Physics-informed neural network modeling reveals global dynamo currents in the Martian ionosphere, indicating strong links to atmospheric circulation and seasonal dust storms.
Key Points
To construct a continuous, locally resolved model of large-scale Martian ionospheric dynamo currents and evaluate their spatial patterns and seasonal variability.
Analyzed magnetic field measurements collected by the MAVEN spacecraft magnetometer across the Martian ionosphere.
Implemented the Neural-Curlometer technique, utilizing a physics-informed neural network constrained by Ampère's law and Gauss's law for magnetism.
Dynamo currents are concentrated between altitudes of 125 and 220 km, displaying a Hall-to-Pedersen transition near 160 km.
Current density correlates strongly with the crustal magnetic field on the dayside (correlation coefficient of 0.89), peaking where field strengths range from 130 to 450 nT at 150 km.
Modeled currents display temporal variations tied to dust storm activity and form hemispheric vortex patterns aligned with wind circulation and polar condensation transport.