ATAlex Teachey

Alex Teachey

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Alex Teachey has a total of 101 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A JWST Transit of a Jupiter Analog. II. A Search for Exomoons

    by David Kipping, Ben Cassese, Quentin Changeat, Daniel Yahalomi, Alex Teachey, Billy Edwards

    We present a search for exomoons around the Jupiter-like exoplanet Kepler-167e using a NIRSpec JWST transit. Our 60 hr time series clearly reveals the enormous impact of long-term trends in NIRSpec data, specifically a gradual flux drift occurring over each of six 10 hr exposures. We weighed the evidence for exomoons by comparing a planet-only model with a planet+moon model for a grid of twelve different analysis choices. Our grid was comprised of three different reduction pipelines and four different models for the exposure-long trends—two using linear models and two using Gaussian processes. Seven grid realizations indicate a strong exomoon detection, typically favoring a Roche-skimming orbit roughly 10% the size of the planet. We find that the only likely real astrophysical feature driving these fits is a syzygy-like event occurring almost exactly mid-transit, which is highly ambiguous with a spot-crossing event. Indeed, we show that a spot of the necessary size is compatible with the earlier Kepler data. Ironically, the fact that JWST is so superior to Kepler means that our fits are effectively driven by a single transit—a regime in which exomoons have enormous freedom to explain non-Gaussian behavior. We thus strongly urge the next transit be observed in 2027 October to break these degeneracies. Our pilot study to seek transiting exomoons with JWST reveals the profound impact exposure-long trends exert—a cautionary tale for future analyses of these data—as well as the need for a deeper understanding of the cause of these systematics and modeling best practices.


  • Open Access

    The democratic detrender : Ensemble-based Removal of the Nuisance Signal in Stellar Time-series Photometry

    by Daniel A. Yahalomi, David Kipping, Diana Solano-Oropeza, Madison Li, Avishi Poddar, Xunhe (Andrew) Zhang, Yassine Abaakil, Ben Cassese, Jeff Jennings, Skylar Larsen, et al.

    Accurate, precise, and computationally efficient removal of unwanted activity that exists as a combination of periodic, quasiperiodic, and nonperiodic systematic trends in time-series photometric data is a critical step in exoplanet transit analysis. Throughout the years, many different modeling methods have been used for this process, often called “detrending.” However, there is no community-wide consensus regarding the favored approach. In order to mitigate model dependency, we present an ensemble-based approach to detrending via a community of models and the democratic detrender : a modular and scalable open-source coding package that implements ensemble detrending. The democratic detrender allows users to select from a number of packaged detrending methods (including cosine filtering, Gaussian processes, and polynomial fits) or provide their own set of detrended light curves via methods of their choosing. It then combines the individually detrended light curves into a single method marginalized light curve. Additionally, the democratic detrender inflates each data point’s uncertainty based on the scatter between detrenders, thereby propagating model-selection uncertainty into the final light curve. This ensemble strategy does not guarantee improvement over the single best-performing detrending method, but it substantially reduces the risk of selecting a detrending solution that is poorly calibrated or overfit to noise.

  • Open Access

    A JWST Transit of a Jupiter Analog. I. Constraints on the Oblateness of Kepler-167 e

    by Ben Cassese, David Kipping, Quentin Changeat, Daniel A. Yahalomi, Justin Vega, Yayaati Chachan, Billy Edwards, Alex Teachey

    In 2024 October, JWST observed a transit of Kepler-167 e, a Jupiter-analog planet on a 1000+ day orbit. These observations, recorded over a long baseline of nearly 60 hr, were designed to search for signatures of planetary oblateness and/or exomoons comparable to Ganymede. In this first in a series of studies analyzing these data, we report on constraints on Kepler-167 e’s oblateness. We explored a large grid of data reduction pipelines and modeling choices, including a new entirely independent reduction pipeline (“ katahdin ”) and two new treatments for limb darkening. We find that under a Bayesian model comparison framework the data are fit equally well by both spherical and oblate planet models, and that our ability to constrain the oblateness is negatively impacted by the influence of exposure-long trends. Using the most conservative of our posteriors, we place a 95% upper bound on the projected oblateness of f < 0.097, which corresponds to a rotation period of P ≥ 7.11 hr if the planet’s spin axis is aligned with the sky plane. We note, however, that the final bound depends on the choice of reduction pipeline and systematics model, and that our suite of end-to-end analyses produced bounds as low as f < 0.065 at 95%. We conclude that leveraging JWST to make tighter constraints on planetary oblateness will require further investigation into mitigating exposure-long trends and correlated noise.


