the events observed on the Earth, Moon, Mars, and Jupiter during the twentieth century, evidence of even greater collisions from the past is easily found on both the Earth and Moon. While the Moon's evidence is obvious to anyone with a small telescope, it takes a bit more searching to find the evidence on Earth. This is due to the presence of a thick atmosphere on Earth, along with the surface and subsurface processes that tend to smooth things over, namely the craters left by the impacting objects. The Moon, lacking such processes, shows nearly all of its craters almost unaltered from their initial formation. The Earth hides most of its craters, or at least has erased most of the evidence, but these craters are available for investigation for those willing to search. In contrast to the Earth’s lack of fresh craters, thousands of craters are visible on the lunar surface, even with low magnification.
Remarkable Collisions
Historic Impacts of Interest
27
Remarkable Collisions
Fig. 3.1. An artist’s rendition of the Shoemaker-Levy 9 comet heading toward Jupiter
One notable exception to the hard-to-find craters on Earth is the Barringer Meteor Crater, located 35 miles east of Flagstaff, Arizona, USA. Of all the known craters on Earth, this one is best preserved, one of the youngest, and looks most like its better-known cousins on the Moon. It was thought to have been caused by a meteoroid of some 50 m (160 ft) across about 50,000 years ago. The crater itself is 1.6 km (nearly a mile) wide, 175 m (570 ft) deep and its rim rises 45 m (150 ft) above the surrounding plains of Northern Arizona.
The AD 1178 Lunar Impact Event Once thought to be a significant impact event (and still is, by most), which created the 22 km (14 mi) crater Bruno, and believed to be only 800-years old, the 1178 June 18 event is considered by some to be a coincidence sighting of an atmospheric meteor superimposed on the Moon’s cusp. What follows is a description of the event by the English monk, Gervase of Canterbury. “[On the evening of June 18, 1178] after sunset when the Moon had first become visible a marvelous phenomenon was witnessed by some five or more men...Now there was a bright new Moon...its horns were tilted toward the east; and suddenly the upper horn split in two. From the midpoint of the division a flaming torch sprang up, spewing out, over a considerable distance, fire, hot coals, and sparks. Meanwhile the body of the Moon which was below writhed, as it were, in anxiety...the Moon throbbed like a wounded snake. Afterwards it resumed its proper state. This phenomenon was repeated 28
There is speculation as to the true nature of this event, for which no other eyewitness reports are known. The event was thought to be the impact of a large meteoroid or asteroid on the Moon's surface, which produced the crater now known as Bruno. If such an event occurred on the Moon, there would have been several million tons of debris escaping the Moon’s gravitational pull, much of this spiraling toward Earth under the influence of Earth's gravity. Spectacular meteor storms, up to 50,000 bright meteors per hour, would have followed and would have been witnessed by the Earth’s population over the subsequent several weeks. With an interesting exception, we have no reports of such spectacular meteor storms. Reports of the event of “Countless stars flew west” on 11 October 1178 by Korean astronomers seem to paint a picture of the expected aftermath of an asteroid impact on the Moon. If this is true, it would seem that the June 1178 event is likely a genuine impact event. As far as the superimposed terrestrial meteor is concerned, the probability of the monks being in the right place at the right time to witness a meteor in the exact location (superimposed on the Moon’s limb) is very, very small, but it is not quite negligible, which could explain why they were the only ones to report this event. You are invited to think about this, consider the evidence, and make a conclusion. The Earth was a target of an impact 730 years later. On the morning of 30 June 1908, a huge explosion near the Tunguska River downed trees within a 20 km (12 mi) radius of the explosion. This event actually occurred in the atmosphere, about 5–10 km (3–6 mi) up, with no meteoroids recovered from the ground. The current, most widely accepted explanation is that a piece of comet entered the atmosphere and exploded under the stress of entry. Based on eyewitness accounts of the fireball’s entry, an orbit was calculated, which roughly matched that of Comet Encke. Thus, it is possible that the Tunguska meteoroid was a piece of the comet that had somehow broken off (perhaps the result of another impact) and was eventually intercepted by the Earth. Even to this day, this explanation is not 100% certain.
Remarkable Collisions
a dozen times or more, the flame assuming various twisting shapes at random...Then after these transformations the Moon from horn to horn...took on a blackish appearance. The present writer was given this report by men who saw it with their own eyes, and are prepared to stake their honour on an oath that they have made no addition or falsification in the above narrative.”3
The Taurid Complex Objects There is some evidence that suggests that the 1178 event may be the source of the extremely slow Corvid meteor shower, observed only in June of 1937. Evidence also points to the 1178 event as being the source of some of the impact-related seismic events observed by the ALSEP (Apollo Lunar Surface Experiment Program). Until a sample of the Bruno crater floor is collected and analyzed, we may never know for sure whether the event witnessed by the Monks in June 1178 was an actual meteor impact on the Moon or something much closer to home. Almost 800 years after the 1178, the ALSEP provided objective evidence of moonquakes and meteoroid impacts. A cluster of 43 recorded impacts from 22 through 26 June 1975 seems to point toward the Corvid meteor shower mentioned above. During the 29
nearly 8 year campaign to record seismic events on the Moon, the ALSEP logged 1743 meteor impact events with 95 being considered major events (potentially observable from the Earth). A man-made impact of a spent rocket stage was used as a reference event against which the other events were gauged. It is possible that the Apollo seismic observations of 17 to 27 June 1975, the 1908 Tunguska event, the Corvid meteor shower of 1937, the 1178 Bruno event, and lesser events have a common ancestor: the Taurid Complex Objects Stream. The “objects that make up the Taurid Complex are survivors of a long history of impacts and collisions that continuously reduce large objects to small ones.”4 This stream consists of objects, from millimeter size up to kilometer size, that appear to come from a common ancestor or set of ancestors. This stream can be considered as a loosely organized meteor stream, yet the closest to an annual stream are the June Taurid meteors. Several additional items may belong to this complex, such as the asteroids 1978 SB and 2201 Oljata, the Farmington meteorite, the Kagarlyk chondrite (which fell on the same day as the Tunguska event, but is not associated with the event itself), the Taurid meteor stream, and others. In addition to watching Earth’s skies during this time of year for increased meteor activity, a useful way to sample this complex is by monitoring the Moon in late June of every year that it is favorably placed in search of impact flashes that may come from these objects. If an unusually high number of these events occur on a regular basis, it is likely due to the members of the complex.
Remarkable Collisions
The “Lunar Flare” Event of 1953
30
The image in Figure 3.2 shows what looks like a very young crater, complete with fresh, bright ejecta (arrowed). This appears to be an ordinary crater with one exception: it lies in the location of an event that was recorded by Dr. Leon H. Stuart, a respected radiologist in Tulsa, Oklahoma, on film on 15 November 1953. This event came to be known as the "Lunar Flare" or "Stuart's Event" and was also observed visually by Stuart himself. Dr. Stuart used his home-built 8-in. f/8 Newtonian telescope to obtain the photographic image of the lunar flare. The flare was determined to last at least 8–10 s and was situated about 15 km (9 mi) southeast of the lunar feature Pallas (Fig. 3.2). This event was once thought (and still is by many) to be the signature of an explosion produced by the collision of a small asteroid with the Moon. This would make Stuart the only individual known to have visually observed and photographed a large meteoroid impact (Figs. 3.3 and 3.4). Nearly 50 years later, Dr. Bonnie J. Buratti of the Jet Propulsion Laboratory reported a 1½-kilometer-wide crater, thought to be the resultant crater from the 1953 event. This was found in data from the orbiter Clementine, which mapped the entire Moon in 1994. This crater matched the position of the flare as seen in the image. The crater appears fresh, bright, and bluish, which are the characteristics of very young craters. Over time, due to the bombardment of cosmic rays, solar radiation, and micrometeoroids, the ejecta degrades from fresh, bright, and bluish, to old, dark, and reddish. This is a process known as space weathering and is considered to take place very slowly, over millions of years. Scientists can estimate the approximate age of a crater by measuring its color and brightness.
Remarkable Collisions
Fig. 3.2. The feature initially thought to be the result of the “1953 Lunar Flare” event. (Photo courtesy JPL, Dr. Bonnie Buratti)
Fig. 3.3. The “1953 Lunar Flare Event” as imaged by Dr. Stuart in November 1953. Courtesy of Lenny Abbey
31
Remarkable Collisions
Fig. 3.4. Close-up of the region containing the Lunar Flare event. Courtesy of Lenny Abbey.
32
Not only did the blueness and brightness of this fresh crater match with the expectations, but the size and shape were also similar to what would be expected by the impact of an object of a particular size estimated based on the brightness of the impact flash. The estimated size was about 20 m (65 ft), producing a crater of one to two kilometers (0.62–1.24 miles) across. In comparison, the crater found in the Clementine images was 1.5 km (0.93 mi.) across. The impact was also calculated to have released energy of about 0.5 megatons (35 times more powerful than the Hiroshima atomic bomb dropped in the final days of World War II). Events of this magnitude are estimated to occur on the lunar surface once every half-century. No likely candidates were found in photographs taken from the Lunar Orbiter spacecraft in 1967. These findings, shortly after being published in the refereed journal Icarus, were contested by Dr. John E. Westfall of the Association of Lunar and Planetary Observers. He had found that the bright feature seen by Clementine also appears in a series of telescopic plates taken decades before Stuart snapped his photo. Dr. Westfall noted that the feature was "pretty obvious" in photographs taken with the Mount Wilson 100-in. (254-cm) Hooker telescope in 1919, on plates taken in 1937 with the 36-in. (91-cm) refractor at Lick Observatory, and in others obtained with Catalina Observatory's 61-in. (155-cm) reflector in 1966. Additional pieces of evidence against this feature being the resultant crater of the 1953 Lunar flare surfaced, including the fact that such young, bright features should be a lot more common on the lunar surface that is what is actually seen. If 20-m-wide objects impact the Moon’s surface once every 50 years or so (which is often enough to make Stuart's sighting statistically plausible), then the lunar landscape should sparkle with tens of thousands of bright, fresh-looking craters. Impact specialist Dr. Alan W. Harris of the Space Science Institute noted (and was also stated by Dr. Jay Melosh, another impact expert) that 8 s is an impossibly long time for a fireball to last which results in such a small crater. Finally, the position of the crater is offset by a full 1°, or 30 km (19 mi), from the center of the flare, based on careful measurements of Mr. Stuart’s image by Sky & Telescope editors Dennis di Cicco and Gary Seronik, and by Westfall himself.
Remarkable Collisions
Based on the available evidence, it seems more and more likely that the 1953 Lunar flare was not an impact of a meteor, but some other phenomenon. One possibility was that the flare was a "point meteor," headed directly at Stuart as he observed and photographed the event, but that is ruled out by the flare's duration, as well as the very small probability that such an event would occur at such a location during such a short exposure (and stay circular for that long a period of time). The perfectly round spot seems to rule against spurious effects related to flaws in the optical system or image processing or nature of the plate used (i.e. a stray reflection or emulsion defect). The best explanation seems to be that the Lunar Flare was another form of LTP, caused by outgassing resulting from the thermal expansion of the surface, releasing gas rapidly. The location near the terminator, as well as event duration argues in favor of this explanation. Alternatively, another explanation, which may be related to the outgassing, is the triboelectric effect, where the gas becomes electrified as a result of contact with the solar wind. This electrified gas then produces an electrostatic discharge which is visible from the Earth. Unfortunately, with the original plate of this event lost forever, it is quite likely that the true nature of the “Lunar Flare” may never be conclusively found. A sample of impact events on three worlds was described above and illustrates the fact that such events are still happening today. While the SL9 impact, and events as large as Tunguska and the lunar event of 1178 AD are quite rare, potentially more common are events such as the Leonid impacts of 1974 (seismic), 1999, and 2001 (both optical), and large bolides that are regularly seen to burn bright in the Earth’s atmosphere. It is possible that a fairly large object may be discovered either by an observatory before a lunar collision or during a collision by a dedicated lunar meteoritic impact observing program. Such a program would have a tremendous potential to reveal scientifically useful information concerning the population of small- to medium-sized objects in interplanetary space in the Earth-Moon region. If a lunar monitoring program is coupled to a systematic atmospheric monitoring program for the same set of annual streams, valuable information about each stream can be obtained. A wide range of sizes of interplanetary objects, from dust specks to large boulders, could be surveyed. In addition, information on stream structure, size distribution, and variations in a parameter known in meteoritics as the population index can be obtained. Also, the physical dynamics of impact events and collisions under a wide variety of conditions could be studied. In addition to all this, knowledge of impact frequencies could provide clues into the dynamical evolution of the interplanetary environment in the present and recent past. Finally, such a comprehensive program could establish just how common these impact flash events really are, and what actual threat they pose to human settlement operations on the Moon and beyond.
33
Beyond the Moon
Beyond the Moon
Chapter 5
We now look beyond the Earth–Moon system to see how impacts are shaped throughout the Solar System.8,9 In Chapter 1, I gave a brief overview of worlds (in the form of images) showing how widespread cratering is from Mercury to Triton, and related this to the most widely accepted theory of the evolution and development of the Solar System. We also saw how this evidence provides clues into the evolution of our local planetary system to its present form. We saw how a huge impact likely led to the formation of the Moon and how impacts shaped the surfaces of the worlds in the early history of the Solar System. Impacts are also considered to be responsible for the retrograde (backwards) rotation of Venus, the high metal content in Mercury, the dichotomy in landforms on Mars, the rings of Saturn, and the extreme axial tilt of Uranus. One of the moons of Uranus, Miranda, appears to have been broken up by a large impact, but something unusual happened after the impact: the object pulled itself together again to become one body. In most cases, and at first glance, the craters on other planets such as Mars and Mercury look just like those seen on the Moon through a backyard telescope. But, upon closer inspection, some subtle differences show up. From world to world, the factors including the composition of the surface and subsurface, the gravitational pull of the world, the presence or absence of an atmosphere, and the physical nature of the impactor, all play their roles in determining the size and shape of the crater produced. There are many pieces of evidence, both observational and theoretical, that show that impacts are still taking place, not only on the Moon and in the Earth’s atmosphere, but throughout the Solar System and in other planetary systems. There is observational evidence of impacts in the form of very fresh craters on Mars, meteors in Jupiter’s atmosphere, disturbances in Saturn’s rings, and spectra of dusty disks around other stars that indicate dustproducing collisions taking place in the disks. I will first talk about the multiple-impact event on Jupiter that was widely observed in 1994, the impact of comet Shoemaker-Levy 9. This was the first time that an impact was definitely observed on a giant planet (or any other planet besides the Moon). It was the first time that such an impact was predicted with a long lead time, allowing many groups to prepare for a thorough observing campaign. After considering this event, the possibility or probability of observing impacts on other worlds is discussed. Meteor showers and impacts on Mars, Venus, Mercury, and other worlds are considered. We also investigate crater morphologies throughout the Solar System, taking into account variables such as surface type and gravity, atmosphere versus no atmosphere, type of impactor, and other considerations. The chapter concludes by presenting a catalog of 20 fresh craters on Mars that occurred between 1999 and 2006, as observed by the Mars Global Surveyor satellite. 71
Beyond the Moon
Target Jupiter: The Shoemaker-Levy 9 Multiple Impact Event On 7 July 1992, a comet approached the gas giant planet Jupiter very closely, passing 0.4 Jupiter radii from the cloud tops. The approach was close enough, such that the tidal forces from the planet caused the fragile comet to break apart. Also, the approach was such that the comet was captured into a temporary orbit around Jupiter as a satellite of the planet and was placed on a collision course with Jupiter. This comet was discovered nine months later as the ninth short-period comet discovered by Eugene and Carolyn Shoemaker and David Levy. The comet was first seen on a photograph taken on 24 March 1993, with a 0.4-m (18-in.) Schmidt telescope at Palomar Observatory in California. The object was soon found to be on a collision course with Jupiter, with the impact predicted to occur during the third week of July 1994. The fragments, shown in the image taken by the Hubble Space Telescope in July 1993 (Fig. 5.1), were two kilometers or less in size. These 22 or so larger fragments (greater than about one kilometer in size) are embedded in a cloud of much smaller fragments and dust. These fragments, over the course of one week (16–22 July 1994), fell into the giant planet one by one in a location just beyond the Jovian horizon as seen from Earth. However, the impact sites were in plain view of the Galileo spacecraft, which was then approaching Jupiter. Galileo captured many images of the impacts by several of the fragments, including the one by Fragment W (Fig. 5.2), taken over a 7-s period on 22 July 1994. These impacts, though not directly visible from the Earth, left visible scars on the planet. The images in Fig. 5.3 (below), taken by the Hubble Space Telescope, show the tell-tale signs of impact in the form of "dirty", brown stains on the planet's cloud tops. Impacting Jupiter at 60 km/s (38 mi/s), each fragment released tremendous amounts of energy, up to a few million megatons of TNT equivalent per impact and produced fireballs that reached altitudes of nearly 3,000 km (1,900 mi). After these hot impact spots cooled, dusty dark scars remained which were the result of cometary and planetary material dredged up and suspended in the Jovian stratosphere. Scientists were able to use the impacts to probe Jupiter’s atmosphere, to learn about its vertical structure and composition. This was done by generating computer
Fig. 5.1. Close-up of the comet Shoemaker-Levy 9 (SL-9) as imaged with the Hubble Space Telescope in January 1994 (lower left and topmost images) and July 1993 (Lower right image). Images courtesy of Dr. H.A. Weaver, Dr. T.E. Smith, STScI and NASA
72
Beyond the Moon Fig. 5.2. Impact sequence showing Fragment W impact flash as it hits Jupiter, taken by the Galileo Space craft en route to Jupiter at the time.
Fig. 5.3. a (left) and b (above). The impact plume of Fragment A; the marks of Fragments A and G on Jupiter’s face. (Images courtesy of NASA and STScI)
73
Beyond the Moon
simulations of the SL-9 impacts with a model of what Jupiter’s atmosphere was thought to be like at the time of the event. They then compared the results of the model with the actual observations and collected information to tweak the model to better represent the real Jovian atmosphere. In addition to probing the physical structure of Jupiter’s atmosphere, and allowing a refinement of atmosphere models, the impacts of SL-9 remind us that collisions in the Solar System happen all the time. Jupiter is probably hit by a comet once every 100 years or so; in 1690 the Italian astronomer Cassini observed a dark spot on Jupiter that appeared and changed in shape over several days. This observation is reminiscent of what was witnessed in the days and weeks following the Shoemaker-Levy 9 impact on Jupiter—dark splotches changing shape as they spread out and dispersed under the influence of winds in the upper Jovian atmosphere. Figure 5.3a shows the evolution of the fireball after one of the impacts. Figure 5.3b depicts the dark stains that remained on the planet following several such impacts. Both the images were taken by the Hubble Space Telescope in low Earth orbit. The dark stains were visible in small ground-based telescopes for up to several weeks after the impact events.
Crater Types and Morphologies Throughout the Solar System The following observations, collected from a variety of planetary geology sources, describe the various impact structures (craters) found on various worlds. We consider not only how the size, surface type, the nature of the projectile, and the presence or absence of an atmosphere affect the form of the crater produced, but also look at features on individual worlds throughout our local planetary system. This provides a perspective that shows craters as a common feature of the Solar System (and likely of any other planetary system that has worlds with solid surfaces), something to keep in mind as you look upon the Moon’s battered surface with a backyard telescope.
Shapes and Features According to Size (Similar Surface Type) Mercury vs. the Moon Considering objects with similar surface properties but different bulk sizes (such as Mercury and the Moon), we observe many similarities and differences between craters of similar sizes and morphologies. First, the gravity fields of objects vary with mass and size. The incoming velocity of an impactor is not only influenced by the impactor’s own orbital velocity, but is also enhanced by the gravitational pull of the target: the bigger the target, the greater the acceleration caused by the target’s gravity. Hence, the velocity of a meteoroid just before impact is the sum of the impactor’s own orbital velocity and the added velocity resulting from the target’s gravitational pull. The higher the combined velocity, the greater the amount of 74
Beyond the Moon Fig. 5.4. Mercury as imaged by the MESSENGER, 14 January 2008, courtesy of Credit: NASA/Johns Hopkins University Applied Physics Laboratory/Carnegie Institution of Washington
energy that the impactor bears just before it impacts. The more the energy, the bigger the crater caused by an impactor of a given mass. A stronger gravitational field implies that an impact ejecta kicked up by the collision does not travel as far from the impact site as the ejecta kicked up on a smaller world. This can be seen by comparing Mercury with the Moon. At first glance, Mercury (Fig. 5.4) looks just like the Moon. It is covered with many craters, and is airless and lifeless. Upon closer inspection we see differences in shades of gray, caused by the various ray fields. The differences in the shades of gray are caused by differences in albedo, or reflectivity. Albedo is measured on a scale of 0–1, with 0 being perfectly black (absorbing all incoming light), and 1 being perfectly white (reflecting all incident light). The smooth plains of Mercury resemble the smooth Maria of the Moon, but Mercury’s “Maria” does not have much difference in albedo as compared with surrounding cratered terrain. As has been seen, the contrast between lunar Maria and highlands is obvious, easily visible with the naked eye. The intercrater plains on Mercury are covered with a high density of superposed small craters (probably secondary craters from larger impacts) 5–10 km (3–6 km) in diameter. The heavily cratered terrain shows clusters of closely packed overlapping craters, with at least 30 of them several hundred kilometers in diameter. The ejecta fields cover a smaller area than those on the Moon due to Mercury’s stronger gravity pulling the debris down sooner than an identical impact on the Moon. We see similar forms of craters and basins on Mercury as we see on the Moon; the shapes and sizes of craters on both worlds are similar. Because Mercury has 2.2 times stronger gravity than the Moon, there are some differences between the craters on the two worlds. The morphology begins changing at smaller crater sizes for Mercury than for the Moon. Those craters larger than 14 km (9 mi) in diameter show flat floors, terracing on the interior walls, central rings or ring complexes. 75
Beyond the Moon
Fig. 5.5. PIA10367: “Phobos from 5,800 Kilometers”; image courtesy of NASA/JPL-Caltech/University of Arizona
Fresher craters on Mercury show raised rims, hummocky terrain that surrounds the feature which transitions to a more radial, ray-like feature. Beyond this lie clusters and chains of secondary impact craters. The depth-to-diameter ratio of Mercury’s craters is similar to that of fresh lunar craters. Some of the youngest Mercurian craters show well-developed bright rayed ejecta systems. Some of these bright-rayed craters are surrounded by a halo of dark materials but others do not. The largest of Mercury’s craters (>130 km or > 80 mi) have at their centers a ringed complex of peaks and are collectively called basins. Mercury’s most prominent crater is the Caloris Basin.
Phobos and Deimos vs. the Moon Mars has two small moons that are irregular in shape and are saturated with small craters. The crater density on each small moonlet is similar to the density in the lunar highlands. The shapes of the Martian moonlets’ craters range from elongated double forms to circular forms with varying degrees of sharpness and freshness. Many have raised rims but none have central peaks, obvious ejecta blankets or ray patterns. Phobos (Fig. 5.5) craters have depth-to-diameter ratios similar to those of the Moon’s craters. The crater Stickney, about 10 km (6 mi) across, is the largest crater on Phobos with a diameter about 40% of the satellite’s diameter. Voltaire is the largest crater on Deimos at about 2 km (1 mi). The low surface gravity of these small objects means that ejecta from impacts that form these craters largely escapes the gravitational pull of the objects, and probably spirals down toward the Martian surface. It is possible that Mars has a collection of meteorites on its surface that come from its satellites, just like the lunar meteorites that are sometimes found on the surface of the Earth. 76
Beyond the Moon Fig. 5.6. Image PIA02142, “Comet Tempel 1”. Image courtesy of NASA/JPL/UMD
Fig. 5.7. Image PIA05571, “Comet Wild 2” Image courtesy NASA/JPL-Caltech
Minor Objects in the Solar System To date about half a dozen asteroids (Eros is shown in Figs. 5.8 and 5.9) and a handful of comet nuclei (Fig. 5.6 shows the nucleus of comet Tempel 1. Fig. 5.7 shows Wild 2) have been imaged by spacecraft at close range. In addition, many 77
Beyond the Moon
Fig. 5.8. Image PIA03134, ”Bright Crater Wall (Eros)”. Image courtesy of NASA/JPL/JHUAPL
Fig. 5.9. Image PIA03141, “The Environs of NEAR Shoemaker’s Landing Site”, Image courtesy of NASA/JPL/JHUAPL
78
Shapes and Features According to Surface Type
Beyond the Moon
asteroid-like planetary satellites have also been imaged by spacecraft, including Phobos and Deimos, four of the minor moons of Jupiter, 10 of Saturn, and two each of Uranus and Neptune. All these objects, ranging in mean diameter from about 12 to 420 km (8–260 mi), show lots of craters, many with soft outlines. Two of the four comet nuclei imaged by spacecraft (Figs. 5.6 and 5.7) show at least a few impact craters.
Rocky vs. Icy Surfaces Comparison of craters on the Moon and Mars shows some interesting differences. Not only does Mars have a thin atmosphere (while the Moon does not), but it also shows evidence of a subsurface permafrost which appears to have contributed to some of the lobate shapes around certain impact features. Also, very fresh craters observed by the Mars Global Surveyor (produced between 1999 and 2003, these are tabulated at the end of the chapter) show dark ejecta, whereas small, fresh lunar craters show bright ejecta. In the outer Solar System, the craters of the planetary satellites may be from impactors that come from a number of sources. Planetary scientists recognize three populations of impactor objects that caused the craters on the satellite surfaces. In the case of the moons of Saturn and Uranus (and possibly for the systems of Jupiter and Neptune as well), rocky Population I objects produce older impacts (which are partly erased on younger surfaces) that are considered to have produced a substantial fraction of the large craters. These objects were possibly swept up from the post-accretion debris added 4 billion years ago and are the outer Solar System’s version of the heavy bombardment that occurred in the inner Solar System around the same time (see Chapter 1 for a summary on the impact history of the inner Solar System). Icy rather than rocky impactors dominate Population II objects, which result in a deficiency of large craters and dominate the resurfacing of younger surfaces, and are considered leftovers of the accretion disks of the giant planets. Two stages in the impact process of the icy satellites of Saturn are evident: a discrete episode of cratering followed by collisions with swarms of debris left over and in orbit around the planet. Finally, Population III objects are cometary impactors that exist to the present day (and are also evident in the inner Solar System) (Figs. 5.10–5.13).
Ganymede and Callisto The two largest Galilean satellites of Jupiter, Ganymede (Figs. 5.10 and 5.12) and Callisto (Figs. 5.11 and 5.13), are each roughly a 50/50 mixture of rock and ice and keep their same sides oriented toward Jupiter at all times. Both may have subsurface oceans underneath a thick, icy crust, and may have similar crater morphology. Callisto, the outermost Galilean moon, is intensely cratered but has very few craters larger than 60 km (38 mi) in diameter. Several multiringed structures are seen on its surface, and the four largest structures are in the leading hemisphere (the side of the planet that faces the direction of its orbital motion); most of the satellites of 79
Beyond the Moon
Fig. 5.10. Image PIA09245: “Ganymede”. Image courtesy of NASA/Johns Hopkins University Applied Physics Laboratory/Southwest Research Institute
Fig. 5.11. Image PIA00362: “Callisto’s Icy Surface”. Image courtesy of NASA/JPL
80
Beyond the Moon Fig. 5.12. Image PIA01617: “Marius Regio, Ganymede”; courtesy of NASA/JPL/Brown University
the outer planets, like Earth’s Moon and Mars’s Phobos and Deimos, all keep their same sides pointed toward their parent planets in a phenomenon known as tidal locking). Valhalla is the largest basin on Callisto and in the Solar System: 4,000 km (2,500 mi) in diameter with a 600 km (375 mi) central bright region. Asgard is the second largest, with a central bright region 230 km (140 mi) in diameter. The topography of most craters in all size ranges shows a subdued profile, with the effect being more noticeable for larger and older craters. Ganymede, the largest natural satellite in the Solar System, is cratered with more relaxed profiles than those of Callisto. The size-frequency profile of the craters is similar to that of the darker terrain of Callisto—up to 100 km diameter. The largest craters are rare and show only pale circular patches of subdued relief in the range 100–300 km (60–180 mi). These “ghost craters” may be traces of ancient major impact structures like those on Callisto. It appears that the same population of impactors bombarded Callisto and Ganymede. However, the dark terrain crater density on Ganymede is three times less than that on Callisto. Overall there are fewer craters on Ganymede (although we would expect more due to Jupiter’s gravity pulling more material in), which implies this moon has a relatively young surface. Ganymede does not show any well preserved multiringed basins like those on Callisto, but there appears to be four sets of remnant furrows that resemble vestiges of impact basins. The bright terrain is younger but the crater density varies from place to place. Overall the brighter terrain has a lower crater density than darker regions. 81
Beyond the Moon
Fig. 5.13. Image PIA01631: “So few Small Craters on Callisto”. Image credit NASA/JPL
One feature that resembles an impact basin is Gilgamesh, a basin 275 km (165 mi) in diameter and appears younger than the basins on Callisto. The morphology of Gilgamesh is similar to impact basins on the Moon, Mars, and Mercury.
Saturn’s Moons Rhea (Fig. 5.14), with a radius of 764 km (460 mi), is a heavily cratered icy satellite. The moon bears large craters of all ages but none of them show signs of viscous relaxing (the post-impact process where the ice becomes soft and results in the features becoming subdued, lowered, or less sharp). This may be due to the lower gravity of the satellite and the rapid cooling and thickening of the satellite’s lithosphere. Rhea’s most densely cratered region is likely from population I objects, with population II cratering near its pole. Iapetus (Fig. 5.15) is of interest: the leading hemisphere has an albedo of 0.1 (like coal) while the following hemisphere has an albedo of 0.5 (like snow). The moon shows an equatorial ridge, heavy cratered surfaces in both hemispheres. Crater densities are comparable to other Saturnian satellites, Mercury, Callisto, Rhea, and the Lunar Highlands. Mimas (Fig. 5.16) has a heavily but not uniformly cratered surface, dominated by Population II type impacts. Large areas of surface lack craters greater than 30 km (18 mi) in size (possibly due to resurfacing), and the most dominant feature 82
Beyond the Moon Fig. 5.14. Image PIA09884: “Soaring Over Rhea”. Image credit NASA/JPL/Space Science Institute
Fig. 5.15. Image PIA08372: “The Himalayas of Iapetus”. Image credit NASA/JPL/Space Science Institute
83
Beyond the Moon
Fig. 5.16. (right). Image PIA08984: “A Major Hit”, image courtesy of NASA/JPL/Space Science Institute
Fig. 5.17. Image PIA09830: “Battered Dione” Credit: NASA/JPL/Space Science Institute
84
Beyond the Moon
on the moon is a 130 km (81 mi) diameter, well preserved crater called Herschel, complete with central peak. Dione (Fig. 5.17) displays a bright, icy surface with the leading hemisphere brighter than the trailing hemisphere. Cratered terrain has numerous craters larger than 20 (6–12 mi); there are less craters at 20 km (12 mi) diameter than the least cratered region on Rhea. Cratered plains show an intermediate crater density while smooth plains show a low crater density. Dione shows signs of resurfacing between Population I and Population II impact epochs. Tethys (Fig. 5.18) shows craters of the same size (almost) as Dione, but with lower density and lower surface gravity. All parts of this moon are densely cratered with Population II craters (smaller craters overall). Most large craters are highly degraded by viscous relaxation; part of the hemisphere is covered by lightly cratered plains. Local resurfacing appears similar to that on Dione. One giant crater called Odysseus dominates one hemisphere; it is 440 km (270 mi) in diameter or about 40% that of the moon. The floor of the crater is relaxed to the shape of the moon and it provides a probe to the interior of Tethys. Several of the largest craters on Tethys have single central peaks (Fig. 5.19). Enceladus (Fig. 5.19), of all the Saturnian system, has the widest variation of crater densities and morphologies. The crater morphologies range from fresh bowlshaped to highly degraded with low rims and bowed up floors. The satellite shows three cratered terrain types: cratered terrain with highly flattened craters 10–20 km (6–12 mi) in size; cratered terrain with well preserved craters 10–20 km (6–12 mi) in size; cratered plains with a lower cratered density and bowl-shaped craters 5–10 km (3–6 mi) across; smooth plains with only a light sprinkling of craters; and smooth plains with no detectable craters. Mimas shows about the same density of craters as the least cratered areas of Saturn’s other major moons.
Fig. 5.18. Image PIA10412: “On the South Side”. Image courtesy of NASA/JPL/Space Science Institute
85
Beyond the Moon
Fig. 5.19. “The North Polar Region of Enceladus”. Image courtesy of NASA/JPL/Space Science Institute.
Uranus’s Moons Oberon (Fig. 5.20) is a heavily cratered moon (at large scales, observations were limited by the 20 km or 12 mi resolution of Voyager 2), with the youngest of the larger craters surrounded by bright ejecta blankets and rays (this appearance implies a dust-rich surface over an icy substrate). Some crater floors have a dark infill. The number of large-impact craters per unit area of the surface is comparable to that of the lunar highlands. These craters are believed to be caused by early Population I impactors, which correspond to the inner Solar System’s late heavy bombardment period. Umbriel (Fig. 5.21) shows a dense population of large impact craters, similar to the lunar highlands. The morphology shows little to no evidence of viscous relaxation. No bright rays or ejecta blankets are visible in Voyager 2 imagery except possibly due to a bright annular feature observed near the limb. Ariel (Fig. 5.22) shows cratered terrain with a lower density than comparable areas on other Uranian satellites. Almost all the craters are population II with a few degraded and relaxed craters in the 50–100 km (30–60 mi) range (these ancient craters have population I origins). Titania (Fig. 5.23) shows abundant population II craters with a few 100–200 km (60–120 mi) basins (relics of population I). This moon has a greater crater density than equivalent regions on Ariel. The largest craters have a somewhat relaxed topography. Titania shows several patches of surface with markedly lower crater densities which indicate resurfacing by volcanic processes. The youngest craters are marked by bright ejecta blankets. Miranda (Fig. 5.24) has a very complex surface, with half of its surface being visible in sunlight, while the other half in perpetual shadow during Voyager 2’s flyby. The moon is densely cratered to the limit of resolution, and is the most densely cratered of all the Uranian moons. A significant fraction of the craters have softened appearances at their rims. The rest appear very fresh (very few show transitional morphology between the very fresh and the degraded). The other half of the visible surface shows a much younger surface with a much lower crater density—much lower than anything else observed in the Uranian system—a fresh appearance overall. 86
Beyond the Moon Fig. 5.20. (left). Image PIA00034: “Oberon at Voyager Closest Approach”. Image courtesy of NASA/JPL.
Fig. 5.21. (above). Image PIA00040: “Umbriel at Closest Approach”, image courtesy of NASA/JPL.
87
Beyond the Moon
Fig. 5.22. (left). Image PIA00037: “Ariel at Voyager Closest Approach”, image credit: NASA/JPL.
Fig. 5.23. (above). Image PIA00039: “Titania - Highest Resolution Voyager Picture”, image credit NASA/JPL
88
Beyond the Moon Fig. 5.24. Image PIA00140: “Miranda Fractures, Grooves and Craters”, image courtesy NASA/JPL
The Influence of Impactor Characteristics Impacting objects come in many shapes, sizes, and densities. Fluffy impactors leave different marks compared to iron-rich dense impactors. Also, impact velocity plays a big role in the size and appearance of craters. Most large impactors come from the asteroid belt, where collisions between asteroids break up the objects and send some spiraling toward the Sun. The Earth, Moon, or another planetary object may get in the way and the result is a collision. If a meteoroid is from the asteroid belt, the typical orbital velocity is about 10–25 km/s (6–16 mi/s). The velocity that a meteoroid attains prior to its impact with a planet depends on a number of factors such as the relative velocity of the objects (that is, how fast the meteoroid moves as seen from the planet’s surface, which includes both objects’ velocities) and the size of the target which leads to a gravity field of a certain strength and a certain acceleration of an impacting meteoroid, resulting in its final impact velocity. The amount of energy carried by the impactor depends on how big and how massive the impactor is. A small impactor will have less kinetic energy than a large impactor of the same material; a small impactor of iron will have more kinetic energy than an equivalent sized impactor of stony material. Finally, if the impactor passes through an atmosphere, it will influence the type of crater that it makes. The friction arising from the motion of the impactor as it transitions from the vacuum of space to a gaseous medium heats up the object and places great stress on it as it decelerates. The heating causes the meteoroid to glow and become a meteor; the stresses of deceleration may be more than what the object can bear, and may cause the object to break apart. The smaller component objects may then burn in the atmosphere before reaching the ground, or they may 89
Beyond the Moon
remain intact until reaching the surface and creating several craters. The internal structure of an impactor, along with its incident velocity, determines whether it will be able to survive the stresses of an atmospheric entry. Small objects tend to burn up well before they reach the surface, which is why planets like Earth, Venus, and Titan show no craters smaller than a certain size. If the impactor encounters an airless world, then the kinetic energy of motion is immediately transferred to the target on impact. It is this transfer of energy that makes the point-like flash visible to several of us as we watched the Moon for Leonid meteor hits back in 1999 and 2001. Since no impeding atmospheric barrier is present, objects of all sizes make it directly to the surface, making craters of all sizes. Hence, the Moon and other airless bodies show craters of all sizes, from microscopic to large size.
Atmosphere vs. No Atmosphere Whether an object possesses a thick atmosphere, such as the Earth or Venus, or whether an object is essentially air-less, like the Moon or Mercury, will determine the types (and frequencies) of craters found on their surfaces. As has already been discussed, worlds with solid surfaces but thick atmospheres show virtually no small craters because of the presence of the atmosphere that protects the surface from meteor impacts. Craters on Venus, Earth, and Titan are few and far between, but they do exist. I will continue with this overview about craters on other worlds by looking at the crater morphologies and coverage on Mars and Triton, the large satellite of Neptune.
Mars Craters Most of the Martian craters are small < 15 km (<9 mi.) and are simple, bowlshaped features with raised rims, smooth-looking walls and floors. Larger craters (>15 km) are complex with complex features and central peaks. Fresh craters > 15 km come with terraced walls, flat hummocky floors similar to large lunar craters. Martian craters, for each size, are shallower than lunar craters of the same size. Rayed craters are extremely rare on Mars and ejecta blankets, and secondary crater fields are much less abundant on Mars than on the Moon. Most fresh-looking Martian craters have lobate ejecta blankets that look like the meteor hit a mud patty; the outer edges are marked with escarpments. The ejecta flow causing these ejecta fields due to gases or water contained in ejecta come from impact melting of ground ice or permafrost. Martian craters with diameters more than 100 km (60 mi) have a central peak that takes the form of a central ring. Those larger still have their central rings replaced by a group of concentric rings. A total of three large and 20 smaller ringed basins are known on Mars. Most of these are close to 200 km (120 mi) in size, but they range in size from 135 km up to 2,000 km (80 up to 1,240 mi), and nearly all of these are found in the highly cratered southern hemisphere. Martian impact basins are much more degraded than their lunar counterparts; three of these are visible from Earth, near the Martian opposition. These include: 90
These are degraded remnants of impact basins which are ringed by mountainous regions—remnants of rim structures and ejecta.
Beyond the Moon
• Argyre-1,400 km (870 mi) diameter-the freshest ringed basin on Mars located near 43°W 50°S • Isidii-located near 270°W 15°N and is 110 km (70 mi) in diameter • Hellas-almost twice as large as the largest lunar basin, 243°W 42°S, 1,600 km × 2,000 km (1,000 mi × 1,200 mi) in dimension, up to 4 km (2 mi) deep
Triton Triton shows a relatively young surface, with regions either crater-free or with a crater density similar to that of the lunar highlands. The crater density drops away from the apex of orbital motion, showing a relatively young surface. This satellite has a thin atmosphere, representing an interesting situation10 that meteoroids possibly make visible meteors as they enter the Triton’s atmosphere. Unlike Earth’s meteors which burn up high in the atmosphere, Triton’s meteors make it all the way to the surface as “shooting stars”, comparable in brightness to what is seen on Earth.
