This document is based on a file supplied by Harris and Warner as documentation of the data files. Since the file formats have been slightly modified to meet PDS requirements, the portions of this document about the specific file formats have been removed. The original document also described several data files supplied by Petr Pravec, which have not been included in this data set. These portions of the document have been removed since they don't apply to the present data set. In addition, a file with a compilation of asteroid spin axes was not included in this data set since it duplicates the data in the asteroid spin axis data set. The part of this document describing the spin axis file has also been removed. Other than the above, the file has not been edited by PDS. LIGHTCURVE DATA BASE (LCDB) Revised 15 February 2008 1.0.0 INTRODUCTION -------------------------------------------------------------------------------- The Asteroid Lightcurve Data Base (LCDB) is a set of files generated from a dBase IV data base that includes information directly and indirectly obtained from observations made to determine the period and/or amplitude of asteroid lightcurves. The information is taken from numerous journals and other sources. It's main purpose is to provide a central location for basic information about asteroid rotation rates and related information that can be used in statistical studies involving a few or many parameters. Some of the data are obtained directly from the observations while other data are inferred or calculated based on orbital characteristics, assumed class, etc. Sections below explain in detail which data are direct and indirectly obtained and, for the latter, their derivation. N.B. Even direct data should be confirmed by reference to the original works whenever possible. Indirect data are provided for information purposes only. They should NOT be used in critical studies. 1.1.0 AUTHOR INFORMATION ------------------------- These data files are maintained by Alan Harris and Brian D. Warner, Space Science Institute, and Petr Pravec, Astronomical Institute, Czech Republic. For further information or updated versions, contact: Alan Harris Space Science Institute Phone: (1) 818-790-8291 4603 Orange Knoll Ave. e-mail: awharris@SpaceScience.org La Canada, CA 91011 USA 1.2.0 DISCLAIMER ----------------- We have made every attempt to keep the data up to date and correct. However, we know that there is the possibility for omissions or errors. Please let us know of any corrections or additions by sending email to one of the below. Brian D. Warner brian@MinorPlanetObserver.com Alan Harris awharris@SpaceScience.org 2.0.0 DATA FILES -------------------------------------------------------------------------------- 2.0.1 AVAILABILITY OF RAW DATA FILES ------------------------------------- The raw dBASE IV data files are not generally available. We are considering a reduced version of the input program, one that would allow running SQL queries against the distributed dBASE files but would not allow modifying the files. 2.1.0 GENERAL FILES -------------------- README.TXT This file of introductory information. LCDETAILS The lightcurve data file, with data from individual references LCSUMMARY Summary data, one line per asteroid, no references. LC_REFERENCES Reference list to all published lightcurve data. LC_AMBIGUOUS Separate summary/detail file for asteroids with ambiguous periods. LC_BINARY Separate summary/detail file for binary asteroids LC_NPA Separate summary/detail file for "tumbling" asteroids 2.2.1 ABOUT BINARY DATA ------------------------ The data in the LC_BINARY file are by no means exhaustive. They are meant to provide a quick overview of the primary period and amplitude as well as a secondary period and, if available, amplitude. The secondary period may be due to rotation of the secondary or the orbit of the secondary if it is tidally locked with the primary. The reader is urged to consult the original journal articles for more complete details. A good summary page with links to those journal references is the web site run by by Wm. Robert Johnston http://www.johnstonsarchive.net/astro/asteroidmoons.html 3.0.0 LCDB DATA -------------------------------------------------------------------------------- The original lightcurve database was a simple text file with a structure that tried to minimize disk space requirements. That served well for many years. However, the rapidly growing number of lightcurves being reported as well as the file's format not being able to accommodate some data prompted a change starting in mid-2006. The foremost change was converting to a relational database that includes numerous tables and has SQL search capabilities. This allows for not only easier maintenance of the database but for generating reports in a new way that are more informative, complete, and constant in formatting. The format of the LCDATPUB, LCSUMPUB and LCREF files has changed from previous years. Very small changes were made again for the 2008 release that may require additional work on parsing programs. The sections below provide the new formatting for each field. This should make rewriting old parsing programs, or new ones, fairly straightforward. 