Rotational Light-curve Recovery and Predictions of the LSST Yield of Hildas

Alexander J. Fleming; David Nesvorny; David Vokrouhlicky; Dmitrii E. Vavilov; Jacob A. Kurlander; Mario Juric; Pedro H. Bernardinelli

arxiv: 2512.00624 · v3 · pith:BVXW4SAAnew · submitted 2025-11-29 · 🌌 astro-ph.EP

Rotational Light-curve Recovery and Predictions of the LSST Yield of Hildas

Alexander J. Fleming , Jacob A. Kurlander , Dmitrii E. Vavilov , David Vokrouhlicky , David Nesvorny , Pedro H. Bernardinelli , Mario Juric This is my paper

Pith reviewed 2026-05-25 07:22 UTC · model grok-4.3

classification 🌌 astro-ph.EP

keywords HildasLSSTlight curvesrotational periodsasteroid surveymean-motion resonancesynthetic populationperiodogram

0 comments

The pith

LSST will discover about 33400 Hildas over ten years and recover light curves for nearly 46 percent of them.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper uses the Sorcha simulator to model LSST observations of a synthetic Hilda population that includes orbital elements, sizes, collisional families, and colors. Three sets of sinusoidal light curves are applied, one drawn from the Lightcurve Database via kernel density estimation and two extreme rotator populations. The simulation forecasts discovery of roughly 33400 Hildas, a fivefold increase over the current catalog, and uses a multiband Lomb-Scargle periodogram to recover periods and amplitudes for 45.96 percent of the LCDB-based sample. Recovery efficiency falls sharply for amplitudes below 0.1 magnitudes while showing only weak period dependence except at the survey's 36-minute revisit cadence. The work supplies the first quantitative test of light-curve recovery from simulated LSST data for this resonant population.

Core claim

Over the 10 yr simulated survey, LSST will discover ~33400 Hildas and confidently recover rotational light curves for ~45.96 percent of the LCDB-based population using a multiband Lomb-Scargle periodogram, with recovery strongly biased against amplitudes below 0.1 magnitudes.

What carries the argument

Synthetic Hilda population model fed into the Sorcha survey simulator, followed by multiband Lomb-Scargle periodogram analysis on the resulting time series.

If this is right

A much larger sample of Hildas will become available for studies of collisional evolution and resonance dynamics.
Light-curve recovery will favor objects with amplitudes greater than 0.1 magnitudes.
Cadence features such as the 36-minute revisit will imprint specific period biases.
The quoted recovery fraction is an upper limit because real light curves are not purely sinusoidal.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Early LSST data releases could be used to test and refine the synthetic population before the full decade is complete.
The same simulation framework can be applied to other Jupiter resonances to compare discovery and recovery statistics.
Improved pole-orientation modeling would tighten the predicted recovery rate for future survey planning.

Load-bearing premise

Light curves are modeled as constant sinusoids that correspond to optimal pole orientations.

What would settle it

The actual count of Hildas found by LSST after ten years differs substantially from 33400, or the fraction of objects with confidently recovered light curves falls well below 46 percent.

Figures

Figures reproduced from arXiv: 2512.00624 by Alexander J. Fleming, David Nesvorny, David Vokrouhlicky, Dmitrii E. Vavilov, Jacob A. Kurlander, Mario Juric, Pedro H. Bernardinelli.

**Figure 1.** Figure 1: Comparison of the known Hilda population from the Minor Planet Center (left) and our synthetic Hilda population (right) in semi-major axis (a) and absolute magnitude (H). The vertical line at a = 3.971 AU marks the location of the 3:2 mean-motion resonance with Jupiter, which defines the Hilda population. Our input sample reproduces the observed distribution in both a and H, while extending to fainter magn… view at source ↗

**Figure 2.** Figure 2: Distribution of our 485,807 simulated Hildas projected onto 2D planes of their proper orbital parameters. (i) (a, e) top left; (ii) (a, sin(i)) bottom left; and (iii) (e, sin(i)) bottom right. Each dark point represents an object from our simulated Hilda population. Approximate locations of the two largest collisional families associated with (153) Hilda (blue marker) and (1911) Schubart (red marker) are i… view at source ↗

