Thermosphere joint observations by TM-1 constellations and Swarm-B during the April 2023 geomagnetic storm

The response of thermosphere density to geomagnetic storms is a complicated physical process. Multi-satellite joint observations at the same altitude but different local times (LTs) are important for understanding this process; however, until now such studies have hardly been done. In this report, we analyze in detail the thermosphere mass density response at 510 km during the April 23−24, 2023 geomagnetic storm using data derived from the TM-1 (TianMu-1) satellite constellation and Swarm-B satellites. The observations show that there were significant LT differences in the hemispheric asymmetry of the thermosphere mass density during the geomagnetic storm. Densities observed by satellite TM02 at nearly 11.3 and 23.3 LTs were larger in the northern hemisphere than in the southern. The TM04 dayside density observations appear to be almost symmetrical with respect to the equator, though southern hemisphere densities on the nightside were higher. Swarm-B data exhibit near-symmetry between the hemispheres. In addition, the mass density ratio results show that TM04 nightside observations, TM02 data, and Swarm-B data all clearly show stronger effects in the southern hemisphere, except for TM04 on the dayside, which suggest hemispheric near-symmetry. The South-North density enhancement differences in TM02 and TM04 on dayside can reach 130%, and Swarm-B data even achieve 180% difference. From the observations of all three satellites, large-scale traveling atmospheric disturbances (TADs) first appear at high latitudes and propagate to low latitudes, thereby disturbing the atmosphere above the equator and even into the opposite hemisphere. NRLMSISE00 model simulations were also performed on this geomagnetic storm. TADs are absent in the NRLMSISE00 simulations. The satellite data suggest that NRLMSISE00 significantly underestimates the magnitude of the density response of the thermosphere during geomagnetic storms, especially at high latitudes in both hemispheres. Therefore, use of the density simulation of NRLMSISE00 may lead to large errors in satellite drag calculations and orbit predictions. We suggest that the high temporal and spatial resolution of direct density observations by the TM-1 constellation satellites can provide an autonomous and reliable basis for correction and improvement of atmospheric models.