3.1.0 DIRECT DATA ------------------ Data that is obtained directly from photometric observations includes 1. Rotation period (usually synodic). 2. Amplitude. 3. Absolute magnitude, H, and phase slope parameter, G, when determined by using reduced magnitude versus phase angle data. 4. Binarity due to mutual events, i.e., occultations and eclipses. In such cases, the rotation period of the primary and orbital period of the satellite and the amplitude of the primary lightcurve are the usual direct results. The size ratio can be computed from the depth of the events. For more details on binary lightcurve analysis, see Pravec et al., 2006, Icarus 181, 63-93. 5. Color indices. 6. Diameter if based on stellar occultation or adaptive optics/radar. Radar diameters can also be considered indirect, depending on how the diameter was determined. 7. Taxonomic class. 3.1.1 SYNODIC VERSUS SIDEREAL PERIOD ------------------------------------- The synodic period depends on viewing aspect and the rate of motion of the asteroid across the sky. An expression for the magnitude of the expected difference between the sidereal period and synodic period based on the Phase Angle Bisector (PAB) is DeltaP = [d(PAB)/dt] * P^2 Where DeltaP difference between synodic and sidereal periods, in units of the rotation period (usually hours). [d(PAB)/dt] angular rate of change in the Phase Angle Bisector in inverse units of the rotation period, e.g., in units of cycles/hour P the rotation period of the asteroid in the same units of time, e.g., hours/cycle For example, assume an asteroid with a rotation period of 8 hours observed when the PAB is changing by 0.05 deg/day (typical for an main-belt asteroid at opposition), then the sidereal-synodic difference is DeltaP = [0.05 / 360.0 / 24.0] * (8 ^ 2) = 0.00037 hr. cycles / hour hours^2/cycles^2 hours / cycle The synodic-sidereal difference can be either positive or negative, and can exceed the value given by this expression for near-polar aspects, but the expression gives a reasonable estimate of the magnitude of the expected difference. In most cases, the period given in the summary and details lines are synodic and not sidereal. An 'S' flag (see notes below) indicates that the period is sidereal. There are many entries in both tables that do not carry the 'S' flag when they should. This is part of the legacy nature of the data after converting the files to the new data base, i.e., the old format did not allow for indicating the period was one type or another. We hope to update these and other legacy values that now have qualifying flags in future versions. For most studies, the difference between sidereal and synodic period is not significant. 3.1.2 INDIRECT DATA -------------------- Indirect data is that obtained by calculation and/or assumption. N.B. THESE DATA ARE PROVIDED FOR CONVENIENCE ONLY, AND SHOULD NOT BE USED FOR STATISTICAL STUDIES, NOR ASSUMED TO BE ACCURATE AND CURRENT. 1. Diameter, H, and albedo (Pv, Pr, etc.). The relationship between H, diameter, and albedo is: D = (1329km) * 10.0^(-0.2*H) / sqrt(albedo) or logD (km) = 3.1235 - 0.2H - (0.5 * log(albedo)). (See Pravec and Harris, 2007. Icarus 190, 250-259). The value of H is usually known, though not always accurately, based on photometric observations. If direct data are available for D and/or Pv, then the above relationships can be used to derive a missing quantity. Data from the SIMPS study (Tedesco, E.F., Noah, P.V., Price, S.D.: 2004, IRAS Minor Planet Survey. IRAS-A-FPA-3-RDR-IMPS-V6.0. NASA Planetary Data System) is used when available and no overriding data are available. If a newer value of H than that used by SIMPS is available, the diameter and albedo are re-computed based on Harris and Harris (1997, Icarus 126, 450-454). If the diameter was based on an assumed albedo and H is revised, the albedo is held constant and the diameter re-computed using the above formulae. If the diameter was determined by radar, resolved imagery, etc. and a new H is available, the diameter is held constant and the albedo is re-computed. 