**Figure 3.** Figure 3: Comparison of simulated and observed Hilda light-curve distributions. This shows a 2D histogram of rotational period versus light-curve amplitude for the combined simulated population, created by sampling periods and amplitudes from the KDE associated with each object’s collisional family. White points indicate Hildas from the LCDB catalog. Our KDE-based sample is consistent with the rotational properties … view at source ↗

**Figure 4.** Figure 4: Comparison of absolute magnitude (H) distributions for the known, input and discovered Hilda populations. The orange histogram shows the currently-known Hildas from the MPC. The blue distribution represents the Hildas discovered in our simulated survey with the gray being the full input population. The synthetic discoveries extend ∼1.5–2 magnitudes fainter than the known population, demonstrating LSST’… view at source ↗

**Figure 5.** Figure 5: Distributions of the reduced inverse power metric R (Equation 1) for the SFR, LCDB and SSR simulated light–curve populations, separated into the correctly and incorrectly recovered rotational periods. The vertical dashed line shows the confidence threshold R = 0.01017, defined as the value below which at least 99% of SFR periods are accurately recovered (including half/double-period harmonics). Correct fit… view at source ↗

**Figure 6.** Figure 6: Discovery fraction of our simulated Hilda bodies for the SFR, LCDB and SSR populations. Each panel displays the per-bin fraction of objects that were detected in our simulated survey, normalized by the overall discovery fraction so that an average bin has a value of 1.0. rates are typically around 65% but have higher variance than the other populations. We see a characteristic dip in recovery efficiency ne… view at source ↗

**Figure 7.** Figure 7: Fraction of light curves meeting the 99% confidence threshold as a function of light-curve amplitude and rotational period. Each bin shows the fraction of objects whose Lomb–Scargle periodogram structure indicates high confidence in the recovered period. Recovery confidence decreases significantly at lower amplitudes, where weaker variability reduces the reliability of period identification. SFR population… view at source ↗

read the original abstract

The Hilda population occupies the stable 3:2 mean-motion resonance of Jupiter and provides a window into solar system evolution, including collisional processes. The National Science Foundation and Department of Energy Vera C. Rubin Observatory will conduct the 10 yr Legacy Survey of Space and Time (LSST). We present a simulation of Rubin's discovery of Hildas with the Sorcha survey simulator and the recovery of their light curves. We constructed a synthetic Hilda population model that includes distributions of orbital properties, sizes, collisional families, and colors. We applied three distinct populations of sinusoidal light curves to this same orbit-size-color model: (1) a Gaussian kernel density estimate fit to rotational periods and amplitudes from the Lightcurve Database (LCDB), (2) a superfast rotator population, and (3) a superslow rotator population. Over the 10 yr simulated survey, we predict LSST will discover ~33,400 Hildas, a fivefold increase over the known population. Using a multiband Lomb-Scargle Periodogram via Astropy we confidently recover ~45.96% of Hildas in our LCDB-based population, higher than typical in observational searches. This suggests our light-curve population model may differ from the intrinsic population. We find strong biases in light-curve amplitude, with recovery efficiency dropping sharply below 0.1 magnitudes, while biases from rotational period are comparatively weak aside from cadence-related features such as LSST's ~36 minute revisit cadence. Our recovery efficiency is likely overestimated due to our assumption of constant sinusoidal light curves, which correspond to optimal pole orientations. These results are the first test of light-curve recovery from simulated LSST observations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper gives concrete LSST yield numbers for Hildas (~33k discoveries) and a 46% recovery rate, but that recovery figure comes from an optimistic sinusoidal model the authors flag as likely too high.

read the letter

The main takeaway is that LSST should find roughly 33,400 Hildas in ten years under their synthetic population, with about 46% of the LCDB-based light-curve sample yielding recoverable periods via multiband Lomb-Scargle. They also show amplitude bias is strong below 0.1 mag while period bias is weaker except at the survey's revisit cadence. That is the new numerical content; the framework itself re-uses Sorcha and Astropy on a Hilda-specific orbit-size-color model with three light-curve variants pulled from LCDB. The setup is transparent and the bias plots are straightforward to read. The authors correctly note that constant sinusoids at optimal poles probably inflate recovery, and they stop short of claiming the 46% is a firm forecast. The limitation is that they do not quantify how much the rate drops under random poles or triaxial shapes, so the headline recovery number stays uncalibrated. No error bars appear on the yield or recovery figures, and there is no direct comparison to existing Hilda light curves. The work is aimed at survey planners and resonant-population modelers who need order-of-magnitude expectations rather than precision forecasts. It is honest about its assumptions and uses public tools, so it is worth sending to referees who can ask for the missing sensitivity runs. I would bring it to a reading group focused on LSST asteroid pipelines.

