Radio Science · 2018 · 11 citations · 31 references
ComputingEarth SciencesForecastingGlobal Navigation Satellite SystemsTime Series AnalysisForecast VerificationTime Series ForecastingForecasting AccuracyForecasting Accuracy EvaluationIonospheric Parameter PredictionIonospheric ScintillationIonospheric Total Electron ContentRegional Ionospheric ModelingScintillationTotal Electron Content Modeling
Abstract We describe a novel empirical technique for the regional, short‐term (from seconds to minutes) forecasting of both TEC (total electron content) and scintillation indices on Global Navigation Satellite System signals. To provide TEC‐forecasted values, the method exploits the continuity equation in the conservative form, while the continuity equation with source term added is used in the case of scintillation forecasting. The performance of the model is investigated, resulting to be satisfactory when applied at equatorial latitudes in Brazil and during postsunset hours, when scintillation phenomena are more likely to occur. Five days affected by strong amplitude scintillation events ( S 4 > 0.7) are used to test the model performance and evaluate the model accuracy. In addition, the performance of the amplitude scintillation forecasting is evaluated by leveraging on larger statistics. On the average, the forecasting accuracy, in terms of standard deviation of the distributions of the differences between forecasted and actual values, is about 5% for TEC and 15–20% for scintillation parameters.
31