A physical background model for the Fermi Gamma-ray Burst Monitor

We present the first physically motivated background model for the Gamma-Ray Burst Monitor (GBM) onboard the Fermi satellite. Such a physically motivated background model has the potential to significantly improve the scientific output of Fermi/GBM, as it can be used to improve the background estimate for spectral analysis and localization of Gamma-Ray Bursts (GRBs) and other sources. Additionally, it can also lead to detections of new transient events, since long/weak or slowly rising ones do not activate one of the existing trigger algorithms. In this paper we show the derivation of such a physically motivated background model, which includes the modeling of the different background sources and the correct handling of the response of GBM. While the goal of the paper is to introduce the model rather than developing a transient search algorithm, we demonstrate the ability of the model to fit the background seen by GBM by showing four applications, namely (1) for a canonical GRB, (2) for the ultra-long GRB 091024, (3) for the V404 Cygni outburst in June 2015, and (4) the ultra-long GRB 130925A.