Study of the Effect of Martian Dust Storms on Ionospheric Electron Density Based on Physical-grid Deep Neural Learning Technology

Abstract

The behavior of the Martian ionosphere during global dust storms has always been a popular topic. Based on many Mars Atmosphere and Volatile Evolution observations, a Martian ionospheric electron density model is constructed herein using a physical-grid deep neural network algorithm. The precision of the model is proved to be high, and the R ^2 value of the test set is 0.979. On this basis, the effects of global dust storms on the electron density and peak height of the Martian ionosphere are explored. The results show that the MY34 global dust storm significantly raised the peak height of the Martian electron layer, and its effects mainly ranged from 20°N to 55°N. The periodicity of the global daily average peak electron layer height following this dust storm was more regular than that before the dust storm.

Keywords

• Neural networks
• Planetary atmospheres
• Planetary ionospheres