Light Curve Database Research Articles

Rotation period estimates for 14 asteroids with the Earth MOID less than 1.1 AU

In the present study, the Lomb–Scargle Scargle (L–S) method was employed for the analysis of brightness measurements from multiple observatories as published in the Minor Planet Center (MPC) circulars, as well as brightness measurements from the Baldone Observatory, to determine the periods of asteroids. The periods of fifteen asteroids were determined. The research yielded results for both asteroids with previously known periods and those without listed periods in the ALCDEF lightcurve database. The results revealed the potential for this approach to be used in the subsequent identification of yet-unknown periods of asteroids. This was demonstrated by the confirmation of periods for four of the five asteroids with known periods from the ALCDEF lightcurve database, namely 1951, 1963, 2134 and 2150. In the case of the fifth asteroid, 2174, for which a previously determined rotation period had been provided, an additional, possible shorter rotation period was identified. As has been shown, the use of the L–S method in conjunction with independent analysis of many series of brightness measurements from different observatories is a suitable approach, even for a small series of samples of brightness observations unevenly spread over a long time, for the determination of periods. Furthermore, this approach is effective even when there are significant intervals between measurements, which is not the case with the classical Fourier method. Additionally, the exclusion of observations that deviate significantly from the linear relationship of the phase diagram, by the three-sigma criterion, markedly enhanced the efficacy and precision of the method. As a result, periods were identified for the first time for asteroids 1779, 1818, 2128, 2318, 2497, 2503, 2538, 2539, and 2583.

The Fastest Rotators: Near-Earth Asteroids Observed with the Arecibo Planetary Radar System

Fast-rotating asteroids (FRAs) are considered to be small bodies having a rotation period (P) faster than the spin barrier of about 2.3 h, starting at diameters of less than 300 m. We selected the 20 fastest Arecibo radar-observed targets, with P<0.13 h (∼8 min). Some key measurements and calculations obtained from radar observations include: the Doppler bandwidth, the circular polarization ratio, radar astrometry, and (with enough signal-to-noise ratio) delay-Doppler images of the object. Rotation period data available from the Light Curve Database for the selected objects combined with the radar observations allow us to constrain the possible diameters and confirm the periods. Of the objects in this sample, the median absolute magnitude (H) is 24.9, and the median calculated diameter is 32 meters.The apparent rotation period indicated by the bandwidth can appear slower than the object’s true rotation, but not faster, providing an upper limit to the true rotation period. Asteroid cohesion required to prevent rotational disruption depends on rotation rate, density, and diameter; we performed calculations for the minimum values for cohesion via the Drucker–Prager (D–P) cohesion criterion. Most of these objects need a few to a few hundred pascals of cohesion; however four cases stand out: 2014 TV, 2015 RF36, 2015 GS2 and 2017 EK, needing a minimum cohesion on the order of a few kilopascals. These are comparable to very weak Earth rocks, and are larger than previously reported values for NEAs.

Distinguishing a planetary transit from false positives: a Transformer-based classification for planetary transit signals

ABSTRACT Current space-based missions, such as the Transiting Exoplanet Survey Satellite (TESS), provide a large database of light curves that must be analysed efficiently and systematically. In recent years, deep learning (DL) methods, particularly convolutional neural networks (CNN), have been used to classify transit signals of candidate exoplanets automatically. However, CNNs have some drawbacks; for example, they require many layers to capture dependencies on sequential data, such as light curves, making the network so large that it eventually becomes impractical. The self-attention mechanism is a DL technique that attempts to mimic the action of selectively focusing on some relevant things while ignoring others. Models, such as the Transformer architecture, were recently proposed for sequential data with successful results. Based on these successful models, we present a new architecture for the automatic classification of transit signals. Our proposed architecture is designed to capture the most significant features of a transit signal and stellar parameters through the self-attention mechanism. In addition to model prediction, we take advantage of attention map inspection, obtaining a more interpretable DL approach. Thus, we can identify the relevance of each element to differentiate a transit signal from false positives, simplifying the manual examination of candidates. We show that our architecture achieves competitive results concerning the CNNs applied for recognizing exoplanetary transit signals in data from the TESS telescope. Based on these results, we demonstrate that applying this state-of-the-art DL model to light curves can be a powerful technique for transit signal detection while offering a level of interpretability.

