Quantifying the Contamination of TESS Ecliptic-Plane Light Curves by Minor Planets

TL;DR

This study addresses foreground contamination of TESS ecliptic-plane time-series photometry by minor planets. It develops a forward-modeling pipeline that uses jorbit to postdict all minor-planet approaches to known high-cadence targets and to generate model Simple Aperture Photometry light curves, incorporating a realistic pixel response and sector-specific apertures. The results reveal pervasive contamination: the majority of targets experience a close minor-planet approach within a single pixel, with many faint targets encountering contaminant flux levels at or above one percent of the target flux over a sector, and interaction timescales spanning a few hours. The work provides a practical framework to pre- or postdict such contamination, enabling robust interpretation of the archival TESS data and guiding mitigation strategies for future time-domain analyses in the solar neighborhood.

Abstract

Though missions devoted to time series photometry focus primarily on targets far beyond the solar system, their observations can be contaminated by foreground minor planets, especially near the ecliptic plane where solar system objects are most prevalent. Crucially, depending on one's choice of data reduction/background estimation algorithm, these objects can induce both apparent brightening and/or dimming events in processed light curves. To quantify the impact of these objects on archived TESS light curves, we used N-body integrations of all currently known minor planets to postdict all 600,000+ of their interactions with stars selected for high-cadence observations during TESS ecliptic plane sectors. We then created mock images of these moving sources and performed simple aperture photometry using the same target and background apertures used in SPOC processing. Our resulting 10,000+ target-specific light curves, which faithfully model the time-dependent positions and magnitudes of the actual solar system objects that approached each target, reveal that $>95\%$ of high-cadence ecliptic plane targets experience a minor planet crossing within 1 TESS pixel of the source. Additionally, 50% of all $T>13$ mag targets experience at least one instantaneous moment where the contaminating flux from minor planets exceeds 1% of the target flux. We discuss these population-level results and others, and highlight several case studies of bright flybys.

Figures (13)

• Figure 1: The distribution of differences in sky position for randomly sampled minor planets as seen from TESS and the geocenter in January 2025. As described in Sec \ref{['sub:jorbit_tess_application']}, we used this biased distribution to set our search radius. The outlier NEO (289227) 2004 XY60 was the only randomly sampled object with a parallax $>15\arcmin$.
• Figure 2: An illustration of our modeling procedure for TIC 81720355 ($T=14.88$ mag), Sector 45. Top left: jorbit's predictions for locations of all minor planets surrounding the target at one specific time. The dotted square has a side length of 3.85, the width of the TPF. Throughout the sector, jorbit flagged and created ephemerides for 174 minor planets that passed within 15 of the target as seen from the geocenter. The postprocessing step revealed that only 42 of these minor planets, including (4628) Laplace ($H=11.27$, $T=14.87$ mag at closest approach), actually contributed flux to the extraction or background apertures set by SPOC. Top right: the conversion from the predicted minor planet locations to a model image. The SPOC target extraction aperture is marked with the red hatched pixel, while the SPOC background aperture is marked with the blue hatched pixels. The pink arrow shows the approximate instantaneous on-sky velocity vector. Note at this time slice (4628) Laplace contributes to the background aperture but not the extraction aperture, and that it is headed for another cluster of background pixels. This panel time corresponds to the moment of lowest predicted model flux in the lower two panels. Middle: the full Sector 45 SAP light curve for TIC 81720355 (retrieved via lightkurve) and our model, overplotted. Note that this is a prediction, not a fit: our baseline, which was determined by converting the reported magnitude of the star to an expected flux, agrees with the average of the sector but not the individual halves. Bottom: A zoom-in of the middle panel around the 16-hour gray shaded region corresponding to the flyby of (4628) Laplace. This time, we add a constant offset to our model for visual clarity but otherwise do not scale the magnitude of the deviations. We see two separated "dips" in the target flux, corresponding to the two times (4628) Laplace crossed a cluster of pixels in the background aperture. Although the target star (likely) did not actually grow fainter during this time, it appeared so relative to the temporarily enhanced background. We do not see a major positive excursion since (4628) Laplace's trajectory never brings the core of its PSF near the red extraction aperture. The small positive excursion corresponds to when the edges of the PSF spill into the red hatched pixel.
• Figure 3: A poor postdiction for the interaction between TIC 388814538 and asteroid (59) Elpis in Sector 43. The offset between peak times is likely the result of slight WCS inaccuracies, while the offset in peak heights could be due to WCS inaccuracies shifting the centroid of the model PSF or inaccurate model flux predictions (see Sec. \ref{['sub:caveats']} and Appendix \ref{['appendix:mag_errs']}).
• Figure 4: Top: the number of minor planets that interacted with either the target or background extraction apertures on each target-specific TPF. The apparent overdensity near RA=$180^\circ$ is an observational effect explained in Appendix \ref{['appendix:overdensity']}. Middle: The brightest cumulative, instantaneous $T$ mag from all minor planets crossing the TPF recorded in either the target or background aperture. Bottom: The smallest distance between the target and a minor planet during a given sector. Note that the colorbar cuts off at 5 although the distribution continues out further, as seen by the subplot to the right.
• Figure 5: Left: The maximum instantaneous magnitude of the sum of all minor planets within TPF apertures compared to their target TIC's magnitude. Note that the contours of constant flux ratio, thanks to the logarithmic definition of magnitude, are logarithmically spaced: faint targets can be overwhelmed by minor planets, (e.g. the $T=17.4$ TIC 610941254, which in Sector 70, was passed by the $T=9.0$ mag minor planet (29) Amphitrite, the lowest point on the panel), while bright targets are comparatively unaffected. The bimodality of target $T$ mags is simply a product of which target stars were accepted as 20 s targets, mostly from General Investigator proposals. Right: The same data as left, but now binned and shown as a cumulative distribution. A vertical dotted line corresponding to 1% contamination (minor planet flux/target flux) and a horizontal dotted line corresponding to 50% of targets within a bin are included as visual aids. Note that this definition of contamination is the brightest moment in either the extraction or background aperture, and that although it was computed by considering contributions from all minor planets, the brightest one typically dominates all others.