Development of algorithms and software for forecasting, nowcasting and variability of TEC

Abstract

Total Electron Content (TEC) is an important characteristic of the ionosphere relevant to communications. Unpredictable variability of the ionospheric parameters due to various disturbances limits the efficiencies of communications, radar and navigation systems. Therefore forecasting and nowcasting of TEC are important in the planning and operation of Earth-space and satellite-to-satellite communication systems. Near-Earth space processes are complex being highly nonlinear and time varying with random variations in parameters where mathematical modeling is extremely difficult if not impossible. Therefore data driven models such as Neural Network (NN) based models are considered and found promising in modeling such processes. In this paper the NN based METU-NN model is introduced to forecast TEC values for the intervals ranging from 1 to 24 h in advance. Forecast and nowcast of TEC values are also considered based on TEC database. Day-to-day and hour to-hour variability of TEC are also estimated using statistical methods. Another statistical approach based on the clustering technique is developed and a preprocessing approach is demonstrated for the forecast of ionospheric critical frequency foF2.