A Detectable Candidate for the YORP Effect of Asteroids

The YORP (Yarkovsky-O’Keefe-Radzievskii-Paddack) effect is one of the mechanisms of the long-term dynamical evolution of asteroids. Compared with factors such as collision and gravitational perturbation, the YORP is of small magnitude, and the short-time scale observation effect is inconspicuous, which brings great difficulties to the direct measurement of the YORP. From the Asteroid Lightcurve Database, asteroids having a high confidence rotation period were selected for this study. Two subsample groups for identifying potential asteroids slowed by the YORP effect are provided by using the kernel density estimation method and the Kolmogorov-Smirnov test to analyze the rotation rate distribution characteristics of near-Earth asteroids and main belt asteroids; a screening model is proposed based on the light-curve data of seven YORP asteroids with YORP rotation acceleration, combined with the YORP intensity estimation method and the detection conditions of the YORP effect. Finally, ten candidates that can directly detect the YORP effect through light-curve data in the future are listed based on the screening model.

New limits from microlensing on Galactic black holes in the mass range 10 M⊙ &lt; M &lt; 1000 M⊙

We searched for long-duration microlensing events originating from intermediate-mass black holes (BH) in the halo of the Milky Way, using archival data from the EROS-2 and MACHO photometric surveys towards the Large Magellanic Cloud (LMC). We combined data from these two surveys to create a common database of light curves for 14.1 million objects in the LMC, covering a total duration of 10.6 years, with flux series measured in four wide passbands. We carried out a microlensing search on these light curves, complemented by the light curves of 22.7 million objects, observed only by EROS-2 or only by MACHO, over about 7 years, with flux series measured in only two passbands. A likelihood analysis, taking into account the LMC self-lensing and Milky Way disk contributions, allows us to conclude that compact objects with masses in the range 10 − 100 M⊙ cannot make up more than ∼15% of a standard halo total mass (at a 95% confidence level). Our analysis sensitivity weakens for heavier objects, although we still rule out the possibility of ∼50% of the halo being made of ∼1000 M⊙ BHs. Combined with previous EROS results, an upper limit of ∼15% of the total halo mass can be obtained for the contribution of compact halo objects in the mass range 10−6 − 102 M⊙.

Rotation periods and shape asphericity in asteroid families based on TESS S1-S13 observations

Here we present the analysis of the distribution of rotation periods and light curve amplitudes based on 2859 family asteroids in 16 Main Belt families based on 9912 TESS asteroid light curves in the TSSYS-DR1 asteroid light curve database. We found that the distribution of the light curve properties follow a family-specific character in some asteroid families, including the Hungaria, Maria, Juno, Eos, Eucharis, and Alauda families. While in other large families, these distributions are in general very similar to each other. We confirm that older families tend to contain a larger fraction of more spheroidal, low-amplitude asteroids. We found that rotation period distributions are different in the cores and outskirts of the Flora and Maria families, while the Vesta, Eos, and Eunomia families lack this feature. We also confirm that very fast spinning asteroids are close to spherical (or spinning top shapes), and minor planets rotating slower than ≈11 h are also more spherical than asteroids in the 4–8 h period range and this group is expected to contain the most elongated bodies.

Asteroid Photometry from the Transiting Exoplanet Survey Satellite: A Pilot Study

Abstract The Transiting Exoplanet Survey Satellite (TESS) searches for planets transiting bright and nearby stars using high-cadence, large-scale photometric observations. Full frame images provided by the TESS mission include a large number of serendipitously observed main-belt asteroids (MBAs). Due to the cadence of the published full frame images, we are sensitive to periods as long as of order tens of days, a region of phase space that is generally not accessible through traditional observing. This work represents a much less biased measurement of the period distribution in this period range. We have derived rotation periods for 300 MBAs and have partial lightcurves for a further 7277 asteroids, including 43 with periods of P &gt; 100 hr; this large number of slow rotators is predicted by theory. Of these slow rotators we find none requiring significant internal strength to resist rotational reshaping. We find our derived rotation periods to be in excellent agreement with results in the Lightcurve Database for the 55 targets that overlap. Over the nominal two-year lifetime of the mission, we expect the detection of around 85,000 unique asteroids with rotation period solutions for around 6000 asteroids. We project that the systematic analysis of the entire TESS data set will increase the number of known slow-rotating asteroids (period &gt; 100 hr) by a factor of 10. Comparing our new period determinations with previous measurements in the literature, we find that the rotation period of asteroid (2320) Blarney has decreased by at least 20% over the past decade, potentially due to surface activity or subcatastrophic collisions.

