Logo image
An Artificial Neural Network for Inferring Solar Wind Proxies at Mars
Journal article   Open access   Peer reviewed

An Artificial Neural Network for Inferring Solar Wind Proxies at Mars

Suranga Ruhunusiri, J. S Halekas, J. R Espley, F Eparvier, D Brain, C Mazelle, Y Harada, G. A DiBraccio, Y Dong, Y Ma, …
Geophysical research letters, Vol.45(20), pp.10,855-10,865
10/28/2018
DOI: 10.1029/2018GL079282
url
https://doi.org/10.1029/2018GL079282View
Published (Version of record) Open Access

Abstract

We present a novel method to determine solar wind proxies from sheath measurements at Mars. Specifically, we develop an artificial neural network (ANN) to simultaneously infer seven solar wind proxies: ion density, ion speed, ion temperature, and interplanetary magnetic field magnitude and its vector components, using spacecraft measurements of ion moments, magnetic field magnitude, magnetic field components in the sheath, and the solar extreme ultraviolet flux. The ANN was trained and tested using 3 years of data from the Mars Atmosphere and Volatile EvolutioN (MAVEN) spacecraft. When compared with MAVEN spacecraft's in situ measured values of the solar wind parameters, we find that the ANN proxies for the solar wind ion density, ion speed, ion temperature, and interplanetary magnetic field magnitude have percentage differences of 50% or less for 84.4%, 99.9%, 86.8%, and 79.8% of the instances, respectively. For the cone angle and clock angle proxies, 69.1% and 53.3% of instances, respectively, have angle differences of 30∘ or less. Plain Language Summary We introduce a new technique for determining solar wind parameter values upstream of Mars using spacecraft‐measured values of parameters in the plasma environment near Mars. This technique involves using an artificial neural network. Data in the solar wind and the plasma environment near Mars measured by the Mars Atmosphere and Volatile EvolutioN spacecraft, which has been orbiting Mars since 2014, were used to train an artificial neural network to simultaneously infer seven solar wind parameters: solar wind density, speed, temperature, and interplanetary magnetic field and its three components. Comparison of the neural network‐inferred values to the in situ measured values reveals that the artificial neural network can infer the solar wind density, speed, temperature, and the interplanetary magnetic field magnitude with high accuracies and the orientation of the magnetic field with moderate accuracies. Thus, this artificial neural network can be successfully used for inferring solar wind parameters at Mars. Since Mars lacks a dedicated solar wind monitor, unlike at Earth, this technique is useful for obtaining solar wind parameters during times when a Mars orbiter does not traverse through the solar wind upstream of Mars. Knowledge of the solar wind parameter values is essential for studying how the solar wind influences Mars' atmospheric escape. Key Points An artificial neural network (ANN) is implemented to infer solar wind proxies from sheath measurements at Mars The ANN can determine solar wind ion density, speed, temperature, and magnetic field magnitude with high accuracies The ANN can determine the magnetic field cone and the clock angles with moderate accuracies
artificial neural networks Mars MAVEN solar wind proxies

Details

Metrics

Logo image