SA015-0001
Revisiting Langevin Modeling of Incoherent Scatter Spectra

Thursday, 10 December 2020
Poster
Brian Humberto La Rosa La Rosa, Pontifical Catholic University of Peru, Lima, Peru, Marco A Milla, Instituto GeofĂ­sico del PerĂș, Jicamarca Radio Observatory, Lima, Peru and Enrique Luis Alfonso Rojas Villalba, Cornell University, Ithaca, NY, United States
Abstract:
An incoherent scatter spectra model for collisional and magnetized F-region plasma, which includes the effect of Coulomb collisions at small aspect angles, has been developed by Milla and Kudeki based on the Langevin equation. The particle trajectories were solved using the Euler-Maruyama numerical approach, and the electron density spectra were constructed using the statistics of the particle displacements. However, the computational cost of these simulations is still too high to use it as a regular tool for parameter estimation. In order to find a more computationally efficient approach, an exhaustive review of various numerical methods for stochastic differential equations has been done. In this work, we will present a new approach to perform these simulations based on higher-order numerical algorithms that avoid the need for adaptive time steps and variance reduction methods that decrease the number of trajectory simulations needed to build the density spectra. We will also present the benchmarks used to evaluate the proposed approach and some of the new developments we are currently working on.