Meteor Showers on Other Worlds Some 417 periodic comets are known to exist as of August 2008, many of which pass through the inner Solar System. When these comets make repeated passes through the inner Solar System, they leave trails of dust and debris along their orbits, with each particle or fragment on its own orbit around the sun. When the orbital paths of the Earth and a comet come close or intersect, the Earth passes through a stream of such particles, and the collision between the Earth (actually its atmosphere) and each particle produces a meteor as the particle burns up in the Earth’s atmosphere. These orbit-crossing events produce what have come to be known as annual meteor showers, the most famous of which is the Perseid meteor shower, which is a stream of debris from Comet 109P/Swift-Tuttle. The International Meteor Organization maintains a working list of 33 annual meteor showers observed visually throughout the year, 10 of which have zenithal hourly rates (ZHR, the number of meteors observed by a single observer with the radiant of the shower at the zenith and the skies with a limiting magnitude of +7.5) of 10 or more. When meteoroids burn up in the Earth’s atmosphere, they do so at heights of 70–120 km (75 mi). Teams of scientists have attempted to predict not only how meteors interact with planetary objects, but also whether annual meteor showers occur on other planets like they do on the Earth.11–14 Those looking for annual showers on other planets have done so by comparing the orbits of planets of interest to those of comets, making assumptions about the width of the debris stream and calculating their maximum ZHR’s. In our Solar System, Venus, Earth, Mars, the four gas giant planets, Titan, and Triton have atmospheres thick enough to produce air-borne meteor trails. On Mars, meteors burn up at altitudes of 50–100 km (30– 60 mi), but are similar in brightness to those seen on Earth. On Venus, the higher density of its atmosphere implies greater heights: 200–300 km (120–180 mi), but are only observable from orbit due to the ubiquitous cloud cover. From the height 91
Beyond the Moon
Table 5.1. Predicted numbers of annual showers at various planets Planet
No. of annual showers
Earth Mars Jupiter Saturn Uranus Neptune
1→4 0→5 28→148 2→24 1→6 0→1
of the International Space Station on Earth (350 km or 220 mi), assuming a comparable orbit height at Venus, meteoroids entering Venus’s atmosphere tend to be 0.2-mag brighter than comparable meteors in Earth’s atmosphere (assuming the same meteoroid mass and velocity in both planets).15 Meteors occur in the atmospheres of the giant planets, much as they do on Earth. Meteoroids burn up in Jupiter’s atmosphere 115–290 km (72–180 mi) above the 100-mb reference level of the atmosphere. In fact, Voyager 1 observed a meteor trail in Jupiter’s atmosphere during its flyby. If the meteors are large enough, however, they have no solid surface to impact, but they merely run into denser layers of atmosphere. We have seen the effects of meteoroids colliding with thicker layers of atmosphere when the pieces of Shoemaker-Levy 9 entered Jupiter’s atmosphere and were abruptly halted deeper down—producing plumes and scars that were observed by space- and ground-based instruments. Meteor showers are predicted to occur at each of the planets mentioned above (Table 5.1). When a planet passes through a meteoroid stream, information about the age, size, composition, ejection (time since release from the comet nucleus), speed, and structure provide clues to the history of the comet itself. There are almost four times as many comets that approach Mars’s orbit than the Earth’s, and most of these are comets belonging to the Jupiter family. At least six annual meteor showers had been predicted at Mars, with observational evidence of one in April 2003 (but not a single shower was observed one Martian year later in 2005). The following numbers of meteor showers per planetary year are predicted to occur; the estimate range depends on the encounter distance from the core of the stream. Triton could be an interesting place to watch meteors. Meteoroids entering Triton’s tenuous atmosphere would enter fast enough (~19 km/s or ~12 mi/s) to be heated to incandescence. To an observer on Triton, the meteors would be seen as bright as what is seen on Earth, but these slow-graceful meteors would continue to glow until they come closer to the ground.10 After considering the possibilities of meteors on other planets, the next question is: can we see any of this activity from large ground-based telescopes on Earth?
Other Historic and Probable Impact Sightings on Other Worlds Apart from the impacts observed on Jupiter during the 1994 impacts of comet Shoemaker-Levy 9 fragments, and the Voyager 1 observation of a meteor on Jupiter during its 1979 flyby of the giant planet, no observations of impacts on other 92
Beyond the Moon
worlds (except the Earth’s Moon) have been reported. There are isolated reports of single dark spots in Jupiter’s atmosphere, including the 1690 Cassini observation mentioned above, the most recent being in October 2003, that resemble the impact markings from the Shoemaker Levy comet impacts. These single spots are probably related to atmospheric phenomena but (as is the case for the 2003 spot) this assertion is far from conclusive. The challenge of finding an impact on another world besides the Earth and Moon is twofold: the distance to even the nearest planets renders almost all impact events invisible. These would be too remote to be seen from ground-based instrumentation. Even if it occurred on the planet’s night side the great distance makes a typical event much too faint to be seen with ground-based instrumentation. Most planets are visible from Earth with most (or all) of its Earth-facing hemisphere in sunlight, impeding the visibility of any impact except for the most rare and largest of impacts. Finally, the timing of the impact is problematic, with virtually no way of telling if and when an impact large enough to be seen from Earth will occur. Rare exceptions do occur, such as the discovery of an asteroid on a potential collision course with Mars in late 2007, with the impact predicted to occur in early 2008. Later refinements in the asteroid’s orbit revealed that the object would miss Mars entirely, which it did. Mercury and Venus, during their crescent phase, present a significant portion of their night sides toward Earth, which is favorable for impact visibility. Mercury is too small, distant, and never gets far from the Sun to enable a useful monitoring program. Venus, on the other hand, presents the best opportunity after the Moon to spot impacts on another world: it sports a large angular size, relatively closer to the Earth, and sufficient (though barely at times) solar elongation visible for short periods of time if the local ecliptic angle (to the horizon) is favorable. Even then, a Venus orbiting spacecraft would be far superior at detecting meteors than any Earth-based attempts. Mars’s proximity to the asteroid belt suggests possibly that impacts would be more common there. In fact, the satellites of Mars, Phobos, and Deimos are thought to be captured asteroids and elongated craters on the surface of Mars might be craters made by previously captured asteroid-satellites. The orbit of Phobos is presently decaying and is expected to impact the planet or decay into a ring system in 50 million years. Although such impacts would be potentially observable from Earth, the long-wait time and the low number of such large events make these nearly impossible to watch from our planet. Efforts have been undertaken to detect very fresh craters on the Moon and Mars. I am not aware of any unambiguous detection of a fresh feature (that is one created in the twentieth or twenty-first century) on the Moon, but 20 of them have been reported on Mars by images taken with the Mars Global Surveyor spacecraft (Table 5.2). The catalog includes position information on Mars as well as the time interval (that is, the dates of images taken of the region) that constrains the impact event, and finally a brief description of the site.16 Almost all the sites show dark material on a brighter background, with the material forming some interesting patterns. With the exception of the ~150 m crater, all the new craters are in the tens of meters of size, with impactors being in the meter size class or less.
93
Beyond the Moon
Table 5.2. Very new impact features, as observed by the Mars Global Surveyor satellite
94
Site
Lat.
Long.
Occurrence window
Notes
1
14.0°N
151.5°W
12 Nov 04 to 6 Jan 06
2
25.8°N
308.0°W
21 Dec 05 to 31 Jan 06
3 4
0.8°S 23.3°N
160.0°W 307.2°W
4 Nov 02 to 25 June 04 4 Feb 01 to 13 Jun 05
5
20.6°N
356.8°W
30 Jun 02 to 7 May 03
6
20.0°N
152.7°W
5 Nov 05 to 13 Nov 06
7
0.03°S
133.2°W
9 May 04 to 17 Feb 06
8
2.5°N
136.0°W
25 Mar 04 to 15 May 04 (51-day interval)
9
11.5°N
156.6°W
11 Feb 02 to 4 Sep 03
10
29.3°N
333.2°W
5 May 03 to 29 Apr 05
11
27.3°N
91.8°W
18 Apr 08 to 17 Feb 04
12
22.2°N
345.5°W
2 Apr 01 to 11 Dec 03
13
5.5°N
135.7°W
2 Jun 99 to 20 Aug 03
14
26.4°N
336.5°W
8 Dec 03 to 26 Nov 05
15
17.0°N
160.7°W
11 Sep 03 to 27 Jan 04
16
13.9°N
84.4°W
17 Feb 04 to 26 Feb 06
17
25.7°S
136.2°W
26 Mar 01 to 9 Jan 03
18
28.7°N
334.9°W
14 Mar 05 to 26 Feb 06
19
5.4°N
136.8°W
20
7.0°N
112.2°W
22 Jan 04 to 22 Apr 04 (88-day interval) 17 Sept 05 to 26 Feb 06
Multiple craters, approximately seven at ~10 m ± 1.7 m diameter. A singular crater of 11.0 ± 1.7 m diameter is offset to the north-northwest from a cluster of craters. Dark markings on brighter background A single crater of 16.0 ± 1.7 m diameter, with several dark spots near the south and east. Dark markings on brighter background Dark markings on brighter background; a single crater of 17.0 ± 3.0 m diameter The impact site exhibits a single crater of 15.6 ± 1.7 m diameter, mostly dark on bright background. The cPROTO view shows nice multitoned ejecta and displays secondary impacts at some distance from the crater The impact site exhibits a single crater of 22.6 ± 3.0 m diameter. Dark central spot with bright ejecta against a dark background The impact site exhibits a single crater of 12.6 ± 3.0 m diameter. Appears as dark material (spot and ejecta material) on a bright background The impact site exhibits four craters (three adjoined and one separate from the group). The diameters are 22.4 ± 3.0 m, 15.4 ± 3.0 m, 12.6 ± 3.0 m, and 12.6 ± 3.0 m. Dark markings on a brighter background The impact site exhibits four craters with diameters of 29.6 ± 3 m, 21.6 ± 3 m, 19.6 ± 3 m, and 18.2 ± 3 m. They appear as dark spots amidst a dark, slightly irregular halo against a brighter background The impact site is a single crater with a 11.2 ± 3.0 m diameter. This includes darker material against a brighter background The impact site is a single crater with a 11.2 ± 3.0 m diameter. Dark material against a bright background A single, dark crater with a 19.8 ± 3.0 m diameter, surrounded by dark ejecta, all of this is on the upper north flank of Ulysses Patera The impact site is a single, dark crater with a 24.0 ± 3.0 m diameter and dark ejecta surrounding it The impact site is a single crater with a 28.2 ± 3.0 m diameter. It appears as a well-defined crater, complete with sharp rim, bowl-shaped profile, and darkish and bright ejecta field The impact site is a single crater with a 22.6 ± 1.7 m diameter. Many small secondary impacts occurred and spread out several kilometers from the crater. A dark spot marks the main crater, and the area is surrounded by a complex tangle of dark filamentary ejecta deposits The impact site is a single crater with a 14.0 ± 3.0 m diameter, a dark ejecta field, with a dark parabola of disrupted surface material downrange The impact site includes two craters, the largest has a diameter of 12.6 ± 1.7 m, the smaller is about 4.2 ± 1.7 m across The impact site has a single crater of 148 ± 3 m diameter. There is uncertainty about whether this crater is new; it is considerably different from the other craters in this study, it lacks a particularly large, dark, blast zone relative to the crater size. This crater is not visible in any images obtained prior to THEMIS I04755006 (9 January 2003), including Viking orbiter data The impact site has two craters, one of 12.0 ± 3.0 m diameter, the other about 6.0 ± 3.0 m in diameter. These are surrounded by dark material against a bright surface The impact site two craters, one of about 10.0 ± 3 m in diameter, the other about 4.0 ± 3 m in diameter. Dark material superposed on bright background The impact site consists of multiple craters of diameters: 16.8, 11.2, 11.5, 6.0, 6.0, and 6.0 ± 3.0 m. A dark wind streak pointed toward the northwest emanates from the site, suggesting that disrupted dust was mobilized by wind sometime after the impact occurred. The overall appearance is one of a dark comet against a bright background
In conclusion, evidence for impact events in our Solar System’s history is widespread, and the physical characteristics of craters on various worlds depend on the physical nature of the target (planet or asteroid) and the impactor (rocky or metallic). For planets with thick atmospheres, most of the impacts occur in the atmosphere as the objects burn up as “shooting stars”, like what is commonly seen on Earth. Large enough impacts produce interesting results, such as the dark stains in Jupiter’s atmosphere after the impact of comet Shoemaker-Levy-9. Although the evidence of past impacts are all around, observations of an impact as it happens (except for terrestrial meteors) is a rarity, as is the detection of craters that were formed very recently (i.e. in the last 100 years or so). But the list of impact observations is growing: recent impacts on Mars, lunar meteor impact observations (129 and counting for NASA, 80 and counting for ALPO-LMIS), extensive observations of the Shoemaker-Levy 9 impact on Jupiter, indirect evidence of collisions of objects with Saturn’s rings, the 9P/Tempel 1 man-made impact, and indirect impact evidence in comets as indicated by sudden outbursts and breakups (although other factors likely play a role, not necessarily the impacts). We have seen how the appearances of craters on other worlds change with different surface characteristics, the presence of an atmosphere, etc. and these compare rather well with the array of craters visible on the Moon. In fact, superficially, most of these objects look like the Moon, whereas in reality there are some striking differences. These differences include the size of the objects being hit (only four of the nearly 160 natural planetary satellites are larger than the Moon, and only a total of 18 of these are round or nearly so in shape), their mass, and their surface composition. All of these factors mentioned, and more, contribute to the final determination of what a particular crater looks like. This chapter also extended the formation history of the Solar System to include impact episodes on the satellites of the giant planets. I introduced three populations (or generations) of impacting objects made of rocky or icy material. These populations contributed to shaping the surfaces and dictating the final appearances of most of these satellites. I conclude that impacts are happening right down to the present day, not only on the Earth’s Moon, but also throughout the Solar System.
Beyond the Moon
Conclusions
95
Part II
An Observer's Guide to Lunar Meteor Impacts, Past and Present
Chapter 6
In addition to the guide (Chapter 9) which shows how to observe impacts as they happen, I provide in this chapter a brief guide for observing existing lunar impact features. Although the impacts observed using ground-based telescopes produce craters that are much smaller than what can be observed through a typical backyard telescope, one can observe their much larger cousins and consider how an impact that resulted in that crater may have looked from the Earth. At first glance, the Moon’s surface shows an overwhelming mix of craters of all sizes. With higher magnifications, details appear in the rims of central peaks of the larger craters and smaller craters become visible. Also, as one looks more closely, one notices that there are areas covered with craters and others that look smooth and have far fewer craters. A closer inspection (at the highest magnification the instrument will allow) reveals that an array of features and shapes change with size. The following sections build upon the background material, along with examples illustrating various descriptions of craters, from Chapter 2 by providing additional information on how to look for the differences apparent in contrasting situations. I will assume that you, the reader, have already selected a telescope with which to make the observations; if not, you are encouraged to look into your favorite reference material on how to buy a telescope for a particular astronomical purpose. For lunar and planetary work, telescopes with long focal length are preferred. The actual telescope selected depends on the tastes and experience of the observer. If you are a beginner then I would recommend a smaller telescope, say a 4.5 in. (114 mm) f/8 Newtonian, which can, with a 4 mm eyepiece, reach magnifications of 225×. If you are more advanced, then more sophisticated equipment may suit your tastes. Binoculars will serve many of the purposes of observing lunar features that are described below, though you will be limited to the largest features. The image in Fig. 6.1, courtesy of NASA photojournal was taken by the Galileo Spacecraft on its way to Jupiter. The image, obtained on 7 December 1992, shows the Earth-facing side of the Moon near its full phase. Major impact features, as well as the Earth-facing Maria are seen in this image. Although this image was made near Full Moon, the phase least favorable for observing the texture of the lunar surface, one can still see many craters and related features on the lunar surface, B. Cudnik, Lunar Meteoroid Impacts and How to Observe Them, Astronomers’ Observing Guides, DOI 10.1007/978-1-4419-0324-2_6, © Springer Science+Business Media, LLC 2009
Guide to Observing Impact Features
Guide to Observing Impact Features on the Moon
99
Guide to Observing Impact Features
Fig. 6.1. “Moon”; Image courtesy of NASA/JPL/USGS
particularly along the northern (top) edge. This is because the perspective of the Galileo spacecraft was a little above the plane of the Moon’s orbit, looking down on the north lunar pole more so than we are able to on Earth-even with favorable libration (tilt). Objects with ejecta blankets that appear brighter than their surroundings can also be seen, particularly in the Maria regions. Note the prominent ray systems of Tycho (just left of center near bottom edge) and Copernicus (large bright spot in dark Maria left of disk center). Some generalities to notice when looking at the Moon include the above changes in appearance with magnification. As you work your way up from low magnification to higher and higher powers, you can see more details, but the amount of detail is limited by a phenomenon known as “atmospheric seeing”, or “seeing” for short. Seeing is the result of the motion and turbulence of Earth’s atmosphere and can have a significant deteriorating effect on the quality of the image. With bad seeing, the image is more blurred, and it becomes necessary to observe at lower powers. With excellent seeing, high magnification can be used since the image is steady and sharp. The first step, then, in attempting to observe lunar features is to gauge the quality of seeing. If the seeing is excellent, you can look at finer details in the form 100
Maria vs. Highland Cratering Just glance up at the first quarter, gibbous, or full Moon and the first thing you notice is that there are two different shades of color visible. The darker gray is the Maria and the lighter gray (or yellowish-white, depending on when you are looking) are the highlands. A look through a telescope at low magnification shows these two distinct regions in more detail: darker gray, smoother regions with few craters (the Maria); and brighter gray, more rugged regions with many craters (the highlands). Based on the considerable differences in the number of craters in each of the regions, coupled with the theory of lunar and solar system history as presented in Chapter 1, there is a difference in the relative ages of each region. The rugged highlands are generally older, while the Maria are generally younger. In addition, one can get an estimate of how old the lunar surface is by counting the number of craters per unit area, which gives the crater density, which is related to the age of the surface. Maria tend to be smoother and darker in appearance – but not all Maria are of the same shade of gray. Look across the Maria and you will see some lighter and darker regions. Also, look at the edges of the Maria and you will notice that they tend to look like arcs of circles. This implies that the Maria are actually large craters or “walled plains” caused by asteroid impacts as was mentioned earlier. Molten rock filled in these large basins, which together form the familiar “man in the Moon” or “woman in the Moon” forms are seen with the unaided eye. Notice also that the “lunar mountains” are not like terrestrial mountains; rather they are formed by tectonic processes as on Earth, many lunar mountain ranges are actually the walls of the large ancient, Maria-bearing impact basins and craters. Also note that most of these mountain ranges occur near Maria-highland boundaries, which add evidence in support of the explanation that lunar mountain ranges are actually the remnant walls of these gigantic ancient craters. At first glance with the naked eye, the Moon appears as a smooth, two-toned sphere, but through binoculars its rugged nature becomes apparent, especially the brighter highland regions. Even a low power telescope view shows the Maria being mostly smooth, with some large craters superimposed. Increase the magnification (if the seeing allows) and notice that there is some texture and lots of small craters that pepper the Maria. Also, observe regions of the Maria during the first and last quarter and compare the appearance of, say, Copernicus, Kepler, Aristarchus, and
Guide to Observing Impact Features
of smaller scale features of larger craters, and you have the ability to pick out the smallest craters at high powers. If the seeing is less than ideal, you are advised to stick with low powers and look at the larger features. Also, it is useful and interesting to look at various features under different sun angles. It is impressive to see the rims of craters catch the first light of a lunar day, then, over the course of a few days, watch the shadows get shorter and shorter, revealing different aspects of a particular feature. While it is best to look at impact features during the days surrounding the first and last quarter phases, it is also interesting to look at the Moon when full, to try and pick out a feature of interest and see if it has a ray system or ejecta blanket. One quick way to gauge whether a crater is young or old is to look at how bright the feature stands out against its surroundings. More information about this is presented later in the chapter.
101
Guide to Observing Impact Features
Lagrenus; then their appearances during a full Moon (also try to identify these features in the earthshine when the Moon is a waxing or waning crescent). Select a handful of larger craters found in the Maria region (those mentioned above will do, or you can use some from the “Top 100” list presented later in this chapter, or you can choose your own) and compare them with similar sized features found in the highland regions (e.g. Tycho, Maginus, Theophilus, etc.) and compare the aspects of their appearance under various illuminations – but try to select highland/Maria crater pairs that have similar longitude so you can compare them under similar lighting conditions. Also, observe them during a full Moon. What are the obvious (and not so obvious) similarities and differences between the two sets of craters? Look closely at them, taking time to really explore the finer aspects of these features and to use moments of excellent seeing to tease out the finest details. You may want to make a sketch or take an image of your craters of choice to help your comparison. In summary, consider the following: ● ● ● ●
Do the craters both show terraced walls and central peaks? Do they have secondary craters both outside and inside their walls? Do they have ray systems or ejecta blankets? Do the Maria craters have dark smooth floors or bright rough floors? Has there been any flooding of the crater? How does the appearance change with changing illumination? How does the appearance change during the full Moon versus during a quarter Moon? ● At full Moon, compare and contrast Mare Crisium and Grimaldi. How are they the same and how are they different? ● ●
At full Moon, can you trace out the host basins that make up each of the Maria? How many tiny bright spots can you pick up on each sea, and how does the number of bright spots for a certain area (say Mare Serenitatis) compare to the number of similar sized highland area? Compare Plato to Aristotle – how do they compare and contrast? All of these questions presented above serve as guidelines and are meant to direct the reader into an observing project that involves looking closely at impact features and making comparisons. In so doing, the observer develops his/her ability to pick out finer details and to get the most out of their observing sessions. Finally, look at several of the smaller craters located in an area of Maria and compare their appearances with similar sized craters in a nearby patch of highlands. Again, look during times when the sun angle varies and during the Full Moon, and ask yourself the above questions to guide you in your search. You may even generate some questions and guidelines of your own based on your own experience and the material presented elsewhere in this book.
Young vs. Old Craters There are several approaches to gauge the relative ages of craters. I discuss them in turn below, using some common principles in geology 101 to show how we can estimate the age of a crater. We can also tell the age of a surface by counting the craters: the older the surface, the more craters. This works for surfaces that are 102
nearly as old as the Solar System itself: for the majority of the history of the solar system the rate of cratering is low, but individual craters, due to exposure to radiation and micrometeoroids, change appearance over time. Fresh material tends to be brighter and bluer; “weathered” material darker and redder. The outline and sharpness of the crater changes over time as well.
We can also get an idea of the age of the crater by context: if the surface is heavily cratered and the crater itself appears to have been disturbed by others, it is true that the disturbed crater is older, since it was first formed, and then disturbed. On the other hand, if the crater stands nearly alone in a flat Maria plain, or if it appears well-defined and undisturbed amidst a background of more degraded craters, or itself appears to have disturbed surrounding craters, then the probability of the feature being relatively young is high. It is this basic law of geology that tells us that the Maria are younger than the highlands: it appears that a number of highland craters are partially or totally flooded by the molten rock that makes up the lunar seas. Something happened that caused the lava to flow on the surface, covering large tracts of Moon, including countless craters. Impacts that likely preceded the laying down of the Maria may have themselves been partially or totally obliterated by the impacts of several asteroid-sized objects that produced the basins that would contain the molten rock. After the Maria solidified, a number of small impacts and a few larger ones occurred, resulting in the widely scattered cratered environment that characterizes the Maria today. It is easy to see some of these geologic laws demonstrated while observing the Moon. Look for buried craters and “ghost” craters near the Maria-highland boundaries. Can you see several craters in different stages of burial? Also, look for instances where craters are superimposed on one another, or where the rim of a larger crater is interrupted by a smaller crater. Look for the rare double-rimmed crater where a second impact made an almost perfect “bull’s-eye” impact in a larger crater. Another example of superposition (in this case, “cross-cutting relationships”) includes rilles (long narrow depressions, possibly collapsed lava tubes) and rupes (or scarps, cliffs, or locations where the crust fractures and slips) that cut across craters, and vice versa. In the case where the long feature cuts across a crater, the crater was present earlier, with the event causing the scarp or rille occurring later. If the reverse was true, the crater would appear intact with the rille or scarp appearing interrupted or disturbed by the event that produced the crater.
Guide to Observing Impact Features
Geologic Law of Superposition and Relative Ages
Crater Counting Crater counting provides an excellent means of estimating the approximate age of a region. The greater the number of craters in a given region, the greater the age of the surface: the younger Maria have fewer craters while the older highlands have more craters. The “Exploring the Moon” column printed monthly in Sky and Telescope magazine provides, each month, guidelines to observe some aspect of the Moon. A few simple guidelines need to be followed to ensure the accuracy of the estimate17: 103
First, do not count craters of volcanic origin, such as the volcanic pits that are found along Hyginus Rille. One may be certain that a feature of interest is volcanic in origin by noting whether it is located along a rille or at one end of it. It is difficult to know whether these types of craters are young or old, but based on current theory they are likely quite old. ● Second, do not count secondary craters which can be hard to distinguish (but easier if located along a primary crater’s bright rays). All the secondary craters of a particular impact were produced at the same time (actually just after, but for the purposes of discussion, and considering geologic time scales, we will say that they were produced simultaneously) and only give the appearance of an old surface by their numbers, whereby they could have all been produced rather recently. ● Third, you need to make counts in areas that formed all at one time, such as a region of Maria. This provides consistency in the comparison of different areas on the Moon; that is, the counting process is not complicated by the use of areas that contain two or more ages of formation (e.g. counting craters in areas that include parts of a Maria and a highland in the same area and in a single count). ● Finally, you need to make the counts over the same sized areas (or at least correct the counts to that area size). This enables consistent comparisons of regions and does not artificially inflate the counts of one area over another, due to the larger size of the first area.
Guide to Observing Impact Features
●
Physical Appearance of Features Young craters show bright ejecta rays, sharp rims, prominent ejecta blankets, secondary craters, and a fresh, bright appearance overall. Older craters show a darker, more degraded appearance: the rays disappear, secondary craters become subdued and disappear, the rough-textured ejecta blanket takes on a smoother texture, the rim sharpness decreases, and any terraces are modified by radial channels. Simple craters, as they age, are partially filled by ejecta from later impacts, and their profiles change over time. Young simple craters have round, bowl-shaped profiles with raised, sharp rims while older simple craters have flat floors and rounded rims. Complex craters also get shallower over time and get filled over time by lava or impact ejecta. Sometimes the central peak is partially covered, other times it is completely buried. Look for these details as you examine features of different ages. As you become familiar with how craters of different ages look, and also consider how their surroundings look in terms of the density of craters (again, being careful not to consider secondary craters), you will be better able to estimate how old or young a crater or its region is by its appearance. Next, we will take a look at how the appearance of craters changes with size, focusing on general patterns seen on the Moon.
Appearance vs. Size Compare the appearance of craters, such as Tycho and Copernicus, with the smallest ones that you can definitely make out at high power, and can see them as more than dots, and you will notice not only the difference in size but also in 104
Guide to Observing Impact Features
complexity. The physics of the impact event dictates the appearance of the resultant crater. The form or morphology of the crater ranges from small and simplebowl shaped to large and complex, complete with central peaks and walled terraces. The important factors that influence the appearance and size of craters include the impactor velocity, density, and size. For the simplest craters, which are the smallest that can be viewed through ground-based telescopes, from about 1 km up to 15 km (0.6 mi up to 10 mi) across, tend to be bowl-shaped, with a smooth, slightly concave floor; sharp, well-defined rims; and steep inner walls; no central peak; and a small debris field. Craters larger than 15–20 km (9–12 mi) in diameter become more complex in appearance. Those ranging in diameters from 20 to 175 km (12–105 mi) usually have central uplifts that look like a central mountain peak or group of peaks, and tend to have flat floors and more complex walls. Larger craters (diameters greater than about 175 km) usually have complex, ring-shaped central peak clusters. Those that are larger still, with diameters greater than 200 km (120 mi), are called basins. There are many examples of each type of crater on the Moon, and they are listed in the “Top 100” list below. I repeat the examples given in Chapter 2, and the reader is encouraged to refer back to Chapter 2 to see images of each example. In addition, in the “Top 100” list, diameters are given for each entry, so that the observer can select features within a certain size range and look for the details outlined in this section. An excellent example of the simple bowl-shaped crater is a feature called the Moltke Crater. This crater is about 7 km in diameter and shows a modest debris or ejecta field. Simple craters (<15 km or <9 mi) generally show a circular outline, a bright inner wall, and a parabolic shadow reflecting a bowl-shaped profile. One example is the 13 km (8 mi) wide feature Mösting A, which is up to 2.7 km (1.7 mi) deep. Look at these particular features at high power and different sun angles; low sun angles show the shadow on the crater floor well, and one can ascertain the profile of the crater by studying the shadow shape. Larger, more complex craters (15–50 km or 9–31 mi) are complex structures with slumping material. The outline is polygonal in shape rather than circular, and it encloses a floor littered with mounds of material that have slid down the crater walls. In some cases, the walls are terraced and the floors flat, with one or more central peaks. The rims are raised and the surrounding terrain shows a radial pattern of rays with secondary impacts up to several kilometers across. Triesnecker is an example of a crater in this size range; it measures 26 km (16 mi) across and is up to 2.7 km (1.7 mi) deep. Craters larger than 150 km (95 mi) exhibit a ring of peaks in the place of the central peak(s). Many of these are classified as basins (large craters with distinct central rings), especially if their sizes are 200–300 km (120–190 mi) across or more. Ringed basins such as the Orientale basin consist of concentric mountain rings and depressions. Rings can extend to 500 km (310 mi) from the center, and mountains that form the rings can have peaks as high as 500 m (1,625 ft). Basins are thought to have formed from the impact of asteroid-sized bodies or comets and their impacts cover most of the lunar surface with their debris and ejecta. There are some 43 basins cataloged on the Moon that have diameters greater than 220 km (140 mi). Roughly equal numbers are on the near side and far side; those on the near side are generally flooded with lava basalt to form the familiar Maria. Several basins are included in the “Top 100” list given below.
105
Guide to Observing Impact Features
How to Recognized Different Types of Features
106
The best times to look for various types of lunar craters is between 3 days after New to 2 days prior to Full, and again from 2 days after Full to 3 days prior to New. This will vary with ecliptic angle, solar elongation, and so forth, but the common feature is the oblique sun angle. The low sun angle produces shadows which identify the physical character of a feature much easily, and adds a 3-D relief to the situation. The full phase (or within one day of Full) is best for observing ray structures and bright spots on the lunar surface. A reliable approach to recognizing different types of features is by studying the morphology of the feature. Features to look for include the presence or absence of a central peak or complex of peaks. The appearance of this central part of the crater formation enables one to gauge the size range of the crater itself. Also look at the structure of the walls of the crater. Is the structure intact or is it broken, and if so, by how much? Are there secondary craters on the floor of the main crater or among the ejecta blankets of the main crater? What does the ejecta blanket itself look like: is it well-developed or quite faded? How does its appearance change with sun angle? How do all of the impact features change with changing sunlight? Where does the crater occur, in the Maria or highland regions? Does it appear to have disturbed its neighboring craters or does it stand out seemingly untouched. Does it interact with any other craters in the region? A useful and interesting exercise would be to observe a handful of features under various lighting conditions, from low sun angle to high sun angle (Full Moon) to see how the feature changes appearance. It would be a good idea to make drawings of features of interest. Drawing a lunar feature helps you focus on the finer details of the feature and to pull out more than what you would have done by merely looking at the crater. Alternately, you can take digital images of craters of interest. There is a number of interesting projects that observers can undertake to increase their ability to recognize different features. Drawing features, as mentioned above, hones the observer’s ability to identify characteristics that make a feature what it is. One can also make a series of drawings of one feature and its immediate surroundings over the course of a lunar day, showing changes in appearance with changing sun angle. The observer is encouraged to do this for a handful of features of different sizes. Also, the observer can select a continuum of craters of different ages, and assemble an “age sequence” of drawings of features of different ages, showing the evolution of a typical crater over time. Some of the features in the list (Table 6.1) have published ages to aid in constructing such a series. Again, if the observer is equipped to do so, and so desires, one can take digital images to accomplish the tasks described above. Additional projects include observing and recording each of the features listed in Table 6.1, being certain to record the date and time of the observation along with instrument and weather data, and a word or two about the appearance of the feature. The observer can work on both the Astronomical League observing clubs involved in observations of the Moon; the “Top 100” list provides an excellent start for working toward membership in the two observing clubs involving the Moon. These projects that have been described in this and preceding sections will give you considerable practice in picking out and identifying various types of craters.
No.
Name
Nature
Est. age (Ma)19
RC #
Lat.
Long.
Size (km)
Size (mi)
1 2 3
Impact structure Crater with ray structure Basin rim
834 138 3,500
31 64 –
9.7°N 43.4°S 18.9°N
20.1°W 11.1°W 3.7°W
93 102 400
58 63 249
4 5 6 5 6 7 8 9 10 11 12 13 14 15 16
Copernicus Tycho and ray system Apennines and Imbrium Structure Theophilus Catharina Cyrillus Aristarchus Proclus Gassendi Petavius Posidonius Pico Messier/Messier A Pickering Schiller Taruntius Sabine and Ritter Hipparchus
2,055 – – 148 – 1,597 1,292 – – 196/376 – – – – –
46 47,57 46 18 26 52 59 14 – 48 45 71 37 35 44,45
11.4°S 11.4°S 11.4°S 23.7°N 16.1°N 17.6°S 25.1°S 31.8°N 45.7°N 1.9°S 1.9°S 51.9°S 5.6°N 1.7°N 5.5°S
26.4°E 26.4°E 26.4°E 47.4°W 46.8°E 40.1°W 60.4°E 29.9°E 8.9°W 47.6°E 47.6°E 39.0°W 46.5°W 19.7°E 4.8°E
110 100 98 40 28 10 188 95 25 11 11 180 56 30 150
68 62 61 25 17 6 117 59 16 7 7 112 35 19 93
17 18 19 20
Davy crater chain Snellius Valley Mosting A Flamsteed P
1,597 – 453 –
43 59,69 43 40
11.8°S 31.1°S 3.2°S 3.0°S
8.1°W 56.0°E 5.2°W 44.0°W
34 590 13 112
21 367 8 70
21
Copernicus secondary craters Humboldtianum basin Atlas dark halos Copernicus H Lambert R Rheita Valley Hesiodus A Plato craterlets Nansen Drygalski Procellarum Basin Marth Orientale Basin Lagrenus Vendelinus Cleomedes Atlas Hercules
Crater and impact melt Crater and impact melt Crater and impact melt Crater Oblique impact Floor-fractured crater Floor-fractured crater Floor-fractured crater Basin ring Oblique impact feature Oblique impact feature Oblique impact feature (?) Floor-fractured crater Twin impacts (?) First crater sketched through a telescope Comet impact (?) Basin secondary crater chain Simple crater Elementary crater (surveyor landing site) Rays and Craterlets near Pytheas Multi-ring basin
834
20
18.9°N
21.2°E
4
2
–
59.0°N
82.0°E
650
404
Dark halo crater Dark halo crater Ghost crater Basin secondary crater chain Concentric crater “Changless crater” Polar crater Polar crater Largest lunar basin known Concentric crater Basin ejecta = Inghirami valley Craters Walled plain Crater
– – – – 453 1,445–3,831 – – – – 3,870 1,826 – 3,822 – –
60, III 15 31 20 68 54 3,4 II 72, VI
46.7°N 6.9°N 23.8°N 42.5°S 30.1°S 51.6°N 80.9°N 79.3°S 23.0°N 31.1°S 44.0°S 8.9°S 16.3°S 27.7°N 46.7°N 46.7°N
44.1°E 18.3°W 20.6°W 51.5°E 17.0°W 9.4°W 95.3°E 84.9°W 15.0°W 20.3°W 73.0°W 60.9°E 61.8°E 55.5°E 44.4°E 39.1°E
87 5 54 445 15 109 104 149 3,200 6 140 132 147 126 87 69
54 3 34 277 9 68 65 93 1,990 4 87 82 61 78 54 43
22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
63 49 60 26 15 14
Guide to Observing Impact Features
Table 6.1. A list of 100 significant impact structures to observe (“Ma” under “EST. AGE” signifies that the ages are given in units of millions of years)
(continued)
107
Guide to Observing Impact Features
Table 6.1. (continued) No.
Name
Nature
Est. age (Ma)19
RC #
Lat.
Long.
Size (km)
Size (mi)
39 40 41 42 43 44 45 46 47 48 49
Endymion Macrobius Piccolomini Fracastorius Aristoteles Eudoxus Cassini/Cassini A Albategnius Aristillus Autolycus Maurolycus
Crater Prominent crater Very prominent crater Walled plain Crater Crater Flooded crater Ring mountains
– 3,650 3,825 – 3,693 1,369 –/3,320 – 1,826 834 –
7 26 58 58 5 13 12 44 12 12 66
53.6°N 21.3°N 29.7°S 21.2°S 50.2°N 44.3°N 40.2°N 11.2°S 33.9°N 30.7°N 41.8°S
56.5°E 46.0°E 32.2°E 33.0°E 17.4°E 16.3°E 4.6°E 4.1°E 1.2°E 1.5°E 14.0°E
125 64 88 124 87 67 57 136 55 39 114
78 40 55 77 54 42 35 85 34 24 71
50 51 52
Archimedes Ptolemaeus Alphonsus
– – –
22 44 44
29.7°N 9.2°S 13.4°S
4.0°W 1.8°W 2.8°W
83 153 119
52 95 74
53 54 55 56 57 58 59 60 61
Arzachel Walter Maginus Clavius Eratosthenes Longomontanus Bullialdus Kepler Grimaldi
– – – – 2,971 – 3,230 1,063 –
55 65 73 72 21 72 53 30 39
18.2°S 33.0°S 50.0°S 58.4°S 14.5°N 49.5°S 20.7°S 8.1°N 5.2°S
1.9°W 0.7°E 6.2°W 14.4°W 11.3°W 21.7°W 22.2°W 38.0°W 68.6°W
62 63 64 65 66 67 68 69 70 71 72
Picard Furnerius Proclus Fabricus Plinius Mitchell Manilius Gemma Frisius Pitatus Billy Clavius Craterlets
1,597 – – – 2,895 – 3,693 – – 3,400 –
26 69 26 68 24 5 23 66 54 40 72
14.6°N 36.3°S 16.1°N 42.9°S 15.4°N 49.7°N 14.5°N 34.2°S 29.8°S 13.8°S 58.4°S
54°E 60.4°E 46.8°E 42.0°E 23.7°E 20.2°E 9.1°E 13.3°E 13.5°W 50.1°W 14.4°W
97 132 × 140 163 225 58 145 61 32 222 (inner); 430 (outer) 23 125 28 78 43 30 39 88 97 46 225
60 83 × 88 101 140 36 90 38 20 139 (inner); 269 (outer) 14 78 17 48 27 19 24 55 60 29 140
73 74 75 76 77 78 79 80
Hippalus Herschel Schickard Reiner Gamma Markov Harpalus Fontenelle Anaxagaras
– – – 1,140 – 1,521 – –
52 44 62 29 1 2 3 4
24.8°S 5.7°S 44.4°S 7.0°N 53.4°N 52.6°N 63.4°N 73.4°N
30.2°W 2.1°W 54.6°W 54.9°W 62.7°W 43.4°W 18.9°W 10.1°W
58 41 227 30 40 39 38 51
36 25 141 19 25 24 23 32
Vast walled plains, central peaks Flooded craters Ringed mountain and central peak Walled plain Vast walled plains Walled plains Walled plain Very prominent crater Very prominent crate Basin with flooded crater Prominent crater w/sharp rim Prominent walled plain Crater with bright ray system Prominent crater Crater with sharp rim Crater Very prominent crater Crater Flooded walled plain Flooded crater Walled plain; craterlets D,C,N,J,JA Remains of a crater Prominent crater Vast walled plain Prominent crater Crater with sharp rim Ray crater Crater Ray crater
(continued)
108
No.
Name
81 82 83 84 85
Goldschmidt Walled plain Kane Flooded crater Strabo Prominent flooded crater Mare Humboldtianum Basin Struve Remains of a flooded walled plain Brayley Crater Krieger Flooded crater Krieger B or van Crater Biesbroeck Rocco (Krieger D) Crater Ruth Small crater Louise Small crater Wallace Remains of flooded crater Beer Circular crater with sharp rim Conon Prominent crater with sharp rim amidst the Montes Alpenninus Bonpland Remains of a walled plain Short Crater Boussingault/ Craters Boussingault A Pitiscus Prominent crater Steinheil Steinheil overlaps Watt Watt Steinheil overlaps Watt
86 87 88 89 90 91 92 93 94
95 96 97 98 99 100
Nature
Est. age (Ma)19
RC #
Lat.
Long.
Size (km)
Size (mi)
– – – – –
4 5 6 7 17
73.0°N 63.1°N 61.9°N 57°N 23.0°N
2.9°W 26.1°E 54.3°E 80°E 76.6°W
120 55 55 160 170
75 34 34 99 106
>(3,850) – 2,132
19 19 19
20.9°N 29.0°N 28.7°N
36.9°W 45.6°W 45.6°N
14.5 22.0 10
9.0 13.7 6
– – – – 3,607 2,055
19 19 21 21 22
28.9°N 28.7°N 28.5°N 20.3°N 27.1°N 21.6°N
45.0°W 45.1°W 34.2°W 8.7°W 9.1°W 2.0°E
4.4 3 1.5 26 10.2 22
2.7 2 0.9 16 6.3 14
– – –
42 73 75
8.3°S 74.6°S 70.4°S
17.4°W 7.3°W 54.7°E
60 71 131
37 44 81
– –
75 76
50.4°S 48.6°S; 49.5°S
30.9°E 46.5°E; 48.6°E
82 67; 66
51 42; 41
Guide to Observing Impact Features
Table 6.1. (continued)
A “Top 100” List of significant Impact Structures to Observe The following is a list of impact structures assembled from several sources and is laid out based on Sky & Telescope’s “Exploring the Moon” column (which includes in its Top 100 all kinds of lunar features, not just craters). The first 35 structures are from the Sky & Telescope list. The next 44 features are from the Astronomical League’s lunar club, to give you a head start on making observations to join this club. The last group of features was selected from Rukl’s lunar atlas.18 Observers are encouraged to use their favorite lunar atlas to locate each of these features, and they are also encouraged to apply what they learn about the nature of the feature in terms of size, age, and appearance. For each listed feature, I include a reference number, the name, the nature of the feature, the selenographic latitude and longitude, and the mean diameter. I also include the chart number where the feature can be located in Rukl’s Moon atlas (labeled “RC #”, unfortunately it is out of print, but comparable atlases can be found in bookstores and websites; I have found a site with an online version of Rukl’s atlas as of August 2008). The Astronomical League currently has two lunar clubs, and more information about these clubs is found in the next section. 109
Guide to Observing Impact Features
A Link to Astronomical League Observing Clubs Related to the Moon
110
The Astronomical League maintains many observing clubs from all areas of amateur astronomy which provides valuable experience in observing various celestial objects. The Lunar Club I and Lunar Club II observing lists may be found at http:// www.astroleague.org; these are listed under Observing Clubs along the left side of the web page. The Lunar Club includes a list of features visible to the unaided eye, binoculars, and small telescopes. These images are listed by whether they can be observed via naked eye, binoculars, or a small telescope; they are also given by feature type (not just impact features but also mountain ranges, Maria features, and rilles) and day, in terms of age since New Moon, when they are best seen. The Lunar II Club includes a list of 100 targets or tasks to perform, which includes making drawings of regions, observing features under different illumination angles, observing the full Moon, and more. The Association of Lunar and Planetary Observers has a significant part of its effort devoted to lunar observation and provides many resources for the lunar observer to use in deepening his/her understanding and appreciation of the Moon. In addition to the Lunar Meteoritic Impact Search section mentioned earlier on several occasions, there are sections which are devoted to Transient Phenomena (in general), Topographical Studies, selected areas, domes (volcanic features), and eclipses.