  • Open Access

    A JWST Transit of a Jupiter Analog: II. A Search for Exomoons

    by David Kipping, Ben Cassese, Quentin Changeat, Daniel Yahalomi, Alex Teachey, Billy Edwards

    We present a search for exomoons around the Jupiter-like exoplanet Kepler-167e using a NIRSpec JWST transit. Our 60 hour time series clearly reveals the enormous impact of long-term trends in NIRSpec data, specifically a gradual flux drift occurring over each of six 10 hour exposures. We weighed the evidence for exomoons by comparing a planet-only model with a planet+moon model for a grid of twelve different analysis choices. Our grid was comprised of three different reduction pipelines and four different models for the exposure-long trends - two using linear models and two using Gaussian processes. Seven grid realizations indicate a strong exomoon detection, typically favoring a Roche-skimming orbit roughly 10% the size of the planet. We find that the only likely real astrophysical feature driving these fits is a syzygy-like event occurring almost exactly mid-transit, which is highly ambiguous with a spot-crossing event. Indeed, we show that a spot of the necessary size is compatible with the earlier Kepler data. Ironically, the fact that JWST is so superior to Kepler means that our fits are effectively driven by a single transit - a regime in which exomoons have enormous freedom to explain non-Gaussian behavior. We thus strongly urge the next transit be observed in October 2027 to break these degeneracies. Our pilot study to seek transiting exomoons with JWST reveals the profound impact exposure-long trends exert - a cautionary tale for future analyzes of this data - as well as the need for a deeper understanding of this systematic's cause and modeling best-practices.


  • Open Access

    On the impact and utility of single-exomoon modelling for multimoon systems

    by Alex Teachey, Garvit Agarwal

    The search for exomoons in time-domain photometric data has to-date generally consisted of fitting transit models that are comprised of a planet hosting a single moon. This simple model has its advantages, but it may not be particularly representative, as most of the major moons in our Solar system are found in multimoon satellite systems. It is critical that we investigate, then, the impact of applying a single-moon model to systems containing multiple moons, as there is the possibility that utilizing an inaccurate or incomplete model could lead to erroneous conclusions about the system. To that end, in this work we produce a variety of realistic multimoon light curves, perform standard single-moon model selection, and analyse the impacts that this model choice may have on the search for exomoons. We find that the number of moons in a system fit with a single-moon model generally has little impact on whether we find evidence for a moon in that system, and other system attributes are individually not especially predictive. However, the model parameter solutions for the moon frequently do not match any real moon in the system, instead painting a picture of a ‘phantom’ moon. We find no evidence that multimoon systems yield corresponding multimodal posteriors. We also find a systematic tendency to overestimate planetary impact parameter and eccentricity, to derive unphysical moon densities, and to infer potentially unphysical limb darkening coefficients. These results will be important to keep in mind in future exomoon search programmes.


  • Open Access

    On the Impact and Utility of Single-Exomoon Modeling for Multi-Moon Systems

    by Alex Teachey, Garvit Agarwal

    The search for exomoons in time-domain photometric data has to-date generally consisted of fitting transit models that are comprised of a planet hosting a single moon. This simple model has its advantages, but it may not be particularly representative, as most of the major moons in our Solar System are found in multi-moon satellite systems. It is critical that we investigate, then, the impact of applying a single-moon model to systems containing multiple moons, as there is the possibility that utilizing an inaccurate or incomplete model could lead to erroneous conclusions about the system. To that end, in this work we produce a variety of realistic multi-moon light curves, perform standard single-moon model selection, and analyze the impacts that this model choice may have on the search for exomoons. We find that the number of moons in a system fit with a single-moon model generally has little impact on whether we find evidence for a moon in that system, and other system attributes are individually not especially predictive. However, the model parameter solutions for the moon frequently do not match any real moon in the system, instead painting a picture of a ``phantom'' moon. We find no evidence that multi-moon systems yield corresponding multi-modal posteriors. We also find a systematic tendency to overestimate planetary impact parameter and eccentricity, to derive unphysical moon densities, and to infer potentially unphysical limb darkening coefficients. These results will be important to keep in mind in future exomoon search programs.

  • Open Access

    Giant Outer Transiting Exoplanet Mass (GOT ‘EM) Survey. IV. Long-term Doppler Spectroscopy for 11 Stars Thought to Host Cool Giant Exoplanets

    by Paul A. Dalba, Stephen R. Kane, Howard Isaacson, Benjamin Fulton, Andrew W. Howard, Edward W. Schwieterman, Daniel P. Thorngren, Jonathan Fortney, Noah Vowell, Corey Beard, et al.

