An improved pipeline to detect astrophysical transients in Atacama Cosmology Telescope time-resolved survey data
PI: Charlotte Ward (Astronomy and Astrophysics)
Post-comp graduate student tuition can be covered by the PI’s start-up funding.
Time-domain astronomical surveys image the night sky at daily to weekly cadences, providing terabytes of imaging data about transient and variable phenomena including supernovae, flaring supermassive black holes, and variable stars. While time-domain optical surveys have well-established imaging processing pipelines, time-domain radio and millimeter surveys have only recently become technologically feasible. Scalable imaging processing and transient photometry pipelines are much harder to build for radio and mm imaging: the background is difficult to model and subtract due to convolution artifacts as well as real structure from the cosmic microwave background, modeling the beam shape is hard for low density fields, and robust source detection amidst the complex background is challenging for faint sources (Biermann et al. 2016). Producing reliable light curves for thousands of transients is therefore both a technical and computational challenge. In this project, the junior researcher will build an improved pipeline to extract light curves of low flux sources from multi-epoch imaging from the Atacama Cosmology Telescope (ACT; Thornton et al. 2016). This will involve a) developing and implementing improved source detection and background subtraction algorithms that will not subtract out faint sources, b) incorporating a more sophisticated model for beam shape to improve photometric precision, c) implementing a large scale photometry pipeline on a cluster that can efficiently extract ACT image cutouts, stack them to improve sensitivity, and produce light curves for thousands of sources.
Expertise/skills of interest:
- Software: Programming experience with python, familiarity with Github, SQL. Useful but not required: some experience with astronomical source detection, PSF modeling, astrometric calibration, and stacking software such Swarp, Scamp, PSFEx, and Astropy.
- High performance computing: Ability to develop software for computationally intensive analysis and processing of large data sets on supercomputing clusters. Use of cluster resource management and checkpointing software.
Expectations:
● Post-comps graduate student or postdoc with at least some experience and/or training in: (1) Astronomy & Astrophysics, Physics or a related field; and (2) Applied Math, Computer Sciences, Data Sciences, IST, Statistics or a related field.
● Write code organized with appropriate documentation that will be implemented on the NERSC Perlmutter cluster and released on Github.
● Weekly meeting and project updates with faculty advisor. Participate in group meetings every month. Goal: Develop an improved pipeline to produce light curves of faint transient and variable sources in multi-epoch imaging from the ACT cosmology telescope.
Specific Objectives:
● Using an existing compilation of variable active galactic nuclei (AGN) with low S/N ACT detections, examine how different source detection and background subtraction algorithms perform, beginning with those previously implemented for bright source photometry with ACT.
● Implement a scalable photometry pipeline on a cluster that creates image cutouts for the source of interest, removes the background, and measures flux in each epoch to generate light curves. Make this code available on github so others can use this on the ACT public data releases.
● As time permits, publish a catalog of variable AGN in ACT using this pipeline.
Engagement:
As a new faculty member looking to build connections with ICDS, Ward will participate in ICDS seminars and workshops, and will explore potential collaborations with ICDS co-hires, particularly those who are part of the Center for Astrostatistics & Astroinformatics.