Referee Report

3 major / 2 minor

Summary. The manuscript simulates LSST discovery of Hildas with Sorcha, building a synthetic population incorporating orbits, sizes, families, and colors. Three sinusoidal light-curve populations (LCDB KDE, superfast, superslow) are injected; the simulation predicts ~33,400 discoveries (fivefold increase) and ~45.96% recovery via multiband Lomb-Scargle for the LCDB-based set, with amplitude biases noted and an explicit caveat that constant-sinusoid optimal-pole assumptions likely overestimate recovery.

Significance. If the overestimation can be bounded, the work supplies the first end-to-end simulation of LSST Hilda yields and light-curve recovery, identifying clear amplitude-dependent biases that would be useful for survey planning and population studies.

major comments (3)

[Abstract / Results] Abstract and main results: the headline 45.96% recovery efficiency is obtained by injecting constant-sinusoid light curves (LCDB KDE plus superfast/superslow populations) that the authors themselves state correspond to optimal pole orientations and therefore overestimate efficiency; no alternative realizations (random poles, triaxial ellipsoids, or non-sinusoidal variability) are run to quantify the size of the overestimate, so the reported percentage cannot be treated as a calibrated prediction.
[Abstract / Methods] Discovery yield: the central prediction of ~33,400 Hildas over 10 yr is presented without uncertainty estimates, sensitivity tests to the free parameters of the synthetic population, or validation against the known Hilda sample, leaving the fivefold-increase claim difficult to assess for robustness.
[Methods] Recovery method: the multiband Lomb-Scargle implementation via Astropy is described at a high level only; the precise detection threshold, handling of the 36-minute revisit cadence, and definition of “confident” recovery are not specified, preventing evaluation of how these choices interact with the acknowledged sinusoidal assumption.

minor comments (2)

The abstract would be strengthened by reporting ranges or standard deviations on the key numerical results (33,400 and 45.96%) to reflect simulation variability.
A brief comparison table or sentence placing the simulated recovery rate against published recovery fractions from existing surveys would help contextualize the claim that it is “higher than typical.”

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for their constructive comments, which highlight areas where the manuscript can be strengthened. We address each major comment below and indicate where revisions will be made.

read point-by-point responses

Referee: [Abstract / Results] Abstract and main results: the headline 45.96% recovery efficiency is obtained by injecting constant-sinusoid light curves (LCDB KDE plus superfast/superslow populations) that the authors themselves state correspond to optimal pole orientations and therefore overestimate efficiency; no alternative realizations (random poles, triaxial ellipsoids, or non-sinusoidal variability) are run to quantify the size of the overestimate, so the reported percentage cannot be treated as a calibrated prediction.

Authors: We agree that the 45.96% figure is an optimistic upper bound, as already stated in the abstract and discussion section. The constant-sinusoid model with optimal poles was chosen to provide a best-case estimate under simplified assumptions. We cannot run the requested alternative realizations (random poles, triaxial ellipsoids) within the scope of a revision, as they would require new large-scale simulations. We will revise the abstract and results to more explicitly frame the percentage as an upper limit and add discussion of how random orientations and non-sinusoidal shapes would reduce efficiency, drawing on existing literature. revision: partial
Referee: [Abstract / Methods] Discovery yield: the central prediction of ~33,400 Hildas over 10 yr is presented without uncertainty estimates, sensitivity tests to the free parameters of the synthetic population, or validation against the known Hilda sample, leaving the fivefold-increase claim difficult to assess for robustness.

