A recurrent neural network approach to quantitatively studying solar wind effects on TEC derived from GPS; preliminary results

摘要

This paper attempts to describe the search for the parameter(s) to representsolar wind effects in Global Positioning System total electron content (GPSTEC) modelling using the technique of neural networks (NNs). A study iscarried out by including solar wind velocity (V_sw), proton numberdensity (N_p) and the B_z component of the interplanetary magnetic field(IMF B_z) obtained from the Advanced Composition Explorer (ACE) satelliteas separate inputs to the NN each along with day number of the year (DN),hour (HR), a 4-month running mean of the daily sunspot number (R4) and therunning mean of the previous eight 3-hourly magnetic A index values (A8).Hourly GPS TEC values derived from a dual frequency receiver located atSutherland (32.38° S, 20.81° E), South Africa for 8 years(2000–2007) have been used to train the Elman neural network (ENN) and theresult has been used to predict TEC variations for a GPS station located atCape Town (33.95° S, 18.47° E). Quantitative results indicatethat each of the parameters considered may have some degree of influence onGPS TEC at certain periods although a decrease in prediction accuracy is alsoobserved for some parameters for different days and seasons. It is alsoevident that there is still a difficulty in predicting TEC values duringdisturbed conditions. The improvements and degradation in predictionaccuracies are both close to the benchmark values which lends weight to thebelief that diurnal, seasonal, solar and magnetic variabilities may be themajor determinants of TEC variability.