Chapter 7
Impacts Today
Lunar Transient Phenomenon (LTP) is defined as a short-lived phenomenon or change observed on the Moon and can take on a number of forms such as red glows, flashes, obscurations, and abnormal albedo and shadow effects. The study of LTP implies the study of such changes that take place on the Moon’s surface. Some forms of LTP are likely triggered by lunar tides and occur within craters and around the perimeters of basins. Many forms of lunar change seem to be localized, occurring or recurring at specific locations and within specific features on the Moon’s surface. In one localized instance, an LTP took place in the crater Alphonsus, and astronomers were able to obtain a carbon spectrum from this event. The following, adapted with the permission from the online Lunar Transient Phenomena Observing Manual by David O. Darling (one can find the website using a search engine and typing the manual’s title), is a coarse classification of the numerous forms of LTP. 1. Gaseous: The surface takes on a diffuse or nebulous appearance and is likely caused by an obscuration (probably due to outgassing from the sub-surface) of some sort. Many observers have seen this phenomenon over the years in areas such as the Mare Crisium basin, the interior of Plato and in the craters Aristarchus and Tycho. 2. Obscuration: Again, outgassing may be the cause of “disappearing surface features”. An example of this is the reported disappearance of the small craters on the floor of the crater Plato. Another example includes the disappearance of large portions of the Mare Crisium basin due to obscuration, with nearby features clearly visible at the same time. 3. Brightening: This effect is observed when high albedo features become even brighter for no apparent reason. This is also seen with unusual albedo variances of the rims of impact craters. Features that have historically exhibited this phenomenon include the rims of the craters Proclus, Censorinus, and Aristarchus. 4. Darkening: Observers have reported this effect on features such as Reiner, Picard, and inside the crater Proclus. This effect can take on the appearance of black ink, which sometimes has the appearance of flowing across the lunar surface.
Impacts Today
Lunar Transient Phenomena and Lunar Meteoroid Impacts
111
Impacts Today
5. Bluish: Bluish colorations have been mostly seen within the darkened or earthlit part of the crescent Moon, and sometimes during lunar eclipses, with individual features such as Aristarchus glowing with a bluish tint or electric blue color. Aristarchus seems to be the feature that exhibits this phenomenon most frequently: it has been reported many times to be glowing bright in the earthshine region of the Moon while the rest of the unlit disk appears dusky and obscured with few other features being visible. 6. Reddish: This effect has been reported on the rim of the crater Aristarchus and inside the crater Gassendi. LTP observers the use a technique called Moon blinking, where the Moon is observed while repeatedly switching between the Wratten #25 (Red) and #38A (Blue) filters. 7. Shadow Effect: Observers report this effect when they notice that the blackness of the lunar shadow has lightened in tone, taking on the appearance of dawn or dusk, becoming more gray than black in appearance. This phenomenon can also be associated with shadows apparently being cast upon the surface when no shadows are expected. For example, a crater’s interior appears to be filled with shadow—during a time when the sun angle is high. 8. Contrast Effects: The effect appears as a graying effect along the zone separating two contrasting regions such as along the terminator or the edge of a shadow. This effect, closely associated with shadow effect and obscuration, can give an unusual appearance to a crater floor in the bright portion of the Moon. 9. Star-like Flashes: The observer is surprised by a brilliant flash that abruptly appears and disappears within a fraction of a second. These are most likely flashes produced by meteoroids impacting the lunar surface. 10. Star-like Lights: This occurs when a stellar point of light is observed on the Moon and lasts from several minutes up to an hour. These events are mostly observed on the dark side of the 3–4 days old Moon, but they have also been reported during total lunar eclipse. 11. Others: All phenomena observed that do not reflect any of the above descriptions.
Some Possible Causes of LTP As can be seen, it is difficult to scientifically determine the true cause of LTP. Some possible explanations, mostly based on remarks from J. Hedley Robinson in the December 1986 issue of the British Astronomical Association, are provided as follows. 1. Tidal: The Moon experiences 32.5 times the magnitude of tidal stress from the Earth as the Earth gets from the Moon. This stress reaches its maximum near perigee, where potential energy can build up and be suddenly released. This release may result in the escape of trapped gasses just under the surface. These trapped gases are subsequently seen as LTP. 2. Albedo Changes: Dust movement, possibly by electrostatic levitation (an assertion which seems to be supported by findings of the Clementine and Lunar Prospector spacecraft) may lead to changes in reflectivity or albedo. Apart from electrostatics, it is unclear how enough dust would be moved, without wind, to account for observed changes in albedo.
112
Impacts Today
3. Thermal Shock: The Moon orbits once in every 29 days, and as it does so, different parts of the surface experience alternating 2-week periods of day and night. The transition between these two leads to a significant temperature change of 200°C. Over a period of two hours centered on lunar sunrise and sunset, surface temperatures vary from 125 to –80°C. This extreme change in temperature over a short time span produces a thermal shock of surface material which could result in LTP. Most LTPs have been observed to occur within three days of local sunrise, and could arise from the different surface materials expanding and contracting at different rates, but these conditions are likely incidental to LTPs rather than a primary cause. 4. Magnetic: Since this effect is much weaker than thermal shock, it is not possibly a cause of observable LTP. It seems that electromagnetic effects are even less likely than magnetic effects to produce LTP. 5. Ultraviolet: With no atmosphere of significance to shield the surface of the Moon, ultraviolet radiation illuminates the surface without being inhibited. This may result in fluorescence as seen at visible wavelengths, but the fluorescence is not strong enough to be observed as LTP and is easily overwhelmed by the visible reflected sunlight. 6. Solar Wind: The solar wind can strike the lunar surface directly during its orbit around the Earth, which can result in an electric discharge from the lunar surface. However, the energy involved here is so low that any resultant effect is unobservable from the Earth. 7. Spectral Diffraction: When the geometry is right, lighting effects may occur, which can take the form of spectral diffraction. This phenomenon, which arises from surface grains or irregularities too small to be seen with ground-based telescopes, may lead to a change in color which is seen all over the lunar disk, but in reality, the color changes are only localized. Lighting effects may arise, sometimes taking the form of spectral diffraction, which may cause change in color. 8. Meteor Strikes: These have been reported frequently, with one of the best observations prior to 1999 being an event near the Apollo 14 landing site on 13 May 1972. The event was a meteor impact that released an energy equivalent of 1,000 tons of TNT. Although meteor strikes cover a tiny area of the lunar surface when compared to the size of most LTPs, they are a likely cause of at least some forms of LTP observed on the lunar surface. 9. Moonquakes: Moonquakes occur deep within the Moon and are often very weak and occur more often near times of perigee or apogee. No obvious connection appears between moonquakes and LTP events. 10. False Color: False color occurs regularly in larger telescopes due to atmospheric conditions. This can be mitigated by comparing a feature or candidate LTP event to the appearance of features elsewhere on the lunar disk. If all of these show the same effect, then the effect is instrumental or atmospheric in origin. 11. Piezoelectric Effects: On the Earth, this effect is well known. Two examples include rock strain generating a strong electric field that ionizes the air above the object, resulting in a glow, which moves with the strain source; and Soviet reports of sudden drops of magnetic field strength at the release of underground tension. This effect may be responsible for at least a small fraction of the reported LTP events, possibly including the 1953 Stuart “lunar flare” event.
113
Impacts Today
Narrowing Down the Causes of LTP to Three: Electrostatic Levitation, Tidal/Thermal Stresses, and Meteoroid Impacts The above discussion indicates that a variety of ideas have been proposed as possible mechanisms for LTP. The author J. Hedley Robinson could not give any one cause any more weight than the other. This is a great dilemma and why many scientists do not consider LTP to be real phenomena, since no satisfactory mechanism has been found that could account for the large energy output of many LTP events. It seems that the documented observations of LTP over time have indicated either the presence of a gaseous medium or the levitation of dust particles being the cause. This assertion appears to be supported by results from the Clementine and Lunar Prospector missions along with in-depth examination of the Apollo and Surveyor data. An electrostatic charging mechanism may be responsible for levitating lunar soil particles above the surface along the solar terminator around the Moon. It has been calculated to reach as high as 10 m above the surface, but observations of the Apollo astronauts seem to indicate elevations extending several kilometers above the lunar surface. They noted streamers and bands of coronal and zodiacal light just before orbital sunrise, a horizon glow that may be explained by a mechanism in addition to the background solar corona and interplanetary dust. Observations of horizon glow phenomena were recorded in images (Fig. 7.1 shows an example) from the lunar Surveyor 7 in which “patchy glows were discovered on the Moon's horizon at dusk that is believed to be caused by individual dust clouds being continually formed above the hills, crater rims, and rocks which remain sunlit while the shadow of night advances over the lower surroundings.”20 Even the more recent lunar orbiting Clementine spacecraft has detected glows just before lunar sunrise and just after lunar sunset.
Fig. 7.1. Glow observed by Surveyor 7 spacecraft 15 minutes after local sunset. Refer to the text for more details. Image courtesy of David O. Darling and NASA.
114
Impacts Today
There may be pockets of gas generated by radioactive decay of non-gaseous elements beneath the surface; this gas eventually works its way to the surface, where it is subsequently released by openings in the crust. These openings may be caused by thermal stress (i.e., the transition from day to night, and vice versa) or moonquakes, or crustal fracturing associated with impact craters. With the release of this gas may come disturbances of local dust or the piezoelectric effect when the gas comes in contact with sunlight. The latter may have been the cause of the famous 1953 Stuart Moon flare. Meteoroid impacts may stir up the dust, release previously trapped gaseous elements, or both, which could lead to LTP events. Daytime impacts, if large enough, may be revealed by one or more of these secondary effects. This was demonstrated to an extent by the Hiten impact, a man-made impact event whose effects were visible in ground-based images, such as the one shown in Fig. 7.2 (A-1). In the case of the satellite, the impact occurred on the night side, but was close enough to the terminator that the rising dust cloud became illuminated by sunlight shortly after the event. Dr. David Allen, at the Anglo–Australian Observatory, Epping, Australia, made the image with an infrared telescope. This image is one of a series showing the evolution of an approximately 5 km (3 mi) diameter cloud caused by Hiten’s impact. The impact of Hiten, and the cloud observed afterward (Fig. 7.2), could serve as a demonstration of what to look for with regard to naturally occurring impact events. More often, however, the actual event observed with the latter is that of a very quick, star like flash of light, such as those observed during the 1999 and 2001 Leonid meteor storms. These are the primary events to look for (although if these other manifestations occur, it is also useful to be at least a bit familiar with them, recording them as they happen), and it is best to look for them on the dark
Fig. 7.2. AATB’s Figure A-1 showing HITEN’s Impact plume.
115
Impacts Today
part of the Moon, since the sunlit part would overwhelm all but the brightest, most rare events. The LTP record contains many such events, and more recently, the Lunar Meteoritic Impact Search (LMIS) section of the Association of Lunar and Planetary Observers (A.L.P.O.) have, over the first eight years of its existence, received many such reports. A catalog of these events, including those documented by NASA, was provided in Chapter 4. Observable manifestations of LTP come in a variety of forms, with perhaps a number of causes, but this observer's guide, and the LMIS ALPO observing program, is concerned about a single cause of LTP: meteor impacts on the Moon. One of the goals of the program is to determine which form of LTP is derived from meteoric impacts. We know the point-flash phenomenon is one such form attributed to LTP, but other forms, visible on the sunlit side of the Moon, may also be due to impacts of meteors. Before any conclusions can be made about the LTP-Lunar Meteor relationship, a number of scientifically confirmed observations of impact phenomena and LTP are needed. The next section expands on the LTP lunar impact phenomena and looks more into how impacts take place and the physics (in layman’s terms) behind what we actually see from afar during an impact event. The section starts with a brief overview of how craters are formed, and is written by Mr. Eric Douglass of the American Lunar Society. One can estimate the kinetic energy of an impactor if one can measure the maximum brightness (which leads to information about the energy output of the collision) and knows about how much of that kinetic energy is changed to visible light as output. If the impact happened during an annual shower, it is likely (thought not 100% certain) that the source of the impact is from the meteor shower, which reveals the likely velocity of the impactor. The mass (and possibly the size) of the impactor can then be found. I then provide information on some ways to indirectly observe impact events through seismology (Apollo 1970s) and changes in the appearance of the Moon’s sodium cloud (1998). Finally, I discuss the recent results by the NASA MSFC Meteoroid environment group and the Mars Global Surveyor, the latter of which, in pairs of images of the same regions of the Martian Surface, saw new impact craters on the second image where none was seen on the first image.
Impact Mechanics: The Physics of Crater-Making By Eric Douglass Meteoroid impacts are described in three phases: compression, excavation, and modification. Since modification happens over an extended period of time and is not associated with the visible impact as seen from the Earth (in the case of a large enough event occurring on the Moon), the discussion that follows is limited to the first two phases, which are most likely to produce a visible signature as observed from the Earth (or in lower lunar orbit). Figure 7.3 summarizes the discussion that follows. Compression Phase. At the point of contact, the kinetic energy of the bolide (impacting object) is transferred to a shock wave. The kinetic energy is proportional to the square of the bolide’s velocity. The velocity of impacts on the Moon is the sum of the escape velocity (2.4 km/s or 1.5 mi/s) and the approach velocity (while these vary considerably, the following are approximations: 10–20 km/s
116
Meteor Impact
Impact!
Incoming
After the Dust settles
(6–12 mi/s) for asteroidal meteorites; 20–25 km/s (12–15 mi/s) for short period comets, and 50–60 km/s (31–38 mi/s) for long period comets). As the velocity of planetary impacts is generally quite high, the energy transferred into the shock wave is also quite high. Indeed, this shock wave produces pressures and temperatures of sufficient energy to vaporize most of the bolide, and a mass of the Moon several times the mass of the bolide. This occurs at pressures in the range of several hundred GPa (1 GPa = 10,000 bars), with temperatures in the range of 5,000–8,000 Kelvin (9,600–15,000°F). After leaving the point of contact, the shock wave expands in a hemispheric geometry (spherical shape) through the lunar surface and rearward through the bolide itself. As the shock wave dissipates, the target materials undergo less energetic phases, such as wholesale melting of the rocks and diaplectic glass formation, as these are still well above the Hugoniot elastic limit (where rocks deform plastically, producing permanent shock changes). What is actually observed from the Earth is the thermal phase of the event, blackbody (or thermal) decay (cooling) from a blackbody temperature of ~3,500 K (~6,800°F). At this point, much of the material is incandescent and is the source of the visible light that we see. The material fades as it cools, much as a large incandescent light bulb fades just after it is turned off. The plasma flash is so short (~10 ms) and mostly hidden from view because it occurs at the bottom of a tunnel drilled by the impactor in the normally loose lunar regolith. Excavation Phase. The shock wave is followed by a rarefaction wave, which begins at the free surface and sets material in an outward motion. This occurs during the 'excavation' stage of crater formation, which means that these vaporized, melted, and fractured rocks are being removed from the impact site, so that a crater (at that stage, called the 'transient crater') is formed. As the shock wave velocity is approximately 10 km/s (6 mi/s) on the Moon, the compression phase for craters of the size discussed in this section, occurs in under 1/100 of a second, while the excavation phase probably lasts for another second or so, depending on the size of the crater.
Impacts Today
Fig. 7.3. The impact processes described in the text: (a) just prior to impact, (b) impact event / compression phase, and (c) after the dust has settled
117
The instrumentation left on the Moon from the Apollo missions (next section) provided data that helped scientists determine the size and frequency of impacts on the Moon. During the years of its operation, it revealed that once a year, bolides with masses of greater than or equal to one metric ton (1,000 kg) strike the Moon. As the 'flashes' observed from the Earth are greater than this number, we can safely assume that most 'flashes' are from bolides that are much smaller. However, making certain approximations (or “educated guesses”) for density and velocity, we can suggest that chondrites of around one metric ton (the largest ones we will likely see) will produce craters in the range of 15–40 m (50–130 ft) in diameter. All such craters would be well below the resolving power of earth-based telescopes. Consequently, the 'flash' will be a point source for these impacts, and is only visualized because of its extreme contrast (much like stars, which are also below the resolution limit of our eyes!).
Impacts Today
Indirect Evidence for Lunar Meteoritic Impact Events: The Apollo Lunar Seismic Program and the Moon’s Sodium Cloud
118
The Apollo Seismometer Experiment was in operation from 1969 to 1977, for a total of 7.9 years. During this time of monitoring, 1,743 seismic events attributed to natural meteoritic impacts were recorded, 95 of which were major events (potentially observable from the Earth). There were nine artificial impacts recorded as calibration events, five of which were classified as major.21 In order to better understand the impact frequency on the Moon and to more adequately address the following problems, more data of high quality are needed. These problems include the risk of meteoroid impacts on astronauts and structures placed on the lunar surface. While it is true that the danger of a 1-g meteoroid hitting an astronaut on the lunar surface is actually insignificant (the probability of such an event is between 1 in 106 and 1 in 108 for each year spent on the lunar surface), the concern is that of the safety of large and critical structures placed on the Moon. These structures include habitats, base support facilities, processing plants, and research instruments. It would be useful for the probabilities of an impact by meteoroids of particular sizes, especially the larger sizes, are fairly well known, so the risks can be properly assessed and dealt with. Prior to the 1999 and 2001 optical detection of the Leonid meteor hitting on the Moon, the Apollo lunar seismic network, detected impacts from the same shower in 1974. The Leonid strikes manifested themselves as a cluster of seismic events at the time of Leonid maximum. Impact clusters were associated with several other annual meteor showers as well. It was not possible to directly measure the masses of the impactors with the Apollo network, but estimates of the masses of the meteoroids range from 0.1 to 1.0 kg (0.21–2.12 lb). Due to lunarLeonid encounter geometry, the Leonids were only detected seismically in 1974 and not in the other years of the survey. The ground-based observed rates of the meteor shower, however, were not enhanced in any of these years. Studies of the impact rate at the Apollo sites (Fig. 7.4) using a limiting mass of 100 g (or 0.2 lb, the lower mass limit for the seismic detectors) reveal a non-storm annual rate of 5 × 10–8 meteor hits per square meter, and it depends on a parameter called the mass index. The mass index is a measure of the mass spectrum (a representation of the mass distribution of particles or how many particles of a given
mass range are there in the stream) of a meteoroid stream, and it seems to vary for the Leonids from about 1.7 to 2.0. This means that the Leonid meteor ribbon has mostly small particles with a few larger ones mixed in (1.7) or that it has almost all small particles (2.0). It has been found that about 10 meteoroids of 100 g and larger should impact the lunar surface during a typical Leonid year,22 assuming a mass index of 1.7. This is consistent with the actual results from the Seismic Network in 1974. If the mass index is 2.0, then the number of expected impacts falls to near zero, which was possibly confirmed with the lack of Leonid impacts recorded by the Seismic Network in 1975 (and conditions were favorable to record impacts). It is possible that this part of the stream sampled by the Earth had a high mass index. This is an illustration of some of the variables involved in determining the visibility of impact flashes from the Earth. The number of impacts observed for a given shower adds evidence for a particular mass index of that shower, and changes in the index may be detected from year to year with further observations. Further evidence for impacts lies with the Moon’s very tenuous atmosphere, composed mainly of sodium. During the time of numerous fireballs during the 1998 Leonid display, this atmosphere showed an enhancement. Although the geometry for observing impact flashes was unfavorable for Earth-based observers
Impacts Today
Fig. 7.4. Location of the Apollo Passive Seismometers. Image Source: http://www.geotimes.org/july06/feature_MoonQuakes.html#bottom and courtesy of NASA
119
Impacts Today 120
at this time, the elevated sodium content did occur at the same time as the Leonid storm, indicating a connection between the meteors and the atmosphere. Thus, the observations of these changes in the lunar sodium atmosphere provide another method to indirectly detect meteor impacts. The undisturbed lunar atmosphere has a density of 2 × 105 molecules/cm3 by night and 104 by day, or about 1014 times less than that of the Earth. In addition to sodium, the primary constituents are neon, hydrogen, helium, and oxygen. Due to the low escape velocity of the Moon, most of these molecules are expected to escape in a relatively short amount of time. However, if they are being replenished by some mechanism, it is possible to maintain the current tenuous lunar atmosphere. The most reliable mechanism to replenish the lunar atmosphere is by meteor impacts, so it is safe to conclude that the Moon's tenuous atmosphere is at least partially maintained by impact-driven vaporization.23 Therefore, close monitoring of the sodium D2 line near the Moon's limb might yield useful information on the overall flux of meteoroids during outbursts. Recently, a sustained, ongoing campaign by the Meteoroid / Space Environment group at the NASA Marshall Space Flight Center has turned up 129 good impact candidates (49 derived from showers, 79 sporadic, between 1 November 2005 and 5 September 2008). The ALPO Lunar Meteoritic Impact Search section receives roughly 1–5 reports of observed flashes per year (refer to the catalog in Chapter 4 for more details), many of which could be genuine impact events. Finally, a press release by NASA on 6 December 2006 revealed the evidence for flowing water on Mars by the Mars Global Surveyor (MGS) satellite. In addition to this evidence, the camera team looked for new impact craters on Mars. The camera aboard MGS photographed some 98% of Mars in 1999 and about 30% of the planet in 2006. The newer images show 20 fresh impact craters, ranging in size from 2 m (7 ft, comparable to craters produced on the Moon by the meteoroids representing the fainter end of confirmed impact flashes) to 125 m (406 ft) in diameter, which were not present seven years earlier. These results, which approximately match the predictions of models, are important for the determination of the impact flux at Mars as well as that of the ages of Martian surface features.16 Based on these numbers, it seems that 9–10 sizable impacts occur on an average at Mars during each Earth year. Back at the Earth, a report published in late 2006 on the spaceweather.com website revealed that an estimated 100 terrestrial fireballs per day occur. If the objects producing the terrestrial fireballs impacted the Moon, many would likely be easily seen through Earth-based telescopes.
Chapter 8
Lunar Impact Observation Programs
As discussed in the Introduction, the Leonid meteor shower of 18 November 1999 produced the first independently confirmed observations and recordings of meteor impacts on the Moon. A number of attempts have been made previously to observe and document lunar meteor impacts, but none have produced scientifically confirmed observations from two or more widely separate (more than a few tens of kilometers or miles) locations. Some of these efforts have met with some success, with a number of probable events recorded. The uncertain and unpredictable nature of this phenomenon, along with insufficient camera sensitivity to record all but the largest events and the difficulty of visually locating very short flashes in the midst of hours of data, has led to its very limited success. With new technologies and techniques, the question has shifted from “Does it happen?” to “How Many?” and “How Big?” The program that follows observations of lunar meteor impacts is characterized by long periods of little activity, punctuated by very brief flashes or clusters of activity. Prior to the twentieth century, there had been little interest in meteors striking other bodies besides the Earth, since the prevailing thought at the beginning of the last century that the craters on the Moon were volcanic in origin rather than meteoritic. However, there has already been several centuries of observations of lunar transient phenomenon, with meteoric impacts being absent from the list of explanations of these events. Walter Haas, founder of the Association of Lunar & Planetary Observers, conducted his own search for lunar meteoritic phenomena in the late 1930s and early 1940s, and again from 1945 to 1965. He recorded at least one good candidate in 1941, but nothing conclusive or confirmed. One of the first large-scale, organized campaigns to perform such observations specifically to study lunar meteor impacts was the Association of Lunar and Planetary Observers (ALPO) programs of the 1950s and 1960s. Other programs have been established to observe LTPs in general, but no specifically organized effort (at least to my knowledge), apart from those noted above, has been made in finding the lunar meteor impacts until November 1999. John Westfall of ALPO summarized the earlier ALPO program in the form of a paper in June 1997 at the 48th ALPO convention in Los Cruces, New Mexico. The program was coordinated by Robert M. Adams from 1955 to 1962, and Kenneth
Lunar Impact Observation Programs
Past Campaigns
121
Chalk, from 1962 to 1965. The project published six “Progress Reports” in ALPO’s journal “The Strolling Astronomer” but it never achieved its goal of simultaneous, independent observations. While it was known that many impacts occurred each year from the Apollo lunar seismic network that operated during the 1970s (Chaps. 3 and 7), it was still unclear how the impacts would appear visually and whether they would even be visible from the Earth (nearly a quarter million miles distant). Some possibilities included bright streaks in the lunar atmosphere, clouds of ejecta being thrown out, newly formed craters, or bright point flashes. In conclusion, Westfall suggested that the Lunar Meteor Search program should be resurrected using the currently available, relatively inexpensive video equipment.24
Current Programs
Lunar Impact Observation Programs
The ALPO Lunar Meteoritic Impact Search program was established in January 2000, in response to the success of the November 1999 Leonid effort by members and observers of the International Occultation Timing Association. This program has since been in operation and has focused mainly on the more prominent annual meteor showers whose maxima occurred when the Moon was favorably placed to enable observation impacts from an Earth-based observatory. In the first eight years of its existence, two confirmed (during the 2001 Leonids), several probable, and many candidate impact events (a total of some 80 events) were recorded. A catalog of these events appeared in Chapter 4. I have served as the coordinator of the program since its formation. Other programs around the world (Table 8.1) are also pursuing lunar impact observations. Three such groups include two from Italy, Unione Astrofili Italiani (UAI) and the Geological Lunar Research (GLR) Group, and one from Brazil, Rea de Astronomia Observacional (RAO). The UAI includes a Lunar Section, directed by Rafaello Braga (directors mentioned in this section were active as of August 2008), which is involved in many aspects of lunar observation, and includes Transient Lunar Phenomenon in general and meteoritic impacts specifically (these comprise two projects within the section). The Lunar Impact program monitors the Moon during annual meteor showers when the Moon is favorably placed. The GLR group, along with the American Lunar Society (ALS) maintains an active and involved Lunar Impact monitoring program, with Rafaello Lena being one of the primary coordinators along with Piergiovanni Salimbeni and Eric
122
Table 8.1. A summary of groups actively monitoring the Moon for meteor impacts • Association of Lunar and Planetary Observers; Lunar Meteoritic Impact Search United States of America coordinated by Brian Cudnik. URL: http://www.lpl.arizona.edu/~rhill/alpo/lunarstuff/lunimpacts.html • Geologic Lunar Research Group/American Lunar Society; Italy/United States of America coordinated by Rafaello Lena and Eric Douglass. URL: http://www.glrgroup.org/impact.htm • Unione Astrofili Italiani Lunar Section coordinated by Guiseppe Sorrentino. URL: http://luna.uai.it/luna.htm • Rea de Astronomia Observacional (REA) coordinated by R. Gregio URL: http://www.reabrasil.org/lunar/impacto_meteoritos.htm • British Astronomical Association, the Lunar Meteor/LTP observations coordinated by Dr. Anthony Cook. URL: http://www.baalunarsection.org.uk/ • NASA/Marshall Space Flight Center Lunar Impact Monitoring Program coordinated by Rob Suggs and Bill Cooke. URL: http://www.nasa.gov/centers/marshall/news/lunar/index.html
Space Missions that Impacted/Will Impact the Moon Since the dawn of the Space Age over 50 years ago, many artificial satellites were sent past, put in orbit around, and landed on the surface of the Moon. But a few over the years actually impacted the Moon (both intentionally and unintentionally), resulting in an artificial lunar meteor. One such event has already been described in the previous chapter: the Japanese Hiten satellite crash-landed into the Moon, producing a plume that became visible when it rose high enough from the shadowed surface into the sunlight (Fig. 7.2). Hiten was one of the 14 spacecraft that have crashed into the Moon; one of the most recent was the SMART-1 satellite
Lunar Impact Observation Programs
Douglass. These groups feature an observation manual, which has been published on the World Wide Web for lunar impact observations, as well as coordinate observations on an international scale. The objectives of these groups include the following: to determine a classification of flashes using standard criteria, to determine whether flashes are cosmic ray strikes on the CCD surface or true impact flashes (Chapter 11, written by Mr. Lena of the GLR, goes into detail concerning this work), to find other ways to confirm that a recorded flash is of impact origin, to coordinate observing activities during times of particularly promising meteor storms, and to coordinate and share information with other groups doing the same work. The ALPO and the GLR/ALS have had much observational collaboration over the years of existence for the ALPO/LMIS. The RAO group in Brazil coordinates lunar impact observations in that part of the world. Their website contains a tutorial (written in Portuguese) as well as information on the activities of the group. Like the Italian groups and the ALPO/ LMIS group, the RAO coordinates observe the Moon during times when annual showers are active and the Moon is favorably placed. The British Astronomical Association has a Transient Lunar Phenomena group that also searches for impact phenomena amidst their efforts to characterize change on the lunar surface. This effort is led by Dr. Anthony Cook, and he coordinates observers in the UK. A team led by Robert Suggs, working at the Marshall Space Flight Center, began a systematic ground-based watch for lunar meteor impacts in November 2005 which continues to this day. The purpose of this program is to establish the impact rates and the sizes of meteoroids impacting the lunar surface. The observations are carried out between new and first quarter phase, and between last quarter phase and new Moon, when the Moon is between 10% and 55% illuminated. Under these conditions, the team is able to observe up to 12 nights per month. The observations are conducted at the NASA Marshall Space Flight Center, Huntsville, Alabama, at the Automated Lunar and Meteor Observatory (ALaMO) facility. The mission statement of the MSFC team: “Use Earth-based observations of the dark portion of the Moon to establish the rates and sizes of large meteoroids (greater than 500 g or 1 pound in mass) striking the lunar surface.” With NASA planning to send astronauts back to the Moon for extended stays on the lunar surface, it is becoming increasingly important to become knowledgeable about the magnitude and frequency of lunar impacts. For the extended stay, spacecraft, vehicles, habitats, and EVA suits must all be designed to withstand the stresses posed by the harsh lunar environment over time.
123
Lunar Impact Observation Programs
Table 8.2. Spacecraft that have impacted the Moon, 1993–200625
124
Name
Launch date
Booster
Weight (lb/kg)
Mission
Hiten (Japan)
24 Jan 1990
Mu-3SII-5
315/143
Lunar Prospector (USA)
6 Jan 1998
Athena II
653/296
SMART-1 (ESA)
27 Sep 2003
Ariane 5
808/366
Earth orbiter that preformed 10 lunar flybys and lunar impact on 10 Apr 1993; sub-satellite Hagoromo entered lunar orbit. Originally named MUSES-A Lunar orbiter. Mapped minerals, magnetic and gravitational fields; discovered evidence of possible water ice at lunar poles. Crashed near Moon’s south pole 31 Jul 1999 First European mission to the Moon; entered lunar orbit 15 Nov 04. Two phase missions to test new technologies and study the Moon. Crashed into Moon 3 Sep 2006
that impacted the Moon in September 2006. The three most recent spacecraft impacts are recorded in Table 8.2. The ALPO-LMIS section coordinated amateur astronomers to make observations of the site of the European Space Agency’s SMART-1 impact. The impact was to have occurred during a time when the major telescopes of Hawaii, Chile, and the American Southwest had access to the Moon above the horizon. The spacecraft impacted the Moon at a speed of 2 km/s (1.2 mi/s) into a volcanic plain called the Lake of Excellence. Amateur astronomers in these regions also participated. Of the roughly two-dozen observers who participated in this event, about half were clouded out and almost the entire other half reported seeing nothing – even with apertures as large as 14-in. (35.6 cm). One team observed the event from the George Observatory (located south of Houston, Texas, USA) 36-in. telescope and video but did not observe anything significant. This was likely due to a combination of low elevation (less than 12°) and haze. The ESA team observed an impact flash and a dust plume from the 3.6-m (140-in.) Canada-France-Hawaii Telescope (CFHT) in Hawaii. The impact flash was quite bright in the infrared region. Of the amateur astronomers, only one, Mr. Peter Lipscomb from Santa Fe, New Mexico, USA, succeeded in observing the impact flash by capturing its image on video. Mr. Lipscomb used a ToUcam Pro on an 8-in. (20.3 cm) LX-90 at f/10; the webcam was set at 5 frames per second with high gain and saturation. The next spacecraft that has been planned to make an impact on the Moon is the Lunar Crater Observation and Sensing Satellite or LCROSS, which comes in three parts: a Lunar Reconnaissance Orbiter (LRO), an upper stage Centaur rocket, and a shepherding spacecraft. What follows about the LCROSS mission is courtesy of Dr. John Westfall. The spacecraft, as of August 2008, is scheduled for launch no earlier than 27 February 2009, which results in an impact that will happen between mid-May and September 2009 in either the North or South Pole. The 2,200 kg Centaur will impact first, producing an estimated 20-m diameter, 3-m deep crater; this is followed four minutes later by the less massive (700 kg) Shepherding Spacecraft (SSC), which will hit 10 km (6 mi) from the first impact site. In fact, as the SSC follows the Centaur in, observing its impact flash then passing through its impact
plume before impacting on the Moon itself. The LRO will observe these impacts from a safe distance in orbit. What can we expect to see and who will be observing this event? The impact flash from the first object will be the brighter of the two, lasting for 0.1–0.2 s. About 10 s later, the rising plume reaches sunlight as it emerges from permanent shadow. And some 20–100 s after the impact event, the plume, some 3″ in apparent diameter, reaches its peak brightness. It may last 10–20 min or longer, by which time it becomes 50 km (30 mi) high and 100 km (60 mi) wide. This process will be monitored not only by LRO from lunar orbit, but the Hubble Space Telescope (in the ultraviolet, looking for water vapor) and ground-based observatories. Professional observatories in Hawaii, Chile, Arizona, California, and New Mexico will be making observations of this event with imaging, photometry, spectroscopy, visible, near infrared, far infrared, and possibly the radio part of the spectrum. Amateur astronomers will be able to help with this event. They have the advantage of numbers, geographic dispersion, and flexibility. However, the disadvantages include small scopes, lack of mid-IR cameras and spectroscopy. Nonetheless, ALPO-LMIS will coordinate the efforts of amateurs in the region favored by the impact time (that is, the time when the Moon is above the horizon when the two spacecrafts impact the Moon) as we provide valuable information (and public outreach) regarding the events.
Watching for lunar meteors involves looking for the visual signatures (i.e. flashes) of things hitting the Moon, a phenomenon that may be more common than once thought. All that is needed is a small telescope (or even binoculars) to reveal evidence of copious amounts of impact events that have happened in the past. However, it is not until this present age that such an event has been witnessed and confirmed as such. Several instances have occurred in the past where single observers have witnessed what were likely meteoritic impacts on the Moon. These events, cataloged in Chapter 4, include reports of lunar transient phenomenon in the form of flashes, lights, and other events. The purpose of ALPO's Lunar Meteoritic Impact Search is to coordinate the observation of a specific family of LTPs: those that occur as a result of meteoroid impact events. The objective is to develop the program and resources to assist with serious professional research of these “lunar flashes.” For the time being, the program utilizes the resources given and coordinates the observations of willing participants to develop a catalog of lunar impact events. Several questions to answer in the immediate future include: Just how common are these “impact flashes” that have been observed on the Moon? (This is really becoming apparent with increasing time.) Is it possible to observe these impacts during the occurrence of every major meteor shower when the Moon is favorably placed? (It is becoming clear that the answer to this question is “yes.”) What can these flashes tell us about the impactors themselves? What sorts of models of size, composition,
Lunar Impact Observation Programs
Mission Statement of the ALPO Lunar Meteoritic Impact Search Program: A Vision of Lunar Impact Research
125
Lunar Impact Observation Programs
kinetic energy to light efficiency, etc. best match the observed flashes? How can systematic observations of lunar meteors best complement ground-based observations of meteors burning up in the Earth's atmosphere? What can both sets of observations (previous questions) tell us about the meteor stream itself? This is more of a long-term goal: is it feasible to position a lunar-orbiting satellite to constantly monitor and observe lunar flashes (during meteor showers and at other times) in various bandpasses (perhaps taking spectra) on the dark side of the Moon? These questions are just the beginning and one would appreciate the possibilities this opens up. We stand on the shoulders of giants, one of whom is the late Gene Shoemaker, who I worked with in my undergraduate college days in Flagstaff, Arizona. Dr. Shoemaker helped to reveal that the vast majority of craters we see on the Moon are actually from impacts rather than volcanoes. We are picking up one of the torches he dropped in his untimely passing by venturing into this field of lunar flash studies. In summary, as of 1 October 2008, the LMIS of ALPO has coordinated many lunar campaign events (cataloged in Chapter 4) and has collected some 80 impact candidate observations. The primary focus has been on meteor impacts during annual meteor showers. Although observations outside of shower times have not been discouraged (but actually encouraged), they had been sporadic at best. One notable exception to this is the “Earthshine Watch” project that Dr. Anthony Cook tirelessly carried out for a three year period. The conclusion of this study was that the lunar meteor phenomena are irregular and rare at best – at least outside of normally occurring showers. However, further evidence from different sources, primarily the NASA group, would later reveal otherwise. Since the mission statement presented above had been originally published on the World Wide Web in 2001, the organization of the ALPO-LMIS has been geared primarily toward amateur observations of lunar meteoritic phenomena with a secondary emphasis on professional-amateur collaboration. This is done primarily through work with the NASA Marshall Space Flight Center Meteoroid Environment Group, whose work has been described and cataloged in this book. Although I had professional research in mind while establishing the ALPO-LMIS and have published a few refereed journal articles and several conference posters on the subject, the primary goal of the ALPO-LMIS is now to provide opportunities for amateurs at all levels to make astronomical contributions of scientific importance, along the lines of ALPO’s general mission.
126
Observing Impacts as They Happen
with Contributions by Many Members of the International Occultation Timing Association
Observing Impacts
Chapter 9
This chapter provides techniques and tips to maximize the chances of obtaining successful observations of lunar meteoritic phenomena. The chapter is divided into four broad sections: general guidelines for observations, techniques for visual observations, techniques for video observations, and highlights of useful equipment and observer setups to guide the novice (and even enlighten the advanced) in the observing process. The two latter sections include a number of elements that have been brought together from dialogues and discussions among members of the IOTA group over several years’ time as well as my own experiences along the way. The members have, over the years, discussed in detail a number of pieces of equipment and their uses in the field of stellar occultation observation, most of which can also be used in the recording of lunar meteoritic impact flashes. However, due to the stringent time requirements inherent to occultation timing observations, a number of items related to equipment and techniques are not applicable in the area of lunar meteoritic observations and have been omitted from this work. Nonetheless, the same equipment used in occultation observation can also be used in the observation of lunar meteoritic impact events. Steve Preston of IOTA provided a tutorial on extracting event times of stellar and lunar occultations (and lunar meteor impacts) from videotape; this tutorial is presented in Appendix C. For the beginner, a very useful aid is the description of setups of experienced observers who have made successful astronomical observations. This chapter outlines a number of such setups and uses these to describe an ideal family of setups that ensures successful observing sessions. These descriptions are not allinclusive, since variations to the theme abound, but are meant to provide ideas for the aspiring lunar meteor observer. New equipment and techniques are constantly coming online. Interested readers are encouraged to join list servers such as
[email protected] and
[email protected] to be a 127
Observing Impacts
part of the discussion on the latest in techniques and equipment. In addition to the family of “standard setups” discussed in this chapter, additional innovative equipment configurations, such as a tri-splitter camera, are described briefly in Chap. 12. The idea behind the tri-splitter setup is to enable up to three different paths or different wavelengths of light to be recorded simultaneously at a single location. In doing so, it becomes possible to rule out most cosmic ray hits as impact candidates, increasing the probability that a given candidate is a true impact event. Please note that websites given herein were active as of August 2008. Inevitably over time, some of the sites may become defunct or inactive; if this becomes the case, the reader is advised to use a reliable search engine and type the name of the topic of choice and use that to get a list of websites that may have more information (i.e. “Google ‘Lunar Meteors’”).