Authors: We accept this criticism. The revised manuscript will add uncertainty estimates based on variations in model parameters, include sensitivity tests for key inputs such as size distribution and family membership, and validate the synthetic population by comparing the number and orbital properties of simulated 'known' Hildas against the real observed sample. revision: yes
Referee: [Methods] Recovery method: the multiband Lomb-Scargle implementation via Astropy is described at a high level only; the precise detection threshold, handling of the 36-minute revisit cadence, and definition of “confident” recovery are not specified, preventing evaluation of how these choices interact with the acknowledged sinusoidal assumption.

Authors: We will expand the Methods section to specify the false-alarm probability threshold, the exact treatment of the 36-minute cadence in the time sampling, and the quantitative criteria used to define a 'confident' recovery (recovered period within a stated tolerance of the input value). These details will clarify the interaction with the sinusoidal light-curve model. revision: yes

standing simulated objections not resolved

Quantifying the precise size of the recovery-efficiency overestimate via new simulations with random pole orientations or triaxial ellipsoids, which would require substantial additional computational resources beyond the current study.

Circularity Check

0 steps flagged

No circularity; forward simulation from external database and tools yields independent outputs

full rationale

The derivation applies an LCDB-derived KDE for periods/amplitudes, plus orbital and color models, as inputs to the external Sorcha simulator to produce synthetic LSST observations; recovery fractions and discovery counts are then computed via Astropy multiband Lomb-Scargle periodogram on those simulated data. None of the headline statistics (33,400 discoveries, 45.96% recovery) reduce by construction to the input distributions or any self-citation chain; the sinusoidal assumption is flagged by the authors as a limitation but does not create a definitional or fitted-input loop.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The simulation rests on fitted distributions for orbits/sizes/colors/families from prior data, three ad-hoc light-curve populations, and the sinusoidal shape assumption; these are the main inputs not independently derived in the paper.

free parameters (2)

LCDB-based light-curve distributions = KDE from LCDB
Gaussian kernel density estimate fit to rotational periods and amplitudes from LCDB
synthetic Hilda population parameters
Distributions of orbital properties, sizes, collisional families, and colors

axioms (2)

domain assumption Hilda asteroids occupy stable 3:2 mean-motion resonance with Jupiter
Standard background fact in solar system dynamics invoked to define the population
ad hoc to paper Light curves can be modeled as constant sinusoids
Explicit modeling choice for the three populations used in recovery simulation

pith-pipeline@v0.9.0 · 5876 in / 1477 out tokens · 56847 ms · 2026-05-25T07:22:31.521118+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages

[1]

P., Tollerud, E

Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f Bowell, E., Hapke, B., Domingue, D., et al. 1989, in Asteroids II, ed. R. P. Binzel, T. Gehrels, & M. S. Matthews, 524–...

work page doi:10.1051/0004-6361/201322068 2013
[2]

2010, in Proceedings of the 9th Python in Science Conference, ed

http://www.MinorPlanet.info/php/lcdb.php Wes McKinney. 2010, in Proceedings of the 9th Python in Science Conference, ed. St´ efan van der Walt & Jarrod Millman, 56 – 61, doi: 10.25080/Majora-92bf1922-00a Wong, I., & Brown, M. E. 2017, in AGU Fall Meeting Abstracts, Vol. 2017, AGU Fall Meeting Abstracts, P33G–07 Yoachim, P., & Jones, L. 2025, lsst-sims/sim...

work page doi:10.25080/majora-92bf1922-00a 2010

[1] [1]

P., Tollerud, E

Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f Bowell, E., Hapke, B., Domingue, D., et al. 1989, in Asteroids II, ed. R. P. Binzel, T. Gehrels, & M. S. Matthews, 524–...

work page doi:10.1051/0004-6361/201322068 2013

[2] [2]

2010, in Proceedings of the 9th Python in Science Conference, ed

http://www.MinorPlanet.info/php/lcdb.php Wes McKinney. 2010, in Proceedings of the 9th Python in Science Conference, ed. St´ efan van der Walt & Jarrod Millman, 56 – 61, doi: 10.25080/Majora-92bf1922-00a Wong, I., & Brown, M. E. 2017, in AGU Fall Meeting Abstracts, Vol. 2017, AGU Fall Meeting Abstracts, P33G–07 Yoachim, P., & Jones, L. 2025, lsst-sims/sim...

work page doi:10.25080/majora-92bf1922-00a 2010