2. Taxonomic Class, orbital class, and albedo These three values can have a complex relationship when the class and albedo are not directly obtained. When spectroscopic or other data are available that can determine the taxonomic class exactly but no direct albedo data are available, the albedo can be assumed. This albedo can, in turn, be used to find the estimated diameter and/or H if those quantities are not directly known. Flags in the Summary table indicate the source of these fields, including if they are assumed based on a combination of available data. The table below shows the assumed values used barring any direct data. The family/group is based on orbital parameters. These are mostly for informational purposes only since the definition for some groups or families is "fuzzy", at best. Note the distinction between a family and group. A family is a set of asteroids with a common parent body. A group is a set of asteroids with common orbital characteristics. Members of a family will almost always be in the same group, but members of a group may not necessarily be of the same family. Group/Family Orbit parameters Class pv NOTES -------------------------------------------------------------------------------------------------- Baptistina Bottke el al C 0.058 Centaur 5.5 < a <= 30 C 0.058 Comet-like orbit Q > 5.0 C 0.058 4 Comet exhibits coma and/or tail C 0.058 Eos 2.99 < a < 3.03, 0.01 < e < 0.13, 8 < i < 12 S 0.14 Erigone 2.32 < a < 2.40, 0.15 < e < 0.22, 4 < i < 6 C 0.058 Eunomia 2.53 < a < 2.72, 0.08 < e < 0.22, 11 < i < 16 S 0.21 Flora 2.15 < a < 2.35, 0.03 < e < 0.23, 1.5 < i < 8 S 0.24 Hungaria 1.78 < a < 2.0, e < 0.18, 16 < i < 34 E 0.3 1 Karin Nesvorny, private communications S 0.26 Koronis 2.83 < 2.91, e < 0.11, i <= 3.5 S 0.26 Main belt - inner a < 2.6 S 0.18 Main belt - middle 2.6 < a < 2.7 SC 0.10 Main belt - outer 2.7 < a < 5.0 C 0.058 Mars crosser 1.3 < q < 1.668, Q < 5.0 S 0.18 NEA (Apollo/Aten/Amor) q < 1.3 S 0.18 Nysa 2.4 < a < 2.5, 0.12 < e < 0.21, 1.3 < i < 4.3 S 0.4 Phocaea 2.25 < a < 2.5, e >= 0.1, 18 < i < 32 S 0.23 Themis 3.08 < a < 3.24, 0.09 < e < 0.25, i <= 3 S 0.1 TNO/KBO a > 30 C 0.1 Trojan - Jupiter 5.05 < a < 5.4 C 0.058 Trojan - Mars similar to Mars S 0.18 Trojan - Neptune similar to Neptune C 0.058 Trojan - Saturn similar to Saturn C 0.058 2 Trojan - Uranus similar to Uranus C 0.058 2 Vestoid 2.26 < a < 2.48, 0.03 < e < 0.16, 5 < i < 8.3 S 0.18 3 NOTES ----- 1 Pv = 0.3 is a compromise value when no taxonomic information is available, since the Hungarias are both a family (common parent, E/X class, pv = 0.4) and group (similar orbits, S class, pv = 0.18). 2 None known and not likely due to perturbations by giant planets, both interior and exterior. 3 Higher albedo (0.4) assigned only if in V class determined (SMASS, etc). Otherwise, class is 'S' and pv = 0.18 on presumption that the object shares orbital characteristics but not parent body. 4 Barring any other classification that meets Q > 5, the orbit is classified as "comet-like." Bottke, W.F., Vokrouhlicky, D., Nesvorny, D. (2007). Nature 449, 48-53, and private communications. 3. Color Index applied to H Color index is not generally assumed or entered into the LCDB. However, some times the value of H was found in a photometric band other than V, e.g., R. In that case, and if the value is used to override the H given by SIMPS or the MPCORB file (Minor Planet Center) in the Summary line, H is transformed to the V band. When the color index is not directly available, these values are used to transform the measured H value V-R 0.45 B-V 0.80 When such a transform is used in the Summary line, whether or not based on an assumed value, the H value has the 'T' (transformed) flag. 4.0.0 LCDB FILE DESCRIPTIONS -------------------------------------------------------------------------------- The following sections describe the specific files that are part of the LCDB release, not including the Binary files from Petr Pravec. A column map is provided for each report that is part of the release. In addition, sub-sections describe the meaning of the flags that qualify various fields in the report. 4.1.0 LCSUMMMARY AND LCDETAILS ------------------------------- These are the primary files in the LCDB release. Both have the same column mapping. They show both direct and indirect data, the most important being the direct data of lightcurve period and amplitude along with our assessment of the quality of the period solution. The latter is expressed by the U code, which is described in detail below. In the LCDETAILS file, a Summary line is followed by one or more Details lines. The Summary line is our best determination of the primary information for a given object based on the data in the Details lines. For example, where several periods are available, the Summary line uses the one that we consider the most likely to be correct. Sometimes that value may be an average of the available values. The Details lines differ in formatting only in that the number and name of the asteroid are replaced by a "short reference" name that will be found in the LC_REFERENCES references file. 4.1.1 MULTIPLE DETAILS TABLE ENTRIES ------------------------------------- In some cases, there is more than one detail line under a given asteroid with the same publication reference. This is deliberate in order to allow statistical studies of lightcurve amplitude versus phase versus class (albedo). For example, if a single publication reports the lightcurve behavior for an asteroid where the synodic period and/or amplitude of the curve changed significantly during the course of the observations, the DETAILS table will include