One of the major and unfortunately unforeseen sources of background for the current generation of X-ray telescopes flying mainly in the magnetosphere are soft protons with few tens to hundreds of keV. One such telescope is the X-ray Multi-Mirror Mission (XMM-Newton). Its observing time lost because of the contamination is about 40%. This affects all the major science goals of XMM, ranging from cosmology to astrophysics of neutron stars and black holes. The soft proton background could dramatically impact future X-ray observatories such Athena and SMILE missions. Magnetospheric processes that trigger this background contamination are still poorly understood. We use machine learning to delineate related important parameters and to develop models to predict the background contamination using XMM and Cluster mission observations. As predictors we use the location of spacecraft, the terrestrial magnetic field geometry and parameters related to solar and geomagnetic activity. We revealed that the contamination is most strongly correlated with the distance, the solar wind velocity and the topology of the geomagnetic field. Based on this analysis, it is recommended for future X-Ray missions in the magnetosphere to minimize observations during times associated with high solar wind speed and avoid closed magnetic field lines.
Where and under which conditions do soft protons affect X-ray observations: space physics meets astrophysics and machine learning