    Discovering and characterizing exoplanets at the outer edge of the transit method’s sensitivity has proven challenging owing to geometric biases and the practical difficulties associated with acquiring long observational baselines. Nonetheless, a sample of giant exoplanets on orbits longer than 100 days has been identified by transit hunting missions. We present long-term Doppler spectroscopy for 11 such systems with observation baselines spanning a few years to a decade. We model these radial velocity observations jointly with transit photometry to provide initial characterizations of these objects and the systems in which they exist. Specifically, we make new precise mass measurements for four long-period giant exoplanets (Kepler-111 c, Kepler-553 c, Kepler-849 b, and PH-2 b), we place new upper limits on mass for four others (Kepler-421 b, KOI-1431.01, Kepler-1513 b, and Kepler-952 b), and we show that several confirmed planets are in fact not planetary at all. We present these findings to complement similar efforts focused on closer-in short-period giant planets, and with the hope of inspiring future dedicated studies of cool giant exoplanets.


  • Open Access

    Giant Outer Transiting Exoplanet Mass (GOT 'EM) Survey. IV. Long-term Doppler Spectroscopy for 11 Stars Thought to Host Cool Giant Exoplanets

    by Paul A. Dalba, Stephen R. Kane, Howard Isaacson, Benjamin Fulton, Andrew W. Howard, Edward W. Schwieterman, Daniel P. Thorngren, Jonathan Fortney, Noah Vowell, Corey Beard, et al.

    Discovering and characterizing exoplanets at the outer edge of the transit method's sensitivity has proven challenging owing to geometric biases and the practical difficulties associated with acquiring long observational baselines. Nonetheless, a sample of giant exoplanets on orbits longer than 100 days has been identified by transit hunting missions. We present long-term Doppler spectroscopy for 11 such systems with observation baselines spanning a few years to a decade. We model these radial velocity observations jointly with transit photometry to provide initial characterizations of these objects and the systems in which they exist. Specifically, we make new precise mass measurements for four long-period giant exoplanets (Kepler-111 c, Kepler-553 c, Kepler-849 b, and PH-2 b), we place new upper limits on mass for four others (Kepler-421 b, KOI-1431.01, Kepler-1513 b, and Kepler-952 b), and we show that several "confirmed" planets are in fact not planetary at all. We present these findings to complement similar efforts focused on closer-in short-period giant planets, and with the hope of inspiring future dedicated studies of cool giant exoplanets.

  • Open Access

    Identifying potential exomoon signals with convolutional neural networks

    by Alex Teachey, David Kipping

    Targeted observations of possible exomoon host systems will remain difficult to obtain and time-consuming to analyse in the foreseeable future. As such, time-domain surveys such as Kepler, K2, and TESS will continue to play a critical role as the first step in identifying candidate exomoon systems, which may then be followed up with premier ground- or space-based telescopes. In this work, we train an ensemble of convolutional neural networks (CNNs) to identify candidate exomoon signals in single-transit events observed by Kepler. Our training set consists of ∼27 000 examples of synthetic, planet-only, and planet + moon single transits, injected into Kepler light curves. We achieve up to 88 per cent classification accuracy with individual CNN architectures and 97 per cent precision in identifying the moons in the validation set when the CNN ensemble is in total agreement. We then apply the CNN ensemble to light curves from 1880 Kepler Objects of Interest with periods >10 d (∼57 000 individual transits), and further test the accuracy of the CNN classifier by injecting planet transits into each light curve, thus quantifying the extent to which residual stellar activity may result in false positive classifications. We find a small fraction of these transits contain moon-like signals, though we caution against strong inferences of the exomoon occurrence rate from this result. We conclude by discussing some ongoing challenges to utilizing neural networks for the exomoon search.


  • The exomoon corridor for multiple moon systems

    by Alex Teachey

    Recently Kipping 2021 identified the so-called “exomoon corridor”, a potentially powerful new tool for identifying possible exomoon hosts, enabled by the observation that fully half of all planets hosting an exomoon will exhibit transit timing variation (TTV) periodicities of 2-4 epochs. One key outstanding problem in the search for exomoons, however, is the question of how well the methods we have developed under the single moon assumption extend to systems with multiple moons. In this work we use N-body simulations to examine the exomoon corridor effect in the more general case of N ≥ 1 moons, generating realistic TTVs produced by satellite systems more akin to those seen in the outer Solar System. We find that indeed the relationship does hold for systems with up to 5 moons in both resonant and non-resonant chain configurations. Our results suggest an observational bias against finding systems with large numbers of massive moons; as the number of moons increases, total satellite mass ratios are generally required to be significantly lower in order to maintain stability, or architectures must be more finely tuned to survive. Moons produced in impact or capture scenarios may therefore dominate early detections. Finally, we examine the distribution of TTV periods measured for a large number of Kepler objects of interest (KOIs) and find the same characteristic exomoon corridor distribution in several cases. This could be dynamical evidence for an abundance of moons in the field, though we caution against strong inferences based on this result.