Geometric Considerations and Preparations Needed for Making Observations Over the years, it has been realized that the simpler we make a process or project; the more likely people will participate and succeed, let alone persevere with the project. A number of us who are involved in lunar meteor observations have developed a rather simple approach in the making of lunar meteor observations that begin with determining when conditions are right for the observations to take place. We start with a brief discussion of the lunar geometry needed to make observations of meteoritic impact phenomena possible. Meteors burning up in Earth’s atmosphere usually happen during the predawn hours, since this hemisphere, the leading hemisphere in terms of the orbital motion of the Earth, is also the hemisphere facing oncoming interplanetary debris that appears as meteors or “shooting stars” when they collide with Earth’s atmosphere. When the Moon is visible in the evening sky to earth-based observers (especially around the time of first quarter), the hemisphere facing into the oncoming debris from meteoroid streams and the general interplanetary population is also visible from Earth, and so we are best placed to witness impacts on the Moon’s surface. However, if the Moon is near the last quarter phase, the side of the Moon receiving the hits is facing away from ground-based observers, thus fewer impacts are seen by ground based observers. However, as has been demonstrated in recent years of lunar meteor observations, impacts on the Moon can occur during a waning lunar phase (in the predawn hours) as well as during a waxing (evening) phase. The probability of an impact flash being observed at a given time from a given shower depends on a number of factors. First, the size and density of the impacting objects have to be large enough to release enough energy in the form of visible light that is bright enough to be seen through ground-based telescopes a quarter of a million miles distant. Second, the number of meteors in a given volume of space also plays an important role in the success of observing an event. The higher the number of meteoroids, the greater the probability that larger objects are included in the stream, objects large enough to produce a visible flash 128
Observing Impacts
as observed through an Earth-bound telescope. Third, the velocity of a meteoroid has to be high enough to add enough energy to the object to produce a visible flash on impact. The energy of a moving body increases as the square of the velocity – in other words, each time the velocity is doubled, the energy carried by the object increases fourfold. For example, a Leonid meteoroid, traveling at 72 km/s (44 mi/s) carries four times more energy than a sporadic meteoroid moving at 36 km/s (22 mi/s). Sporadic meteoroids cover the entire range of velocities, including the rare extra-solar meteoroids with velocities possibly exceeding that of the Leonids. To successfully document these events, some degree of preparation is necessary. There are a number of approaches that one can take to observe lunar meteor impacts, from visual to video. If done properly, visual has the advantage of seeing the event as it happens, having an almost immediate record of the flash, whereas video observations involve reviewing hours of videotape in search of a transient flash. This used to be a limiting factor but with the public release of Lunarscan in 2007 (more about this software appears in Chap. 10), the review factor has become much less of an obstacle. Video also has the great advantage of having a permanent record of an event, one that can be analyzed in detail and compared to other observations for definitive confirmation. Video does not have the problems with fatigue, blinking, being able to only focus in a small area at high resolution, etc. In all cases, it is vital that observations be made in pairs, with each observer separated by at least 30 km (18 mi) such as to maximize the probability that an event documented by both observers is actually an impact event as opposed to a point meteor, a satellite glint, or a similar, near-Earth phenomenon. The following sections describe both methods of making lunar impact observations and provide guidelines to maximize the chance of success.
Techniques for Visual Observations The visual observer needs only a few pieces of equipment to get started. A telescope of just about any aperture with clean optics, a hand-held tape recorder (a digital voice recorder is preferred), a radio equipped with shortwave receiver (again, digital is preferred to enable quick signal acquisition), and lots of patience is all that is needed for the visual observer. It is also helpful for the observer to have a simple but accurate Moon map on which the observer can make note of the locations of impact candidates for later verification. One such type of map can be found on the ALPO LMIS website. It is suggested that a visual observer be paired with a video observer, so that any event observed by the visual observer can be unambiguously confirmed by video (such as my very own Leonid impact event was confirmed by video in November 1999). The setup for the visual observer is simple: a telescope with a wide field of view such as to encompass as much of the unlit portion of the Moon as possible. It is recommended, though not absolutely necessary, that the scope be clock driven at lunar rate. It is also recommended that the observer be familiar with the appearance of stars at various magnitudes through the same telescope at the same magnification as would be used for lunar meteor watch. By doing so, it becomes possible that a rough estimate of the magnitude of a candidate impact event be made. Just before an observing session is to begin, the observer should draw in the 129
Observing Impacts
location of the terminator and a few major craters along the terminator on the Moon map and note the time this drawing was made. The observer centers the telescope’s field of view on the part of the Moon centered on the core of the projected meteoroid stream and begins to watch, being careful to keep the bright part of the Moon out of the field of view. The observer should have the tape recorder in hand, with WWV playing in the background, so as to be recorded by the tape recorder. The observer will speak into the tape when the observing session starts and stops, as well as any breaks in the session, interruptions by clouds, the observing conditions (and changes of these), properties of the telescope system, and any other pertinent details. Finally, if an impact event is observed, the person immediately calls out the event, along with the estimated magnitude and color of the event, and the location on the Moon, if known (at this point, the observer notes the location on the Moon map; this can later be compared to a more detailed map to get latitude, longitude, and the name of the nearest feature) and any other descriptions (such as duration of flash, rate of brightness change, afterglow, etc) that the observer may have of the event. The time, to the nearest second, can be obtained from the audiotape, and the position can be inferred by comparing the sketched location with a detailed lunar atlas. With these details, other observers can subsequently check their data to confirm the impact candidate. Depending on stamina or fatigue, the visual observer should wisely choose the time interval used to stare through the eyepiece. Ten to fifteen minute sessions, with a three-to five-minute break between sessions, is recommended. Too much time at the eyepiece will cause fatigue, increasing the likelihood of a missed event. A degree of alertness is required to maximize the chance of seeing the very brief flash that occurs without warning. To date, all the confirmed meteor impacts on the Moon have durations of 1/30 s, easily missed by a blink of an eye. Certain websites, such as http://topendsports.com/testing/reactiontest.htm in which the background color of the user’s monitor changes color at a random time, allow the user to measure his or her reaction time. As soon as the screen color changes the person clicks his/her mouse, and the reaction time is displayed. To find one’s true reaction time, the experiment can be repeated between 20 and 40 times and the average taken from the measurements. This information is especially useful to have in order to correct event times for reaction time.
Techniques for Video Observations Video observations have the advantage over visual observations in the fact that the camera never blinks or becomes tired, the observer has a permanent record of the event (assuming one hits the “record” button on one’s device!), the frames containing the event can be analyzed for event brightness/magnitude and selenographic position, and the event can be replayed repeatedly to study the light curve and other details. Before one starts videotaping for the first time, one must select the components of the setup carefully (a number of specific examples are provided later in this chapter as a guide) then practice with the setup to become familiar with it before beginning an observation program. Important considerations to take into account when assembling a system are described as well. One of the challenges first encountered by new CCD video users is focusing 130
Observing Impacts
the video camera. Useful recommendations for that are provided in the next section. As better equipment and software come online for lunar meteor detection and analysis, the list of equipment for a viable setup grows; a checklist of such equipment (including both software and hardware components) is provided near the end of Appendix D. This section also contains information on locations to purchase some of these components. The first step toward securing scientifically useful videos of lunar meteor impacts is to assemble and test such a system (again, several examples are given toward the end of this chapter). Once the system is assembled and the user becomes familiar with the setup, then the user may proceed with the steps outlined from this point forward, starting with focusing the video camera.
Making the Observations: Focusing the CCD Video Camera One of the challenges of getting started in CCD astronomy is focusing the camera. This is especially true with a Schmidt–Cassegraine Telescope (SCT) system. Since the field of view (FOV) of an SCT system is rather narrow for most lunar meteor work, a focal reducer becomes necessary to get adequate coverage. This added element to the system also adds to the challenge of focusing the camera. The functionality is that the spacing from the focal reducer to the focal plane of the CCD chip is essentially determined by the focal length of the focal reducer lens. To “focus” the SCT system with a particular spacer behind the focal reducer, results in altering the effective focal length (and effective f/ number) of the resulting SCT + focal reducer lens system (because the primary mirror has been moved to bring the focal point of the SCT into coincidence with the focal plane of the focal reducer lens). For the SCT, the primary mirror moves to focus the telescope. The secondary amplifies the focal length by several times, so the motion of the primary is multiplied by the same factor. This results in the primary mirror moving a little while the focus moves a lot. Moving the primary mirror in an SCT alters the effective focal length and the effective f/ number. Also, the focal reducer is designed to be screwed onto the back of the SCT. If a spacer is interposed (for example, the JMI NGF electric focuser) between the SCT back and the focal reducer – one may or may not get the system to focus, especially with a video camera, due to the nonstandard “flange focal depth” of the typical video camera (compared to the flange depth value for a 35-mm film camera, which is evidently the design standard for the commercial focal reducers.). A brand of focal reducer commonly used is the Meade f/3.3 focal reducer. Focal reducers are positive power lenses designed to go inside the focal plane of the telescope. If the focus adjustment is set where the focal plane of the instrument is inside of the reducer position, the reducer will project the image a considerable distance behind the normal focus position (200 mm is roughly the distance). An effective way to approach the process of focusing is as follows: First, refocus the scope visually. Put a diagonal and an eyepiece back onto the scope without the reducer and focus on the Moon (or a star if the Moon is not available). Remove the diagonal and eyepiece and install the reducer with the camera, resetting the focus with this system in place. By focusing with a diagonal and eyepiece in place, one should get
131
Observing Impacts
closer to the proper focal position to enable the reducer to work properly. In turn, this will help ensure that the focal reducer is ahead of the SCT prime focus plane. The f3.3 focal reducing lens works fine with an 8-in. Celestron. The horizontal width of the field of view is almost 25¢, enough to take in most of the Moon’s dark side, and a little more of the area recorded with a PC-23C on a C-5 with f6.3 focal reducer (this setup was used by David Dunham, President of IOTA, for the 1999 lunar Leonids). The focal reducer also helps for asteroidal occultations (larger field of view) and lunar occultation reappearances. The f6.3 focal reducer is available in the range of $99–$140 or so (as of mid-2008) from many telescope dealers. Probably even better, for a larger field of view to image virtually all of the Moon’s dark side is the Meade Series 4000 CCD f3.3 focal reducer available for around $150 from Meade dealers and from Focus Camera, Inc. in Brooklyn, NY (phone orders 888221-0828). With the Moon, things are greatly simplified because the Moon is such a large object (as opposed to point-like stars). One simply needs to point the scope to the Moon and focus on the lunar image. We suggest that the number and direction of focus turns be recorded. To do this, focus the video system to the Moon. Then remove the video camera, focal reducer, and any other component and replace this with the visual back/diagonal/eyepiece holder and low power eyepiece. Count the exact number of focuser turns as well as direction of turns and record these. Having this information will speed up the setup by enabling the user to make the exact number of turns to focus the system, leaving out time-consuming guesswork.
Keeping Accurate Time, Time Insertion, and Recovery Having successfully focused the system to be used in the observations, the next step is to find a reliable time source. The radio station WWV is commonly used in the United States to obtain an accurate time signal. In some cases, problems arise in the reception of WWV. If this happens, the Canadian time signal from CHU in Ottawa is also available. From Melbourne, Australia, there are times that CHU on its “non-standard” frequencies of 3.300, 7.335, and 14.670 MHz can be received with a usable signal for a few hours. Observers in Canada and the USA should have a much better signal. Propagation for 3.300 and 7.335 MHz could be different (better or worse) than that for 5 or 10 MHz, depending on observer location and the ionosphere. Shortwave radios to receive these time signals are widely available. Appendix E provides a listing of time sources for various parts of the world. For comparative purposes, the time of a lunar meteor impact candidate should be determined within one-half of one second to greatly increase the probability of confirmation of the nature of impact of the event. While the time requirement is not quite as rigorous as it is for occultation work, nonetheless without a good time record, an observation of a potential lunar meteoritic impact event is of little use. For visual observers, the setup is simply a tape recorder, shortwave radio, and telescope, and the impact event candidate is called out into the recorder, from where a time can be extracted by listening to where the call was made against the background WWV signal, then correcting for human reaction time. Done carefully, visual observations can be key in confirming the sighting of an event. For video observations, many cameras are equipped with a microphone that enables an audio signal to be saved along with the video signal. The minimum 132
Observing Impacts
amount of equipment needed to recover the times of lunar meteoritic impacts and other time-critical phenomena from videotapes may include simply the audio–video components, and a stopwatch to use during playback to pin down the exact time of an event. For easier and more accurate determination of event occurrence time, a time insertion setup is required. Video time inserters (VTIs) are commonly used in astronomical and nonastronomical work. Such devices imprint the time in one corner or edge (chosen by the user) of the image, time that is synchronized to that given by the nearest shortwave radio station. VTIs are widely available in various forms and names, and cost as little as $69.95 (US dollars as of August 2008) from companies such as Supercircuits (the product is referred to as a “Time and Date Generator”). Increasingly, users are synchronizing their VTIs with GPS. GPS-based time inserters are not so widely available, but two individuals, whose websites are listed below (and are active as of August 2008), produce GPS-based VTIs for interested entities. The cost may be higher than the simpler commercially available products, but the added convenience and accuracy of GPS synchronization is there. http://www.geocities.com/kiwi_36_nz/kiwi_osd/kiwi_osd.htm http://www.mcafeeastrometrics.com/ These websites contain more information about GPS VTIs including technical specifications and wiring diagrams. A few years ago,we purchased a KIWI OSD Video Time Inserter with GPS for a total of $247 from PFD Systems in Bethesda, Maryland, USA (their website is http://www.pfdsystems.com/kiwiosd.html). The equipment came with a printed user manual, and one can access the manual online as well. The OSD VTI is integrated with the system (in series) and stamps information such as UT, latitude, longitude, elevation on each frame. The GPS unit (a Garmin 18 LVC) received signals from GPS satellites (including UT) and uses this information to make the time stamp on each video.
Twelve Examples of Observer Equipment Setup and Use Introduction Vital to the success of any lunar meteor impact-monitoring program is an effective system setup. The setup may range from a simple visual setup (scope + tape recorder + WWV signal and SW (short-wave) radio + blank Moon map to record impact candidate locations) to a sophisticated video setup with video time insertion and automated impact-detection software. In any case, it is important to be thoroughly familiar with your setup to maximize your ability to make useful observations. This can be done by practicing with your system on a nonshower Moon prior to making actual observations. In so doing, you can discover problems before they have a chance to interfere with your ability to take observations. Some considerations to take into account when starting an observation program and setting up a system includes your experience at astronomical observations, the time you have to devote to lunar meteor observations, your budget, and how you will look for candidates, among others. If you are an absolute beginner, we suggest starting with a visual setup; or if you have the experience and the budget, assemble 133
Observing Impacts
a setup similar to those described below. The examples that follow serve to provide a starting point and ideas for a setup but in all cases, what is needed is a telescope (4″ or larger recommended), a detector/recorder, and a timing source, which can give accurate time to the nearest second at least. The examples build upon this basic framework. We begin the examples by posing the question below. Several examples of setups and systems (such as the portable example shown in Fig. 9.1) used in the observations of stellar occultations and lunar meteors are provided, and in several cases, illustrative images are provided. These setups, by extension and with slight modifications in a few instances, can also be used in the observations of lunar meteor phenomenon. In almost all cases, the type of impact manifestation is the point like impact flash, such as those confirmed in the 1999 and 2001 Leonid events. Further information about many of these products can be found by going online to the website(s) of the product vendor(s) (Fig. 9.1). “Can you give a detailed description of the equipment you are using, from the telescope to the recording device?” 1. One setup consists of a C11 telescope on a G11 mount, an f/3.3 focal reducer, a modified Supercircuits PC164C lowlight CCD video camera, an STVASTRO and Garmin GPS35 (for time insertion into the video stream). These feed data into a JVC SVHS recorder. The tape is analyzed on a Panasonic PV-VS4821 VCR because it has field-by-field advance, which is very useful for determining precise times and for looking for multiframe appearances of an event. If the event appears in three frames or more, this increases the likelihood that the event is a true impact event as opposed to a cosmic ray.
Fig. 9.1. Richard Nugent’s setup including all the elements needed for a complete system, courtesy Richard Nugent.
134
Observing Impacts
2. Another observer uses an LX200GPS 12-in. telescope with an f3.3 focal reducer mounted directly behind a microfocuser. The detector used is a Supercircuits PC164C video camera, which feeds the video directly to a Canon ZR50 Digital Video Camcorder, which has analog-to-digital circuitry, so the video is recorded digitally. Also in use is a Radio Shack DX-396 AM/SW/FM Receiver. The WWV radio signal is fed directly to the DVC via the same cable that provides the video feed (three part cable feeding one jack into the camera, two audio channels, and one video). 3. One uses a Sony DTR-510 camcorder as a recorder for a monochrome video camera. This camcorder records digitally on Hi8 tape with a run time on a 2 h Hi8 tape of 1 h with this system. The camcorder also has the low-light feature, which slows the integration time down to 1/15 and 1/4 s. (LL1 and LL2 respectively). The IR rejection filter is also pulled out of the way with these exposures to provide more sensitivity. Those shutter speeds do not work while on “VCR” setting, which is where it has to be to recorded from the B&W camera. The camera has a 4-in. LCD screen, which is useful. A battery can also be used to power the B&W camera (which runs on 9-12V anyway), so the system is compact and portable. 4. A Sylvania A/C-D/C 9″ TV/VHS Combo unit makes up another system, which features simplicity: all one piece which simply is plugged into the cigarette lighter socket of a vehicle, or run from a 12v deep cycle marine battery. The G-11 mount for the system is run from a separate source, keeping cabling to a minimum (no converter; no wiring from monitor to recorder). This setup was used successfully with 3 different telescopes: 6″ f/5 Newtonian (advantage of portability); 10″ f/5 Newtonian, and an 80 mm f/5 refractor. The 6-in. Newtonian’s short focal ratio is optimized for the PC 23C camera and it can easily reach 10th magnitude stars. The short wave radio providing time input to this setup is a small digitally tuned Radio Shack type, with a handy wrist strap. The strap gets looped over the finder scope or focuser, which sets the radio close to the camera’s built-in microphone. If necessary, a 25-ft wire is clipped to the radio’s antenna and the other end is tossed over a tree, fence, or other handy object. The cable requirements for this system are as follows: (a) BNC-to-RCA co-axial cable for camera video-to-TV/VCR. (b) RCA-to-RCA for camera audio-to-TV/VCR. (c) Auto-socket-to-co-axial plug for camera power (supplied with camera). (d) Auto-socket-to-co-axial plug for TV/VCR power (supplied with TV/VCR). (e) If possible, we suggest that white cables are used; they are visible in the darkness, whereas black ones are not. Doing so could prevent an accident that can ruin an observation session.
5. Another setup uses a Supercircuits PC23C camera whose output is fed into the video input on a Sony digital camcorder. Viewing recorded video on the digital camcorder allows excellent signal to noise ratio, and the ability to stop on individual frames without the annoying frame jitter present on many analog video recorders. Time code is automatically written onto the recording by the Sony digital camcorder. A battery holder for the PC23C 6xAA completes the system, making everything portable. The PC23C fits snuggly into the eyepiece drawtube on an 8″ f/5 Dobsonian and reaches Newtonian focus. This system was tried under nonideal conditions through an insect screen (2004 Aug 25 09:50-10:16UTC) in an apartment. It was able to pick up the lunar limb in Earthshine and a faint star (of undetermined magnitude) off the south cusp of the Moon. 135
Observing Impacts
6. A SHARP camcorder with a built in color LCD screen is used in another setup. Most camcorders have a built in screen for direct viewing of what is being recorded. With no significant disadvantages, the advantages of having the built in screen include:
(a) Convenience of the screen being portable and present (b) See exactly what you record (c) Portable – battery powered, small, lightweight, easy for travel (d) 8-mm and Hi 8-mm tapes (forget VHS – its on the way out) (e) Built in screen allows positioning at any angle, no need to “crunch” your neck to look thru a viewfinder (f) Cost: $300+, but you may find a used one. (g) Easy to copy to VHS for time insertion (h) Easy to show to groups, the camcorders are complete portable VCRs 7. The video gear of this setup consists of a Magnavox TV/VCR combo unit. A red filter for the front of the screen helps especially if the observer is with other people or making other observations and needs his or her night vision. Otherwise, the red filter makes seeing faint stars on the monitor more difficult. Also plugging an earphone into the jack to disable the speaker can also be helpful. A camcorder with a video input jack is also an option, but would be more costly. Another recommendation is a three-outlet car outlet adapter for power, one for the camera, and one for the TV/VCR and the third for the scope, dew heater or whatever. With the Supercircuits camera, the car outlet to coax plug cable is extra, but highly recommended, as is a 12V extension cord. If using a battery, one may want to get a car outlet that clamps on and also has reverse voltage protection to protect one’s equipment. Scopetronix sells such an item, or it can be made on one’s own. 8. The NASA group, observing from the NASA Marshall Space Flight Center in Huntsville, Alabama, uses a Stellacam EX with a 14″ RCX400 operating at f/2.6 with a Meade f3.3 focal reducer and a 10″ f/4.7 Newtonian with a focal length of 1,200 mm. They make their video recordings with a SONY Digicam or use its firewire output to capture the data directly to the PC. The Stellacam has the advantage in that the exposure can be set at 1/120 s with low enough gain to image the terminator for focusing. The MSFC site includes an automated setup, the Automated Lunar and Meteor Observatory (ALaMO). More details about their setup and the program itself can be found at this internet address (valid as of August 2008): http://www.nasa.gov/centers/marshall/news/lunar/program_ overview.html#link2. 9. My own setup (Fig. 9.2a, b) is similar to all of the systems profiled above. These images add the “diagrams” needed to make sense of the description. The setup depicted in Fig. 9.1 at the beginning of the present section is Richard Nugent’s portable setup, with each component identified. In both pictorial cases, a Cassegraine-type telescope is used as a collector of light, with a low-light CCD camera (in my setup’s case, a PC 23C) at the prime focus. The camera is attached directly to the visual back, which in turn is attached to an f/3.3 focal reducer, attached to the back of the telescope. The PC 23C includes a microphone, which records WWV 5.0 MHz (easily obtained in the United States of America) along with any voice comments. Since my work is done in conjunction with ongoing research, the telescope was provided by Prairie View A&M University, Department of Physics. A KIWI-OSD with GPS receiver was later added to the system
136
Observing Impacts Fig. 9.2. a & b Mr Cudnik’s setup. A KIWI-OSD with GPS receiver (not shown here) completes the setup; the video feeds through the KIWI, which adds a time stamp that shows up on screen. The blue and white 10-inch (25.4-cm) Dobsonian in the background is sometimes used for backup visual observations.
and it provides a time stamp to the video. I use this along with the audio signal, so if one signal fails, the other provides a backup. Both audio and video are fed from this scope to a TV/VCR combination appliance, with the tape play/recording speed set on SP (LP and EP introduce more noise to the videotape). 10. The sophisticated components of Derek Breit’s “Occultation Overkill Setup” are 137
Observing Impacts
highlighted in pictures below (Figs. 9.3a–f). Mr. Breit uses a Supercircuits PC 180 with a Meade 12 in. scope, a TV/VCR combo, and a time inserter equipped with GPS. Again, the basic components are displayed in this setup, with the time inserter to take the place of an audio signal. As indicated by the location of the setup, a certain amount of mobility is implied.Fig. 9.3 11. Robert Spellman uses a system that consists of two telescopes, a 200 mm Newtonian (with a focal length of 1,000 or f/5) a 113 mm Newtonian (focal length 450 mm or f/4.5) both on a single mount. The 200 mm system is equipped with a focal reducer, which cuts in half the focal length and f-ratio of the instrument. The instruments are managed by a control center that includes a DVD recorder for the Newtonian video stream, a VHS recorder for the refractor video stream, 2 monitors, telescope tracking and focusing controls, and a short wave radio receiving a time signal from station WWV. More information about his setup, including images, can be found at his website: http://www.angelfire.com/space2/robertspellman/ 12. George’s (Varros) Lunar Impact Detection System consists of an 8″ Celestron f/5.0 NGT Newtonian reflector with CG-5 mount and NexStar controller; and a Watec 902H2 Ultimate, which replaced the StellaCam II, is much more sensitive, and is connected to a Kiwi OSD GPS video time stamp generator, recorded in Digital format. A focal reducer is used at times to help include as much of the dark portion of the Moon in the field of view. Mr. Varros uses Canopus ADVC-55 to convert the video signal from NTSC to 1394 firewire DV-AVI, for input into a laptop; and uses Blue Star Bluetooth adaptor for guiding the telescope from the laptop.
Making the Observations: One Observer’s Example George Varros describes a typical observing run on his web site, and the description is adapted with his permission, for this book, to serve as an example. The run is executed seamlessly over an entire evening by capturing the AVI files (a video format, pronounced “Ay-vee-eye”) live to a Western Digital MyBook Tera-byte drive (with USB2, 1394-a or 1394-b), which eliminates the need to change tapes. This prevents time, and possible impact events, from being lost during the tape change. The AVI files are processed as follows: use Virtual Dub to break the files into segments then process the segments with LunarScan (written by Pete Gural) to detect and extract lunar impacts images. More information about his equipment and observation runs, as well as images of the setup, a link to download LunarScan, and images of his impacts can all be found on his website http://www. lunarimpacts.com/lunarimpacts.htm. Here is how the observing run proceeds: 1. Setup the mount. Level and polar align mount. Add counterweights; realign mount. Mount telescope, connect focus controller cable, video cable, and power for camera. Secure any loose cables. Realign mount to Polaris. Set telescope to home position. Connect power supply to mount and turn on the controller. After initial setup, manually acquire Moon, set controller to lunar tracking. 2. Focus and align image on CCD. Initially focus the camera and align it so that the illuminated portion of the Moon is just at the edge of the top of the FOV and is aligned with the width of the screen so as to maximize the Moon’s darkened portion on the CCD. Focus camera accurately and slew from Moon to a star to refine focus as precisely as possible. Reacquire the Moon. 3. Recording the run. Turn on KIWI OSD GPS video time stamp unit; turn on video deck, plug in fire wire cable to laptop. Launch video capture software and once 138
Observing Impacts Fig. 9.3. Mr. Breit’s setup. The images in the top row show the overall setup, to include the Meade 12-inch telescope. The middle row of images depicts the detector (the Supercircuits PC 180 camera) in two views: the first view shows the assembled product and the second shows the disassembled product. The final row of images shows the TV/VCR setup and the time insertion device.
139
Observing Impacts
Fig. 9.3. (continued)
GPS time stamp is running, start recording as 720 × 480 DV AVI without compression. The video is recorded to an external 1394 Tera-byte MyBook drive. The drive must be capable of recording without dropping frames. It is also very important to have as accurate a time reference as possible. Audio WWV is an 140
Observing Impacts Fig. 9.3. (continued)
option or possibly, a very well timed recorder time stamp may suffice. 4. Alternately, you can record to a digital video recorder and at a later time, record it to a computer. Always run the KIWI OSD time inserter. However, one can turn off the time stamp after starting to record. The software used to analyze the AVI files will keep accurate time and does not need to have a time stamp. 5. Chop up the huge AVI files into manageable segments with Virtual Dub. After the run is over, use Virtual Dub to divide the AVI files into manageable sizes. Instructions on how to save these files are in the LunarScan software manual (to download, go to the website given above, and where you can find a link to the LunarScan software for download). These files are saved to a laptop computer. It is recommended that the computer has at least 80 GB of hard drive space for these files. 141
Observing Impacts
6. Run LunarScan 1.3. After chopping the AVIs into a series of small files, run LunarScan to analyze the files. After LunarScan is done, review the log file for impact candidates. It will take a few runs to get comfortable with all of these steps working with these monster AVI files. 7. Review and report!! Review the log file and extract candidate events from AVI files and use other softwares to analyze the resulting frames. Cosmic Rays cause impact-like events on the CCD and need to be eliminated. It may take a confirmed impact or two, to be comfortable with being able to identify an event worth reporting. Typically, the light curve for an impact goes to maximum brightness on the first video field (odd or even field extracted from an interlaced video frame). Subsequent video fields should then be dimmer and drop off in brightness. However, there is no guarantee this will be the case with short duration events. Due to atmospheric turbulence, any one star observed using video rapidly fluctuates in brightness. Dim stars sometimes disappear for an individual odd or even video field. 8. Report impact candidates to Danielle Moser,
[email protected] at the Meteoroid Environment Office, NASA Marshall Space Flight Center. These examples serve as general guidelines to help the novice (even the intermediate observer) assemble a functional setup and use that setup to collect actual data. Several points to remember when assembling your system are as follows: It is essential that the scope be motor-driven, preferably at lunar rate, but equatorial should suffice. The narrow field of view makes it impossible (or at least very difficult) to videotape with a nonmotor driven mount. Two systems make use of an 8″ F10 SCT and F6.3 and F3.3 focal reducers. For impact recording, the tracking should be at “lunar” rate, but it may still be necessary to make adjustments periodically. If a setup is to be used in a permanent observatory with AC power, 4 head VCR and quality monitor would probably be the best bet, but for field use, it is recommended that everything be 12V-powered gear. Always record in SP mode and use a premium quality tape. Finally, when assembling the setup for the first time, it is very important to practice with the setup, becoming intimately familiar with it before using it in the field. This familiarization saves time and frustration and increases the probability of obtaining good data from the first run of the setup.
Some Examples of Products and Resources Introduction and Frame Stacking This section investigates some of the technical aspects of lunar meteor video observations and presents information about some products available to enhance the collection of lunar meteor data. We start with a brief discussion about a practice commonly used in planetary imaging, frame stacking (that is, taking several frames of images and stacking them together to make a better, final image). While this may prove useful in making still images even better and for enhancing the visibility of targets in a video stream, it may not be such a good idea for lunar meteor work. After discussing why this is so, we consider some questions of CCD 142
Observing Impacts
video cameras and their use in video astronomy. Next, infrared filters and transmission gratings as well as their potential in gathering real and useful scientific data on lunar meteor impacts are considered. Finally, frame grabbers and automated detection software are considered. Someone may wonder whether stacking images would improve the detectability of lunar impact events. Occasionally observers may come across techniques to improve the sensitivity of their systems. If integration is done on the CCD chip and involves only one readout (at the end of the integration period) and the readout noise dominates, as will be the case if the integration is not too long, then the noise is constant (the noise comes from the output amplifier on the CCD chip). The noise corresponds to one readout operation, independent of the time of integration (within limits). There is no actual stacking operation involved. The star signal increase by t, and the S/N is proportional to t, where t is the integration time in units of the time for one frame. If the integration is done by stacking (i.e., adding) video data in n sequential frames, each obtained by doing a readout of the CCD video data, then the star signal increases by a factor of n, while the noise from n readouts increases only by a factor of the square root of n. Thus, in this case the signal to noise ratio increases n . At the high sensitivity end, where we see individual electronic events, the as n noise is purely random. In this case 4× stacking gives 2× gain in signal/noise ratio. On the other hand, if the electronic amplification is not sufficient to see the individual electronic events on the chip but there is still noise it may be from the electronic circuitry. In this case, the noise is not proportional to the square root of the signal but constant in the background. In this case 4× stacking gives 4× gain. For lunar meteor events, which typically last 2–3 frames, stacking the frames increases the noise but not necessarily the signal. Therefore, with a typical lunar meteor event, the S/N ratio is worsened, thus making the practice of image stacking not very useful to lunar meteor work.
CCD Video Camera Testimonials and Comparisons The following section is a collection of reports from individuals who have used various brands of CCD video cameras in the field for lunar meteor impact searching and asteroid occultation observations. It is meant to assist in making a decision about the selection and purchase of a camera, not necessarily to endorse one product over another. The individuals providing the reports are mostly members of IOTA, and their experiences with CCD video cameras show that the items is of great benefit in the work of lunar meteor observations. As of mid 2008, the PC 164C, sold by Supercircuits, retailed at $115 for the 420 line resolution CCD and $165 for the EX model, which has 600 lines of resolution. This product is an excellent compromise between quality and price. The only inconvenience is that the camera does not come with a microphone, but that can easily be purchased at a nominal price (or, as a number of users suggest, use a PC-23C camera as a microphone). One observer used the PC164C during his lunar earthshine patrols. He used a 4-in. f/6.4 refractor with the camera; this setup was able to include about 80% of the lunar disk. Compared with the PC-23C camera, the PC 164C camera showed the earthlit portion very clearly against the black sky. It had a bit of noise but much 143
Observing Impacts 144
less than the PC23C. The major albedo features were clearly visible on the earth-lit portion. The observer also recorded an occultation of a triangle of stars clearly seen during the observing session, with the faintest of the three (AB Aquarii) easily visible at magnitude 9.15 (average). Based on the results of this trial, the PC 164C camera is very suitable for lunar meteor strike patrols. A modification to make the gain manually controllable would be helpful, as when the bright portion of the crescent intruded in the field about 20%, the gain was cut and the earthshine disappeared. Comparisons have been made with a Watec 902H camera, which has similar specifications. The PC164C is half the price, and nearly as sensitive and has a 1/3in. pixel array – the same pixel area as the PC23C. The 902H Watec has a 1/2 in. pixel array and 570 line resolution, which will give a sharper image than the 420 line resolution of the PC164C (the PC23C also has 420 line resolution). If one is on a budget, the PC164C is highly recommended, but if one wants better resolution along with the low-light sensitivity, the Watec 902H is superior. One observer used the Watec 902H camera to image the 10% sunlit Moon’s dark side, recording four reappearances of stars. The output showed incredible detail on the Earthlit dark side, and the 9th-mag. stars showing brilliantly, using an 8-in. Schmidt–Cassegraine with an f6.3 focal reducing lens. The Watec camera will clearly be able to record lunar impacts about 2 magnitudes fainter than the PC-23C, and similar gains are obvious for lunar and asteroidal occultations as well. Its performance is comparable to some image-intensified systems, but at a fraction of the cost. The Watec 902H camera is made in Japan and widely used by observers there. The Watec 902H is very small (32 mm on a side) and operates essentially the same way as the common Supercircuits PC-23C (but it has just an RCA output rather than BNC video output, a BNC-to-RCA adaptor is not needed) and is powered the same way, with 12V DC, with the same power cord available from Radio Shack – and like the PC23C, the Watec does not have a microphone. As for prices (as of mid-2008, contact the dealer of choice for current pricing information, including any deals), the 902HS sells for approximately $320 and the 902H was under $300. (The 902H is recommended; tests show that the 902HS is noisier, but not really more sensitive for detecting stars, than the 902H). One caution concerning the WAT-902H camera is this: the standard high sensitivity setting is essentially useless, as the noise background is very high, greatly reducing the dynamic range, and the sensitivity is no better than with the low setting. The cover on the back of the camera needs to be removed (micro-size Phillips screws, use a number 0 or 00 Phillips screwdriver) to get to the tiny white ultra-micro switch to change the sensitivity to low (the camera is extremely small and light weight). This microswitch (not labeled) is mounted on the right side of the right side card (looking down into the camera with the RCA video output jack up), about one-fourth the size of the shutter on/off switch on the left side that is accessible through a hole in the back plate. The sensitivity microswitch comes in the up (high sensitivity) position; just use a small object, or your fingernail, to push it down and then reattach the back plate. When observing lunar occultations, it behaves in the same way as does the PC23C, so that the gain is reduced automatically if too much of the bright side of the Moon is visible (or also if there is a lot of glare from the invisible bright limb). Like the PC23C, it has a switch to control the exposure time (either 1/60 s or automatic). For dark images, these two settings give the same results.
The Near-IR Filter To improve the ability to detect lunar meteor impact flashes, especially during twilight, Dr. Anthony Cook of the University of Nottingham, UK, suggested a W 87C near-IR filter. (Dr. Cook also provided the instructions on how to fit the filter to one’s system and gave suggestions for its use, as written below.) Since most of the light output of a typical lunar meteor impact flash is in the infrared part of the spectrum, and the CCD chip of a typical low-light video camera is very sensitive to light beyond 700 nm, it seemed logical to take toward improving the detectability of meteoritic flashes on the Moon. One can (very carefully) mount a Wratten 87C filter in a spare C-ring – the gelatin filter can be cut into shape with a sharp knife, but is easier to trim with sharp scissors. Gelatin is slightly brittle and cutting with a knife can make it tear or fracture. The following is a suggestion based upon experience. Cut the filter to shape and apply a very small amount of “Glue Stick” (solid glue) to the back end of the C-ring and stick this centrally onto the 25 × 25 mm of W 87C filter. After a few minutes, place the filter/C-ring filter-side-down on a clean sheet of paper, and then cut “small” sections off with a knife following the edges of the C-ring. Press down on the C ring while doing this to avoid fractures going inside the diameter of the C-ring. Actually you might want to practice cutting blank 35 mm negatives – to get a feel for things. Once cut to size – remove the filter (use a knife to lift it off from the base of the C-ring) and gradually trim it with sharp scissors to where the residual ring of glue is on the filter. You can hold the filter with cotton gloves, or with a clean handkerchief. Once you think you have cut it down to the right size, try placing it inside the C-ring. If it doesn’t fit, (don’t force it or else it might fracture on the edges), take it out and trim the edges slightly with scissors, then try again. Once you have the filter fitting inside the C mount you can fix it there with a thin layer of glue, or better still place a washer or ring over it and glue this in. Avoid getting any glue on the center of the filter. Hold the C-ring/filter up to the light to make sure there are no light leaks around the edges. If so, cut off a small section of filter off-cuts and glue this over the offending hole. Finally, find a clean box to store the C-ring/filter in! Apart from its potential use during daylight/twilight hours (prior to Earthshine becoming visible to the eye) to capture 3rd to 4th magnitude flashes during the Leonids (or other annual meteor showers), it can also be used for:
Observing Impacts
The Usefulness of Wrätten 87C (Near-IR) Filters and Transmission-Type Diffraction Gratings
1. Imaging Venus, Mercury, Mars, the Moon, and Jupiter in daylight. 2. Viewing the Moon in the near-IR – things look slightly different. 3. Observing K & M spectral type star occultations in daylight (probably down to Vmag = 5?) and much fainter in civil twilight. 4. Observing K and M spectral type star occultations under near Full-Moon conditions or on the illuminated limb, to possibly a mag or more fainter than you would normally expect to see due glare. 5. Used with a C mount lens at night – you get to see a near-IR view of the sky – red K and M spectral type stars will show up very strongly 145
Observing Impacts
6. Used with a C mount lens in daylight you will see that vegetation is very bright and the sky is very dark (by ~1/10th). Clouds, aircraft & contrails and the Moon will show up easily. 7. Used in a dark room – anything over a temp of ~300°C will glow in the near IR and be detected by this camera. The transmission is 0% up to 800 nm, ramps up, to about 80% transmission at 890 nm, and then plateaus out to 90% transmission until the end of the spectral plot at 1,100 nm. To find the central wavelength one needs to multiply the spectral curve by the spectral response of one’s CCD. When one does this, one may wonder about the loss of area under the curve (apparent loss in sensitivity), however, this is not as bad as it seems because: (a) if being used in daylight or next to a bright limb of the Full Moon, background scatter is reduced by ~10× & (b) if one observes red K or M type occultation stars (or hopefully red? impact flashes) the black body radiation peaks up in the red/IR end where the sensitivity is at maximum. When reporting such observations, remember to include the fact that the W 87C was used. In addition, as is the case when unfiltered observations were made, it is important to make standard star observations with the W 87C in place, so as to provide calibration and atmospheric extinction data for later calibration of any lunar impact candidate events. If an impact event is observed in both filtered and unfiltered systems, it becomes possible to obtain temperature and emission area information for the brighter impacts.
The Diffraction Grating One of the problems with videotaping meteor impact flashes is the common occurrence of cosmic ray hits on the CCD that may mimic impact flashes. In fact, cosmic ray “impacts” seem to be far more common than actual lunar meteor impacts. The best way to get around this is to have two observers stationed far enough apart, observing at the same time and the same part of the Moon, so they can verify each others’ observations. However, that is not always possible, or it does not always work so favorably. One of the several solutions (others are offered in Chap. 11) to get around this problem is to add a diffraction grating to the system. Genuine impacts will show spectral signatures (if bright enough), cosmic ray hits will not. To utilize the grating, mount it in a similar fashion as described above for the near-IR filter at <1 cm from the CCD. Unlike the near-IR filter, the diffraction filter is only to be used when it is dark, and not during twilight; it would be a good idea to practice a few nights before on stars to determine your magnitude limit. Another drawback is the limiting magnitude – spectra can usually be obtained to mag. 4 and possibly 5 – but this depends upon sky transparency and altitude above the horizon. The diffraction grating will cause a 1–2 mag loss in sensitivity – so use it only for observations if you want to risk it or have a spare CCD and telescope that can view the event unfiltered simultaneously. Glare from the Moon will be a problem, so point the scope at the dark limb furthest from the terminator. It is also important to get calibration spectra at some point – so point the telescope at a distant metallic object showing specular (coherent) reflection from a Sodium or Mercury street lamp. Aircraft strobe can also provide spectra if you can move the scope fast enough to capture one. Other uses for the diffraction grating could include capturing the spectra of bright occultations and 146
Observing Impacts
attempt to detect the wavelength dependent Fresnel diffraction bands. Also try to capture the spectra of meteor showers in the sky. Again, as was the case with unfiltered and any sort of filter, make observations of several standard stars, to ascertain a spectral response for the central (zeroeth) order of the spectra, for possible photometric use. Also, the limiting magnitude and atmospheric extinction may be obtained with suitable observations of standard stars and standard star clusters. Information about standard and comparison stars and their use in calibrating data is presented in Appendix G. This information is a brief summary about standard and comparison stars and how they are used to calibrate data; more information about these things are provided at the web sites given in Appendix G. Also provided is a short list of such stars and links to websites that have many more standard stars and information on how to use them in photometric work.
