Ionosphere-thermosphere energy budgets for the ICME storms of March 2013 and 2015 estimated with GITM and observational proxies

The ionosphere-thermosphere (IT) energy partitioning for the interplanetary coronal mass ejection (ICME) storms of 16\textendash19 March 2013 and 2015 is estimated with the Global Ionosphere-Thermosphere Model (GITM), empirical models and proxies derived from in situ measurements. We focus on auroral heating, Joule heating, and thermospheric cooling. Solar wind data, F10.7, OVATION Prime model and the Weimer 2005 model are used to drive GITM from above. Thermospheric nitric oxide and carbon dioxide cooling emission powers and fluxes are estimated from TIMED/SABER measurements. Assimilative mapping of ionospheric electrodynamics (AMIE) estimations of hemispheric power and Joule heating are presented, based on data from global magnetometers, the AMPERE magnetic field data, SSUSI auroral images, and the SuperDARN radar network. Modeled Joule heating and auroral heating of the IT system are mostly controlled by external driving in the March 2013 and 2015 storms, while NO cooling persists into the storm recovery phase. The total heating in the model is about 1000 GW to 3000 GW. Additionally, we intercompare contributions in selected energy channels for five coronal mass ejection-type storms modeled with GITM. Modeled auroral heating shows reasonable agreement with AMIE hemispheric power and is higher than other observational proxies. Joule heating and infrared cooling are likely underestimated in GITM. We discuss challenges and discrepancies in estimating and global modeling of the IT energy partitioning, especially Joule heating, during geomagnetic storms.