Study on the Spin Characteristics of the Flora Asteroid Family

Asteroid families are the remnants of catastrophic collisions, and their fundamental physical properties provide us the information of their parent bodies and thereafter dynamical evolutions. Especially, the orbit and spin characteristics can reveal the influences of the Yarkovsky effect and the Yarkovsky-O’Keefe-Radzievskii-Paddack (YORP) effect on the evolution of the asteroid family, respectively. Based on the Asteroid Lightcurve Database (LCDB), the spin rate distribution of the Flora asteroid family is studied, and a tendency that the spin rates of the small Flora family members concentrate primarily in the range of 3–5 d−1 is found. The analysis on the spin states of the Flora family asteroids tells that most of these asteroid family members are in the prograde spinning state. However, for the Flora family members with an orbital semi-major axis smaller than 2.2au, the ratio between the number of prograde spinning members and that of retrograde ones is close to that of the near-Earth asteroids, namely 1:3. Furthermore, for those prograde spinning Flora family asteroids with an orbital semi-major axis larger than 2.2au, a portion of them exhibit the aggregation in the distribution of orbital semi-major axis against the absolute magnitude, and in which nine members show the features similar to the Slivan state.

Study on the Spin Characteristics of the Flora Asteroid Family

Asteroid families are the remnants of catastrophic collisions, and their fundamental physical properties provide us the information of their parent bodies and thereafter dynamical evolutions. Especially, the orbit and spin characteristics can reveal the influences of the Yarkovsky effect and the Yarkovsky-O’Keefe-Radzievskii-Paddack (YORP) effect on the evolution of the asteroid family, respectively. Based on the Asteroid Lightcurve Database (LCDB), the spin rate distribution of the Flora asteroid family is studied, and a tendency that the spin rates of the small Flora family members concentrate primarily in the range of 3–5 d−1 is found. The analysis on the spin states of the Flora family asteroids tells that most of these asteroid family members are in the prograde spinning state. However, for the Flora family members with an orbital semi-major axis smaller than 2.2au, the ratio between the number of prograde spinning members and that of retrograde ones is close to that of the near-Earth asteroids, namely 1:3. Furthermore, for those prograde spinning Flora family asteroids with an orbital semi-major axis larger than 2.2au, a portion of them exhibit the aggregation in the distribution of orbital semi-major axis against the absolute magnitude, and in which nine members show the features similar to the Slivan state.

PELICAN: deeP architecturE for the LIght Curve ANalysis

We developed a deeP architecturE for the LIght Curve ANalysis (PELICAN) for the characterization and the classification of supernovae light curves. It takes light curves as input, without any additional features. PELICAN can deal with the sparsity and the irregular sampling of light curves. It is designed to remove the problem of non-representativeness between the training and test databases coming from the limitations of the spectroscopic follow-up. We applied our methodology on different supernovae light curve databases. First, we tested PELICAN on the Supernova Photometric Classification Challenge for which we obtained the best performance ever achieved with a non-representative training database, by reaching an accuracy of 0.811. Then we tested PELICAN on simulated light curves of the LSST Deep Fields for which PELICAN is able to detect 87.4% of supernovae Ia with a precision higher than 98%, by considering a non-representative training database of 2k light curves. PELICAN can be trained on light curves of LSST Deep Fields to classify light curves of the LSST main survey, which have a lower sampling rate and are more noisy. In this scenario, it reaches an accuracy of 96.5% with a training database of 2k light curves of the Deep Fields. This constitutes a pivotal result as type Ia supernovae candidates from the main survey might then be used to increase the statistics without additional spectroscopic follow-up. Finally we tested PELICAN on real data from the Sloan Digital Sky Survey. PELICAN reaches an accuracy of 86.8% with a training database composed of simulated data and a fraction of 10% of real data. The ability of PELICAN to deal with the different causes of non-representativeness between the training and test databases, and its robustness against survey properties and observational conditions, put it in the forefront of light curve classification tools for the LSST era.