Frame Grabbers and AstroStack One useful item to have in the lunar meteor observation toolkit is a frame-grabbing product, such as the AVerMedia USB2.0 frame grabber, which can be bought for less than $70. Similar products are available on the market to suit one’s needs and range from $199 on up to nearly $5,000. The usefulness of a frame grabber is that one can select and “grab” the frames of choice if one notices a lunar impact candidate event. One can save these frames separately and later use them to analyze the event. One can use AverMedia or another product of choice to do the grabbing of AVI video frames (short video clips) and then use AstroStack to process the data (see: http://www.astrostack.com/ for a free demonstration copy of that software). However, if the video camera transmits analog video, one would also need an analog-to-digital converter, and these devices are discussed a bit more in the following lines. Having a converter would make possible editing of video, or grabbing the exact frame (or frames) to stack the frames and produce a high quality image. Astrostack allows you to load in a series of bitmaps or an AVI file, then align them and stack them, as well as play with the resulting sharpness, and more. This works great in general for getting a fantastic photo from a series of frames off of your video. Unfortunately, it does not work as well for Lunar Meteor Impacts because you want to isolate a particular frame, not stack several. But if you grab several frames around the one you want, Astrostack will let you look through them quickly, and so you can identify the individual bitmap frame that you want. It would be very useful if the program loaded one single frame to play with the image, but this is unfortunately not the case. However, it can be useful to stack several frames around an impact event to improve the contrast of the earthshine features and better locate the impact flash. MaximDL is another useful program for this task. But while videotaping the Lunar Phases, it was discovered that when one plays with the gain and shutter, one can get a very detailed view of the Earthshine-lit portion of the Moon. Of course, the greater the sunlit portion, the greater the light bloom and lens flare. Up until about a four day old Moon, one can compensate by moving the sunlit portion off of the field of view. This should work nicely for impacts, given the entire Moon is visible and surface features are as clearly identifiable as they would be on a Full Moon video. But as the Crescent waxes, the flare increasingly interferes. A more effective method of countering the stray light problem is to fit a black baffle about one to two millimeters smaller than the inside diameter of the camera 147
Observing Impacts
adaptor (c-mount or t-mount) and this will eliminate much of the glare from the bright terminator. It would be a good idea to test this out before use, with a lamp or other artificial light source, to make sure no bright spots appear in the adaptor from the source. Look at this from all angles, and when it passes the test it is ready to use in your system.
Analog-to-Digital Conversion Devices (Composed of Information from George Varros and Members of IOTA) Canopus ADV-55/110 Video Converter vs. Datavideo DAC-200 Mr. George Varros compared two analog-to-digital converters (Canopus versus Datavideo) and has written a review of each, which are adapted for this section. It is necessary to convert video from analog to digital before streaming it to a computer and preparing it for use with LunarScan, the automated impact detection software described in more detail in the next chapter. The Canopus ADVC-55 video converter takes NTSC or S video in (only) and has as output pure DV-AVI 720×480 28,771 kbs video/1,536 kbs 16-bit audio via Firewire/1394/iLink (only). It checks out well and works without doing any bad things to the drivers on an XP based laptop. The unit is powered by a 6-pin Firewire port; however, it does not come with a DC power adapter, something that needs to be purchased separately. The setup is compatible with the KIWI OSD unit in place and the KIWI was reset without it terminating the Firewire feed to the converter. The ADVC-55 unit is small, 3″ × 5″ × 1″ and has a few dip-switches for NTSC/PAL and 0 db/+15 db gain for audio and is a simpler implementation of the ADVC-110, which converts inbound Firewire, NTSC or S video to any of the three as an output. The ADVC-110 also allows a second viewing or recording device to be used concurrently. However, it is not clear whether it is defaulted to what the ADVC-55 defaults to. The ADVC-55 A/D Converter retailed for $156.97 plus $6.98 ground shipping for a total of $163.95 (prices as of April 2008). The Datavideo DAC-200 FireWire Analog-Digital video converter was tested and had initially passed the test, but some issues for consideration were raised in the testing. The device is a video signal converter or video “pass through” device that converts an NTSC, S Video, or a 1394 (iLink Firewire) input to a different flavor for the output. It costs $184.00 and is a rather inexpensive alternative if you do not have a device that outputs a FireWire signal from which a computer can record a raw DV AVI (Mac and PC compatible.. .not sure about Windows Vista) It comes with a power supply, which its competitor did not (see more on that below). Here are some things to consider before purchasing a DAC-200 video converter: 1. The DAC-200 has two different settings, one for DV or AVI and another for 12-bit audio or 16-bit audio. The user must manually push both buttons in order to set them to AVI and 16-bit audio. This allows your on-board video drivers to recognize the attachment as an AVI-DV capable device. After recording a movie, the PC sends a “stop” signal to the device telling it to “stop.” This resets the setting back to the default of DV/12-bit audio and the PC pauses as it “burps” and “wonders” what happened to the device. To set the buttons back to AVI/16-bit
148
George Varros’ GV-D800 Watchman and DTRV-740 do not turn off when using them as a pass through device like the DAC-200 does because it is not “playing.” So, in summary, the DAC-200 implementation has lots to be desired. Despite his mention earlier about having to buy an “optional” power supply and the higher price from Canopus, Mr. Varros returned the Synchrotronic DAC-200 after ordering and testing the Canopus ADVC-110.
Observing Impacts
audio, the button for the audio needs to be pressed repeatedly until it sets. It then takes several seconds for the video to stream through to the viewing window; a reset takes time. 2. With the device configured to Windows XP/ME (i have XP), the KIWI OSD reset button, when pressed, interrupts the feed into the computer and resets the settings, killing the video entirely. No saved file or part of a saved file. By configuring the device to be just a NTSC Video converter, the KIWI does not kill the device.
Other Analog-to-Digital Conversion Methods Other products that have been used to convert the video stream into digital (which is necessary for use with LunarScan and other analysis software) include a simple video to USB converter (ADS VideoXpress), which has worked well. The uncompressed video stream can be captured with VirtualDub using the HuffYUV codec for hours without dropping frames. This is possible by having a second hard drive dedicated to video files (like one of the many high-capacity external hard drives commonly available today). One pitfall in capturing the video straight from a Windows laptop is that Windows maintenance activities interrupt the disk writing. If you do not have an external or second hard drive, and would like to use your laptop as a primary capture device, it helps to have a checklist of Windows services to deactivate prior to the video capture. Some examples of what to turn off include: wireless Internet, virus checker, screen saver, power saving mode, etc; and when capturing, exit all unnecessary applications. Another option is to record to mini-Digital Video recorder (DV), which has the advantage of being self-contained and can be quite reliable. A video stream from a video camera such as a Supercircuits or Watec is streamed to the DV, instead of a TV/VCR combo unit, and the stream can be recorded to digital video for later porting to a laptop. It is important to protect the DV by placing the camcorder in an insulated bag (which also doubles as the camera’s carrying case). However, not all mini-DVs are created equal and not all are useful for this process. We have found that the Canon products, such as the ZR-90 (no longer widely available except perhaps on eBay) and all the model numbers up to ZR-300, are best for analog-to-digital conversion. The ZR-700 also is able to do this task, but all the other models are not. One is encouraged to make certain that the DV camera of choice (not just the Canon line) is able to take analog and save it as digital.
USB vs. IEEE 1394 One concern about using USB interface for video recording is that USB relies on the computer resources and may easily get interrupted, especially if many programs and processes are running at the same time. One way around this is the 149
Observing Impacts
use of IEEE 1394 (firewire), which uses few internal PC resources and is much better in throughput, as far as continuous data stream is concerned. Two possible configurations that can be used include the following: a video converter (which accepts composite VIDEO, S-video or 1394 inputs, and delivers 1394, S-video or composite video outputs). This box proved very versatile when used with Ulead Video Studio to capture the AVI file. A second configuration uses a 1394 video converter from Imagesource. It is small, accepts video (2 inputs) and S-video and outputs through a 1394 port. It comes with IC Capture which allows single frame or sequence frame capture, as well as AVI capture.
Putting it All Together: A Lunar Meteor Observing Plan By Peter Gural During each lunar cycle there are optimal sun angles for viewing the nonilluminated surface of the Moon and having the near side of the Moon running head on into the sporadic meteoroid environment. The illuminated portion of the lunar surface is too bright to produce sufficient contrast to see the flash events, thus the observations are centered on the dark face with the bright limb out of the field of view. This occurs during the time period centered on first quarter (greater than two days but no more than ten days past new Moon) where a substantial portion of the dark limb is visible for telescopic inspection and monitoring. Closer to the new phase, the Moon is too close to the horizon or sets too quickly after nightfall, whereas near the full phase, very little dark edge is visible. During the first quarter of the lunar cycle, an observer would collect and record imagery on digital videotape, playing it back through a detection software package at a later convenient time. The camcorder clock would be set to the correct time at the start of each observing session resulting in a time stamp being imprinted on the video record to synchronize detections with other observers. For spectrographic work, the period of effective observations should be curtailed to be no later than first quarter as glare from a brighter Moon would significantly corrupt the spectra. An observation session would typically commence one-half hour after sunset and continue until the Moon gets within fifteen degrees of the horizon. Observations should also be targeted for special nights during major meteor showers possessing good geometry as well as during total lunar eclipses when the entire lunar face could be observed for over an hour. Working from a list of known meteor showers and selecting only those with significant zenith hourly rates, a single table of observable events that covers predictions through the year 2061 (see Table 9.1) can be constructed. This table provides an easy means to determine the favorable observation geometries for the Quadrantids, Eta Aquarids, Capricornids, Perseids, Southern Taurids, Leonids, and Geminids. The table takes advantage of the repeatability of the lunar phase to within a day every 19 years. Best observable periods for meteor showers fall within two ranges: a week near first quarter (Moon phase is 3 to10days old) and again near last quarter (Moon phase is 19.5 to 26.5-days old). In the table, favorable geometries are shown as bold numbers (days past new Moon) within gray boxes. All the showers listed in Table 9.2 have observable impact 150
Year
Year
Year
QUA; Jan 4
ETA; May 6
CAP; July 30
PER; Aug 13
STA; Nov 6
LEO; Nov 18 GEM; Dec 14
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023
2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042
2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061
22 4 15 26 6 18 29 10 21 3 14 24 5 17 28 8 19 1 12
27 8 18 1 12 22 3 14 26 7 17 28 10 21 2 13 24 5 16
24 5 15 27 9 19 29 11 23 3 14 25 7 17 28 10 21 2 12
7 19 0 11 21 4 14 25 6 18 28 9 20 2 13 23 5 16 27
4 15 26 7 18 29 10 21 2 14 24 5 16 28 9 19 1 12 23
16 27 7 19 1 11 22 4 15 25 6 18 29 9 20 2 13 23 5
13 24 4 16 27 8 19 1 12 22 3 14 26 6 17 29 10 20 2
Observing Impacts
Table 9.1. Favorable viewing geometries of the major meteor showers through the year 2061 are shown in gray boxes with bold values indicating number of days past new Moon
This table was constructed by Peter S. Gural and has given permission for its reproduction Table 9.2. Best regions on the nonsunlit surface of the lunar near side to direct a telescopic impact survey (avoiding the bright limb in the field of view) Meteor shower
Maximum date
Near first quarter
Near last quarter
QUA ETA CAP PER STA LEO GEM
January 4 May 6 July 30 August 13 November 6 November 18 December 14
Northeast quadrant Central region Eastern half Northeast/central region Eastern half Eastern half Eastern half
Northwest quadrant Not observable Western half Northwest limb Western half Not observable Western half
regions for both the first and last quarter except the Leonids and the Eta Aquarids (first quarter only). As the geometry changes are not very significant around each meteor shower’s observable dates, one can generalize the focus regions on the nonilluminated face for telescopic video observations (see Table 9.2). More detailed plots of the observable regions for each major meteor shower are presented in Appendix H. How does one know if an observed flash is a lunar meteoroid impact? A pair of observers with potential detections at the same time can verify the reality of a lunar impact by comparing their respective selenographic (lunar) coordinates of the impact site. If the times and locations agree, then there is a very high probability that the flash originated on the Moon’s surface. A ground site spacing of at least 30 km (18 mi), timing coincidence to within a second, and positional accuracy of an arc minute is sufficient to ensure that the flash was not due to sun glint (out to about 80,000 km or 50,000 mi) off a near Earth orbiting satellite, geosynchronous satellite, 151
Observing Impacts
or space debris. Note that cosmic rays can excite several localized pixels on a CCD causing potential false alarms but would affect only a single sensor at any given time, therefore having coincidence observations are critical to rule these out. With finer angular positioning and/or wider spacing between observing sites it becomes simpler to verify the lunar origin of the events as the parallax effect is greater on objects closer to the Earth than at the distance of the Moon. The automated detection software LunarScan includes a utility to make your own impact plot, specific to when you plan to observe the Moon. It has information for most of the significant meteor showers in memory and constructs the impact plot for a given date and time. This is especially useful if one wishes to observe a shower not included above as well as dates not listed in the general listing. Specific details on the use of LunarScan can be found in the documentation that can be downloaded with the software.
Conclusion This chapter went into considerable detail to describe techniques that observers of a wide range of experience levels and equipping can use in setting up and maintaining their own lunar impact observing programs. I have included basic techniques as well as detailed explanations by three individuals on how to set up and run an observing program. Also, twelve examples of video setups were included as well as information on some of the products used to enhance a lunar meteor observing program. Appendices D through H have observing resources designed to enhance the observer’s abilities to capture and record lunar meteoroid impacts. Appendix D presents a simple method for timing occultation (and lunar meteor impact) observations by Steve Preston of IOTA. Appendix E has a listing of much of the equipment used in lunar meteor work, including vendor information. Appendix F contains a list of sources of WWV timing signals from around the world, and Appendix G has information on how to estimate the dimmest stars that your setup can see. Finally, Appendix H includes impact plots to go with the last section by Peter Gural, to visualize the orientation of the Moon during times of favorable shower occurrences. You, the reader, are encouraged to visit websites associated with various organizations, including the Association of Lunar and Planetary Observers (ALPO) and the International Occultation Timing Association (IOTA) to learn more, join listservers, and keep up to date with the changing technologies of video astronomy. Details about these organizations were presented in Chap. 8.
152
Chapter 10
Finding Collisions
The obvious need for automated impact detection software has been met almost from the beginning of the era of digital impact observation. LunarScan was available (though not to the public) as early as late 1999 and was used to add candidates to the list of impact events observed during the Leonid storm of that year. Nearly 8 years later, the software was made available to the public, with the agreement that information pertaining to any impact candidates detected by the software be shared with the NASA-Marshall Space Flight Center Meteoroid Environment Office. More information about this software is provided later in this chapter. The function of the software is to look for changes in successive video frames with the fact that a change in the scenery corresponds to an event of some sort. When nothing happens, the image looks static, unchanging, but when an impact or cosmic ray flash occurs, that introduces a change which may or may not be detected by a watching human. The human is only able to effectively focus on a small area of the video at any instant, and the attentiveness may drop after a short period of staring at an unchanging scene. Automated software does not suffer from any of these effects and is able to effectively monitor the entire frame for changes attributed to impact events. Cosmic rays and a drifting Moon (with the bright portion drifting into view) can trigger positive detections. It is up to the human to decide whether a positive detection is a genuine lunar event or something artificial or otherwise. Prior to LunarScan’s release, Mr. Roger Venable and others found a way to use other freeware for automated detection. The freeware is Registax and, depending on the need of the observer and the system at the observer’s disposal (LunarScan requires some specific elements to be included in one’s system), may be more convenient for the observer to use. Information on its use in automated detection is provided next.
Finding Collisions
Automated Impact Detection Software
153
The Use of Registax for Automated Lunar Meteor Detection
Finding Collisions
By Roger Venable
154
Although there has been talk in the amateur astronomy community about laborious review of videotapes by frame-by-frame step-through, this task proved overwhelming and few events have been detected by that technique. Attempts are being made to detect events fainter than can be seen on a visual review of the videotape, by the use of computer programs to screen the digitized video for bright flashes that stand out, statistically, from the background. The first such program, LunarScan, was used by Peter Gural to detect a number of the flashes reported in the 1999 Leonid meteor shower. Peter screened the videotape made by David Dunham with a 125-mm Schmidt–Cassegrain telescope and a highly sensitive black and white security-type video camera. The program he used was initially designed to detect meteor streaks in an all-sky video, as an alternative to visual meteor observation. He modified the program for its use in lunar impact detection. He has modified and updated LunarScan for optimization of the impact detection routines, in coordination with the NASA lunar meteor impact detection program operating in Huntsville, Alabama. In a similar vein, Cor Berrevoets is working on a modification of his Registax program so that it can be adapted to the detection of lunar meteor impacts. The program has long been used for alignment and stacking of multiple images, such as digitized video images, in order to yield a high resolution astronomical image. About 3 years ago, Cor added a “sigma clipping” routine as one of its image processing functions, to remove from the video image files those pixels that were outliers in brightness, according to an adjustable number of standard deviations, or “sigma.” Roger Venable recognized the relevance of this function to lunar meteors. He wants to list and display the bright pixels rather than “clip” them, and he has been working with Cor on the development of an event detection routine that is being incorporated into the next major update of the Registax program. Cor has always allowed this program to be freely distributed. Even with the use of a computer, the detection procedure is tedious. A number of programs must be used during the process, to deal with various issues. The first issue is file type. The most widely used video file type is the avi format. “avi” files are a Microsoft Corporation standard file type. The initials stand for “audio–video interleave” and it is usually pronounced as “ay-vee-eye.” If you decide to use the Registax program to search for impact events, you will have to record your video in the avi format so that Registax can read it. The issue of file size is immediately evident. Video files on a computer are huge files. Black and white avi files with high resolution typically contain 120–165 MB/ min of video. The use of Virtual Dub video filters will enlarge the files by a factor of three or four. Registax cannot handle files larger than a gigabyte. Consequently, it is necessary to limit the size of each avi file to 1 or 2 min of video. If you use Registax to search these unmodified avi files for impact events you will get a list of innumerable video noise events, so that you will not be able to find the real impacts amid the noise. It is necessary to process the avi files to decrease the noise in a way that will not alter the brightness of impacts. Analog-to-digital (A-to-D) conversion results in characteristic noise events that are short horizontal
Finding Collisions
bright or dark streaks that are only one pixel in vertical dimension. In contrast, impact events have dimensions of more than two pixels, typically five or so, both horizontally and vertically. Averaging each pixel’s brightness with that of its surrounding pixels will thus decrease the prominence of the A-to-D conversion noise but have relatively little effect on the prominence of the impact flash. Using the freeware program by Avery Lee called Virtual Dub, results appear good with the “2:1 reduction, high quality” video filter provided with the current release of the program (release 24415.) This filter will cause the avi file to be much larger than the original but the frames will be reduced in size by a factor of 2. That is, the 1-min avi file size will increase from the original ~150 MB to the final 456 MB, while the frame dimensions are decreased from 720 × 480 to 360 × 240. Another type of noise in the video file is the random noise on the original videotape. The blurry and rough appearance of individual still frames of the video is caused by such noise fluctuations. They are characterized by the fluctuating brightness of every pixel, including the dark background, with diameters of the individual bright and dark fluctuating areas equal to the resolution of the video system. In general, the fluctuations last only for a single video field. Since meteor impact flashes have a duration of greater than one video field, their prominence will be increased by temporally averaging the video pixels to smooth this background noise. A gentle temporal filter is needed, averaging only two or three frames together, so as not to decrease the prominence of real impacts. Preliminary results with the “motion blur” video filter of Virtual Dub appear good, decreasing the number of noise events detected by Registax, but the “temporal smoother” filter results in an increased number of noise events and thus should not be used. Experimentation with these and other video filters that run with Virtual Dub is ongoing. Users should understand that the use of Registax for event detection is very different from its use for image processing, and there is no need to align and stack the avi frames. In fact, doing a typical Registax alignment will result in numerous false events due to hot pixels in the video camera, and these cannot be completely eliminated by calibrating each image with the dark frame and flat field functions of Registax. Registax version 4 (http://www.astronomie.be/registax/) can be used in the following way to detect events. After selecting and loading the avi file you wish to process, select “none” on the “Alignment options” page of the “Align” section. Then click on the “Limit” button at the left, and the “Stack” section will immediately appear. Select the “Sigma Clipping” page, and without making further adjustments, click on the “Search Events” button. The event dialog box will immediately appear. Set sigma in the event dialog box to a low value, such as 4.00. Experience will tell you how low to set the “Low pixel value” number, and I suggest a value of 20 to begin with. Using the controls at the lower part of the dialog box, adjust the edges of the yellow-outlined search box to include as much area of the Moon as possible but leave a small margin at the edges of the field (e.g., 5 or 10 pixels) to eliminate the many noise events that would otherwise be detected there. You should also exclude from the yellow search box any letters written by a video time inserter. Then, click on the “Start” button, and the program will take a few minutes to search and list the outliers in brightness. The resulting list can be scrolled up and down in the dialog box. There will be more events in the list than you can inspect individually. By adjusting sigma upward, the number of flagged events decreases, so that you can adjust it to a manageable number of events. The advantage of setting sigma low initially is that it allows you to compensate for
155
Finding Collisions 156
the differing noise levels of different files by adjusting the sigma specifically for the file you are examining. By clicking on a frame in the event list, you can display the frame in the image window, and a yellow circle will appear around the aberrantly bright pixel. Then, the “Next Frame” and “Previous” buttons nicely display the surrounding frames for review. The greatest issue in the computerized detection of meteor impacts is the separation of noise events from true meteor impacts. Even though you have processed the file to decrease noise, many or most of the selected events will be noise. You must inspect them individually to ascertain whether they have the features of meteor impacts. The simultaneous use of Registax and Virtual Dub in the multitasking Windows environment will allow you to inspect, with Virtual Dub, the original avi file’s fields at the pixel selected by Registax. To do this well, you will have to manipulate the view with Virtual Dub. After loading the original avi file into the program, select the “de-interlace” video filter, and then select “Unfold fields side-by-side.” Then drag the bottom edge of the filtered field downward to give the view its full original height. Also, drag the right edge of the input frame to the left, compressing the input frame horizontally, so that more of the output frame can be seen. If the candidate impact flash is in the far right portion of the video frame, it may still be off the computer screen in the right-hand field. To remedy this, rightclick on the output field and a dialog box will appear that will let you select a smaller display size. A meteor impact flash has dimensions on the video equal to the size of a star image of the same brightness. For faint flashes, the size is small and will be comparable to the resolution limit of the video system. This limit, even on the filtered, reduced-size video frames, is greater than one pixel. A real impact looks somewhat blurry, with central brightness greater than peripheral brightness. Also, it is brightest on the first field in which it is visible, and it is visible in more than one 0.0167-s video field. Blurry flashes that appear in only one field are likely cosmic ray hits. A flash that satisfies all these criteria merits reporting to other observers in order to confirm that it is of impact origin. When you report a suspected impact flash, you will want to report its location amid the lunar landscape so that other observers will know exactly where to look on their videotapes. You can bring out the albedo features on the lunar dark side by stacking and processing one of your unfiltered video files using the regular Registax alignment, stacking, and wavelet processing routines. By saving the stacked file in bitmap (bmp) format, it will be accessible to other image processing routines. Meanwhile, the video fields from the original avi file that contain the suspected impact flashes that were selected by Registax and confirmed by inspection in Virtual Dub, can be saved using Virtual Dub by selecting the fields and saving the image sequence as bitmap files. Having done all this, you can use the Windows Paint program to open the image of a candidate impact. Then, mark your computer screen to indicate exactly where on the screen the flash occurred. This mark can be made with a nonresidue sticker such as masking tape or a Post-it Note. Alternatively, it can be made with an ink mark on a piece of transparent elastic food wrap that will adhere loosely to the computer screen. Once the location is marked, paste the stacked and processed image of the lunar dark side onto the field, obscuring the flash field, and align it so that the Moon’s edges coincide with the lunar edges as seen on the flash field. Then the screen marker will be pointing to the albedo feature at which the flash occurred. Use a lunar map to identify this area by lunar latitude and longitude and by description. The stacked, processed image of the lunar dark side will have the albedo features of the full Moon,
Finding Collisions
because it is lit only by earthshine, so that a full Moon albedo map may be useful to identify the flash’s location. You may wish to measure the brightness of the flash on the original avi file. This can be done with the Limovie program (for an information sheet, go to this website http://www005.upp.so-net.ne.jp/k_miyash/occ02/limovie_en.html), which is freeware. If the program’s “Field Measure” function is checked at the lower right of the user interface, then the flash’s peak one-field brightness will be reported in the readout from a brief measurement run spanning the flash’s frames. Take care to adjust the aperture radius to the smallest size that includes the entire flash, and to use a large background measurement area. Unfortunately, the field-to-field noise will be high and any measurement taken from a single field will be an approximation. Limovie can also be used to measure the video brightnesses of comparison stars, so that the flash’s brightness reading can be calibrated. Be sure that the comparison stars’ brightnesses are measured with the “Field Measure” function selected, as with the flash’s brightness measurement. In contrast to the impact flash, the brightness of comparison stars can be averaged over many frames, by using the CSV file saving function of the Limovie program and importing this file into Microsoft Excel for processing. Limovie measures brightness linearly, so that some exponential math will have to be done to compute the magnitude of the flash. Thus, on the night of the lunar impact observation, the observer may wish to make a video record of comparison stars near the Moon, using the same equipment he used to record the lunar dark side. Choose comparison stars that are not so bright as to appear saturated on the video screen, because saturation is data loss. In a nutshell, the process is as follows: 1. Record the lunar dark side on videotape. 2. Capture the lunar dark side video onto a computer using a standard, high resolution video recording device at its highest resolution setting, to make files such as avi files that can be read by Registax. A firewire or USB2 port, not USB1, is needed for this. 3. Divide the computerized video file into small files of 1–2 min each. This division of files can be done with the video capture program. 4. Stack one of the video files with Registax, in the regular way, and process it to bring out the surface features of the lunar dark side. This is your impact-locating reference frame, and it can be used to locate all the impacts you detect on the same night with the same equipment and camera orientation. 5. Use Virtual Dub with video filters to do a “2:1 reduction, high quality” and a “motion blur” filtering of the avi files, and save them. Do not discard the original avi files. 6. Process each filtered file with the Registax event detection routine. 7. Increase the sigma in the Registax event dialog box so that perhaps 10 or 20 candidate events per minute of avi file are listed. 8. On the original, unfiltered avi file, use the Virtual Dub de-interlace filter, unfolding the fields side by side, to inspect the selected candidates on de-interlaced individual fields, to see whether they have the features of impacts. 9. For those events that appear to be impacts, use Virtual Dub to print the field sequence to a bmp file that can be read by an image processing program such as Windows Paint.
157
10. Open these saved images in Windows Paint, mark the flashes on the screen, and paste the lunar albedo image over them so as to identify the location of the flash on the lunar surface. 11. Report your flashes with the exact time and location to the ALPO lunar meteor impact coordinator.
Increasing the Probability of Detection with LunarScan By Peter Gural and Brian Cudnik
Finding Collisions
An Automated Flash Detection Program
158
One of the most daunting parts of a comprehensive lunar meteor observing program has been the need to review hours of nearly featureless video tape in searching for the brief, transient spark that marks the spot of a lunar meteor impact. In fact, this very obstacle has long prevented such a program from taking place. One is likely to miss a candidate impact event due to fatigue if one watches the video for too long. LunarScan was developed shortly after the first confirmed Leonid impacts, and its purpose is to increase the probability of detecting an actual impact event. A key to the improvement of detection efficiencies for transient lunar flashes is the use of automated detection software. A prototype version of software entitled “LunarScan” has already been developed and used on the lunar impact videotapes recorded by David Dunham during the 1999 Leonid meteor storm. The software was highly successful in detecting the six impacts found through visual inspection of the tapes plus an additional five events that were not previously identified in the manual scans. This occurred despite the poor tracking quality of the imagery and periodic realignment of the telescope’s pointing direction. Note however that those characteristics required constant vigilance by the operator to avoid false alarms, which can be prevented by the proposed fully automated system through better telescope tracking (lunar rate) and proper polar alignment. A display of the eleven detections and their computed magnitudes are shown in Fig. 10.1 for the observations made during the Leonid meteor storm on 18 November 1999. The LunarScan software is based on another highly successful real-time processing code used for video meteor detection called “MeteorScan.” That software was successfully deployed in several remote sites around the world in 1998 and 1999 to detect and report on meteor statistics in real-time. The lunar flash variant requires far less image processing capability, but is not as far advanced in post-detection analysis capability at this time. These eleven lunar impact events ranged in apparent visual magnitude from +4 to +8 and are easily within the range of small telescopes equipped with inexpensive CCD cameras generating frame rate video imagery. The duration of these events is not well defined but each was typically less than 17 ms, having appeared in a single interleave frame of standard NTSC video. Some professional modeling efforts have indicated that the visible flash duration could be as short as 3–4 ms. Because of the very brief nature of the transient flash, coupled with the Earth– Moon distance making it appear quite small in angular extent, the flash was visible in only one to two video frames, and usually appeared as little more than a brief
159
Finding Collisions
Fig. 10.1. Eleven lunar impact flashes seen between 1:30 and 5:30 UT on November 18 1999 recorded with David Dunham’s 5” telescope and CCD video camera (the Supercircuits PC 23C camera). Note the banded appearance of the flash in the odd or even rows of the image due to the flash’s short duration (<17 mec), thus showing up only in a single interleave video field. Images identified with letters A–F were independently verified as lunar impacts.
4:46:15.5 UT 4:51:24.9 UT 5:14:12.9 UT 5:15:20.2 UT 5:26:43.2 UT Cosmic Ray “A” mv = 5.1 mv = 6.3 “B” mv = 6.2 “C” mv = 5.3 mv = 5.3 Detection
1:46:09.7 UT 2:52:19.7 UT 3:05:40.4 UT 3:49:40.4UT 4:08:04.1 UT 4:40:26.8 UT mv = 8.3 “F” mv = 6.2 “D” mv = 4.9 “E” mv = 5.8 mv = 6.3 mv = 6.7
Finding Collisions
noise spot on the television screen. In addition, for visual observers, the very short flash was easy to miss, if the observer happened to blink at the time. Such was the case for me, Mr. Cudnik while I was visually monitoring the Moon during the 2001 Leonid events. I was watching during the time one of the confirmed impacts occurred, but did not see the actual event. I later concluded that the event likely happened while I blinked. As a result of this miss, it was also concluded that, in general, only the brightest flashes were discernable by human observers reviewing video tapes, and in many cases because they had been cued from another observer as to time and location from an independent observation. The flash detection algorithm used in LunarScan is based on a search for a bright cluster of pixels in the interleave rows (separately processing odd or even rows) that exceed a dynamically tracked noise level (i.e., the “exceedance”). A mean and standard deviation is periodically updated for each pixel in the field of view and combined into a threshold to identify pixel exceedances. If a cluster of three adjacent pixels in a row plus one pixel in an interleave row above or below, have all generated an exceedance, then a potential flash candidate is declared, and a time sequence of several frames before and after are saved. Later in a user playback confirmation mode, the user can examine the flash in greater detail and either retain it as a true detection or throw it away as a false alarm. The confirmed flashes from the Leonid data tapes (both 1999 and 2001) all show a distinctive compact spatial structure and have provided the only useful criteria to their detection besides coincidence observations. The temporal light curve response is far too short lived to be of any use in detection at the 30 Hz frame rates of NTSC video. As seen in Fig. 10.2, the flash duration is typically very short, barely spanning a second interleave frame in only 25% of the cases known to date. The detection scheme tries to avoid the numerous false alarms that would arise out of cosmic ray events that cause single pixels to flare with bright noise spikes. This scheme allowed for the discovery of six new flashes in Dunham’s 1999 Leonid tapes, doubling the number found from multiple human inspections of the tapes. Unfortunately only one of these was confirmed, owing to lack of coverage by a second telescope or because the second telescope pointed to a different part of the lunar disk. The impact flashes are also spatially extended in the imagery from CCD blooming, which makes it possible to detect them relative to CCD pixel noise spikes. This ultimately allows for the determination of their location on the surface of the Moon.
A Comprehensive Observation Program with LunarScan For a comprehensive observing program, the following should be carried out: During each lunation cycle, from early evening crescent (3–4 days after new Moon)
Fig. 10.2. Examples of lunar impact flashes lasting two interleave video fields. Shown are the alternating odd and even video fields broken out separately for a 1/60 s frame-to-frame resolution. The top row is leonied "A" at 4:46:15.52 UT and the bottom row an unconfirmed flash at 4:51:25 UT recorded by Dave Dunham Nov. 18, 1999.
160
Finding Collisions
to 10 days after the new Moon, an observer should attempt to detect lunar impact flash events given favorable weather conditions with the Moon at least 15° above the horizon. Data should be collected over a large number of nights throughout the year to obtain statistics of impact frequency, magnitude distribution, and a rudimentary estimate of the average duration of the visible light curve. The latter can be attempted because nearly 25% of the impacts seen in the past have shown a faint trace in a preceding or trailing interleave image frame (see Fig. 10.2). Using a theoretical light curve for hypervelocity impacts one can estimate the percentage of impacts one should see spanning multiple image frames for a given hypothesized flash duration. To reliably estimate the flash duration from statistics, however, will require obtaining a sufficient number of events. To obtain accurate magnitude estimates (if one is so inclined to do so, and only if one is able to disable the auto-gain control on one’s camera) the observer should train the instrument on a set of standard stars at the start and end of each telescope session recording several seconds of video data. After background subtraction and integration of a star’s bloomed pixels, a calibration curve of log pixel intensity to V-magnitude can be made. This has been found to be linear in functional form from past experience with video CCDs in meteoric work. If an absolute flux value is to be determined (not merely a magnitude estimate), K and M class spectral type stars should be avoided due to their near-IR bias and sensitivity of the CCDs in that wavelength range. Ideally, stars of spectral class A, flux standard stars, along with G-type stars for solar analogs should be used. Stars should also be observed at several elevations to provide the means to correct the observations for absorption by the Earth’s atmosphere, which increases with decreasing elevation, and can vary from night to night. At a minimum, a video sequence of a handful of standard stars near the Moon throughout the observing session (or at least at the beginning and end of the session) should be made. A list of a sample of standard stars for calibration purposes is provided in Appendix G, along with the web addresses of more complete lists of standard stars. In addition to the normal watch during the early phases of each lunation cycle, special targeted nights should be highlighted for observations corresponding to periods of major meteor shower activity and total lunar eclipses. Chapter 9 presented information and instruction on how to use the impact plots of Appendix H to plan observations. These impact plots are for seven major annual meteor showers throughout the year and each shows the geometry of the event for the start and end of an observable period. From this information, the observer can note roughly the region where impacts would most likely be seen. The first (leftmost) column of figures corresponds to an observation approximately 3 days after new Moon. The second column is for the end of the first quarter period with the geometry of the corresponding meteor shower 10 days after the new Moon. The third and fourth column corresponds to meteor showers that would have observable impacts on the Moon from the beginning to the end of the last quarter observable period of the lunar phase. The pinpoints on each displayed plot represent the potential impact area on the lunar disk visible from the Earth that is facing into the oncoming meteoroid stream. The “+” symbol represents the sub-radiant point on the Moon for the given shower, that is the point where the meteor stream radiant is overhead for a lunar observer. A white “+” signifies that this point is on the Earthfacing hemisphere of the Moon; a black “+” means that the sub-radiant point is on the hemisphere facing away from the Earth (i.e., the far side of the Moon). Some of the plots show a curved line, which marks the edge of the extent that the impacting
161
meteoroids striking the Moon can be seen from the Earth and be associated with the respective meteor shower. The plots assume a meteoroid stream with a uniform particle density, and they are plotted with lunar north up. The denser spotted regions have a higher probability of experiencing an impact. The lighter gray region is the sunlit part of the Moon, where impacts are generally regarded as unobservable, owing to the brightness of the lunar surface.
Finding Collisions
Observation System Configuration The equipment needed for the observation program described above consists of a lunar-rate-tracking telescope, low-light sensitive B/W CCD video camera (30 Hz frame rate), computer with digital video interface board, and the flash detection software (see Fig. 10.3 and also refer back to Chap. 9 for more details on general observer setups). The CCD video chip should be mounted at prime focus producing an image field size that should be somewhat smaller than the diameter of the Moon. The mirror size and CCD camera chosen should be selected to push the limiting magnitude of detection as deep as possible without generating flash-like noise spikes (typical when using image intensified systems which is why these should be avoided). The telescope design requirements are that it would be easy to set up and align for tracking, possess a lunar rate drive system, and be of an appropriate focal length to image a good fraction of the Moon. One good example of a commercially available telescope meeting these requirements is Orion’s 10² f/4.7 Newtonian. For a 5.0 × 6.8 mm CCD chip size (WATEC-902H) the effective field of view is 17 × 23 arc minutes which gives good lunar coverage without sacrificing resolution. The Newtonian design, less expensive than a Schmidt–Cassegrain telescope (SCT), utilizes fewer optical elements than an SCT and is recommended since a Newtonian would suffer from less light scatter due to the bright lunar limb that would be nearby and positioned just barely off-axis. An inexpensive but very sensitive CCD imager, model 902H (0.0003 Lux) supplied by the WATEC Corporation, could serve as the primary video sensor. The video signal from the CCD would be recorded on digital videotape and played back through a computer later. The desired computer setup would contain an IEEE 1394 “firewire” video board that comes as a standard feature on PCs sold today. The purpose of the digital interface is to stream the video playback to hard disk storage which the software can then access and process. This has been demonstrated with meteor detection software on a Leonid storm videotape study carried out at NASA/Marshall. The use of the camcorder adds field
Fig. 10.3. Lunar Impact Monitoring System Configuration
162
Conclusion-Further Notes on the Methodology for Observing Lunar Meteoroid Impacts
Finding Collisions
portability as there is no need for the computer and AC power at remote observing sites and also allows the capability of saving the entire night’s recording for archival purposes. The computer system requirements are minimal for the current level of technology in the PC industry. A PC system with 2 GHz CPU, 128 MB of memory, and 40 GB hard drive can be obtained relatively inexpensively by today’s standards and has more than enough computing power to perform the necessary image processing functions for flash detection. The remaining hardware components of this system include a digitizing frame grabber and the digital or analog video camera attached to a telescope. With the more modern computers, coupled with digital recording cameras and IEEE 1394 interface boards, it is possible to stream the video data directly to the computer hard drive for later processing. One of the most important requirements of the setup is a stable equatorial telescopic mount, properly aligned, and driven at lunar rate. This configuration is such so as to prevent the bright lunar limb just outside the field of view from “flashing” into or across the field of view causing a cascade of hundreds of detections (“false positives”) in the software.
Basic requirements that apply to visual and basic video observations of the lunar meteor phenomena also apply to automated systems. A recorded timing signal, whether WWV on the audio track or a time stamp (GPS-based or otherwise) injected into the video stream prior to (or during) the recording, is necessary for accurate timing of the flashes and confirmation by independent observers. There should be at least two sites operating simultaneously and separated by at least 30 km (18 mi), observing the same region of the Moon. Having a second site providing a second video not only helps to rule out reflections caused by low earth orbiting or geosynchronous satellites or space debris, but it also helps to eliminate cosmic ray signatures since they cannot be coincident on two separate CCDs. Finally, high quality digital recording is preferable to VHS, 8 mm or other analog videotape media because the digital record/playback is free of the read head noise that can notoriously masquerade as impact flashes. Manual viewing of the video recordings can lead to missed events owing to the short-term nature of the flash and inattentive attention gaps from human fatigue. It is easy to miss the 1/60 s flash in the “blink of an eye” whereas automated detection software operates with a more uniform probability of detection across the entire image continuously. Without automated detection software there will undoubtedly be many unseen flashes and an underestimate in the number density of impacts visible from the Earth. More information on each of the software products featured in this chapter (Registax and LunarScan), including links to download the software itself and users manuals for the software can be found on their respective websites (listed below). The zip file of the lunarscan software provided from the second link includes the manual. http://www.astronomie.be/registax/ http://www.gvarros.com/ If these websites become unavailable, Google (or use your favorite search engine) “registax” or “lunarscan” to find websites that host these programs. 163
Finding Collisions
I anticipate these programs will continue to be available and even be upgraded or superseded in the coming years. One drawback of automated impact detection software is the detection of many “transients” or “false positives,” i.e., cosmic ray signatures that need to be sorted through to find genuine events. One approach of sorting through these events is laid out in the next chapter, which features a thorough analysis of cosmic ray signatures and other electronic noise. The chapter, written primarily by the Geologic Lunar Researches group of Italy, is the result of their extensive work on the subject. They conclude the chapter by giving some suggestions on how to separate genuine impact events from spurious effects.
164
Chapter 11
Spurious Flash or True Impact Event?
Written by Raffaello Lena with the Geologic Research Group and the American Lunar Society
One of the active groups involved in lunar meteoritic impact research is the Geological Lunar Research (GLR) of Italy. Working in conjunction with the American Lunar Society, they have assembled an online manual that provides guidance on how to make successful observations of lunar meteor impact events. (The complete copy of the manual is available, as of October 2008, for download at the following web address: http://www.glrgroup.org/lunarimpact/manual.htm) In their case, the manual emphasizes the detection and mitigation of spurious flashes and offers solutions on how to distinguish true impact candidates from cosmic ray flashes. The spurious flashes segment of their manual is reproduced, with permission, below. In order to further examine and characterize spurious flashes, so-called dark tests were used, where two video cameras at different locations recorded dark images through telescopes (that is, with the telescope cover in place). Since these were dark tests, any signal received would be defined as a spurious flash. Next, the two video images were compared to ensure that the flashes did not occur on both tapes.
