Global Structure of Magnetotail Reconnection Revealed by Mining Space Magnetometer Data

The location of magnetic reconnection in Earth’s magnetotail has been a central topic since the original conjectures of Dungey and Hones regarding the global structure of the magnetosphere and the role of reconnection in plasma convection and substorms. The X-line location, length, and duration determine the total magnetic flux and energy deposited by magnetic reconnection during storms and substorms. In 2015, the Magnetospheric MultiScale (MMS) mission was launched to fly through these regions to understand the microscale processes that govern magnetic reconnection. During the 2017–2020 tail seasons, MMS observed 26 reconnection sites in the form of Ion Diffusion Regions (IDRs), but their global structure remained elusive. Here we show that mining archives of space magnetometer data from 20+ spacecraft covering more than a quarter century, organized by global activity indices and the solar wind electric field parameter together with the basis function representation of the geomagnetic field (SST19 algorithm), reveals the global structure of these X-lines. Of the 26 observed IDRs, 24 are consistently reconstructed using this data mining approach. This success is remarkable given that only ~0.03 % of the swarm of measurements used to reconstruct the magnetic field picture was from the given event, with the other ~99.97 % coming from other times when the magnetosphere was in a similar configuration. This finding not only demonstrates the power of the modern machine learning tools, it also shows that X-lines are a robust feature of the magnetosphere and their size and position strongly correlate with storm and substorm activity.