| Speaker: | Chris Storfer (U Hawaii) (H Hawaii) |
|---|---|
| Title: | From Discovery to Science: An End-to-End Pipeline for Analyzing Strong Gravitational Lenses |
| Date (JST): | Thu, Aug 06, 2026, 11:00 - 12:00 |
| Place: | Seminar Room A |
| Abstract: |
Strong gravitational lensing has proven to be one of the most powerful tools for cosmological and dark matter investigations, but realizing its full potential requires large samples with high-quality data and robust analysis pipelines. In this talk I will detail my dissertation work, which seeks to measure the mass density profile slope across a population-level sample of elliptical galaxy lenses at 0.6 < z < 1.0, a relatively unexplored regime in which the slope traces galaxy assembly and merger history. To enable this we have built an end-to-end pipeline spanning discovery, spectroscopic follow-up, and robust lens modeling. We construct a sample of thousands of new strong lenses with diverse properties by conducting systematic searches in wide-field imaging from the Ultraviolet Near Infrared Optical Northern Survey (UNIONS) and Dark Energy Spectroscopic Instrument (DESI) Legacy Surveys using machine learning. Critical spectroscopic follow-up and redshift determination for these systems with DESI has been critical and highly successful, complemented by near-infrared (NIR) observations from Keck. Looking ahead, the Prime Focus Spectrograph (PFS) uniquely offers optical and NIR coverage with multiplexing which is extraordinarily powerful in measuring redshifts for many strong lensing systems. For modeling, we rely on Euclid whose homogeneous high-resolution imaging is essential for extracting accurate physical parameters from strong lenses. With hundreds or even thousands of systems to model, we use GIGA-Lens, a gradient-informed and GPU accelerated lens modeling software which makes large-scale modeling tractable. This pipeline will yield a large sample of spectroscopically complete strong lenses with uniform models. The sample will enable not only our science goals, but a broad range of downstream lensing science, serving as a lasting legacy data product. |
