Abstract

Contributed Talk - Splinter Learning

Tuesday, 22 September 2020, 11:55   (virtual room G)

Using realistic synthetic observations to improve observational techniques

Christine Koepferl
University of Applied Sciences Munich

There exist a variety of great observations of many objects in the universe. To the contrary there are many theoretical simulations trying to explain these objects. There is a large potential in these simulated and observational datasets if we can trust the compared properties. Unfortunately, techniques which infer fundamental properties from the observations such as the star-formation rates, gas mass, dust properties, ... have rarely been tested sufficiently. I created a large data set of realistic synthetic observations of star-forming regions in the infrared. In the past (Koepferl et al. 2017 abc) we used this synthetic observations to test commonly used observational techniques to infer measured properties. We found that most techniques do not work sufficiently. In this talk I will outline my ideas of how the data set can be used to create better measurement techniques and hoping I am attempting lively discussion.