PI: James Nightingale
The COSMOS-Web Lens Survey (COWLS) uses deep imaging from the James Webb Space Telescope (JWST) to discover and study strong gravitational lenses across the 0 . 54 deg 2 COSMOS-Web field. In our recent work (arXiv:2503.08777) , we present a sample of over 100 strong lens candidates identified in JWST/NIRCam data. These systems include some of the highest-redshift lens galaxies known and sources extending into the epoch of reionisation, demonstrating the unique capability of JWST to reveal gravitational lensing at high redshift and with exceptional image quality. High performance computing resources on the DiRAC Data-Centric system and CSD3 were used to run large numbers of automated strong lens models using the PyAutoLens software. Each candidate lens is modelled by fitting the mass distribution of the foreground galaxy and reconstructing the lensed background source. Running these analyses across the survey required many independent Bayesian inference jobs, making DiRAC essential for efficiently analysing the full lens sample and producing reproducible modelling results.
Links.
- COWLS I (paper): https://arxiv.org/abs/2503.08777
- COWLS II (paper): https://arxiv.org/abs/2503.08782
- Public release repository: https://github.com/Jammy2211/COWLS_COSMOS_Web_Lens_Survey
Figure 1: RGB images of some of the most visually striking strong grav- itational lenses discovered in the COSMOS-Web field (from COWLS II, arXiv:2503.08782).