Exploring pitch-angle diffusion during high speed streams with neural networks

Abstract

This summary paper reports progress on an ongoing development of a neural network (DNN) model of quasi-linear pitch-angle diffusion coefficients of radiation belt electrons caused by whistler-mode hiss waves during high speed streams (HSS). We newly explore pitch-angle diffusion for the HSS of 26 January 2013, discussing the simplicity of performing exploration with a neural network model and comparing how mean diffusion coefficients behave compared with individual ones. The neural network is also newly used to explore the variations of diffusion coefficients at fixed Kp. The model is finally used to predict pitch-angle diffusion during sustained intense HSS yet unobserved.