Spurious Flash or True Impact Event?
How to Identify True Lunar Meteoritic Impact Events
GLR Dark Test Examination of our tapes revealed seven flashes that were recorded during a dark test, and were not recorded on the other videotape. These are recorded in Table 11.1. Also recorded in this chart is the signal to noise ratios of the significant flashes. The S/N ratio was measured graphically with the aid of an intensity profile plot of a flash. The distance between the peak and the mean noise level was considered the Signal. The S/N ratio was then calculated by dividing the measured
165
Table 11.1. Typology of spurious flashes Test
Time (min)
Significant flashes
Multiple points
Segments
Faint flash
S/N ratio of significant flashes only
A B C Da Ea F G
60 128 30 60 60 128 60
1 0 2 0 1 2 1
100 60 20 80 2 5 3
0 1 10 0 1 0 1
7 0 5 10 4 1 0
6.6 5.6 and 5.5 7.0 6.4 and 5.1 4.6
Results obtained with observers in different locations at the same times
Spurious Flash or True Impact Event?
a
166
Fig. 11.1. Image of a single-point spurious flash
Signal by the measured Noise. Examination of the images revealed four types of spurious flashes: single point flashes, multiple point flashes, segment flashes, and faint flashes. These are characterized as follows: Single Point Spurious Flashes: These are flashes that occur at a single point in the dark test tapes. Their profiles are all similar. Causes for this type of spurious flash include cosmic ray interaction with the chip, as described below in 'Segment Spurious Flashes'. A photographic image of one such flash is shown in Fig. 11.1 and its brightness profile in Fig. 11.2. A second image of a similar flash is shown in Fig. 11.3 and its brightness profile in Fig. 11.4. Multiple Point Spurious Flashes: These flashes appear as a series at separate points in a horizontal line. This kind of spurious flash occurs when the first pixel is found to be more luminous with respect to the next pixel (it is the sum of the two). The following pixel is found black because it is empty. This defect can happen more times in a defective cycle of reading. The photographic image of one such flash is shown in Fig. 11.5 Segment Spurious Flashes: These are flashes that are connected in a single, horizontal line. The photographic image of one such flash, along with its profile,
Fig. 11.3. Image of a spurious flash
Spurious Flash or True Impact Event?
Fig. 11.2. Intensity profile of the flash in Fig. 11-1
Fig. 11.4. Intensity profile of the flash in Fig. 11-3
is shown in Fig. 11.6. We suspect that this type of flash is caused by a cosmic ray. Cosmic rays represent high-energy particles generated outside of our solar system. They normally interact with molecular species in our atmosphere, producing showers of secondary particles. The same effect is due to the decay of radioactive elements in nearby rocks, such as granite, or in structures made of concrete. Faint Flashes: These are flashes that have low S/N ratios, and probably represent a noise peak. Their extreme faintness suggests that they are due to effects from 167
Fig. 11.5. Image of a multiple-point spurious flash
Spurious Flash or True Impact Event?
Fig. 11.6. Image of a segment spurious flash
electronic components, tapes, or the recording VCR. Examination of faint flashes reveals that their S/N ratio is almost always equal to or less than two standard deviations of that of the measured noise. Given this, we suggest that the definition be adopted that significant flashes are greater than five standard deviations. This should rule out the faint spurious flash. Unfortunately, this standard will also rule out faint 'true' meteor flashes, unless they are confirmed by a second, distant observer. The multiple point spurious flashes and segment spurious flashes -all have significantly different profiles that can easily be separated from the lunar impact flash. It is significant that the profiles of a star, of an actual lunar impact flash, and of the single point spurious flashes are similar enough to be indistinguishable! The photographic image of spurious signals probably due to cosmic rays in deep Sky CCD images (taken by L. Comolli) is shown in Fig. 11.7.
Validation of Lunar Flashes: A Network of Observers for Simultaneous Patrols This preliminary work suggested that the reality of the record could be ascertained by the criterion, consolidated by W. Haas, alone, i.e. the confirmation by a remote observer (separated by a minimum of 30 km or 18 mi) who has obtained an independent and contemporaneous identical record. It is clear that a videotape of a flash, which is restricted to a single event on a single frame, is insufficient evidence to prove that a lunar impact has occurred. The image of a star, and presumably that 168
Spurious Flash or True Impact Event?
Fig. 11.7. Spurious signals (arrowed) likely due to cosmic rays on deep sky CCDE images
of a lunar flash, recorded by a CCD placed at the focus of a telescope is all but distinguishable from the image of a spurious flash caused by: (1) cosmic ray, (2) local radioactivity, (3) CCD thermal noise (the video cameras are not cooled), (4) other CCD noise. This conclusion was reached by comparing star images (i.e. sigma Scorpii and 44 Ophiuchi) obtained with a telescope-CCD-recorder with the flashes recorded during the so called “dark tests,” i.e., videotape records obtained by capping the telescope (Figs. 11.8–11.10 below). There are several points to keep in mind: (a) A video event with an inaccurate, or no recorded time signal, is of little value for attempting to confirm lunar meteor events. Timing will be preserved using a receiver tuned to station HBG, or a WWV signal, or an accurate (to-the-second) time display on the video. 169
Spurious Flash or True Impact Event?
Fig. 11.8. Flash recorded during a ”dark test” and its intensity profile
170
(b) A recorder must check the positions of all geo-stationary satellites in order to assure none was close to the Moon as seen from his location at the time of the candidate event (Refer to the USNO website to obtain positions of the satellites during the observing sessions, at http://tycho.usno.navy.mil/) (c) Accurate locations of the flashes detected on the dark side of the Moon must be obtained. This can be done by measuring their Limb Angle (PA) and their distance (D) from the lunar center (expressed in lunar radius unit). The Limb Angle (PA) is measured in direction N→W, so the north region and the mean west region correspond at 0° and 90° respectively. (d) Look if a flash is present in a stationary position in every frame, which again rules out artificial satellite glints (these are always trailed and appear on many frames). (e) A recorder should also calculate the signal-to-noise (S/N) ratio for flashes detected. The S/N ratio is calculated by first measuring the distance between the peaks and troughs in the non-flash sections of the tape to obtain the mean noise level. The signal then is the height of the flash peak above this mean noise level. The S/N ratio is then calculated by dividing the measured signal by the measured noise. (f) Wait for an independent confirmation before you claim an impact. However, if you have confidently ruled out all spurious sources, you may conclude that the event is a likely impact in nature.
Spurious Flash or True Impact Event?
Fig. 11.9. The star sigma Scorpii and its intensity profile
Identification of the Flash Profiles Dark tests have shown that certain profiles are characteristic of cosmic rays (e.g. Lorenzo Comolli Hi-SIS CCD camera cooled at –40°C fitted to a 20 cm Schmidt Cassegraine). Figures 11.12 and 11.13 are the sum of 23 and 49 dark tests respectively (duration of 10 min). Flashes and segments connected in a line or having an “S or L” shape were detected. Most often, cosmic ray flashes appear round or nearly round on the screen, but occasionally a short streak will be picked up. Two flashes are often seen in cosmic ray flashes (Fig. 11.11). We have also seen them as triplets. This is because, at ground level, cosmic rays occur in little showers precipitated by a single upper atmosphere event. Although the term 'cosmic ray' is the generic terminology for these phenomena, in this context, it means any type of ionizing radiation, including ground sources (trace amounts of uranium and other unstable elements). 171
Spurious Flash or True Impact Event?
Fig. 11.10. The star 44 Ophiuchi and its intensity profile
Fig. 11.11. cosmic rays (Two flashes are often seen in cosmic ray)
172
Spurious Flash or True Impact Event? Fig. 11.12. Dark tests ( Lorenzo Comolli Hi-SIS CCD camera cooled at -40° C fitted to a 20 cm Schmidt Cassegraine)
173
Fig. 11.13. Cosmic rays
174
Spurious Flash or True Impact Event?
A decay of any of these nuclei can produce a “cosmic ray” indication (“flash” or “hit”) in a frame. A “cosmic ray” hit can come from any direction. They aren't usually seen on the same horizontal video line, unless by chance. Flashes of a type that usually appear on the same horizontal video line are instrumental noise that lasts longer than a cosmic ray hit.
Single Observer Validation of Lunar Meteor Impacts
1. Defocus the image slightly – any impact flash will be seen as a small blurred disk or ring, but cosmic rays will still appear sharp. If the CCD camera is slightly defocused, e.g., some tenths of a millimeter, both the star image and the lunar flash image, will be defocused. Each will appear as a doughnut several pixels wide if the telescope is a reflector with a secondary mirror, or a disk of comparable diameter if a refractor is employed. The defocusing will have no consequences on the images of spurious flashes, which will remain limited to a few pixels, thus allowing them to be distinguished from lunar flashes (Fig. 11.14). Defocusing will spread the flash’s (or the star’s) light on a larger CCD surface, and it will lower the content of each pixel involved in the image so that it will raise the CCD threshold limit. We find that, enlarging the doughnut image of a star up to 10 pixels of a Cookbook 245 camera fitted to a 305 mm diameter Newtonian
Spurious Flash or True Impact Event?
(For more detailed information on the following, refer to Refs. 26 and 27) A single observer may (tentatively) validate a flash as follows:
Fig. 11.14. Defocused star image with cosmic ray signature (legend: “raggio cosmico” means cosmic rays, “rumore” is noise and “stella sfuocata” is defocused star)
175
Fig. 11.15. Optical configuration to produce spectra of impact flashes
Spurious Flash or True Impact Event?
having 2,000 mm of focal length, requires an exposure 17 times longer than for that of a focused image. On this basis, we can foresee about a 3-magnitude loss on the limiting magnitude reached by the instrument, which will record only the brightest flashes. In the case of small meteors most of the flashes will probably be lost, but in the case of large meteors – e.g. the Leonids – most of the flashes will probably remain detectable. 2. Place a wedge shaped prism in front of the CCD – a true impact flash will show a spectrum, cosmic rays will not. If the optics depicted in Fig. 11.15 are placed between the telescope objective and the CCD, a spectrum of the flash light will be recorded. The negative, plane concave, lens acts as a Barlow lens and produces a parallel beam when its focus is coincident with the original focus of the telescope. The parallel beam traverses a dispersing element, a prism or a grating (the zigzag line), and then it is refocused by a positive, plane convex, lens. If the two lenses have the same focal lengths and are built of the same glass, they behave as a plane parallel plate, which will not introduce any noticeable aberration on the beam of an objective having F/D > 5. All of these optics are available from science dealers such as Edmund Scientific. The image of a lunar flash focused on a CCD fitted at the focus of the spectroscope will appear as a spectrum several pixels long. This length increases with increasing dispersing power of the prism or of the reticule employed. This spectrum may give some interesting information regarding the composition and the temperature of the light source, i.e. the material vaporized by the impact. 3. Hartmann Mask This method may give interesting information about a real lunar impact. Usually, the Hartmann mask is sold with 3 holes. You can buy a ready-made; metal Hartmann Mask called Kwick Focus. These are available for 8, 10, 11, 12, 14, and 16 in. scopes with prices ranging from $40 to $73. Pocono Mountain Optics sells these devices. You may build one yourself if you have some cardboard. Simply cut a circular mask the same size as your telescope outer diameter and make three (or two) circular holes (as shown in Fig. 11.16a, b) in it. The hole size is not critical. When the telescope is not in focus, each star in the field will appear three (or two) times, once for each hole in the mask. As the telescope is brought into focus, the images will move closer together and will finally merge into one image. By implication, then, a lunar flash will also appear three (or two) times 176
Spurious Flash or True Impact Event? Fig. 11.16. Hartmann Masks
when the CCD camera is slightly defocused. Using this method should avoid the need to have more than 2 observers; however, similar to defocusing the image, this method does considerably lower the detection sensitivity because the light grasp of the telescope is reduced. By using the Hartmann mask, the images of spurious flashes will remain as a single spot, limited to a few pixels, thus allowing them to be distinguished from a lunar flash which will appear as two or tree spots of light (see Figs. 11.17 and 11.18, next page). 177
Spurious Flash or True Impact Event?
Fig. 11.17. Image taken by Marco Fiaschi with a webcam fitted to a 41 cm (F/4.84) Newtonian telescope. The image (scale of 0.59” per pixel) was obtained using the star alpha Bootis.
Fig. 11.18. Image taken by Marco Fiaschi with a webcam fitted to a 41 cm (F/4.84) Newtonian telescope and a Hartmann mask with 2 holes (10 cm). Compare with Figure 11-17. The image (scale of 0.59” per pixel) was obtained using the star alpha Bootis.
Conclusion In conclusion, the GLR group has presented the results of their dark tests to characterize many of the manifestations of cosmic ray hits. Several techniques were also presented to assist the observer in telling the difference between cosmic ray signatures and true impact signatures. Although it is best for an impact candidate to be confirmed by a second, independent observer, the probability of an observed 178
Spurious Flash or True Impact Event?
impact candidate being a true impact event is greatly increased if spurious sources producing the candidate can confidently be ruled out. In addition to the techniques, a close examination of a putative impact event may indicate whether it is a true impact event or not. If the event lasts longer than two video frames, and it has a slightly “soft” appearance, it is likely an impact. Cosmic ray signatures tend to have a sharper profile, but these can also appear “softened” by bleed-overs of the affected pixel to adjacent pixels, as was shown earlier this chapter. Additional techniques and technologies to help the single observer obtain valid impact observations are introduced in the next chapter.
179
Professional and Amateur Collaboration
Professional and Amateur Collaboration
Chapter 12
Introduction: Pro-Am Collaborations Up to this point, for the most part, we looked at lunar meteor impact observing from an amateur astronomer’s perspective. We also surveyed the products of meteoroid impacts – craters – on the Moon and throughout the solar system. Background information of these, as well as observing techniques, has been presented. This chapter changes the focus a bit, brings in a little professional perspective, and discusses how amateurs and professionals can work together to make a significant contribution to lunar impact research. The value of professional astronomers working alongside amateur astronomers had begun to be appreciated by the end of the twentieth century. Amateur astronomers have acquired better, more sophisticated equipment, such as CCD cameras, low-resolution spectrographs, photoelectric photometers, and filters. This coupled with the extra time that amateurs have to do such work, make the observations they produce comparable to those of some professional astronomers. However, the amateur astronomers do not have the assistance of a technician if something goes wrong, nor do they have the funding to expand their resources; both these are advantages of the professional. In addition, professional astronomers are equipped with the experience (through formal academic training) to do high-quality science.. The ability of the professional to carry on extended observing programs is limited by lack of available observatory time (and having to compete with their colleagues and the weather for that time), bureaucracy, peer review, budget and funding problems, lack of facilities for long-term projects, academic obligations, among other factors. With these things in mind, it seems that a partnership between the two communities should naturally evolve out of both groups’ desires to see the science of astronomy advance to its fullest. Since the Leonid impact observations of 1999 and 2001, an array of new opportunities has opened up in observational and theoretical astronomy. These opportunities may help bridge the gap between the atmospheric sampling of the smallest meteoroids and the ground-based telescopic observations of the larger asteroidal objects. In addition, the Moon provides a laboratory for the study of impacts with collision velocities that cannot yet be attained on Earth, even in the best laboratories. These factors, along with the professional-amateur astronomer collaboration mentioned above and the development of automated detection software (eliminating the time-intensive activity of reviewing videotapes for 181
Professional and Amateur Collaboration
impact event signatures, freeing the observer to engage in other activities) promise to advance the study of lunar meteoritic phenomenon considerably. To further encourage the pro-am partnership, various astronomical organizations are working to make data collected by the amateur easily accessible to the professional. Examples of such organizations include the Association of Lunar and Planetary Observers (ALPO), the International Astronomical Union (IAU), the International Occultation Timing Association (IOTA), the American Association of Variable Star Observers (AAVSO), the International Meteor Organization (IMO), the Astronomical Society of the Pacific (ASP) and the American Astronomical Society (AAS).
Examples of Professional Research in Lunar Meteor Impacts Introduction What follows is but a few examples of how amateurs can help contribute to the science of Astronomy.28 An excellent partnership between professionals and amateurs occurred in August and September 2006 with the SMART-1 impact experiment. SMART-1 was a lunar orbiter launched by the European Space Agency which had completed a successful 3-year mission to test a European built ion-propulsion engine, as well as acquired thousands of high resolution pictures and mineral maps. Since the spacecraft was running low on fuel, it was decided to place the spacecraft on a trajectory that will produce a controlled impact event visible from the Earth. It was hoped that this impact event would provide much needed information about the physics behind high speed collisions between meteoroids and the Moon. While nearly all of the amateurs involved in the project were unable to observe anything definitive during the impact (and afterwards, in their searches for post-impact plumes and ejecta lofted into the sunlight by the impact), their observations provided an upper limit to the visible output of the impact flash. At the same time, observers at the Canada-France-Hawaii telescope in Hawaii, observing with a 3.6 m telescope and an IR camera, recorded a bright flash generated by the impact. This demonstrated that the impact of the spacecraft, traveling at 4,500 miles per hour at an approach angle of 1° to the horizontal, produced almost all the impact flash in the infrared. Another campaign that is ongoing and was began in October 2005 is being carried out by a team at the NASA Marshall Space Flight center with a CCD video camera and a 10-in. f/4 reflector has documented many impact flashes, each of which appears to be a genuine impact event, showing itself in multiple frames (information about these impacts is documented in the catalog in Chap. 4). The frequency of these flashes coupled with the type of instrumentation used demonstrates that impact flashes are reasonably accessible (that is, they occur regularly enough) to be worth pursuing on a regular basis. Dr. John Westfall of ALPO sees three opportunities for amateur astronomers to get involved with the LCROSS project (see Chap. 8 for more information on this mission). The first is public outreach, where amateurs located in the visibility zone of the impacts videotape and web-cast the events in real time for the public to 182
Professional and Amateur Collaboration
see. This would certainly stir up public interest and excitement about the mission and would certainly tie in nicely with the International Year of Astronomy 2009. The second way amateur astronomers can approach this is in the form of an observing challenge. This would be especially applicable for amateurs that are not motivated or equipped to make observations that are comparable to professionals, but perhaps after meeting the challenge, they may be motivated to become more involved in lunar meteoritic impact observing. This leads to the third avenue: amateur science. Amateurs can make valuable contributions by imaging the impact plumes. This imaging, if done properly and with complete documentation, may provide useful information on photometry and size and altitude measurements of the plume.
Meteoric Flux Estimates, Luminous Efficiency, Hypervelocity Impact Studies, and the Frequency of Lunar Impact Events Another area of research useful in the work of lunar meteor phenomena includes the flux (the number in a given amount of time) of meteoric impacts at the Moon, since knowledge of the impact flux determines the probability of whether an event will be seen in a given time or not. One way to get an estimate is observe the occurrence of terrestrial fireball events, which would likely be more frequent for the Earth than the Moon due to the Earth’s stronger gravitational influence. Impact flux at the Moon is given by the expression log N (E) = –0.99 log E + 11.38 m where N = number of impacts per year with energies from 3 × 1011 J to 1012 J, and m is the mass of the impactor. Another research group used the Apollo seismic data to derive an impact flux on the Moon The Moon was thought to be useful as an impact counter29 and a study of the relationship between crater diameter and frequency was done, and this led to an estimate of the present day flux of objects in Earth’s part of the solar system. This enables an estimate of the frequency of collisions of the Earth with objects of various sizes. A similar approach may be taken with the search for meteor impacts on the Moon, an extended search could yield important statistical properties of the near-Earth environment of space, bridging the gap between objects observed through telescopes and objects seen burning up in the Earth’s upper atmosphere. The main difference between objects impacting the Earth’s atmosphere and those impacting the Moon is that the latter makes a direct impact on the surface, without having its energy dissipated by friction with a layer of gas. In both cases, the energy dissipation is partially manifest as visible light, but while the former may last several seconds or more, the latter lasts only 1/30th to 1/20th (or slightly more) of a second. In both cases, one is concerned with the luminous efficiency of the impact; that is, how much of the impact energy is converted to visible light versus other forms. In the case of lunar meteoritic impacts, a fraction of the energy of collision goes into providing the visible light evidence of the impact flash (both the plasma flash not observable from Earth and the visible light emission from the thermal decay which is manifest as the optical flash to ground-based observers), another fraction goes into seismic energy, another fraction into heat energy, etc. The energy is released suddenly during the collision and re-emitted by the region of the Moon that is affected during a period of time that it cools, 183
Professional and Amateur Collaboration 184
depending on the energy absorbed. Radiation from an expanding fireball also contributes to the visible flash, but the duration of the explosion is so short (about ~10–3 ms), that it is unlikely that the actual observed flash (usually lasting about 30 ms) is a result of the expanding fireball. The cooling of the affected lunar surface most likely is what one sees during an impact flash event.30 The afterglow (and the fluctuations of the afterglow) of longer duration events (greater than 100 ms) is attributed to thermal radiation from a plume of hot droplets ejected from the lunar surface during the high velocity impact event. High-speed study of the lunar impact phenomena would provide valuable insight into the characteristics of impact dynamics and would place limits on the parameters used in modeling hypervelocity collisions in space. If a lunar meteoroid impact has enough brightness, low-resolution spectra of the impact flash could be obtained, providing compositional information of the lunar surface and of the impactor, along with insights into the physics of the collision. The actual visible flash part of the impact arises from optically thin line emission from a vapor plume31 that rapidly evolves, faster than standard video can capture the event. This is where a high-speed study becomes useful. High-resolution intensity versus time data would allow collision modelers to more accurately simulate the phenomena, which could provide information on many of the unknown parameters such as luminous efficiency (how much of the impact energy goes into making a visible flash). Luminous efficiency itself is not well defined, with32,33 and34 quoting values of 10–4 to10–5, whereas35 and36 estimate an order of magnitude higher for the Leonids in order to make these impacts more easily visible from the Earth. Alternatively, if observations of a sufficiently large number of impact flashes are made during a meteor shower whose components have a known velocity and mass distribution, perhaps with supplemental data from a global return of ground-based terrestrial meteor observations, it is possible to work back to an estimate of luminous efficiency. Over time, this can be done across various meteoroid speed ranges with multiple meteoroid streams. With improved parameters used in collision modeling and refining the number density of boulder sized objects, the resurfacing behavior of asteroids and other small bodies would also be better understood. Of interest to anyone involved in the observations of lunar meteor events is the probability of seeing an event at any one time, which is directly related to the flux of meteoroids. Several investigators have calculated the expected frequencies of observable impacts outside of shower events, but even so, the actual frequency is not well known. Based on the records of the seismographs that the Apollo missions left on the Moon, at a given location on the Moon, on average, 70–150 events per year in the 100 g to 1,000-kg range were recorded by the instrumentation.30 Based on this flux estimate, about one object greater than 1 kg in mass impacts the Earthfacing hemisphere of the Moon every 140 h, possibly producing an observable impact. It is not yet known what constitutes the smallest mass that can produce a visible flash, although this depends on factors such as size, velocity, and density. Additional analysis of the lunar seismic data resulted in sporadic meteoric fluxes of one impact every 35 and 7 h for objects over a kilogram.37,38 The Leonid meteors that impacted the Moon in 1999 were estimated to have a mass of around one kilogram,39 but their frequency of occurrence is still largely unknown due to the limited data set. The Moon, at the time of the observations, was passing through a very dense filament of cometary debris, which would have produced the equivalent of a ZHR of at least 50,000 or roughly 5,000 times the
Professional and Amateur Collaboration
average sporadic background levels. Based on this limited data set, it is estimated that about 715 h of observations would be needed before an impact could be detected during nonstorm periods.40 In contrast to this, other estimates place this number at a more optimistic 200 h.41,42 Estimates based on the 2001 observed Leonid lunar impact events place this number at roughly 50 h of observing time between impacts (based on the Leonid flux level being some 100 times above the average background meteoroid flux). Based on both 1999 and 2001 Leonid lunar events, it would seem to be a reasonable estimate that the rate of impacts observed that are comparable in brightness to the Leonid flashes, is one every 100 h for nonshower periods.7
Families of Meteoroids One of the outcomes of the Apollo Passive Seismic Experiment of the 1970s is the finding that many of the impacts observed seismically come in groups or clusters. These may be related to streams of meteoroids, like the annual meteor showers experienced by Earth-based observers. However, unlike the typical meteor shower stream, comprising mainly dust particles and objects the size of sand grains and pebbles, the seismically detected meteoroid streams may be composed of larger, denser bodies. Of the 1753 natural impact events recorded by the Apollo Experiment, from objects estimated to be 500 g to 500 kg in mass, recorded over the eight years of operation of the seismic program, 20% of the events occurred in clusters of 2, 3 or more events in close temporal proximity. Some 25 such clusters in the seismic data have been identified, and many of these have been associated with the known showers of the Leonids, Taurids, Geminids, and Perseids. Many of these clusters correlate well with radiants of known meteor showers. In addition, the large (meteoroids greater than 1 kg or 2.2 lb in mass) meteoroid impact distribution that was observed agrees well with the distribution expected from fireballs that occur from time to time at Earth. That is, the frequency of their occurrence on the Moon is similar to that experienced on the Earth. In terms of clustering and meteoroid size relationship, 28% of the “small” (that is, less than 1 kg) meteoroids that impacted formed clusters and are thought to be mostly cometary in nature and origin. Of the “large” meteoroids, 15% occurred in clusters, and are thought to come mostly from near-Earth asteroids and short period comets. It is interesting that two impact clusters that include high-density meteoroids took place in June 1975 and January 1977. Meteoroids greater than 5 kg in mass usually are near the plane of the ecliptic, have aphelia between 2 and 5 AU, and have a mass distribution that is described by the equation log n = B + G log m, where n = the number of meteoroids of mass m (grams) or more, B = 0.9, and G = –1.5. Amateur astronomers can contribute to this effort by monitoring the Moon closely at favorable times during enhanced activity of these meteoroid families.
The Moon’s Sodium Atmosphere The famous fireball Leonids display in 1998 also resulted in another interesting outcome: the Moon’s sodium tail was enhanced considerably. All sky imaging of the lunar sodium tail at new Moon revealed this enhancement that coincided 185
Professional and Amateur Collaboration
with the Leonid storm. In addition, enhancements were observed in January and March of 2000, both near New Moon. A number of factors contribute to the enhancement of the very rarefied cloud of sodium that surrounds the Moon, but most are dependent on the location of the sub-solar point, that is, the location on the Moon where the sun is directly overhead. One of the possible causes is meteoroid impacts. Apparently the Leonid fireballs of 1998, like the 1999 Leonids, struck the Moon in large numbers, but in the former case, the portion of the Moon being showered upon was on the side facing away from Earth-bound observers, so no impact events could be directly observed. The first process that contributes to the lunar sodium cloud is direct photon sputtering: the transfer of energy by incoming photons releasing sodium atoms from the lunar surface. The second is similar: ion sputtering involving charged solar wind particles (protons, electrons, ionized helium) impacting the lunar surface and releasing sodium atoms. The third process, similar to the two, is known as chemical sputtering: the liberation of sodium atoms from excess energy being deposited in the surface by chemical reaction with the solar wind particles that have settled on the surface without impact. The fourth and final nonimpact process is referred to as thermal desorption: the evaporation of material from the surface. These processes create a source of sodium that varies with solar zenith angle and reaches a maximum at the sub-solar point. The fifth process involves meteoric impacts: meteors impacting the lunar surface vaporize the upper layer of the lunar regolith, along with the meteoroid, and it does not depend on solar zenith angle. The relative importance of each of these five processes is uncertain, and so the relationship between impact events and sodium atmosphere changes would be extremely helpful in determining the role of this process versus the other four processes. Thus, careful monitoring as long as possible around new Moon of the sodium tail of the Moon, both inside and outside meteor shower activity and over extended period will add to the knowledge of the physical processes that affect the lunar regolith, and contribute to the sodium atmosphere. The process of monitoring the Moon’s sodium tail is a difficult one, but not impossible for amateurs to attempt. One needs to look at the time of new Moon in the direction of the nighttime sky directly opposite the Moon’s location (or where the full Moon would be located on that day). It is only during this time that the sodium tail is very faintly visible (its density just above the Moon’s surface is only a few atoms per cubic centimeter or less, compared with the molecular density of 1019 per cc in Earth’s lower atmosphere), under the most transparent and darkest of skies, and with the most sensitive detectors. This could possibly be done with a wide field, large aperture telescope equipped with a filter that can see the wavelengths of light emitted from sodium atoms. This would need to be done from a dark sky site, well away from the ubiquitous sodium lights found in towns and cities. Thus far, the only observed enhancement was noted on 17 November 1998 after the Leonid meteor shower peak (which featured a large number of fireballs).
Meteoroid Impacts on Other Worlds As has already been discussed in detail in Chap. 5 not only is the Moon and the planet Jupiter the target of objects from interplanetary space, but all the other worlds, plus the Sun, are “fair game” for meteoroids, asteroids, and comets. 186
Professional and Amateur Collaboration
The SOHO satellite has documented hundreds of “impacts” of comets into the Sun (in reality, most, if not all comets vaporize before reaching the photosphere or Sun’s visible disk). Recent papers published in professional journals describe the many possibilities concerning meteoroids impacting other worlds: the nature of meteors on Neptune’s satellite Triton, impacts on Mars, Titan, and Venus, and Saturn’s rings. Planets with atmospheres of any significance will cause any incoming meteoroid to become incandescent as it enters that atmosphere, planets and Moons without atmospheres result in the impactor colliding directly with the surface, resulting in pinpoint impact flashes such as what has been observed on the Moon. Close studies of the impact rates of meteoroids at various locations in the Solar System have the potential to add valuable knowledge about the distribution and inventory of small objects in these locations. Watching such impacts in a variety of settings provides a variety of laboratories, as it were, to study hypervelocity impacts under different conditions. Some of the professional literature even describes missions and instruments to observe such phenomena from space, such as the Smart Panoramic Optical Sensor Head (SPOSH).43 In addition to SPOSH, there are plans in the works for the development of systems and missions to detect meteor impacts (and other transient events such as lightning) in Earth’s atmosphere and on the dark sides of the Moon and the planets. Mars and Venus are especially valuable targets since both have atmospheres and Mars has many missions in queue in the coming years to study various aspects of the Red Planet. Some spacecraft are in the planning stages and include the studies of meteoric phenomena as listed in Table 12.1.44 Currently, the state of research on meteor impacts on the moon and other worlds is one in transition as the theory describing hypervelocity impacts begins to be tested by valid observational evidence. Three measurements are outlined for progress in the field of meteoroid-planet interactions for the near future.42 1. Existing instrumentation is used for other purposes, namely to look for evidence of meteoric impacts in the atmospheres and surfaces of other worlds. Current technology allows flexibility to be built into the software that runs the instrumentation on board spacecraft, making it possible to redirect these to look for evidence of meteoric interactions. Increasingly, predictions are being made for annual meteor showers in other worlds, and these can be used to target searches for evidence of meteor impacts. Finally, data from these measurements and observations will enable the development and design of next-generation meteor detection instrumentation. 2. Next-generation planetary instrumentation designed specifically to watch meteors on other worlds, whether the streak of an alien “shooting star” in an atmosphere or the flash of an impact on an airless surface. 3. Long-lived instruments to watch the effects of meteoroids over the course of a planetary year to validate predictions of annual meteor showers and characterize additional activity. Impact phenomena are the primary objectives to mission design and operations. Observing networks would operate over the long term which would be able to respond to meteoroid-related events on short notice. One key question concerning the feasibility of this work is whether it can be done with minimal cost.
187
188 Lander on Phobos equipped with seismic detectors that will be able to pick up the seismic signatures of meteoroid impacts on the Martian Moon Long distance Mars rover equipped with a MastCam, able to take up to 4 minutes of high definition video at 10 frames per second, perhaps motion triggered. A fixed station on Mars with seismometer capable of detecting meteoroid impacts. Also equipped with an atmosphere/electronic probe capable of detecting VLF/ELF signatures of meteor showers
03/06→03/08 05/08→10/08 08/10 09/10 2015 04/06→05/09 12/10→06/11 10/08→11/08 2019
Mars Reconnaissance Orbiter (NASA)
Phoenix Lander (NASA)
PHOBOS-GRUNT (RSA)
Mars Science Lab (NASA)
EXO-Mars (ESA)
Venus Express (ESA)
Planet C (JAXA)
Lunar Reconnaissance Orbiter (NASA)
BepColombo (ESA/JAXA)
The world of interest is highlighted in bold
Surface stereo imager on the lander at the Martian arctic, possibly able to detect trails left by dense, slow meteoroids
01/04→
Mars Exploration Rovers (NASA)
Two Lunar orbiters equipped with neutral and ionized particle analyzer. Two impactors will also be deployed to strike the lunar surface
This Lunar satellite will be re-imaging, at high resolution, areas covered by Apollo 15-17 hi-res. Imagery to detect craters that have been formed since 1972 to help determine current impact rate
Venus orbiter equipped with a lightning and airglow camera and a lightning detector at 50kHz, should be able to pick up large meteors
The Venus orbiter’s camera may be able to detect fireballs in the atmosphere; the radio science experiment may be able to detect enhancements caused by meteors in the meteoric ion layer.
Equipped with SHARAD radar capable of detecting individual meteor echoes
A possible atmospheric meteor was detected by the surface rover Spirit; the meteor may be associated with the comet 114P/Wiseman-Skiff
This orbiter is able to detect layers of meteoric metal ions, indirect evidence of meteor activity in the Martian atmosphere
Mars Express (ESA)
How it can be involved in meteoric research
Duration 12/03→05/09
Mission (agency)
Table 12.1. Current and near-term space missions with the potential to contribute to meteor research
Professional and Amateur Collaboration
Introduction Nowadays, the amateur astronomer has become well-equipped technologically speaking. High-Tech camera and telescope systems and limited spectroscopy capabilities, along with less constraint on time and budget (as compared with their professional counterparts) have positioned the amateur to make valuable contributions to the study of lunar meteor phenomena. More than ever, they are able to provide valuable support data to the projects described above. An excellent example of this is the collaboration between the Meteoroid Environment Office of NASA-MSFC and the ALPO Lunar Meteoritic Impact Search sections. The well-equipped amateur, working alongside professional astronomers, and the ability to coordinate efforts has the potential to answer some of the questions that were posed earlier in this book. We have already considered the mitigation techniques of spurious flashes, the use of the W87 and diffraction filters, and automated detection software. In addition to manned observations, robotic observations are on the rise, producing additional coverage without the human fatigue factor. Another idea to increase the efficiency of monitoring programs includes the use of a 2-pass filter, where a single instrument is used, but the light beam is passed into two detectors. An advantage of this setup is the elimination of cosmic ray events: cosmic ray strikes would not likely show up in both detectors at the exact same time in the exact same location on the image. Another advantage would be the use of two different filters for each detector, to enable a black-body temperature of the impact flash to be determined. Also, one detector could house an IR filter while another can host a spectrograph. A disadvantage is that the light intensity per detector is reduced, which may filter out faint impact events. This may be compensated for with the use of a large instrument, image intensifiers, and/or ultra-low-light video cameras.
Professional and Amateur Collaboration
Advances in Amateur Lunar Meteor Observations
New and Innovative Designs: The Tri-Splitter Camera One project that has been developed is the three-beam splitter camera. This innovative device is designed to allow the observation of the Moon with up to three cameras simultaneously, which provides a means to eliminate spurious cosmic ray signals masquerading as impact candidates. In addition, the nature of the setup would make it possible to put filters in front of each of up to three CCD chips in addition to a single filter upstream in the 1.25-in. C mount adapter. This makes it possible to do two-color photometry to derive some science from the observations. The following description is for the older two-beam model; a third beam now can be added for additional analysis. The design (Fig. 12.1) is for an f/3 system. It has provision for focusing one detector independently of the other and all of the components can be contained in one small unit. The two analog video streams would be digitized without compression by two video-to-FireWire converters and fed into one computer that has dual FireWire busses. For Lunar impact detection the computer would run some security 189
Professional and Amateur Collaboration
Fig. 12.1. The Bi-splitter configuration of the Bumgamer setup.
software that combines the video, motion detects, and then writes snapshots of any detected changes on one or the other video stream. For photometry almost any FireWire video capture would do – or perhaps two video capture cards could be used. Better still is to record the entire session to DVD-RAM, thus preserving all raw data. For analysis later, feed the recorded video back into whatever video cards and computer setup one uses. One can get a good recording Panasonic DVD VCR for under $200. Nowadays, the video stream can be digitized and sent to two computers (or one now and again later) equipped with LunarScan to look for impact flash candidates. Since faint impacts are expected to be more common than bright impacts, and are likely beyond the reach of most amateur instruments, it would be useful to secure time on a large telescope. Although surface area is generally sacrificed to light sensitivity, such a scope, equipped with an IR video camera, could shed some 190
Professional and Amateur Collaboration Fig. 12.2. Robotic telescopes at the University of Nottingham.
light on the frequency of smaller impacts and provide a more complete census of the size spectrum of a given meteoroid stream (in addition to regular monitoring and terrestrial meteor counts). An ideal instrument would be a deep survey type instrument with a very small (fast) f-ratio and large aperture to incorporate most of the unilluminated Moon. Infrared sensitivity would be a big plus for this system. Also, a high-speed, ultra-low light video camera, successfully capturing events, makes it possible to obtain light curves of these events. Typical high-speed video cameras can capture 1,000 frames per second, providing high time resolution of an event. A multispectral series of light curves in high resolution of the same event could reveal much about the physics of high-velocity collisions, a phenomenon that presently cannot be duplicated in ground-based laboratories.
Remote Control Impacts-Robotic Telescopes: The University of Nottingham Robotic Telescope Project (By Dr. Anthony Cook, University of Nottingham, School of Computer Science and IT) Two telescopes were available, a Celestron CPC11˝ GPS and a Meade 10˝ LX200GPS. These were sited on top of the library at the Jubilee Campus of Nottingham University (Fig. 12.2). The telescopes were operated from 2006 to 2007. These have now been relocated to the University of Aberystwyth and at the time of writing (December 2007) are being re-installed. Video was recorded to hard disk in raw mode – no compression. Observing on a given night was on an accumulated time 191
Professional and Amateur Collaboration
basis, namely the Earthshine was monitored for a few minutes, to keep file sizes small, followed by a 1–2 min gap for archival, and then the process would be repeated. Each scope had a PC next to it and the dome and the telescope were operated remotely from another PC at an operator’s home, several miles away. No timing information was provided from shortwave on the audio track, but instead relied upon file creation and modification time set by the GPS clock on the telescope. An Osprey-100 TV frame capture card was used on the PC in each dome. While at Nottingham the instruments were used in several modes of operation listed below: 1. Both scopes operating through focal reducers (Celestron = f/6.3 and Meade = f/3.3) in white light mode with Watec 902H and 902HS cameras respectively – these could capture stars down to just fainter than magnitude 10 (visible). Both cameras were even more sensitive in the near-IR. 2. The Meade scope (without focal reducer) was sometimes used with a 589 nm (FWHM = 10 nm) narrow band interference filter – this was to look for Sodium flares for impacts on the day side of the Moon. The Moon is known to accumulate a temporary Sodium atmosphere during major meteor showers. It is just possible that a concentrated release of Sodium ions might be visible in the vicinity of a meteoroid impact site, and perhaps there is a chance of recording this? 3. The Celestron scope was equipped with a low resolution transmission diffraction grating of 70 lines per mm. This was placed in between a Watec camera and focal reducer and used to monitor the Earthlit side of the Moon in an attempt to record point-like impact flash spectra. However, because the photons from a potential impact are now spread across many pixels, this lowers the sensitivity of the system to about magnitude 6. 4. The Celestron scope was operated in white light with an integrating CCD camera capable of exposing 0.5–2 s in Earthshine, without glare problems. This was used to capture animated images of Earthshine to look for signs of potential dust clouds/ejecta debris kicked up into sunlight near the terminator. Such an effect was seen in thermal IR from Hawaii during the SMART-1 spacecraft impact – it is doubtful whether the cloud of debris would be seen in the near IR, but it was thought to be no harm in trying as an experiment. 5. Although not part of the robotic telescopes, a thermal imaging camera was used on a few occasions during major meteor showers to look for thermal IR signatures from the heat of impact debris and subsequent crater temperature decay. A FLIR Systems Indigo Thermovision Micron/A10 thermal IR camera working in the range of 7.5–13.5 microns was used. Because the lens on the front could not be removed, a spare Germanium lens was used to form an eyepiece projection system on an 8″ f/5 Newtonian telescope. Image resolution was very poor, at worse than 20 s of arc, and so it is doubtful if the system could detect typical impact flashes. Nevertheless the system has functioned well during lunar eclipses in picking up thermal IR from craters such as Tycho that retain heat long after entering the Earth’s shadow. Software by an undergraduate student, Andrew Bullock, was written to check video sequences for flashes, however, much of the captured video data and images remain to be analyzed as this was not part of funded research at Nottingham. However, with the public release of the LunarScan software in early 2007, this becomes less of an issue.