Building the Evryscope: Hardware Design and Performance

The Evryscope is a telescope array designed to open a new parameter space in optical astronomy, detecting short-timescale events across extremely large sky areas simultaneously. The system consists of a 780 MPix 22-camera array with an 8150 sq. deg. field of view, 13″ per pixel sampling, and the ability to detect objects down to in each 2-minute dark-sky exposure. The Evryscope, covering 18,400 sq. deg. with hours of high-cadence exposure time each night, is designed to find the rare events that require all-sky monitoring, including transiting exoplanets around exotic stars like white dwarfs and hot subdwarfs, stellar activity of all types within our galaxy, nearby supernovae, and other transient events such as gamma-ray bursts and gravitational-wave electromagnetic counterparts. The system averages 5000 images per night with ∼300,000 sources per image, and to date has taken over 3.0M images, totaling 250 TB of raw data. The resulting light curve database has light curves for 9.3M targets, averaging 32,600 epochs per target through 2018. This paper summarizes the hardware and performance of the Evryscope, including the lessons learned during telescope design, electronics design, a procedure for the precision polar alignment of mounts for Evryscope-like systems, robotic control and operations, and safety and performance-optimization systems. We measure the on-sky performance of the Evryscope, discuss its data analysis pipelines, and present some example variable star and eclipsing binary discoveries from the telescope. We also discuss new discoveries of very rare objects including two hot subdwarf eclipsing binaries with late M-dwarf secondaries (HW Vir systems), two white dwarf/hot subdwarf short-period binaries, and four hot subdwarf reflection binaries. We conclude with the status of our transit surveys, M-dwarf flare survey, and transient detection.

Hubble Space Telescope Far-UV Spectroscopy of the Short Orbital Period Recurrent Nova CI Aql: Implications for White Dwarf Mass Evolution

Abstract An Hubble Space Telescope Cosmic Object Spectrograph Far UV spectrum (1170 Å to 1800 Å) was obtained for the short orbital period recurrent novae (T Pyxidis subclass), CI Aquilae. CI Aql is the only classical Cataclysmic variable (CV) known to have two eclipses of a sensible depth per orbit cycle and also to have pre- and post-outburst light curves that are steady enough to allow estimates of mass and orbital period changes. Our far-ultraviolet (FUV) spectral analysis with model accretion disks and non-LTE high-gravity photospheres, together with the Gaia parallax, reveal that CI Aql's FUV light is dominated by an optically thick accretion disk with an accretion rate of the order of 4 × 10−8 M ☉ yr−1. Its database of light curves, radial velocity curves, and eclipse timings is among the best for any CV. Its orbit period (P), dP/dt, and reference time are rederived via a simultaneous analysis of the three data types, giving a dimensionless post-outburst dP/dt of (−2.49 ± 0.95) × 10−10. The lack of information on the loss of orbital to rotational angular momentum leads to some uncertainty in the translation of dP/dt to the white dwarf (WD) mass change rate, dM1/dt, but within the modest range of to +7.8 × 10−8 . The estimated WD mass change through outburst for CI Aql, based on simple differencing of its pre- and post-outburst orbit period, is unchanged from the previously published . At the WD's estimated mass increase rate, it will terminate as a Type Ia supernova within 10 million years.

Fast in-database cross-matching of high-cadence, high-density source lists with an up-to-date sky model

Coming high-cadence wide-field optical telescopes will image hundreds of thousands of sources per minute. Besides inspecting the near real-time data streams for transient and variability events, the accumulated data archive is a wealthy laboratory for making complementary scientific discoveries. The goal of this work is to optimise column-oriented database techniques to enable the construction of a full-source and light-curve database for large-scale surveys, that is accessible by the astronomical community. We adopted LOFAR's Transients Pipeline as the baseline and modified it to enable the processing of optical images that have much higher source densities. The pipeline adds new source lists to the archive database, while cross-matching them with the known cataloged sources in order to build a full light-curve archive. We investigated several techniques of indexing and partitioning the largest tables, allowing for faster positional source look-ups in the cross matching algorithms. We monitored all query run times in long-term pipeline runs where we processed a subset of IPHAS data that have image source density peaks over $170,000$ per field of view ($500,000$ deg$^{-2}$). Our analysis demonstrates that horizontal table partitions of declination widths of one-degree control the query run times. Usage of an index strategy where the partitions are densily sorted according to source declination yields another improvement. Most queries run in sublinear time and a few (<20%) run in linear time, because of dependencies on input source-list and result-set size. We observed that for this logical database partitioning schema the limiting cadence the pipeline achieved with processing IPHAS data is 25 seconds.

