Minimizing NeQuick TEC prediction error via data ingestion at three African equatorial stations

Abstract

NeQuick model is a three dimensional and time dependent ionospheric electron density model that gives as output electron density profile N(h) and via integration the slant TEC (sTEC) for any particular locations on the globe. The model requires as input solar ionizing index in addition to the coordinates of the ray points. The index being used as a proxy to the EUV which is responsible for the ionization of the upper atmosphere is either the smoothed sunspot numbers (R12) or the 10.7 cm solar radio noise flux (F_10.7 ). The closer the values of the proxy index to the EUV value, the better the performance of the NeQuick model. We have therefore investigated the improvement of NeQuick via optimization of the F_10.7 index using GPS sTEC derived data from three stations in the equatorial region of the African sector. In order to achieve this, the root mean square of the difference between the modeled and the observed sTEC values were calculated for a range of values of F_10.7, and via optimization, the effective F_10.7 index (i.e. the one that gives the minimum root mean square error or the one that minimizes the error between the observed and the modeled) was obtained for a particular day and for a particular station. The value obtained (i.e. F_10.7 effective ), which is a station based value, is then used as ionization index into the NeQuick model for the next two consecutive days. The results obtained showed an improvement in the model performance for these two days although that of the first day, in most cases, is better than that of the second day.