Response of data-driven artificial neural network-based TEC models to neutral wind for different locations, seasons, and solar activity levels from the Indian longitude sector

TLDR: In this article, a set of observations carried out in the Indian longitude sector have been reported in order to find the amount of improvement in performance accuracy of an ANN-based Vertical Total Electron Content (VTEC) model after incorporation of neutral wind as model input.

Abstract: The perturbations imposed on transionospheric signals by the ionosphere are a major concern for navigation. The dynamic nature of the ionosphere in the low latitude equatorial region and the Indian longitude sector has some specific characteristics such as sharp temporal and latitudinal variation of Total Electron Content (TEC). TEC in the Indian longitude sector also undergoes seasonal variations. The large magnitude and sharp variation of TEC causes large and variable range errors for satellite based navigation system such as Global Positioning System (GPS) throughout the day. For accurate navigation using Satellite Based Augmentation Systems (SBAS), proper prediction of TEC under certain geophysical conditions is necessary in the equatorial region. It has been reported in the literature that prediction accuracy of TEC has been improved using measured data driven Artificial Neural Network (ANN) based VTEC models, compared to standard ionospheric models. A set of observations carried out in the Indian longitude sector have been reported in this paper in order to find the amount of improvement in performance accuracy of an ANN-based Vertical TEC (VTEC) model after incorporation of neutral wind as model input. The variations of this improvement in prediction accuracy with respect to latitude, longitude, season and solar activity have also been reported in this paper.