Forward modeling of dust and transients - a method for the generation of synthetic Light Echoes

Light Echoes (LEs), the reflection of cosmic explosions on interstellar dust, are rare, faint, and extended variable sources that exhibit a variety of shapes, making them one of the hardest astrophysical transients to detect. Their detection offers unique opportunities for the study of transients and dust, but there faintness, and variate, evolving morphology are a challenge. The cadence of the upcoming Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST) combined with its large amount of data per night can enable the detection of rare and faint transients and variable objects. But the LSST transient detection pipeline is designed for point sources and will miss extended variable sources entirely, Artificial Intelligence (AI) techniques are suitable to deal with the challenge of detecting LEs. However, a large amount of data is needed to train a successful AI model. I will introduce the theory behind the generation of LE observable properties with the final goal of producing a realistic collection of simulated LE images that will be part of the training data set of AI models for the detection of LEs.