an appropriate number of entries. Those entries will "split out" the results so that the period and/or amplitude can be tied to a specific (though maybe only approximate) set of PAB or Phase values. A good example would be a paper reporting observations of an NEA asteroid over several weeks where the amplitude of the curve when from 1.1 to 0.3 magnitudes over the range of observations. In most cases, splitting the results into distinct sets was not difficult, e.g., the asteroid was observed on one night at one-week intervals. In some cases, the split was not so distinct. In this case, compromises were made in order to avoid having an excess of multiple entries while still retaining sufficient resolution of the variations versus time. A variation on the above is if the author(s) forced the data from several blocks of dates to fit a fixed period solution. Here, the period will be the same for all entries, though the amplitude may change. In this case, the period and U code rating are left blank, leaving only the amplitude for the reasons given above. Other information that was derived based on the given block of data, e.g., a value for H, G, or a color index, will be included within that Details record as well so that it's clear which block of data was used to derive the given values. 4.1.2 U (QUALITY) CODE ----------------------- The U code provides our assessment of the quality of the period solution, not necessarily of the data per se. Depending on the specifics for a given asteroid, a good period solution can be obtained by using a large amount of lesser quality data about as well as using less data that is of higher quality. Many factors come into play making the assessment. The table below gives the general outline of the criteria used. 0 Result later proven incorrect. This appears only on records of individual observations (see below). 1 Result based on fragmentary lightcurve(s), may be completely wrong. 2 Result based on less than full coverage, so that the period may be wrong by 30 percent or so. Also, a quality of 2 is used to note results where an ambiguity exists as to the number of extrema per cycle or the number of elapsed cycles between lightcurves. Hence the result may be wrong by an integer ratio. 3 Denotes a secure result with no ambiguity and full lightcurve coverage. 4 Indicates that a pole position is reported. N.B. The meaning of the U=4 will change starting after this release (February 2008), at which time, the highest U value will be 3 and reflect only the quality of the period solution. In the past, there have been cases where a 4 was assigned because there was a pole solution given but the best available period solution was no better than 2. The change means that the period solution will be isolated from any pole solution, which will have its own quality rating. The delay in making this change is because each U=4 rating will have to be revisited to decide if it should be changed to 3 or to 2. The "Pole" flag in the Summary report will be use to indicate if there is a pole solution available, be it in the SPINAXIS report and/or on the Poznan or Durech web sites. In some cases, the numerical quality code may be followed by a plus or minus sign, indicating that the reliability is judged somewhat better (+) or worse (-) than implied by an un-signed number alone. Assignment of the refined ratings is a work in progress as we catch up with almost 10 years of data entry. Therefore, not all U code ratings will match what we would give under current rules and are subject to change. Some Details lines, and even some Summary, may not contain a U code rating. This is deliberate and can be for several reasons. 1. The available data does not include a lightcurve, therefore, it is not possible to give a rating to the curve. In some cases, where the results are reported by observers whose standard of work is known to be of sufficient quality, we may assign an interim U code, usually 2, until a lightcurve or the data are available. 2. In the case where several results are published for a given object in the same reference, we will assign a U code rating for the "best" available data and include only new information for that given Details record, e.g., see section 4.1.1, "Multiple Details Table Entries." There are instances when a lightcurve is available but the available data cannot reasonably define a period or even constrain a range in which the period lies. In addition, the data may not be able to provide any reasonable indication of the amplitude. In these cases, the entry in the Details record will have the reference to the work and a U code rating of 1 but no period or amplitude. If the Details line is the only one for the object, the Summary line has only the number (if any), name or designation, a U code of 1 and, again, no period or amplitude. This is done to show that there is some data available for the object but that it may be of very limited use. 4.1.4 FIELD (FLAG) CODES USED IN SUMMARY