New Asymptotic Giant Branch Carbon Stars in the Galactic Halo

For the first time the data on the eight confirmed or candidate carbon (C) stars found mainly from objective-prism plates are presented. By using the Catalina database of lightcurves, we find that all these stars are pulsating, allowing a distance to be estimated through the K-band Period-Luminosity (PL) relation. This relation does not depend on spectral type (M or C) and distances are reliable even for C candidates. Seven stars are more than 10 kpc from the galactic plane, suggesting they do not belong to the galactic disk. We also find one star located at about 180 kpc from the Sun, being one of the most distant star in the Galaxy. Many of these new C stars are relatively blue. Some comments are also provided on seven other known halo carbon stars for which either a pulsation period is obtained, or because they were not included in previous works on halo C stars.

Apparent rotation properties of space debris extracted from photometric measurements

Knowledge about the rotation properties of space debris objects is essential for the active debris removal missions, accurate re-entry predictions and to investigate the long-term effects of the space environment on the attitude motion change. Different orbital regions and object’s physical properties lead to different attitude states and their change over time.Since 2007 the Astronomical Institute of the University of Bern (AIUB) performs photometric measurements of space debris objects. To June 2016 almost 2000 light curves of more than 400 individual objects have been acquired and processed. These objects are situated in all orbital regions, from low Earth orbit (LEO), via global navigation systems orbits and high eccentricity orbit (HEO), to geosynchronous Earth orbit (GEO). All types of objects were observed including the non-functional spacecraft, rocket bodies, fragmentation debris and uncorrelated objects discovered during dedicated surveys. For data acquisition, we used the 1-meter Zimmerwald Laser and Astrometry Telescope (ZIMLAT) at the Swiss Optical Ground Station and Geodynamics Observatory Zimmerwald, Switzerland. We applied our own method of phase-diagram reconstruction to extract the apparent rotation period from the light curve. Presented is the AIUB’s light curve database and the obtained rotation properties of space debris as a function of object type and orbit.

The spin-barrier ratio for S and C-type main asteroids belt

Asteroids of size larger than 0.15 km generally do not have periods P smaller than about 2.2 h, a limit known as cohesionless spin-barrier. This barrier can be explained by means of the cohesionless rubble-pile structure model. In this paper we will explore the possibility for the observed spin-barrier value to be different for C and S-type Main Asteroids Belt (MBAs). On the basis of the actual bulk density values, the expected ratio between the maximum rotation periods is PC/PS≈1.4±0.3. Using the data available in the asteroid LightCurve Data Base (LCDB) we have found that, as regards the mean spin-barrier values and for asteroids in the 4–20 km range, there is a little difference between the two asteroids population with a ratio PC/PS≈1.20±0.04. Uncertainties are still high, mainly because of the small number of MBAs with known taxonomic class in the considered range. In the 4–10 km range, instead, the ratio between the spin-barriers seems closer to 1 because PC/PS≈1.11±0.05. This behavior could be a direct consequence of a different cohesion strength for C and S-type asteroids of which the ratio can be estimated.

ASTEROID SPIN-RATE STUDY USING THE INTERMEDIATE PALOMAR TRANSIENT FACTORY

Two dedicated asteroid rotation-period surveys have been carried out using data taken on January 6-9 and February 20-23 of 2014 by the Intermediate Palomar Transient Factory (iPTF) in the $R$~band with $\sim 20$-min cadence. The total survey area covered 174~deg$^2$ in the ecliptic plane. Reliable rotation periods for 1,438 asteroids are obtained from a larger data set of 6,551 mostly main-belt asteroids, each with $\geq 10$~detections. Analysis of 1751, PTF based, reliable rotation periods clearly shows the "spin barrier" at $\sim 2$~hours for "rubble-pile" asteroids. We also found a new large-sized super-fast rotator, 2005 UW163 (Chang et al., 2014), and other five candidates as well. Our spin-rate distributions of asteroids with $3 < D < 15$~km shows number decrease when frequency greater than 5 rev/day, which is consistent to that of the Asteroid Light Curve Database (LCDB, Warner et al., 2009) and the result of (Masiero et al., 2009). We found the discrepancy in the spin-rate distribution between our result and (Pravec et al., 2008, update 2014-04-20) is mainly from asteroids with $\Delta m < 0.2$ mag that might be primarily due to different survey strategies. For asteroids with $D \leq 3$~km, we found a significant number drop at $f = 6$ rev/day. The YORP effect timescale for small-sized asteroid is shorter that makes more elongate objets spun up to reach their spin-rate limit and results in break-up. The K-S test suggests a possible difference in the spin-rate distributions of C- and S-type asteroids. We also find that C-type asteroids have a smaller spin-rate limit than the S-type, which agrees with the general sense that the C-type has lower bulk density than the S-type.

