Systematic evaluation of ground and geostationary magnetic field predictions generated by global magnetohydrodynamic models

Abstract

In this work a systematic evaluation of ground and geostationary magnetic field predictions generated by a set of global magnetohydrodynamic (MHD) models is carried out. The evaluation uses four geospace storm events and ground magnetometer station and geostationary GOES data for comparisons between model output and observations. It is shown that metrics analysis of two different geospace parameters, i.e., geostationary and ground magnetic field, show surprising similarities, although the parameters reflect rather different properties of geospace. More specifically, increasing the spatial resolution and inclusion of more realistic inner magnetospheric physics successfully made the model predictions by the BATS‐R‐US model more accurate. Furthermore, while the OpenGGCM model had a tendency to have larger differences to observations than BATS‐R‐US in terms of the prediction efficiency, the model provided more accurate representation of the observed spectral characteristics of the ground and geostationary magnetic field fluctuations.