192
A number of considerations go into determining what makes up an effective observing program. One component, and an important one at that, is the telescope. One has the tradeoff of mirror diameter versus field of view: it is best to encompass the entire dark portion of the Moon, yet have enough apertures to collect as much light as possible. To go really faint, one approach is to ask for telescope time at professional observatories. These sites stand a better chance of clear nights during the target time and there should not be too much competition from other astronomers since they generally do not prefer to observe with the Moon in the sky or shortly after sunset. Another approach involves the use of Focal Reducers which have proven quite useful in stellar occultation and lunar impact observations. As for spectroscopy, it is entirely within the realm of amateur astronomers to obtain spectra of a lunar impact. It is important, however, to go for a low-resolution spectrum since light gets dispersed over a number of pixels and intensity is lost. One possible setup involves a holographic diffraction grating placed over the CCD; this yields spectra from stars as faint as magnitude 6 with an 8 in. Newtonian (that’s with a 1/60th sec video frame exposure and a Watec 902HS CCD camera – an image intensifier would work just as well). The main problem is to find a calibration source in the sky (a nearby mercury vapor or sodium vapor lamp is adequate for the task). What is needed is an imaging spectrograph that is fast (both from the optical and time resolution ends), something like a grism in front of a CCD running in high time resolution (e.g., focus) mode. While this will not have too much spectral resolution (like ~50 Å), it could get decent S/N spectra of “flashes” of meteoric impacts. Conventional spectrographs using slits and reflection gratings probably will not work, but a project is under development at Rice University, in Houston, Texas, to make portable spectrographs available for telescopes in the 16–36 in. range. Moreover, photoelectric data on lunar impacts (with high speed low-light video cameras that are capable of obtaining temporal resolution of up to 10–5 s of accuracy) should give a unique light profile that may eliminate the need for cross detections: one detection could validate the observation. In addition, details may be captured in the light curve that would provide valuable information for hypervelocity impact modelers. This would shed new light on the physics of such collisions. In fact, the Moon is a large area collector for meteors and would very likely enable us to learn more about the physics of hypervelocity and massive impacts as well as the frequency of such events. The very beginning of the light curve could give to us an indication of the dimensions of the object, the spectroscopic data, the composition and consequently the mass, and the luminous efficiency. Attempts to document lunar meteor impact flashes by professional groups to record meteor impacts from other showers such as the Perseids had yielded nothing conclusive, either due to the absence of significant objects in the stream or the extremely low occurrence of such objects. These attempts had been isolated but not ongoing and systematic; campaigns of the latter nature stand to have some success in obtaining valuable information about lunar meteor impacts. In fact, collaboration between many different communities will help to enhance any program to characterize this phenomenon. Small professional observatories working with individual amateurs and advanced amateurs with observatories could produce a multiwavelength program to study the phenomena in greater physical detail. Observations at specific wavelengths in the Johnson visible (UBVR) and
Professional and Amateur Collaboration
Looking Ahead: Some Closing Thoughts
193
Professional and Amateur Collaboration 194
infrared (IJK, 5.0 and 10.0 microns) would provide temperature information of the observed impact events. High-speed, low-light cameras observing the events in multiple wavelengths would provide valuable multispectral light curve and thermal data. Analysis of the measured peak intensity could yield information about velocity and angle, target and projectile type. Complementing the light curve study would be low-light spectroscopes that would enable spectral data to be obtained to complement existing radiometric catalogues and provide additional information about the lunar subsurface and the impacting object. The existing data about the Moon’s surface can also be used to “subtract” the known lunar features from such spectra, leaving spectral information about the impacting object and /or the lunar sub-surface. Unambiguous detection of water in the polar regions of the Moon and the use of the Moon as an impact detector to determine modern impact rates are two areas where well equipped amateur astronomers can participate. The Lunar CRater Observing and Sensing Satellite (described in Chap. 8) will deliberately crash a projectile into the polar regions of the Moon to search for water sometime in 2009. The LCROSS spacecraft will follow 15 min later, its impact being observed by the Lunar Reconnaissance Orbiter (with which it will launch). Supporting observations by amateurs will be beneficial to this mission. Finally, the Moon provides a second chord to probe the structure of annual meteor structure (the Earth being the first chord). Serious studies of the LTP problem would benefit from close studies of impacts as one source of the phenomena (and the study may reveal additional sources). Seismic waves induced by meteor impacts aid in the study of the lunar interior and core and impact processes may make changes to the crustal magnetic field and magnetism. To include all of the above would be quite ambitious, but the information that it would yield, even of a single impact event, would be of immense value.
I would like to thank the following people for their contributions to this manual, without which this work would not be possible.
Acknowledgments
Acknowledgments
Walter Haas who pioneered this visionary area of lunar astronomy more than 60 years ago and wrote the foreword. l David Dunham, President of IOTA, for coordinating the lunar Leonid watch of 1999, which started this whole process and for providing the images and information used in Chaps. 1 and 4. l Bill Porter and Wesley Swift for proofreading the manuscript and offering lots of helpful suggestions. l John Watson, for helping to restructure the book into a more interesting and wide-reaching format. l David O. Darling, for allowing me to use his Lunar Transient Phenomena classification scheme. l Eric Douglass, for helping with Chap. 1 and writing the Impact Mechanics section in Chap. 7. l Rafaello Lena, the Geologic Lunar Researches group, and the American Lunar Society, for proofreading the entire manuscript and providing all of the material for Chap. 11. l Steve Preston for his contribution entitled “A Simple Method for Timing Videotaped Occultations (and Lunar Meteor Impact Flashes)” in Appendix D. l Roger Venable for his contribution “The use of Registax for Automated Lunar Meteor Detection” in Chap. 10. l Anthony Cook for providing material about the W87C filter and the transmission gratings (Chap. 9), as well as his many useful suggestions. l Peter Gural for his helpful suggestions and contributions to this work, including the introduction, the “Lunar Meteor Observing Plan” in Chap. 9, “Increasing the Probability of Detection with LunarScan” in Chap. 10, and the helpful tables and impact plots that appear in Chap. 9 and Appendix H. l Sandy Bumgarner for providing details of his innovative Bi-Splitter and TriSplitter camera systems. l The many members of
[email protected] who provided very helpful experience and advice, many of which was used in Chap. 9. I want to especially thank Derek C. Breit and Richard Nugent for their images and information. l The many observers who contributed observations for the Lunar Meteor Catalog in Chapter l
This work is truly the result of the efforts of many individuals and groups, I merely have the editorial role of assembling and organizing the material in a coherent manner to maximize its usefulness for as wide an audience as possible. 195
Appendix A: References 1. Spray JG (2005) Earth Impact Database, Planetary and Space Science Center, University of New Brunswick, http://www.unb.ca/passc/ImpactDatabase/CIDiameterSort.html 2. Grier JA, McEwen AS, Lucey PG et al (1999) The optical maturity of Ejecta from large rayed craters: Preliminary results and implications. In: Workshop on new views of the Moon II: Understanding the Moon through the integration of diverse datasets, Flagstaff, Arizona, 22–24 September 1999, abstract no. 8057 3. http://www.planetary.org/html/news/articlearchive/headlines/2001/1178noimpact. html, the Planetary Society. This quote originally appeared in the following reference: Hartung J (19756) Meteoritics 11:187–188 4. Hartung JB (1993) Giordano Bruno, the June 1975 meteoroid storm, Encke, and other Taurid complex objects. Icarus 104:280–290 5. Middlehurst BM, Burlee JM, Moore P, Welther BL (1968) Chronological catalog of lunar events. NASA Technical Report R-277 6. Cameron WS (2006) Lunar transient phenomena catalog extensions NSSDC/WDC-AR&S vol 78-03 7. Cudnik BM, Dunham DW, Palmer DM et al (2003) Ground-based observations of lunar meteoritic phenomena. Earth Moon Planets 93(3):145–161 8. Descriptions of craters on other worlds adapted from: Glass BP (1982) Introduction to planetary geology. Cambridge University Press, New York 9. Descriptions of craters on other worlds adapted from: Hamblin WK, Christiansen EH (1990) Exploring the planets. MacMillan, New York 10. Pesnell WD, Grabowsky JM, Weisman AL (2004) Watching meteors on triton. Icarus 169:482–491 11. Christou AA, Oberst J, Koschny D et al (2007) Comparative studies of meteoroid-planet interaction in the inner solar system. Planet Space Sci 55:2049–2062 12. Campbell-Brown MD (2004) Optical observation of meteors. Earth Moon Planets 95:521–531 13. Di Martino M, Carbognani A (2006) Detection of transient events on planetary bodies. Mem Societa Astronomica Italiana 9:176–179 14. Larson SL (2001) Determination of meteor showers on other planets using comet ephemerides. Astron J 121:1722–1729 15. Korycansky DG, Zahnle KJ (2004) Atmospheric impacts, fragmentation, and small craters on Venus. Icarus 169:287–299 16. Malin MC, Edgett KS, Posiolova SM et al (2006) Catalog of new impact sites on Mars formed May 1999–March 2006. Malin Space Science Systems, Inc., San Diego, California 17. Wood C (2008) Exploring the Moon-counting craters July 2008. Sky Telescope 69–71 18. Rukl A (1990) Atlas of the Moon. Kalmbach Publishing, Wisconsin 19. Baldwin RB (1985) Relative and absolute ages of individual craters and the rate of infalls on the Moon in the post Imbrium period. Icarus 61:81–91 20. As reported in Sky and Telescope magazine, September 1973, 146 21. The classification of “major” events was made by the Lunar Source Book – A User’s Guide to the Moon, 41, 46, and 47 22. Oberst J, Nakamura Y (1991) Icarus 91:315 23. Potter AE, Morgan TH (1988) Science 241:675 24. http://iota.jhuapl.edu/lunar_leonid/alpoproj.htm, from ALPO Monograph 7
Appendices
Appendices
197
Appendices 198
25. http://www.braeunig.us/space/lunar.htm, and The Illustrated Encyclopedia of Space Technology, Kenneth Gatland, Orion Books, 1989, both of which lists not only these but all spacecraft that have been sent, or attempted to be sent, to the Moon 26. Lena R, Favero G (2002) Validation of lunar flash observations. Selenology 21.1:3–4 27. Lena R, Favero G (2002) Ricerca di impatti lunari caratteristiche dei flash registrati da una telecamera e nuove metodologie di osservazione, Astronomia UAI, n.3 maggiogiugno, 25–28 28. Cudnik BM, Dunham DW, Palmer DM et al (2003) The observation and characterization of Lunar meteoroid impact phenomena. Earth Moon Planets 93(2):97–106 29. Öpik EJ (1960) The Lunar surface as an impact counter. Mon Not Roy Astron Soc 120:404–411 30. Latham G, Dorman J, Duennebier F et al (1973) Moonquakes, meteoroids, and the state of the lunar interior. Proc Fourth Lunar Sci Conf 3:2515–2527 31. Shuvalov et al (1999) Can we observe Frank’s comets impacting the Moon? 30th Vernadsky-Brown Minisymposium, Moscow, Russia, pp 103–104 32. Melosh HJ (1993) Remote visual detection of impacts on the lunar surface. Lunar and Planetary Science Conference XXIV, pp 975–976 33. Nemtchinov IV (1998) Lunar and planetary science XXIX, Houston, Texas, paper 1032 34. Crowe D, Franken P (1997) 26th Microsymposium on comparative planetology, Moscow, Russia 35. Bellot-Rubio LR et al (2000) Luminous efficiency in hypervelocity impacts from the 1999 Lunar Leonids. Astrophys J 542:L65–L68 36. Artemieva et al (2000) Lunar leonid meteors – numerical simulation, lunar and planetary science conference XXXI, Paper 1402 37. Duennebier et al (1975) Meteoroid flux from passive seismic experiment data. Lunar Planetary Sci VI:2417–2426 38. Oberst J, Nakamura Y (1989) Monte Carlo simulation of the diurnal variation in seismic detection rate of sporadic meteoroid impacts on the Moon. Lunar Planetary Sci XIX:615–625 39. Sigismondi C, Imponente G (2000) The observation of lunar impacts, Part II. J Int Meteor Organ 28(6):230–232 40. Sigismondi C, Imponente G (2000) The observation of lunar impacts. J Int Meteor Org 28(2/3):54–57 41. Ortiz JL, Sada PV, Bellot Rubio LR et al (2000) Optical detection of meteoroidal impacts on the Moon. Nature 405:921–923 42. Ceplecha Z (1994) Impacts of meteoroids larger than 1 meter into the Earth’s atmosphere. Astron Astrophys 286:967–970 43. Koschny D, Marino A, Oberst J (2006) A camera for observing meteors from space – The Smart Panoramic Camera Head (SPOSH). In: Bastiaens L, Verbert J, Wislez J-M, Verbeeck C, International Meteor Organisation (eds), Proceedings of the International Meteor Conference, Oostmalle, Belgium, 15–18 September, 2005, ISBN 2-87355-016-3, pp 99–104 44. Christou AA, Oberst J, Koschny D et al (2007). Comparative studies of meteoroid–planet interactions in the inner solar system. Planetary Space Sci 55:2052
Accretion disk
According to the most widely accepted theory of solar system formation, a cloud of gas and dust provided the raw material for the formation of the Sun and the planets. The cloud eventually collapsed into a disk surrounding the proto-sun; out of this disk formed the planets and material that make up the modern solar system.
Albedo
Appendices
Appendix B: Glossary
A measure of reflectivity of a surface where 1.00 means purely reflective (that is, 100% of the incident light is reflected) and 0.00 means perfectly dark (that is, all of the incident light is absorbed). The Moon is quite dark, with an average albedo of 0.07 (this ranges from 0.05 to 0.08 for the darker maria, and from 0.11 to 0.14 for the brighter highlands).
ALPO
Association of Lunar and Planetary Observers, a group that promotes collaboration between amateur and professional astronomers. ALPO collects observations of the planets and the Moon and makes them available for professional researchers to use.
ALSEP
Apollo Lunar Surface Experiment Package was the array of connected scientific instruments that were left on the lunar surface by the Apollo missions. Among the instruments was a seismometer, which measured the seismic vibrations caused by moonquakes and meteoroid impacts.
Annual meteor shower
A regularly occurring meteor shower that happens at (or nearly at) the same time each year. The Perseids in August and the Geminids in December are two of the best examples of annual meteor showers.
Aperture
The diameter of a telescope’s objective lens or primary mirror.
Apogee
The point in the Moon’s orbit that is farthest from Earth, which can be as great as 406,700 km (254,200 mi).
Apollo
The manned lunar program of the United States which included six missions that landed on the lunar surface between 1969 and 1972. These missions allowed 12 individuals to explore the lunar surface.
Asteroid
A rocky body found in the orbit around the Sun. Most asteroids are found between the orbits of Mars and Jupiter, but are found anywhere from inside the orbit of Mercury to beyond the orbit of Jupiter. 199
Appendices
Basalt
A dark, fine-grained volcanic rock which fills the lunar basins, mostly on the near side; this is the primary constituent of the lunar maria.
Basin
A very large circular impact structure, sometimes with many concentric rings of mountains that were thought to have formed by the impact of asteroids early in the geologic history of the Moon. Many of the basins on the Moon, mostly on the near side, are filled with basalts almost to the outer edges.
Capture hypothesis
The Moon origin theory which asserts that the Moon was once a separate planet that got too close to the Earth and was captured into orbit around our planet.
CCD video camera
An instrument, widely used in astronomy, that uses an electronic chip as its “film” and is capable of capturing low-light “movies” of astronomical events.
Central peak
An elevated structure, found at the center of an impact crater, usually formed by elastic rebound of the crust after impact.
Chondrite
A stony meteorite that has not been altered by melting or differentiation of its parent body.
Co-accretion hypothesis
Lunar formation theory that states the Moon was formed from an accretion disk or cloud of debris surrounding earth.
Collision hypothesis
The theory of the Moon’s origin, currently the most widely accepted, that postulates a grazing collision between the Earth and a Mars-sized object produced a cloud of debris which came together to form the Moon.
Cosmic rays
Energetic particles that impinge upon the Earth’s atmosphere. These may have their origins from distant galaxies and frequently show up on CCD images and videos as point-like spots of light.
Crater
A circular, bowl-shaped depression most of which is the product of an impact by a meteoroid or asteroid. The larger craters have all been produced by asteroid impacts, and a few of the smaller craters have volcanic origins.
Crescent Moon
The period between New Moon and First Quarter, when the lunar disk is less than 50% illuminated. 200
During the impact process, as the initial shock wave dissipates, the process may melt rocks and produce diaplectic glass, a product of the impact process.
Domes
A low, rounded mound-shaped elevation with shallow sides, thought to be formed by volcanic processes
Dark halo crater
Appendices
Diaplectic glass formation
A crater surrounded by a dark ring of material. The dark material, in some cases, is likely the ejecta produced by an impact dredging up darker material. In other cases, the feature is volcanic in origin.
Dark side
The hemisphere of the Moon not in sunlight. This also refers to the hemisphere that faces away from the Earth.
Dichotomy
The phase (first and last quarter) when the Moon is exactly 50% illuminated.
Earthshine
Reflected earthlight illuminating the dark side of a waxing or waning crescent Moon.
Eclipse
A phenomenon where the Moon passes partially or entirely into the Earth’s shadow. This also describes an event where the Moon appears to pass in front of the sun, partially or totally blocking its disk.
Ecliptic
The apparent path of the sun on the celestial sphere over the course of a year; this also is the plane of the Earth’s orbit as it is projected onto the celestial sphere. The ecliptic is tilted 23.5° with respect to the Earth’s equatorial plane, and the paths of the major planets and the Moon are found within a few degrees of the ecliptic.
Ejecta
The sheet of material thrown on the surrounding terrain which is produced by the impact of a meteoroid or asteroid. The blanket of this material can be brighter than the surroundings and can include systems of rays and secondary craters.
Elongation
The angular distance of an object from the sun as seen from the earth, measured between 0° and 180°, east or west.
Era of Heavy Bombardment
The first 600 million years of solar system history when large impacts were commonplace as developing planets and satellites swept up material from the accretion
201
Appendices
disk from which they arose. This is the era where the heavily cratered surfaces of these worlds originated.
Eta aquarid meteor shower
The annual meteor shower, derived from Comet Halley, which peaks in early May of each year.
Far side
The hemisphere of the Moon that always faces away from the Earth.
First quarter
Half-phase (between New and Full) that occurs one quarter of a lunar orbit after New Moon.
Fission hypothesis
The old (now dismissed) theory that states the Moon formed as a result of a rapidlyspinning Earth throwing off a chunk of material which became the Moon.
Focal reducer
A special lens used in telescope-video systems which works to reduce the overall focal length of the system. The purpose of this action is to widen the video field of view without losing light sensitivity.
Frame stacking
A technique in image processing used to bring out fainter details in an object. This is done by stacking two or more images of the object to increase the signal photons while keeping the noise at bay.
Full Moon
The point in its orbit when the Moon’s disk is completely illuminated by the sun, as seen from the Earth’s surface.
Geminid meteors
One of the best annual meteor shower for Northern hemisphere observers, which occurs each December and is derived from the asteroid 3200 Phaethon.
Gibbous
The phase of the Moon between half and full.
Hartmann mask
An observing aid which assists in focusing. This consists of an opaque over-theaperture screen with two or three circular openings.
Highlands
Heavily cratered, lighter toned areas on the Moon which are generally higher in elevation.
202
During an impact, this is the limit where rocks that are deformed are irreversibly changed. This is a form of shock metamorphism that arises from the impact of a meteoroid with a rocky surface.
Impact crater
A pit or depression in the Moon’s (or a planet’s or asteroid’s) crust resulting from an impact of a solid object at high velocity.
Appendices
Hugoniot elastic limit
Integration time
The amount of time a CCD chip is exposed to light; the longer the integration time, the more photons available to build up an image.
IOTA
International Occultation Timing Association
Kelvin
A unit increment of temperature whereby the zero point is set at Absolute Zero, the theoretical absence of all thermal energy.
KIWI OSD VTI
KIWI (the brand name) On Screen Display Video Time Inserter, one of a number of time inserter devices available on the market. The purpose of this device is to stamp the exact time of each video frame of a videotape to enable precise timing of an astronomical event.
Kuiper belt
The region beyond Neptune, between 30 and 55 AU distant, where there exists over 1,000 small, icy objects similar to Pluto. This is a region similar to the asteroid belt, but much wider and more massive, and its component objects are made of icy materials rather than rocky material as is the case for the asteroid belt.
Leonid meteors
The annual meteor shower associated with Comet Tempel-Tuttle, which occurs during the middle part of November. Normally a minor shower (ZHR of 10–20), the Leonds have been quite active during its most recent storm phase (1998–2002), producing more than 1,000 h at its maximum.
Low-resolution spectra
Spectra data that cover a broad range of wavelengths but are limited to how fine the detail that is discernable.
Lunar
Anything pertaining to the Moon.
203
Appendices
Lunar cataclysm hypothesis
The theory that states a spike in impact activity occurred toward the tail end of the Era of Heavy Bombardment some 3.9–4.0 billion years before present.
Lunar eclipse
The phenomenon where the Moon passes partially or totally into the earth’s shadow.
Lunar geology
The study of the ancient history, materials, and processes pertaining the formation and evolution of the Moon.
LunarScan
An automated transient-event detection program, written by Peter Gural, which enables a quick, positive identification of lunar meteoroid impact flashes.
Lunation
The length of time that it takes for the Moon to complete one cycle of phases, which corresponds to one orbit around the Earth as measured with respect to the Sun.
Maria
Large, dark volcanic plains on the Moon that are lightly cratered and occupy 17% of its surface area. Most of these occur on the nearside (Earth facing side) of the Moon.
Meteor shower
See Annual meteor shower
New Moon
The lunar phase where the entire Earth-facing hemisphere is not illuminated.
Occultation
The celestial phenomenon where a solar system object, such as a planet, asteroid, or the Moon, passes in front of a star, temporarily blocking its light.
Obscuration
A type of transient lunar phenomena where a small region of the Moon appears fuzzy or obscured by haze while the rest of the surface of the Moon appears clear.
OMAT
Optical MATurity, an index used by lunar and planetary geologists to gauge the age of an ejecta blanket and ray system.
Perigee
The point in the Moon’s orbit where the Moon is closest to the Earth, which can be as close as 356,400 km (221,400 mi).
204
The annual meteor shower which is derived from the periodic comet Swift-Tuttle which occurs each August. The meteors appear to radiate from the constellation Perseus, hence the name.
Piezoelectric effect
The production of electricity from certain materials in response to a mechanical stress. A variant of this effect applies to the Moon and involves the release of gas and an electric discharge as the gas expands; this is thought to be a possible cause of the 1953 Lunar flare observed and photographed by Stuart.
Appendices
Perseid meteors
Planetesimals
In the accretion theory of the formation of the solar system, it is thought that an accretion disk formed around the forming Sun. Within the disk is material that collected together to form planetesimals, asteroid-like objects that would later combine to form the planets.
Point flash
A zero-dimensional flash of light; all observed lunar meteor impacts took on the appearance of point flashes.
Protosun
The Sun in its pre-main-sequence stage, prior to the onset of thermonuclear reactions in the core.
Ray
A bright feature that emanates from many craters, best seen under full Moon illumination. This is part of the crater’s ejecta blanket system and its brightness and color are used to gauge the age of the feature.
Registax
A free software tool that registers, stacks, and processes sequences of images.
Regolith
The upper layer of the Moon’s surface consisting of bits of rock resulting from eons of meteoroid erosion activity.
Rille
A narrow valley on the Moon resulting from either faulting (linear or straight) or the collapse of underground lava tubes (sinuous or winding).
Satellite
A smaller object in orbit around a larger object.
Secondary cratering
Impact craters produced by the ejecta and debris thrown out by a larger impact.
205
Appendices
Seeing
A measure of the quality of an image produced by an astronomical telescope. “Good seeing” signifies a crisp, clear image whereas “bad seeing” describes an image that is fuzzy and shimmering. The steadiness of the atmosphere determines the quality of seeing.
Selene
Another name for Moon; the ancient Greek goddess for Moon.
Selenology
The study of the geology of the Moon.
Signal-to-noise ratio
The quantitative measure of intensity of a signal of interest expressed relative to the average level of background noise. Usually expressed as a ratio of signal level divided by noise level.
Spectral diffraction
An optical effect thought to be a minor cause of transient lunar phenomena (TLP) where localized changes in color appear to arise as a result of the scattering of sunlight.
Taurid complex object stream
A stream of objects whose radiant appears in the constellation of Taurus. Members of this stream include source objects of several notable events such as the Tunguska event of 1908, the 1178 Bruno crater impact event, and others, all of which are thought to have come from a common ancestor object.
Telescope-Schmidt cassegraine
A telescope that uses both lenses and mirrors in a configuration that features a folded optical path that allows a compact design. The system consists of a corrector plate lens, a spherical primary mirror (with a circular opening in its center to allow the light reflected back from the secondary to pass to the detector) and a convex secondary mirror. This system ends at a final focal plane behind the primary where an eyepiece or camera is placed.
Terminator
The boundary between day and night on the Moon or other planetary object.
Time insertion
The act of inserting time stamps on frames of a video stream as they are recorded; the time is usually derived from GPS or from an official source of time, such as radio station WWV or other such stations.
Transient lunar phenomena (TLP)
Lunar change: the observation of a sudden or gradual change on the Moon, which is mostly regional or localized in nature. 206
A type of contact electrification where certain types of material become electrically charged after coming in contact with other materials then separated. This has been used as an explanation of a cause of TLP where gas released from beneath the Moon’s surface becomes electrified after coming in contact with the solar wind (which itself is made up of charged particles).
Tunguska event
A huge explosion, probably caused by the disruption of a comet fragment or meteor as it entered the atmosphere, which occurred near the Tunguska River in present day Krasnoyark Krai in Siberia, Russia. The event occurred at 7:14 am local time on 30 June 1908, between 3 and 6 miles (5–10 km) in altitude.
Appendices
Triboelectric effect
Walled plain
Impact basins or large craters with flat floors, bordered by mountain ranges. This includes the prominent maria or seas that dominate the near side of the Moon.
Waning Moon
The period between Full Moon and New Moon where the illumination of the Earth-facing hemisphere of the Moon decreases.
Waxing Moon
The period between New Moon and Full Moon where the illumination of the Earth-facing hemisphere of the Moon increases.
WWV
The call sign of the National Institute of Standards and Technology (NIST of the United States of America), based in Fort Collins, Colorado. It continuously broadcasts official US Government time signals, with the time in Universal Time given every minute on the minute.
ZHR
Zenithal Hourly Rate
207
Appendices
Appendix C: Impact Candidates Observed by ALPO/LMIS Likely to be Cosmic Ray Hits or Other Spurious Phenomena Introduction What follows is a listing of impact candidates received in the early years of ALPO/ LMIS that are most likely cosmic ray hits or other spurious effects. I made this determination based on the confidence level of the observations, as well as the cumulative experience (see next paragraph) of all observers. I selected, for inclusion in the chapter itself, the impact candidates that were either confirmed, provincially confirmed, or best resembled true impact events, based on past confirmed events. It is possible that the candidates below are actual, but fainter impacts that are visible on a single video frame. Early on in the campaigns, many people observed and recorded many candidates which ended up being cosmic ray hits or other spurious signals. But as experience is gained by observers and they are able to tell what is a cosmic ray hit and what is a true impact hit, observers report less and less events. Unfortunately, as the years continued, involvement became less and less, with the greatest obstacle being the need to review hours of near-blank videotape to identify impact events. This resulted in the gap between events lasting from 2004 through 2007. With the advent of the sustained observing program by the Meteoroid Environment Office of the NASA Marshall Space Flight Center in November 2005, and the public release of LunarScan in early 2007, interest in lunar meteor impacts increased once again, resulting in more reports of impacts of higher quality and more genuine nature.
1999: Lunar Geminid Impact Candidates Many impact candidates were reported by many overzealous observers as is indicated by the number of events reported. This came on the heels of the excitement generated by the recently confirmed Leonid impact events. Many of the events below are visual events, possibly caused by electrochemical responses of the eye, the visual version of cosmic ray hits.
209
Appendices
Table B.1. Event information UT time and date
Confidence
Observer and comments
22:58:02, 11 Dec 00:47:48, 13 Dec 06:30, 13 Dec 02:54:40, 14 Dec 03:17:52, 14 Dec 14:10:46, 14 Dec 00:55, 15 Dec 01:07:56, 15 Dec 02:12:29, 15 Dec 02:14:51, 15 Dec 02:16:57, 15 Dec
3 3 3 3 3 3 3 3 3 3 3
Palmer Palmer English; low middle of dark part Hendrix; near the center of Mare Imbrium, mag ~4.5 Hendrix; approx. mag 5 Li Cudnik Cudnik, mag 7 Cudnik 50% confidence, mag 7 Cudnik 50% confidence, mag 7 Cudnik 50% confidence, mag 7
Table B.2. Observer information Observer
Location
Equipment
Steve Hendrix
Cameron, MO; Lat.: 39d 44¢ 40″; Long: 94d 14¢ 30″W
Dave English Brian Cudnik Binyang Li
Oceanside, CA Houston, TX Xinglong, Hebei, China; Long: 117d 34¢ 42″ E; Lat: 40d 23¢ 47″ N Greenbelt, MD
Unspecified telescope with short-wave receiver and tape recorder, visual Fixed Table top telescope, visual 14 inch f/11 Schmidt Cassegraine at ~110 power, Visual Binoculars, visual
David Palmer
C-5 and PC-23C video camera
Source: http://iota.jhuapl.edu/lunar_leonid/Gemtab2.htm
May 2000: Lunar Eta Aquarid Impact Candidates The eta Aquarid meteors were observed as they impacted the Earth and the Moon. As was the case with the Geminids 5 months earlier, many likely spurious events have been video recorded or observed visually. It is interesting to note the cluster events observed by two separate observers at two separate times.
Table B.3. Event information UT date and time
Confidence
Observer, location, comments
07 May, ~1:32UT 08 May, 2:42:00 08 May, 2:04–2:08 08 May, 2:22:22.0
3 2 2 2
08 May, 2:42:30
2
9 May, 05:04:26 ± 0:01
3
(NS) (location not noted) (BMC) north of Plato, on the north “shore” of Mare Frigoris (near W. Bond Crater) (6 events, BMC) throughout the Mare Vaporum and points South region (BMC) just southeast of Mare Nubium (about an arc minute southeast of its “shore” in lunar coordinates), 6th mag. (BMC) just west of the Theophilus crater, Southwest of Sinus Asperitany (lunar coordinates again), 7th mag. (5 events, FRA) Mare Imbrium near the Archimedes crater
Observer Abbreviation and Information: FRA = Frank and Ragini Anet, Valencia, CA; 8-in. Cassegraine equipped with Sony digital camcorder SF = Sam Falvo, Utica, NY, CCD video camera, 10-in. Dobsonian NS = Ned Smith, Trenton, GA; 10-in. Dobsonian equipped with video camera BMC = Brian Cudnik, Houston, TX; visual, 14-in. f/11 Cassegraine at ~110×
210
It was determined that Earth-based observers would have a second good chance of observing a lunar meteor impacts during the 2001 Leonid meteor shower, thus a global campaign was launched to have observers watch the waxing crescent Moon for impacts. Many reports were received from various observers around the world, with a handful of confirmations. The number of confirmed events this time was less than that of the Leonids 2 years earlier, but a number of good candidates were noted. The confirmed events as well as the excellent candidates are listed in Chap. 4. In addition to the confirmed and probable meteor impacts listed for the November 2001 Leonid Meteor event, several more good candidates have been reported from a team in India. Dr. T. Chandrasekhar leads the team, which had recorded a large number of candidates. Dr. Chandrasekhar’s team and the process by which their observations were made are listed and described in Chap. 4. Their best impact candidates are also reported in Chap. 4, with many additional candidates listed in Table B.4. It is likely that most of the events listed below are electronic noise in the digital detector or cosmic ray signatures (Table B.4), but it is conceivable that some of these may be genuine impact events, but none have been confirmed. In the early days of impact videotaping, a cosmic ray was regarded as a potential impact candidate and was usually catalogued, but toward the middle of the decade, as experience grew, such events were dismissed more readily (see Chap. 11 for more information about spurious events and how to mitigate them).
Appendices
The Leonid Meteor Storms of 2001
The Perseid Meteor Shower: 12–14 August 2002 The Moon was favorably placed over a 3-day period in August 2002 for observations of lunar Perseid impact events, so that another campaign was carried out to observe the Moon as much as possible during this time. As a result, a large number of candidates were recorded during this period of elevated meteoroid activity. The best events, most likely to be actual impact events are presented in Table B.4; the rest are listed in Table B.5 All reported events are listed in the following table, along with observer information and other information as is available. As was the case for the lunar Leonid candidates of November 2001, it is likely that most of the following were cosmic rays.
211
Appendices
Table B.4. Additional impact candidates reported by various observers around the world UT date and time, 2001
Observer
Comments
Nov 17, 12:50 Nov 17, 13:02 Nov 17, 13:13 Nov 17, 13:32 Nov 17, 13:33 Nov 17, 13:35 Nov 17, 13:46 Nov 17, 13:47 Nov 17, 13:53 Nov 17, 13:57 Nov 17, 13:57 Nov 17, 14:00 Nov 17, 14:06 Nov 17, 14:14 Nov 17, 14:15 Nov 17, 14:28 Nov 17, 14:28 Nov 17, 14:28 Nov 17, 14:33 Nov 17, 14:37 Nov 17, 14:49 Nov 17, 22:53:30 Nov 18, 00:56:11
Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Shah, Gogte, and Ganguly Tony Cook, Alexandria, VA, USA David Palmer, New Mexico, USA
Northwest quadrant (celestial coordinates) of the Moon, near Mare Crisium Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Northwest quadrant of the Moon Flash somewhere N Serenitatis Possible but faint: “It looks to be about the same South latitude as Mare Nectaris and the same meridian as Mare Serenitatis. Near the south-center of the Moon’s disk, and near the edge of the Leonid-exposed region.”
Nov 18, 01:18:47 Nov 18, 17:08:30 Nov 18, 17:09:20 Nov 18, 17:12:20 Nov 18, 18:09:50 Nov 18, 23:19:15.3 Nov 18, 23:25:20 Nov 18, 23:25:33 Nov. 18, 23:52:52 Nov 19, 00:18:47 00:39:58.6 Nov 19, 00:46:14 Nov 19, 00:48:13 Nov 19, 00:58:17.2 Nov 19, 01:22:37 Nov 19, 01:24:13
David Palmer Juan Carlos Echaniz Martin Stangl Graz, Austria Juan Carlos Echaniz Juan Carlos Echaniz Roger Venable Tony Cook Tony Cook Sam Falvo David Palmer Roger Venable David Palmer David Palmer David Palmer David Palmer David Palmer
Visual observations from Heider, Barcelona, Spain Southern “shore” of Mare Tranquilitatis Visual Visual at Lacus Gaudi (SW edge of Serenitatis) SW Mare Vaporum SW of Kepler “Flash recorded approx. 3¢ arc south from cusp, Along the terminator line” Faint, possibly electronic noise Location not determined
±1 s uncertainty
As mentioned in the text, the majority of these are likely electronic noise or cosmic ray interactions with the CCD detectors used a“Shah, Gogte, and Ganguly” represent the observers Kiran Shah, Pradyumna Gogte, and Surabhi Ganguly of Pune, India
212
UT date and time 2002
Observer
Comments
Aug 12, 01:19:25 Aug 12, 01:26:33 Aug 12, 01:38:28 Aug 12, 01:48:00 Aug 12, 01:53:50 Aug 12, 02:58:58
David Dunham David Dunham David Dunham David Dunham David Dunham Brian Cudnik
Aug 12, 19:27 Aug 13, 00:46:18 Aug 13, 00:53:53 Aug 13, 00:55:53 Aug 13, 01:01:26 Aug 13, 01:02:02 Aug 13, 01:03:31
James Cook Kevin Wigell Kevin Wigell Kevin Wigell Kevin Wigell Kevin Wigell Peter Gural
Video, these likely are mostly cosmic rays Video, these likely are mostly cosmic rays Video, these likely are mostly cosmic rays Video, these likely are mostly cosmic rays Video, these likely are mostly cosmic rays Visual, 10-in. f/6 at 86×, near the southernmost extreme of Mare Frigorus (just west of the Eudoxus crater, lunar coordinates) Writtle, Chelmsford, Essex, UK, 8-in. f/5 Dob, 100× (visual)
Appendices
Table B.5. Lunar Perseid candidates, which are likely cosmic ray hits
Two closely spaced flashes seen at this time
Each of Peter Gural’s Candidates were scanned with LunarScan on a tape produced by an observer named Ron with his video camera and Schmidt-Cassegraine telescope Bright
Aug. 13, 01:09:43 Kevin Wigell Aug 13, 01:10:33 Kevin Wigell Aug 13, 01:18:56 Kevin Wigell Aug 13, 01:24:12 Kevin Wigell Aug 13, 01:27:44 Kevin Wigell Aug 13, 01:27:55 Kevin Wigell Aug 13, 01:28:32 Kevin Wigell Two closely spaced flashes seen at this time Aug 13, 01:28:45 Kevin Wigell Two closely spaced flashes seen at this time Aug 13, 01:32:11 Kevin Wigell Aug 13, 01:31:19 Peter Gural Aug 13, 01:36:32 Peter Gural Four additional candidates reported by Peter Gural have the following disclaimer: “The following covered a single interleave line and was likely a segment of hot pixels but were brighter than the usual noise groupings I was seeing from the rest of the tape.” 13 August, 01:14:31, 01:18:21, 01:21:43, 01:43:04 Aug 13, 01:51:58.901 Roger Venable Eastern Mare Imbrium – (one video frame only) Aug 13, 02:09:46 Peter Gural Aug 13, 02:19:51 Peter Gural Aug 13, 02:26:10.934 Roger Venable NE Mare Imbrium – (one video frame only) Aug 13, 03:00:06 Brian Cudnik Visual, 10-in. f/6 at 86×, north edge (Polar region) of the Earthlit part of the Moon (lunar coordinates) Aug 13, 03:09:34 Jim Stoffaire Plato area (each of his events “should be taken with a grain of salt”) Aug 13, 03:56:55 Jim Stoffaire Area around Aristoteles?? Aug 13, 04:07:30 Jim Stoffaire Mare Sinus Roris Aug 13, 04:12:44 Jim Stoffaire Mare Imbrium Aug 13, 04:17:50 Robert Spellman (Possibly cosmic ray hits) Aug 13, 04:18:46 Robert Spellman (Possibly cosmic ray hits) Aug 13, 04:20:47 Jim Stoffaire Right on the limb. Near Crater Langley?? Aug 13, 04:35:49 Jim Stoffaire Area around Archimedes Aug 13, 04:38:53 Jim Stoffaire Area around Archimedes Aug 13, 04:40:00 Jim Stoffaire Area around Archimedes Aug 13, 04:43:09 Jim Stoffaire Mare Sinus Roris area Aug 14, 00:25:15 Phil Dombrowski “Be advised that I have taken one hour of video of the Moon’s northern dark region…from 00:08UT Aug 14, 00:59:50 Phil Dombrowski – 01:14UT using my 12″ Meade SCT and an AstroVid video camera. I have yet to examine the tape but had a visual hint of a flashes at 00:25:15s and 00:59:50s.” In conclusion, it seems that the vast majority of these events are cosmic ray or otherwise spurious events. There may be two or three probable events (more likely to be lunar in nature rather than cosmic ray) but it is difficult to tell this from what is available as of late. Further efforts to observe lunar meteors after the August 2002 Lunar Perseids are described in Chap. 4. Since then, more observer experience and greater care in reviewing and reporting impact events have resulted in most of the spurious cosmic ray events being weeded out
213
Appendices
Appendix D: A Simple Method for Timing Videotaped Occultations (and Lunar Meteor Impact Flashes) By Steve Preston
Introduction This document describes a simple method for achieving good accuracy (less than 0.1 s) in timing occultation videos. I have adopted this methodology as a replacement for the “stopwatch method.” To use this methodology you must have recorded WWV1 on the HiFi audio track of the videotape while recording the video of the occultation. For more information on recording WWV and videotaping occultations in general visit IOTA’s web site (http://www.lunar-occultations.com/iota/ asteroids/astrndx.htm).
System Requirements PC or Mac with a sound card and good audio recording software. I use a PC so my examples are based on the PC. For the PC, I highly recommend the software package Cool Edit 2000 from Syntrillium. The shareware version is free and works fine for the timing methodology described here even though you will not be able to save your recordings after 30 days. I decided to pay the $49 registration fee and my examples in this document are from Cool Edit 2000. HiFi VCR (VHS, SVHS) Camcorders are OK as long as they have HiFi audio (most do). I cannot resist the opportunity to add a plug here for Panasonic VCRs. If you are looking for a new VCR which you will be able to use for occultation timing, get a Panasonic – they are the only manufacturer whose VCRs will show both fields in a video frame during frame advance mode. (At least, the only one I know of until you move into the much more expensive professional broadcast equipment). Push button Audio/ Video selector switch (I use a small $20 version from Radio Shack, part no. 15-1952). Audio Cable to connect VCR audio output to your PC: 1/8² stereo PIN to dual RCA phono jacks. I am using one from Radio Shack, part no. 42-2483. Stereo Audio cable (dual phono to dual phono) of the type used to connect your VCR’s audio output to the TV’s audio input.
