A targeted search for strongly lensed supernovae and expectations for targeted searches in the Rubin era

Résumé : Gravitationally lensed supernovae (glSNe) are of interest for time delay cosmology and SN physics. However, glSNe detections are rare, owing to the intrinsic rarity of SN explosions, the necessity of alignment with a foreground lens, and the relatively short window of detectability. We present the Las Cumbres Observatory Lensed Supernova Search, LCOLSS, a targeted survey designed for detecting glSNe in known strong-lensing systems. Using cadenced $r^\prime$-band imaging, LCOLSS targeted 112 galaxy-galaxy lensing systems with high expected SN rates, based on estimated star formation rates. No plausible glSN was detected by LCOLSS over two years of observing. The analysis performed here measures a detection efficiency for these observations and runs a Monte Carlo simulation using the predicted supernova rates to determine the expected number of glSN detections. The results of the simulation suggest an expected number of detections and $68\%$ Poisson confidence intervals, $N_{SN} = 0.20, [0,2.1] $, $N_{Ia} = 0.08, [0,2.0]$, $N_{CC} = 0.12, [0,2.0]$, for all SN, Type Ia, and core-collapse (CC) SNe respectively. These results are broadly consistent with the absence of a detection in our survey. Analysis of the survey strategy can provide insights for future efforts to develop targeted glSN discovery programs. We thereby forecast expected detection rates for the Rubin observatory for such a targeted survey, finding that a single visit depth of 24.7 mag with the Rubin observatory will detect $0.63 \pm 0.38$ SNe per year, with $0.47 \pm 0.28$ core collapse SNe per year and $0.16 \pm 0.10$ Type Ia SNe per year.