Enabling Science Program 2021 Award Recipients

“We will develop a machine learning framework to identify the faint, slow-moving objects in the presence of systematics from artificial satellites and reconstruct their trajectories in order to reveal their dynamics and bulk characteristics. Our machine learning pipeline will enable the characterization of the underlying occurrence of long-period SS bodies using LSST.”

This award partially funded the virtual research program RECA Internship 2022, which seeks to give research experience to undergraduate students from Colombian institutions in astronomy, physics, engineering, and related areas to help further their careers and make them more competitive for graduate school in astronomy internationally. The students worked on 5 projects related to LSST science and their results presented in a report and a Jupyter notebook that will serve as tutorial for future LSST users. In addition to gaining research experience, the students  also participated in technical development activities (e.g., a one-week Python bootcam, remote observing at telescopes in the Canary Islands) as well as professional development lectures (including talk as about equity, diversity, and inclusion in astronomy and the state of astronomy in Colombia). This astronomy research internship for undergraduate students is one of the first of its kind (paid) in Colombia, if not the first, and served to promote the development of astronomy in Colombia and South America and strengthen the links between Rubin/LSST and the South American astronomical community in preparation for Rubin operations and LSST science.

“We will test methods to robustly detect and classify EBs in LSST light curves.  We will first adapt an existing pre-classifier that detects and flags EB candidates in TESS light curves, using the Photometric LSST Astronomical Time Series Classification Challenge (PLAsTiCC) dataset to optimize the code's metrics and thresholds for LSST data.  We will also use the synthetic EBs in the PLAsTiCC data to build out the architecture for a Convolutional Neural Network that can detect EBs in the LSST dataset, to be ready for training with LSST commissioning data.”

"We are working to develop new tools which jointly model the intrinsic alignment of galaxies, photometric redshift uncertainties and galaxy bias using shared parameters. This will enable more robust cosmological constraints with LSST data"

"Ismael works on two complementary approaches to mitigating the blending problem in LSST DESC analyses, namely: (1) community software to train and evaluate deblending algorithms, and (2) a fully probabilistic algorithm that enables uncertainty propagation in galaxy measurements. The proposal encompasses a v1.0 code release for the Blending Toolkit (a DESC tool), the journal publication and code release of BLISS (a deep neural network approach to probabilistic deblending), and a DESC note illustrating BLISS incorporated into the Blending Toolkit."

"We propose to explore how the Rubin survey can be used to search for microlensing events,   articularly free-floating black holes in the MIlky Way when they gravitational lens a background star.  We will estimate the impact of various survey scenarios on microlensing searches, using population synthesis models to estimate candidate black hole yields with Rubin, and testing microlensing event identification pipelines from ZTF on simulated Rubin data."

"We will adapt and use LSST-DESC tools such as TXPipe, Core Cosmology Library in order to extract intrinsic alignments of galaxies from precursory data sets such as DES and KIDS surveys. Removing this astrophysical systematic will provide more accurate measurements of cosmic shear from LSST."

"Characterizing the performance of the LSST Science Pipelines in crowded stellar fields is critical to Milky Way and Local Volume science with Rubin Observatory. We are simulating and fitting images of pairs of point sources with varying separations, flux ratios, and background levels to produce fitting formulae for impact of blending in crowded stellar fields"

"An LSST Observing Strategy Metric for Photometric Redshifts Melissa Graham and Murtaza Jafry, University of Washington. This student research project will build on existing software to provide an analysis of the photometric redshift quality expected from various LSST observing strategies that is robust, detailed, and presented in a journal publication. This work will also produce PZ-related software that will be available for all to use and should ultimately help optimize the extragalactic science impact of the LSST."