A GPU Code for Finding Microlensing Critical Curves and Caustics

TL;DR

This paper introduces a GPU-accelerated code that efficiently locates caustics in gravitational microlensing simulations, aiding in the analysis of microlensed quasars and supernovae.

Contribution

The paper presents a novel GPU-based computational tool for finding microlensing critical curves and caustics, improving simulation speed and accuracy.

Findings

The code efficiently locates caustics in microlensing maps.

It helps determine the number of caustic crossings.

It constrains parameters like source position and velocity.

Abstract

Advancements in analyses of caustic crossing events in gravitationally microlensed quasars and supernovae can benefit from numerical simulations which locate the caustics in conjunction with the creation of magnification maps. We present a GPU code which efficiently solves this problem; the code is available at https://github.com/weisluke/microlensing/. We discuss how the locations of the microcaustics can be used to determine the number of caustic crossings and the distances to caustics, both of which can be used to constrain the space of nuisance parameters such as source position and velocity within magnification maps.