Global empirical model of TEC response to geomagnetic activity

Abstract

AbstractA global total electron content (TEC) model response to geomagnetic activity described by the Kp index is built by using the Center for Orbit Determination of Europe (CODE) TEC data for a full 13 years, January 1999 to December 2011. The model describes the most probable spatial distribution and temporal variability of the geomagnetically forced TEC anomalies assuming that these anomalies at a given modified dip latitude depend mainly on the Kp index, local time (LT), and longitude. The geomagnetic anomalies are expressed by the relative deviation of TEC from its 15 day median and are denoted as rTEC. The rTEC response to the geomagnetic activity is presented by a sum of two responses with different time delay constants and different signs of the cross‐correlation function. It has been found that the mean dependence of rTEC on Kp index can be expressed by a cubic function. The LT dependence of rTEC is described by Fourier time series which includes the contribution of four diurnal components with periods 24, 12, 8, and 6 h. The rTEC dependence on longitude is presented by Fourier series which includes the contribution of zonal waves with zonal wave numbers up to 6. In order to demonstrate how the model is able to reproduce the rTEC response to geomagnetic activity, three geomagnetic storms at different seasons and solar activity conditions are presented. The model residuals clearly reveal two types of the model deviation from the data: some underestimation of the largest TEC response to the geomagnetic activity and randomly distributed errors which are the data noise or anomalies generated by other sources. The presented TEC model fits to the CODE TEC input data with small negative bias of −0.204, root mean squares error RMSE = 4.592, and standard deviation error STDE = 4.588. The model offers TEC maps which depend on geographic coordinates (5° × 5° in latitude and longitude) and universal time (UT) at given geomagnetic activity and day of the year. It could be used for both science and possible service (nowcasting and short‐term prediction); for the latter, a detailed validation of the model at different geophysical conditions has to be performed in order to clarify the model predicting quality.