Revisiting Langeving Modeling of ISR Spectra and Applications to Parameter Estimation

An incoherent scatter spectral model for collisional and magnetized F-region plasma was developed by Milla and Kudeki (2011), based on single-particle statistics and a nonlinear Langevin equation that captures the physics of Coulomb collision at small aspect angles. The stochastic differential equation resulting from this model was solved using a first-order approximation known as Euler-Maruyama. Furthermore, the statistics needed to perform the autocorrelation function were estimated by Monte Carlo experiments. The numerical stiffness of this system requires adaptive time-stepping.

In this work, we propose a higher-order numerical algorithm to solve this nonlinear Langevin equation. We can assess previous estimates and avoid adaptive time-stepping using this approach. Benchmarks, like computational execution time and weak convergence, are described. Finally, we describe preliminary results using non-Gaussian distributions to fit the simulated particle statistics and outline possible applications for ISR estimation.