The LOFAR Transients Pipeline

Current and future astronomical survey facilities provide a remarkably rich opportunity for transient astronomy, combining unprecedented fields of view with high sensitivity and the ability to access previously unexplored wavelength regimes. This is particularly true of LOFAR, a recently-commissioned, low-frequency radio interferometer, based in the Netherlands and with stations across Europe. The identification of and response to transients is one of LOFAR's key science goals. However, the large data volumes which LOFAR produces, combined with the scientific requirement for rapid response, make automation essential. To support this, we have developed the LOFAR Transients Pipeline, or TraP. The TraP ingests multi-frequency image data from LOFAR or other instruments and searches it for transients and variables, providing automatic alerts of significant detections and populating a lightcurve database for further analysis by astronomers. Here, we discuss the scientific goals of the TraP and how it has been designed to meet them. We describe its implementation, including both the algorithms adopted to maximize performance as well as the development methodology used to ensure it is robust and reliable, particularly in the presence of artefacts typical of radio astronomy imaging. Finally, we report on a series of tests of the pipeline carried out using simulated LOFAR observations with a known population of transients.

Rotational properties of the binary and non-binary populations in the trans-Neptunian belt

We present results for the short-term variability of Binary Trans-Neptunian Objects (BTNOs). We performed CCD photometric observations using the 3.58 m Telescopio Nazionale Galileo, the 1.5 m Sierra Nevada Observatory telescope, and the 1.23 m Centro Astronomico Hispano Aleman telescope at Calar Alto Observatory. We present results based on five years of observations and report the short-term variability of six BTNOs. Our sample contains three classical objects: 2003MW12, or Varda, 2004SB60, or Salacia, and 2002 VT130; one detached disk object: 2007UK126; and two resonant objects: 2007TY430 and 2000EB173, or Huya. For each target, possible rotational periods and/or photometric amplitudes are reported. We also derived some physical properties from their lightcurves, such as density, primary and secondary sizes, and albedo. We compiled and analyzed a vast lightcurve database for Trans-Neptunian Objects (TNOs) including centaurs to determine the lightcurve amplitude and spin frequency distributions for the binary and non-binary populations. The mean rotational periods, from the Maxwellian fits to the frequency distributions, are 8.63+/-0.52 h for the entire sample, 8.37+/-0.58 h for the sample without the binary population, and 10.11+/-1.19 h for the binary population alone. Because the centaurs are collisionally more evolved, their rotational periods might not be so primordial. We computed a mean rotational period, from the Maxwellian fit, of 8.86+/-0.58 h for the sample without the centaur population, and of 8.64+/-0.67 h considering a sample without the binary and the centaur populations. According to this analysis, regular TNOs spin faster than binaries, which is compatible with the tidal interaction of the binaries. Finally, we examined possible formation models for several systems studied in this work and by our team in previous papers.

THE EB FACTORY PROJECT. I. A FAST, NEURAL-NET-BASED, GENERAL PURPOSE LIGHT CURVE CLASSIFIER OPTIMIZED FOR ECLIPSING BINARIES

We describe a new neural-net based light curve classifier and provide it with documentation as a ready-to-use tool for the community. While optimized for identification and classification of eclipsing binary stars, the classifier is general purpose, and has been developed for speed in the context of upcoming massive surveys such as LSST. A challenge for classifiers in the context of neural-net training and massive data sets is to minimize the number of parameters required to describe each light curve. We show that a simple and fast geometric representation that encodes the overall light curve shape, together with a chi-square parameter to capture higher-order morphology information results in efficient yet robust light curve classification, especially for eclipsing binaries. Testing the classifier on the ASAS light curve database, we achieve a retrieval rate of 98\% and a false-positive rate of 2\% for eclipsing binaries. We achieve similarly high retrieval rates for most other periodic variable-star classes, including RR Lyrae, Mira, and delta Scuti. However, the classifier currently has difficulty discriminating between different sub-classes of eclipsing binaries, and suffers a relatively low ($\sim$60\%) retrieval rate for multi-mode delta Cepheid stars. We find that it is imperative to train the classifier's neural network with exemplars that include the full range of light curve quality to which the classifier will be expected to perform; the classifier performs well on noisy light curves only when trained with noisy exemplars. The classifier source code, ancillary programs, a trained neural net, and a guide for use, are provided.