AND DETAIL LINES ---------------------------------------------------------- The flags appear in the data field immediately before the value they qualify. In most cases, they are a single character. EntryFlag --------- Blank Pre-existing record. * New or updated record since the last release. PFLAG (Period Flag) ------------------- Blank NONE < Less than > Greater than D No numerical value, see P DESC field description S Sideral period, default is no flag and synodic period U Uncertain, not the same as ambiguous where one or additional periods are reported. For example, the data did not allow finding a definite period and so the author(s) reported a "best guess." AMPFLAG (Amplitude Flag) ------------------------ Blank NONE < Less than > Greater than NOTES (single letter flag(s)) ----------------------------- Blank NONE ? Usually tied with 'T' to indicate uncertainty 1 Monomodal curve (one min/max per rotation) 3 Trimodal curve (three min/max per rotation) 4 Quadrimodal curve (four min/max per rotation) A Ambiguous period (see LCAMBPUB.TXT for details) D Period determined by us that differs from that given in the original publication H Space telescope observations N No lightcurve published O Adaptive optics observation P Photographic photometry R Radar observation T Tumbling (NPA rotation - see LCNPAPUB.TXT for details) V Visual photometry 4.2.0 LC_AMBIGUOUS (AMBIGUOUS PERIODS) -------------------------------------------------------------------------------- This file includes any record where the Notes flag for a summary or detail line indicates an ambiguous period. There is not always a direct cross-connection between the Summary and Details entries. For example, it's possible to have a Summary line without the ambiguous period flag but one or more of the Details lines to have the flag. In this case, we judge that the ambiguity has been resolved by subsequent observations, but retain the ambiguous flag in the detail line for historical accuracy. 4.3.0 LC_BINARY (BINARY ASTEROIDS) -------------------------------------------------------------------------------- This file includes those asteroids that are known or suspected binaries. This is not meant to be a comprehensive compilation of data for binary asteroids. Visit the URL given in section 2.2.0 for a page that provides more details as well as links to the original journal articles. For asynchronous systems, the secondary period and amplitude will be blank in most cases, indicating the independent rotation period and amplitude of the satellite were not determined. POrb indicates the orbital period of the satellite about the primary. For synchronous systems, the primary, secondary, and orbital periods are identical and so all three values will be the same. The primary and secondary amplitudes are also indentical and will be the same. The amplitude of mutual events (eclipses/occultations) is not given. See the original references and those in section 2.2.0 for that and additional data. 4.5.0 LC_NPA (NON-PRINCIPAL AXIS ROTATION - TUMBLING) --------------------------------------------------------------- 4.5.1 PAR RATING ---------------- The PAR rating is adopted from Pravec et al, Icarus 173, 108-131. Those asteroids where we believe the claims of NPA are not justified are not included in the NPA table, nor are they given an entry in the NOTES field of the Summary and/or Details tables. Instead a special entry is made in the "free-form" NOTESEX field of the appropriate table and so appear in the LC_NOTESEX file. Following is a brief description of the PAR codes. See the Pravec paper for a more detailed explanation. -4 Physical model of the NPA rotation constructed -3 NPA rotation reliably detected with the two periods resolved. There may be some ambiguities in one or both periods. -2 NPA rotation detected based on deviations from a single period but the second period is not resolved. -1 NPA rotation possible, i.e., deviations from a single period are seen, but not conclusively. 0 Insufficient data to determine if rotation is PA or NPA +1 PA rotation is consistent with the data but coverage is insufficient. +2 PA rotation likely, or deviations from PA are small with some overlapping data fitting a PA rotation period. +3 PA rotation quite likely +4 PA spin vector obtained. Entries with a positive number are rare and used when the asteriod was thought to be tumbling but futher examination showed it was likely in PA rotation, or when the damping time to PA rotation is sufficiently long that the given asteroid would more likely be in NPA than PA rotation. 4.6.0 LC_REFERENCES (REFERENCES) ---------------------------------- The LCREF.TXT file contains the complete literature citation for each reference in the data file. The left column gives the "shorthand" form as it appears in the table. On the right is the proper full citation, listing all authors, year, journal, volume and page numbers. N.B. Initials for names are packed, e.g., Warner, B.D. and not Warner, B. D.