System Setup Basically, you connect the VCR’s video output to a TV (so you can watch the video) and the VCR’s audio output to the PC via the A/V selector switch. You route the VCR’s audio output through one input of the A/V selector switch and leave the other A/V input unused. In this manner, you can cutoff the audio from the VCR by selecting the unused A/V input. The PC is setup to record the output from the VCR (via the A/V selector). Here is a description of how I do this setup on my PC.
214
Video Connection Connect the VCR’s video output to the TV. There are two possibilities for most TVs: via a video input or via the antenna/cable input. You probably already know how to do this part of the setup, but I will write it down anyway. Video IN: If your TV is a recent model, it probably has a video input. In this case, connect the VCR’s Video OUT to the TV’s Video In. To view the VCR output, you set the TV to Video IN instead of using the channel Tuner. Antenna/Cable IN: If your TV does not have a video IN connector, you will need to connect the VCR’s Antenna/Cable OUT (coax) to the TV’s Antenna/Cable IN. In this case, you view the VCR output by setting the TV to channel 3 or 4 (depending on the setup of your VCR).
Appendices
Hook-Up
Audio Connection Connect the VCR’s Left and Right Audio Out to input A of your A/V selector switch using a standard stereo audio patch cable. Then connect the A/V switch output to the LINE IN of the PC’s audio card (the microphone input will probably work but the line input is better) using the 1/8² stereo PIN to stereo RCA phono cable. Do not forget to set the A/V switch to position “A” at this point.
Setup PC for Recording Next you will setup the PC for recording the occultation audio track. For most PC audio recording applications, this step is fairly simple as long as your remember to setup the Windows Mixer properly. The Windows mixer is sometimes listed as “Volume Control” in the Entertainment or Multimedia section of Programs®Accessories. Before trying to record, you should setup the recording section of the Windows mixer. I will give a brief description of the Windows mixer setup here.
Windows Mixer Settings Start the Windows Mixer/Volume Control via Programs ® Accessories ® Entertain ment®Volume Control. The mixer actually has separate controls for recording and playback, but it will show only the playback controls on startup. You should check the settings for both playback and recording. Since the starts up with the Playback controls, check to make sure the sections labeled Volume Control Balance (overall volume control), Wave Balance, and Line-In Balance are active (the Mute checkbox should be empty) and the volume levels are somewhere near the middle range (mid-way up). Next, you should setup the recording controls. To get to the recording volume controls, you must use the following menu item to change to the Recording controls view: Options→Properties. In the Properties dialog, you will see a section
215
Appendices
labeled “Adjust volume for” that has three options: Playback, Recording, and Other. Select Recording, then OK to see the Recording control. The first column of the recording control, Recording Balance, sets the overall recording volume and balance. Start by setting the overall recording volume to the middle position (mid-way up the slider). Next, notice that each of the other columns has a little checkbox in the lower left corner labeled “Select.” These “Select” boxes determine which inputs are fed into the audio card for recording. For the occultation recording, you should select only the Line-In Balance section. In all the other sections the checkbox should be empty (deselected). Set the Line-In Balance volume level to the middle level (mid-way up) for now. Now you are ready to setup your PC audio recording software (e.g., Cool Edit 2000).
PC Recording Software Settings Setting up the recording software varies from program to program, so I will just mention that you should try for 16-bit, 44.1 Khz, stereo PCM recording. In Cool Edit 2000, this dialog pops up when you select File→New. Find the “RECORD” button and read the next section to learn how to proceed from here.
Timing the Occultation To time an occultation, you record the VCR’s audio track (WWV) with the PC while watching the videotape. When the occultation event happens, you “blank” the audio by selecting the other input (e.g. input “B”) on the A/V switch. To determine the time of an event, you will review the audio recording on the PC and measure the time from a WWV minute mark to the position on the audio recording where the audio signal (WWV) stops (at the “button” press). Since you pressed the “B” button when you saw the occultation event, this process gives you the time of the occultation.
Position Videotape for Recording
Now review the videotape on the VCR and find the first audible WWV minute mark PRIOR to the occultation event you are timing. Make a note of the UT minute for this minute mark as announced by the voice prior to the start of the minute. Note: if you hear a female voice announcing the time, you are likely to be listening to WWVH out of Hawaii instead of WWV out of Fort Collins, CO (this will be important later). You might also find it useful at this stage to play the tape to the occultation event and note the approximate time of the occultation event. This will make it easier to “be ready” for the event when you do the actual “timing run.” Now position the videotape approximately 20 s before the UT minute mark prior to the occultation event.
Record Start recording audio on the PC (hit the “record” button in Cool Edit). Start the video (play) and watch the video on the TV for the occultation event.
216
You “mark” the occultation event on the audio track by switching to the “empty” A/V input when the event happens. If your VCR audio input is connected to input “A” and input “B” is not connected to anything, you will hit the “B” switch after you see the occultation event. Do not try to “guess” and hit the switch at the same time as the event, just try to react quickly and hit the switch as soon as possible after you see the event – this should provide more consistent results. The “B” switch is stop button on your “PC stopwatch.” After you hit the “B” switch to mark the event, you can stop recording with the PC (stop button on Cool Edit) and you can stop the VCR.
Appendices
“Mark” the Event
Calculating the Time for the Occultation Event Review the recording with Cool Edit 2000 to find the minute mark for WWV and note the time within the recording for the start of the tone that marks the beginning of a minute. Go to the end of the recording and you will see where the data ended as you switched to input “B.” Note this time – the time of the “button press.” Subtract the minute mark time from the button press time and you have the time from the WWV minute mark to the event. This gives you the time of at which the button press stopped the audio to the PC. Since the audio actually cutoff after you saw the occultation event on the video, you must now subtract some time to account for this time lag. This time lag between the event on the video screen and the audio cutoff is your “personal equation.” It varies from person to person and the variations can be several tenths of a second, so you really should “calibrate” yourself to determine how much time to subtract for your “personal equation.” Obtaining a good estimate of your personal equation is probably the most challenging part of this method. To do a good job of estimating your personal equation you really should do a couple of test runs with a videotape of an occultation event for which you already have an accurate time. You try the “PC stopwatch” method of timing several times and compare this result against the real time to determine your personal equation. My personal equation is about 0.2 s and my times varied by about 0.02 s over ten trials (since many people do not have a videotape of an event for which they know the accurate time, feel free to contact Steve Preston if you would like him to make a copy of his “calibrated” occultation event for you). Alternatively, you can test your personal equation against the WWV minute mark – listen to the tape and stop the recording when you hear a WWV minute mark. Your personal equation will vary from day to day and depending on various aspects of an event (disappearance, reappearance, contrast of star, etc.). So, even if you have been careful in determining your personal equation, I recommend that you account for these variations in two ways. First, report your timings as less accurate than your estimates show. I measured my time lag at 0.2 s, with a variation of plus or minus 0.02 s. However, for hand timings, I will only report my times as accurate to 0.1 s. Second, always do a couple of timings (trials) for each occultation event and watch for large variations. Of course, if you want to go one step further, you can check your personal equation against the “calibration” video before every session with the “PC Stopwatch.” Personally, I plan to trust the 0.2 s and increase
217
Appendices
my ±0.02 error estimate to 0.1 s to account for the daily variations. I will probably only recheck my personal equation once a year. OK, once you have determined your personal equation, you subtract the personal equation (e.g., 0.2 s) from the “button press” time. And, you are almost done. You have now determined what I call the “videotape” time for the event. However, to get the “real” time, you must make two more adjustments. You must account for a delay within the CCD video camera and account for the radio propagation delay of the WWV radio signal. To factor in the radio propagation delay of WWV, you must add time. Specifically, you must add 3.335 microseconds per kilometer distance from the WWV transmitter in Fort Collins, Colorado. Note: if you were actually recording WWVH out of Hawaii you must compute the distance from the transmitter location in Hawaii rather than Fort Collins. To factor in the internal delay of the CCD video camera, you must subtract the “personal equation” of the CCD camera – the time difference between the exposure of the CCD and the signal on the video output of the camera. I have measured this at approximately 17 ms for the PC-23C and the AstroVid 2000. If you do not have one of these, 16 ms is probably a safe bet – it will not be any larger. After including these two factors, you should have the real event time – finally!
Conclusion So far, this has been much easier than using a stopwatch to time events off the videotape in play mode. Currently, I only use this method for the dim occultations, which I cannot follow in field-by-field review on the videotape. For brighter events, I have used an LTC time code inserter to identify the video fields and apply a variation of the audio recording methodology of this document to establish the times for video fields. Using LTC time code and field-by-field review provides more accuracy and eliminates the observer’s personal equation. Recently, I purchased a GPS-based device to overlay time directly on the video. This is a more expensive solution (~$300) but much easier for reducing the data for bright stars.
218
Introduction: The following features most of the equipment that would be useful for meaningful lunar meteor work. Most of the material is courtesy of the IOTA Observers Manual, in press, with permission. Telescopes and dew/light shields: A flexible dew shield is preferred that extends well beyond the telescope tube. This doubles as a light shield from passing cars and streetlights. A battery powered hair dryer is recommended in case of dew formation. Orion sells both of these for Schmidt-Cassegraine telescopes.
Appendices
Appendix E: Equipment Checklist and Vendors
Meade Instruments®http://www.meade.com Celestron®http://www.celestron.com Orion Telescopes and Binoculars®http://www.telescope.com (P.O. Box 1815, Santa Cruz, CA 95061, 1-800-676-1343) Focal Reducers: T-C adapters: Check your local photography/telescope outlet The Meade f3.3 is for video use only. Most Meade telescope retailers carry this focal reducer, which runs $150-$175. http:///www.meade.com Celestron f6.3: Most Celestron telescope retailers carry this focal reducer, which typically costs about $130. http://www.celestron.com Brackets/Adapter: Scopetronix®http://www.scopetronix.com 1423 SE 10th Street, Unit 1A, Cape Coral, FL 33990, 239-945-6763 Image Intensifiers: Collins Electro Optics has come up with an easy to use, ready to use image intensifier, the I3 Piece for small telescopes. This product was reviewed in Sky and Telescope, February 1999, page 63. This device can be used as a visual eyepiece adding 2 magnitudes to the visible limit and adding 3–4 magnitudes to a video camera’s limit. Collins Electro-Optics®http://www.ceoptics.com (9025 Easy Kenyon Avenue, Denver, CO 80237, 303-889-5910) A word of caution here: one may want to contact someone who owns and uses an image intensifier, as they may be a bit noisy for use in lunar meteoritic work. Video Cameras: The PC-164C is the recommended low light black and white camera for faint-light astronomical applications. See also the PC-180XS, which uses the same 1/3˝ CCD chip as the PC-164C but is much smaller overall, allowing the video attachment to the telescope to easily swing through the fork arms of a Schmidt-Cassegrain. The PC-164 retails for about $120, the PC-180XS for about $90. Supercircuits (PC-23C, PC 164)®http://www.supercircuits.com (One Supercircuits Plaza, Liberty Hill, TX 78642, 1-800-335-9777) 219
Appendices
Watec 902H2: Several electronics and security company suppliers carry this camera, which retails for about $300. The Watec 902H2 has the same sensitivity as the Supercircuits PC-1t64C, but it uses a larger ½˝ CCD chip, providing a larger field of view, which may be beneficial specifically to lunar meteor work. Watec Cameras®http://www.wateccameras.com (60 Dutch Hill Road, Orangeburg, NY 10962, 888-567-4294) Red Flashlights: Orion offers a red LED flashlight (Cat no. 5755) and one with a dual red/white beam (Cat no. 5756). Orion Telescopes and Binoculars®http://www.telescope.com(89 Hanger Way, Watsonville, CA, 1-800-447-1001)Shortwave Radios: Grundig manufactures a high quality digital shortwave radios. They are sold at many electronics retailers in the USA including Radio Shack. An excellent choice for occultations is the Grundig digital tuning radio model YB300PE, its photo appears as in Fig. 6.6. Grundig digital shortwave radios: http://www.etoncorp.com, 1-800-872-2228 USA; 1-800-637-1648 Canada Radio Shack: http://www.radioshack.com Grundig SW radios at Radio Shack Catalog 20-233, retails for around $80 (digital tuning radio) 12-830, retails for around $30 (non digital tuning radio) Radio Shack DX model shortwave radios are no longer made; however, they can be bought used on internet sites such as Astromart.com, and ebay.com. Used Equipment: http://www.Astromart.com – Astromart has links to other telescope and equipment suppliers. http://www.ebay.com Voice Recorders: These can be found at most electronics stores such as Circuit City, Best Buy, Office Depot, OfficeMax, Wal-Mart, Fry’s, and Radio Shack. Brands to look for are Olympus, Sony, Panasonic, and Sharp. Olympus: http://www.OlympusAmerica.com Panasonic: http://www.Panasonic.com/consumer_electronics Sharp: http://www.SharpUSA.com Sony: http://www.Sony.com KIWI Precision Time Stamp Utility: KIWI is a “freeware” DOS program that uses a PC and GPS (with 1pps) to timestamp an event to millisecond accuracy to UTC. The program uses the GPS to first calibrate the PC timing and then every 5 s resynchs the PC timing to the GPS (to track thermal changes in the PC quartz crystal). This method enables any old PC to be used to timestamp an event over long periods (days or months even) and yet retain the 1 ms accuracy to UTC. The event to be time stamped can trigger KIWI via a logic signal or by a manually operated switch connected to the PC printer port. For astronomical events using a camcorder, a LED (optional) is flashed for 50 ms to identify the video at the time of the “trigger” for subsequent field/frame accuracy determination.
220
Appendices
The PC timing acts as a “flywheel” to the GPS 1 pulse per second drumbeat. Even if the GPS briefly loses synch with the satellites, the software will retain its accuracy. One of the functions of the program shows how the PC timing compares to the 1PPS from the GPS – 5 microseconds RMS for most PC’s common (http://www. geocities.com/kiwi_36_nz/kiwi_osd/kiwi_osd.htm). Checklist of minimum system requirements for a viable lunar meteor monitoring program: The following is a list of minimum requirements posted by the NASAMSFC group on their website and contains most of the equipment that is listed above. The example of their full setup was presented in Chap. 10 along with several other setups. 8˝ telescope, ~1.0 m effective focal length Equatorial mount or derotator tracking at lunar rate Astronomical video camera with adapter to fit telescope Digitizer – for digitizing video and creating a 720 × 480 .avi (could take the form of a digital video camera such as a Canon ZR 500) Software to segment .avi to files less than 1 GB (8,000 frames or approximately 150-s clips of video) Time encoder/signal: GPS timestamp or WWV audio PC compatible computer with ~80 GB free disk space LunarScan Software for detecting flashes (the freeware Registax has been used with limited success)
221
Note: Most of the material is courtesy of the IOTA Observers Manual, in press, with permission. Much of the information on short wave time signals has been obtained from the Minor Planet Center’s website and the following website: http://longwave.bei.t-online.de/. The accuracy of the data is not guaranteed. CHU, Ottawa, Canada Frequencies: Call sign: Location: Operating hours: Power: Modulation:
Appendices
Appendix F: Details of Shortwave Time Signals for Astronomical Timings
3,330, 7,335, 14,670 kHz CHU Ottawa, Canada, 45° 18¢ N, 75° 45¢ W Continuous 3 kW at 3,330 and 14,670 kHz, 10 kW at 7,335 kHz AM (USB only), tones and voice
The first minute of each hour commences with a 1 s pulse of a 1,000 Hz tone, followed by 9 s of silence, and then the normal pattern of 0.3 s tones of 1,000 Hz at 1-s intervals. The normal pattern for each of the next 59 min starts with a 0.5 s 1,000 Hz pulse. The pulse in second 29 is omitted. Following the normal pulse at 30 s, for 9 s period, 1,000 Hz pulses of 0.01 s occur, each followed by the CHU time code. The pulses between 40 and 50 s are of normal length. Identification Signal: Alternating French/English station identification in the last 10 s of each minute, followed by UTC time announcement, valid for the following minute. During the announcement period, the 1,000 Hz second pulses are shortened to “ticks.” Further information: NRC time services: http://www.nrc.ca/inms/time/whatime. html DCF77, Mainflingen, Germany Frequencies: Call sign: Location: Operating hours: Power: Modulation: Identification Signal:
77.5 kHz DCF77 Mainflingen, near Frankfurt, Germany, 50° 01¢ N, 9° 00¢ E continuous 50 kW. Estimated radiated power 25 kW Amplitude keying The call sign is transmitted twice in Morse code in minutes 19, 39, and 59 seconds 20–32 in AM; the amplitude is switched between 85% and 100% with a 250 Hz waveform. The signal may be omitted in the future
For Further information: Physikalisch-Technische Bundesanstalt, http://www.ptb.de JJY, Japan Frequencies: Call sign: Location: Operating hours: Power: Modulation:
40 kHz and 60 kHz JJY Mt. Otakadoya, Fukushima prefecture, 200 km (120 mi) N of Tokyo, 37° 22¢ N, 140° 51¢ W Continuous Radiated power > 10 kW Top of each minute has a 0.2 s tone. Seconds 9, 19, 29, 39, 49 and 59; Have 0.2 s tones
For further information: Communications Research Laboratory (CRL), Tokyo. Their website has a lot of information in Japanese only: http://jjy.crl.go.jp/
223
Appendices
LOL, Buenos Aires, Argentina Frequencies: Call sign: Location: Operating hours: Power: Modulation: Identification Signal:
5,000, 10,000, 15,000 kHz LOL Buenos Aires, Argentina, 15° 09¢ S, 50° 09¢ W 1100–1200, 1400–1500, 1700–1800, 2000–2100, 2300–2400 2 kW am 440 Hz and 1,000 Hz tones and voice. The beginning of each second is marked with a 5 ms long tick (5 periods of 1,000 Hz), except second 59 Call sign in Morse and announcement. Different minutes after the full hour have different transmission contents. Full details on website
Further information: Observatorio Naval Buenos Aires. http://www.hidro.gov.ar MSF, Rugby, United Kingdom Frequencies:Call sign: Location: Operating hours: Power:
60 kHzMSF Rugby, England, 52° 22¢ N, 01° 11″ E Continuous 15 kW
Further information: National Physics Laboratory Time and Frequency Service (http://www.npl.co.uk/npl/ctm) RWM, Moscow, Russia Frequencies: Call sign: Location: Operating hours: Power: Modulation: Identification Signal:
4,996, 9,996, 14,996 kHz RWM Moscow, 55° 48¢ N, 38° 18¢ E Continuous 5 kW at 4,996 and 9,996 kHz, 8 kW at 14,996 kHz On-off keying (A1B) Call sign in Morse in minutes 09 and 39 00m00s–07m55s, 30m00s–37m55s Unmodulated carrier 08m00s–09m00s, 38m00s–39m00s No transmission 09m00s–10m00s, 39m00s–40m00s Morse Code 10m00s–19m55s, 40m00s–49m55s 1 Hz pulses with UT1-UTC code. Pulse duration 100 ms, minute pulse 500 ms 20m00s–29m55s, 50m00s–59m55s 10 Hz pulses. Duration 20 ms. Second pulse = 40 ms, minute pulse = 500 ms
Further information: http://longwave.bei.t-online.de/
224
Frequencies: Call sign: Location: Operating hours: Power: Modulation: Identification Signal:
2,500, 5,000, 10,000, 15,000, 20,000 kHz WWV and WWVH Fort Collins, Colorado, 40° 41¢ N, 105° 02¢ W Kekaha, Hawaii, 21° 59¢ N, 159° 46¢ W Continuous Radiated power: 2.5 kW on 2.5 MHz (WWVH: 5 kW), 10 kW on 5/10/15 MHz 2.5 kW on 20 MHz am. Various tones and voice announcements. Top of each minute 1,000 kHz, 800 ms tone. Second pulses 01–28, 30–59 are 1,000 Hz, 5 ms duration. Second 20 pulse is omitted. Top of each hour tone: 1,500 Hz 800 ms duration. WWV – male voice announcement seconds 52–58 WWVH – female voice announcement seconds 45–52. During each hour various voice announcements are made concerning storm information, GPS reports, Geomagnetic alerts, Station ID. Announcement in minutes 00 and 30 (WWV), minutes 20 and 59 (WWVH)
Appendices
WWV, Fort Collins, USA and WWVH, Kekaha, Hawaii
Further information: National Institute of Standards and Technology, NIST, Time and Frequency Division, http://tf.nist.gov/ YVTO, Caracas, Venezuela Frequencies: Call sign: Location: Operating hours: Power: Modulation: Identification Signal:
5,000 kHz YVTO Caracas, Venezuela, 10° 30¢ N, 66° 56¢ W Continuous 1 kW am, tones and voice. Each seconds starts with a 1,000 Hz tone of 100 ms duration, except second 30, when the tone is omitted. A 800 Hz tone of 500 ms duration is emitted at the beginning of a minute. Time announcement in Spanish in seconds 52–57. Announcement in seconds 41…50: “Observatorio Naval Cagigal – Caracas, Venezuela”
Further information: Observatorio Naval Cagigal. http://www.dhn.mil.ve/
225
Finding Limiting Magnitudes for Visual and Video Camera Observation Most of the following material is courtesy of the IOTA Observers Manual, used with permission. To best determine the magnitude of an event that was recorded, it is important to know what the limiting magnitude of the system one is using, as well as how bright stars of a range of magnitudes appear in the same system. This is true whether the system is a video or a visual system and it can vary from night to night or within a given night. One suggestion is to use the charts published in the RASC Observer’s Handbook, as a standard reference to determine these limits. These charts are copyrighted material and are not being published on line for those who do not have or use the RASC Observer’s Handbook. However, Guy Nason of Toronto, Canada contacted the necessary individuals to obtain permission to place these charts on the IOTA website. The links are given below and these enable the user to obtain these charts, but before accessing these charts, please read the following email correspondence (on the next page) http://lunar-occultations.com/iota/videolimits.htm The chart and information can also be found in the Observer’s Handbook, published annually. In the chapter entitled “Optics and Observing” there are many useful pieces of information for general observing programs. In the 2007 edition, the limiting magnitude information can be found on pp. 62 and 63. There are two figures, the first being in the region of Polaris showing stars of magnitudes 2.0–7.4. The second is a pair of maps, a “Left-Right Correct” view and a “Mirror Reversed” view. The stars presented in these figures are within an 8¢-wide segment in the northwest quadrant of M67 and range in magnitude from 10.60 to 21.03. Alternatively, one can go to the AAVSO to obtain charts of variable stars that meet the needs of one’s desire to find the limiting magnitude of one’s system. The advantages of the AAVSO approach is that (1) the charts are freely available to anyone with an Internet connection, (2) they cover a large number of fields throughout the entire sky so that one can find a chart for one’s season. However, in many cases, the magnitudes of comparison stars are not convenient for use with video cameras, and the stars are more widely scattered than in the case of the open cluster. One recommendation around these obstacles is to select charts that are used by beginners and/or binocular users. In so doing, one has a wide range of brighter comparisons that one can readily use, and one would only need to go a short distance from one star to another. The website to access the charts is http://www.aavso.org. Note that if one is unable to disable the auto gain control of one’s video camera, then it does one very little good to make these observations. One would simply need to be content with a sighting that would locate the position of the impact candidate and its brightness relative to the surrounding earthshine-lit lunar surface, if visible. It would be possible to obtain a magnitude value by making this comparison, if one knew in advance the mag/arcsecond of the earthshine at a given time. At the very least, though, one could report the peak flash magnitude in terms of mean earthshine intensity; or one can provide a light curve in terms of CCD counts as a function of time. Guy Nason’s original suggestion on using the RASC Observer’s Handbook follows below.
Appendices
Appendix G: Stellar Resources for Comparison and Calibration
227
Appendices
The Observer’s Handbook 2003, published by the Royal Astronomical Society of Canada, has a very good article and chart on pages 58 and 59. It includes charts of the Northwest quadrant of M67 that identify the visual magnitudes of >30 component stars ranging from 10.60 all the way down to 21.03 (!). The spring would be a good time to test your system using this technique, since M67 culminates at approximately midnight to early evening local time when its altitude is around 60° for mid-northern observers. This is high enough in the sky to remove the effects of atmospheric extinction. The next time you are out videotaping the Moon and/or an occultation event, and if the object is above the horizon, please take a few minutes afterwards to train your scopes on M67 and record a minute or so of the NW corner of M67 (centre on R.A. 8h51.1m; Dec. +11d53¢). Then compare your recording to the RASC charts in comfort later. If you do not want to hang around for several hours waiting for M67 to culminate, try it anyway. There is a bit on Page 59 (Page 62 of the 2007 edition) that describes how to compensate for atmospheric extinction encountered at various altitudes. Guy's request for publication to the IOTA website granted with conditions: From: “Guy Nason” Subject: Fw: A request wrt the Handbook Date: Thu, 17 Apr 2003 21:24:06 – 0400 Here is Rajiv Gupta’s reply to my request to post pp58 and 59 of the RASC Observer's Handbook to the website and/or Yahoo list. As you will see, there are four conditions that must be met, none of which I see as a problem. In fact, no. 4 (permission of the originator of the item, Doug Pitcairn) has already been met in a separate e-mail that was forwarded to me today. It appears below as “Second forwarded message,” for the record. – Guy From: “Rajiv Gupta” To: “Guy Nason” Cc: “Douglas Pitcairn” Sent: Thursday, April 17, 2003 1:53 am Subject: RE: A request wrt the Handbook We do occasionally grant complimentary permission for noncommercial reproduction of Handbook material, and your request is a valid one. I will grant your request on the following conditions: 1. You scan the entire page, including the footer that indicates the source of the material 2. When you post the scans, you announce the source and also give the URL http:// www.rasc.ca/publications.htm as a place where more information on the Handbook can be found 3. You indicate that the material is reproduced with permission of the editor 4. The author of this section, Doug Pitcairn, cc'd here, also agrees to the request I hope there will not be a problem meeting these conditions, and that the scans prove useful to IOTA members. – Rajiv
228
The following charts are reproduced by permission of the editor(s), of the RASC Observer’s Handbook. More information on the handbook can be found at: http:// www.rasc.ca/publications.htm Those interested in such information are encouraged to go to the above website to obtain copies of the charts. Alternatively, one can obtain a copy of the RASC observer’s handbook, which has been including the charts and information each year.
Appendices
From: Douglas Pitcairn Sent: April 17, 2003 6:47 am To: Rajiv Gupta Subject: RE: A request wrt the Handbook OK by me.....the more that use it, the better....Doug
Standard Stars for Extinction Correction and Flux Calibration To get an absolute value of the amount of energy released from an impact on the Moon, it is essential to observe calibration standard stars. In a nutshell, here is how this works. If an observer makes an observation of a star at a certain elevation, the amount of atmosphere the starlight has to pass through to get to the groundbased detector is what is known as airmass. By definition, airmass has a value of 1 at the zenith and about 38 at the horizon. This value increases approximately as the secant (or 1/cosine) of the angle to the zenith. Ideally one wants to get the extinction profile of the atmosphere the night that observations were taken, so as to provide the most accurate correction possible. The aim is to correct the observation so that it would be the same as if the observation were taken above the atmosphere. The extinction profile can vary from night to night and even within a single night of observing. Flux calibration takes stars that are known as standards and uses their flux to calibrate the flux of an unknown object. Once the flash observations are corrected for atmospheric extinction, then the flash image is measured. After this, the standard stars, which serve as calibration targets, are corrected for atmospheric extinction, and then measured as well. The flux, or magnitude of the impact flash is divided by the flux of the standard star magnitude, then the resulting value is multiplied by the flux of the standard star (in watts/cm2/sec). The vast majority of observers of lunar impact phenomena, apart from the serious professionals, will likely not be concerned with the flux of the impact, let alone correcting the observation for airmass, etc. However, if one is interested in and equipped to delve deeper into this, then a number of websites exist that provide excellent tutorials on the subject. One example can be found in the first link, which provides a useful background to astronomical calibration procedures. The following is a short list of photometric standard stars. Two web links, currently as of September 2008, provide resources on how to make flux corrections using standard stars, and many lists of thousands of standards are available through the second link.
229
Appendices
STAR
RA
DEC
MAG
SPECT
19 δ Sag (Kaus Meridianalis) 27 ϕ Sag 37 ξ⃞2 Sag 38 ζ Sag (Ascella) a Sct γ Sct 35 η Oph (Sabik) 64 γ gamma 5 a1 Cap (Al Giedi) 6 a2 Cap (Al Giedi) 23 θ Cap 18 ϖ Cap 96 Her 16 Cyg A 16 Cyg B 51 Peg
18h20m59.5s 18h45m39.2s 18h57m43.6s 19h02m36.5s 18h35m12.1s 18h29m11.7s 17h10'22.5" 17h59m1.4s 20h17m39s 20h18m03s 21h05m56.6s 20h51m49.1s 18h02m23.0s 19h41m48.9s 19h41m52.0s 22h57m28.0s
–29d49¢42″ –26d59¢27″ –21d06¢21″ –29d52¢49″ –08d14¢39″ –14d33¢57″ –15d43¢30″ –09d46¢25″ –12d30¢30″ –12d32¢42″ –17d13¢58″ –26d55¢09″ +20d50¢01″ +50d31¢31″ +50d31¢03″ +20d46¢07″
2.70 3.17 3.51 2.59 3.85 4.70 2.43 3.34 4.24 3.57 4.07 4.11 5.27 5.96 6.20 5.47
K2 B8 K1 A2 K3 A2 A2 K0 G3 G9 A0 K5 B3 G1.5 G2.5 G2.5
http://spiff.rit.edu/richmond/snap/snap.html is the web link with the how-to guides on flux correcting astronomical data. The link http://sofa.astro.utoledo.edu/SOFA/domains.html contains many lists of thousands of standard stars for a large variety of astronomical applications.
230
Appendices
Appendix H: Impact Plots
231
232
CAP
ETA
QUA
3 Days Past New Moon
Appendix H (continued) 10 Days Past New Moon
19 Days Past New Moon
26 Days Past New Moon
Appendices
233
LEO
STA
PER
3 Days Past New Moon
10 Days Past New Moon
19 Days Past New Moon
continued
Appendices
26 Days Past New Moon
234 10 Days Past New Moon
19 Days Past New Moon
26 Days Past New Moon
These impact plots, compiled by Mr. Peter Gural, are intended to be used for general reference in conjunction with the material given toward the end of Chap. 9 “Observing Impacts as they Happen: Guidelines for Observations of Lunar Meteoritic Phenomena.” For the six annual meteor showers selected, the geometric configuration between the Moon as is illuminated by the sun as seen from Earth, and the area covered by the impacting meteoroid stream. The “+” indicates the location on the Moon where the shower’s radiant is directly overhead as viewed by an astronaut at that location
GEM
3 Days Past New Moon
Appendix H (continued)
Appendices
Index
B Basalts, 8, 24, 105, 200 Beer, 43, 109 Bessel, 23 Blackbody (radiation), 18, 30, 113, 117, 146, 184 Bolides, 33, 116–118 Bombardment, 4, 5, 17, 20, 30, 79, 86, 201, 204 Boussingault, 109 Bowl-shaped (craters), 8, 21–23, 85, 94, 104, 105, 200 Bradford, 42 Brayley, 66, 109 Bullialdus, 65, 108 Byrgius, 44, 67 C Callisto, 7, 79–82 Caloris, 76 Camcorder clock, 150 recording with, 135, 136 setup component, 136, 138 verified with, 56 Campaign Apollo lunar surface experiments package, 29–30
earthshine watch, 27, 126 to learn about LTP, xiv lunar meteor watch, 126 NASA/Marshall Space Flight Center, 120, 182 Cassegrain, 51, 52, 57, 60, 62, 64, 131, 136, 144, 154, 162, 171, 206, 210, 213, 219 Celestron, 69, 132, 138, 191, 192, 219 Chondrite, 30, 118, 200 Circuitry, 135, 143 Clark, D., 69 Clausius, 68 Cleomedes, 107 Clive, B., 48 Clouds, 3–5, 27, 36, 37, 39–42, 54, 60, 72, 91, 114–116, 118–120, 122, 124, 130, 146, 186, 192, 199, 200 Cognitum, 65, 67 Comets, 3, 16, 27, 29, 43, 71, 72, 74, 77, 79, 91–95, 107, 188, 202, 203, 205, 207 Confirmation, 211 Copernicus feature, 19, 101–102 location of LTP, 111 physical data, 19, 46 source of ray system, 100 Crater Barringer Meteor, 28 counting, 18–20, 102–104 features, 19, 101, 102 LCROSS, 124, 194 LTP, 111 morphologies, 10, 23, 71, 74–92 Crescent earthlit portion LTP, 41 glare from, 144–146, 148, 150, 192 lunar meteor report, 45–46, 53 Mercury and Venus, 93 Crisium (Mare) how to observe, 44, 58, 102 LTP location, 111 lunar meteor impact location, 44 Cyrillus, 107 D Darling, D., 43, 44, 47, 111 Depth-to-diameter ratio of craters, 76 Detector digital, 56, 211 seismic, 118, 188 video cameras as, 135, 189 Differentiation, 5, 7, 200
Diffraction, 46, 113, 145–147, 189, 192, 193, 206 Domes, 110, 111, 192, 201 Dubyago, 66, 69 DV-AVI, 138, 148 DVD, 138, 190 E Earth-based (observing) attempts, 93 observers/observations, 53, 123, 128, 185, 211 telescopes/observatories, 15, 118–120, 122 Eclipses ALPO, v, 44 lunar, 44, 47, 112, 150, 161, 192, 201 meteor impact, 112, 150 Endymion, 41, 108 Eratosthenes, 41, 108 Exposure, 18, 33, 59, 64, 103, 135, 136, 144, 176, 193, 218 Extinction, 146, 147, 228–230
Index
A Agrippa, 39 Albedo, 75, 82, 111, 112, 144, 156–158, 199 American Lunar Society (ALS), 116, 122, 123, 165 Apogee, 113, 199 Aquarids, 49–51, 150, 151 Ariel, 86 Aristillus, 19, 108 Asteroids bombardment of, 7, 17 component of Taurid complex, 30 component of the solar system, 3 solar system formation, 3, 17, 95 sources of meteoroids, 7, 17 surface features, 29, 105 Astrostack, 147–148 Atmosphere extinction, 146, 147, 228, 229 of Jupiter, 74 lack of, 23 meteor, 17, 29, 33, 91, 92, 120, 122, 126, 187, 207 seeing, turbulence, 100, 142
F Farmington, 30 Fecunditatis (Mare), 66 Filters, 145–148 Fireballs, 15, 29, 32, 72, 74, 119–120, 183–185, 187 Firewire, 148, 189–190 F-ratio, 138, 191 Frequency, 7, 15, 27, 46, 81, 118, 123, 161, 182–185, 191, 193, 224, 225 Frigoris, 60, 210 G Garmin, 133, 134 Geminids, 49–52, 57, 63–64, 66, 185, 199, 202, 209–210 Geologic evolution, 118 laws, 19, 103 planetary, 74 principles, 19, 102 theory of solar system history, 74 Geo-stationary, 170 Gilgamesh, 82 GLR group, 60, 122, 178 Goldschmidt, 109 Goldstein, 50 Grating, 46, 143, 145–148, 176, 192, 193
235
Gravitational, 4, 8, 23, 29, 71, 74–76, 124, 183 Gravity formation of solar system, 4–5, 71 impacts, acceleration of impactor, 4, 74, 89 influence in crater formation and structure, 10, 75, 85 Grimaldi, 38, 39, 41, 43, 46, 67, 69, 102, 108 Guidelines, 19, 102, 103, 127, 129, 142, 234
Index
H Hedley, 112, 114 Heraclides, 38, 41 Herigonius, 65, 69 Herodotus, 43 Highland, 17–19, 26, 44, 75, 76, 82, 86, 91, 101–104, 106 Hortensius, 69 I Imbrium (Mare) basin rim, 107 impact site, 59, 60 LTP site, 111 Impactor energy of, 74–75, 89, 90, 116 man made, 30, 36, 95, 115 physical nature of, 10, 71, 95 size, 71, 89 sources of, 79 Infrared, 36, 61, 115, 124, 125, 143, 182, 191–194 Instruments, 48, 62, 63, 92, 93, 99, 113, 118, 131, 138, 161, 175, 176, 182, 184, 187, 189–192, 199, 200, 219 Integration, 133, 135, 143, 161, 203 Intensity detector response, 189 impact light/flash profile, 165–166 of lunar features, 19 versus time/light curve, 184 K KIWI (On Screen Display) as an observing system component, 136–138, 149, 203 information about, 136–138, 141, 148, 149, 203, 220, 221 L Luminous efficiency, 183–185, 193 Lunar Crater Observation and Sensing Satellite (LCROSS), 124, 182, 194 Lunar impact automated detection of, 143, 152, 153, 158, 163, 181–182 events, candidate events, 125, 143, 146, 147, 158, 185 frequency of, 123, 183–185 light curve, 46, 142 observations of, 121–126, 129, 152, 157, 193
236
phenomena, 116, 184, 229 structures (top 100), 26 surface features, 111, 120, 147 verification of, 159 Lunar leonid, 56–59, 118, 132, 185, 210, 211 Lyrid, 50, 63, 67 M Martian, 16, 76, 90–92, 116, 120, 188 Meade, 53, 59, 131, 132, 136, 138, 191, 192, 213, 219 Meteoric, 116, 121, 161, 183–188, 193 Moon craters, 17, 19, 21, 22, 75, 76, 79, 85, 86, 89, 90, 99, 101–111, 117, 126, 130, 181 fresh craters, 93 lunar meteor phenomena, 183 lunar transient phenomena, 44, 111–113 observing procedures, 175–177 observing programs, 116, 121–126 polar ice, 124 research, 125–126, 181–182 space missions, 123–125 surface evolution, 71 Moore, P., 41, 42 O Observatory, 32, 33, 53, 58, 62, 72, 115, 122–125, 136, 142, 181, 193, 224, 225 Occultations focal reducer for observing, 132, 193 observations, 127–152 setup of observing equipment, 133–138 timing of, 127–152, 214–218 Orbiter, 30, 32, 43, 94, 124, 182, 188, 194 Orionids, 45, 49–50, 63 P Palmer, D., 51, 56, 58, 210, 212 Particles, 4, 91, 114, 118, 119, 162, 167, 185, 186, 188, 200, 207 Perseids ALSEP, 29, 30, 199 lunar meteor candidates, 60 lunar meteor observing opportunity, v, 51, 53 Physics, xiii, 35, 55, 105, 116–117, 136, 182, 184, 191, 193, 224 Pickering, W.H., 40, 41, 107 Plato, 38, 39, 41, 43, 102, 107, 111, 210, 213 Plume, xiii, xiv, 36, 92, 123–125, 182–184 Products, 133–134, 142–150, 152, 163, 181, 200, 201, 219 Programs, xiii, xiv, 29, 33, 37, 46, 47, 62, 93, 116, 118–126, 130, 133, 136, 147, 149, 152, 154–161, 163, 164, 181, 185, 189, 193, 199, 204, 209, 215, 216, 220, 221, 227
R Rings. See Saturn Royal Astronomical Society of Canada (RASC), 227–229 S Saturn annual meteor showers experienced at, 91 formation of, 79 impacts (in rings), 71, 76, 95 moons of, their surface features, 82–86 rings of, 71, 95, 187 Scientists, xiii, 23, 27, 30, 72, 79, 91, 114, 118 Shadow crater shadow profile, 23, 105 day/night boundary, 206 earth’s, 192, 201, 204 Miranda, 71, 86 permanent, 125 Shoemaker-Levy (comet), xiii, 16, 27, 71–74, 92, 93, 95 Sony, 51, 52, 58, 62, 64, 135, 136, 210, 220 Spectrographic, 150, 181, 189, 193 Spectroscope, 176, 194 Spectrum, 40, 111, 118, 125, 145, 176, 191, 193 Spellman, R., 63, 70, 138, 213 Stoffaire, J., 60, 213 Storm (meteor), 29, 53–57, 115, 123, 158, 196, 203, 211 Structure impact, craters, 8, 15, 16, 74, 81, 90, 200 Jupiter’s atmosphere, 74 man-made on lunar surface, 36, 115 meteoroid stream, 33 ringed, rimmed, rayed, or walled, 79, 103 Stuart Leon, H., 30, 32, 41, 205 the lunar flare, xii, xiii, 27, 113, 115 STVASTRO (Video Time Inserter), 133, 134 Supercircuits cameras, 46, 134–136, 144, 149, 159, 219, 220 the company, 133, 220 PC 164 camera, 134, 135, 143, 144, 219 PC 23C camera, 134, 135, 143, 144, 159, 219 Surveyor, 16, 71, 79, 93, 94, 107, 114, 116, 120 T Telescope Canada-France-Hawaii 3.6 meter, 124, 182 earth-based/ground-based, 15, 118, 120
Tycho comparison with other craters, 104 the crater, 18, 19, 47, 48, 102, 104, 107, 111, 192 LTP location, 111 ray system, 100, 107 thermal emission during eclipses, 183, 192 top 100, 19, 102, 107
V Venable, R., 48, 58, 60, 153, 154, 212, 213 Volcanoes, 38–40, 126 W Walled (impact features), 101, 105, 107–109, 207
Index
how to observe with, 128–129, 138–142 Hubble, 72, 74, 125 innovations, 189 Palomar, 72 small, backyard, 71, 74, 99 used in making lunar meteor observations, 134, 191–192 Tempel-Tuttle (comet), 203 Triton, 8, 71, 90–92, 187
237