Sea level prediction using meteorological factors

Climate change directs coastal and ocean engineering studies depending on the unpredictable effects on meteorological factors. Long-term sea-level change prediction is of great importance for the design of coastal structures, while short term prediction is essential for the handling operations and preventing potential problems in the harbors. To this end, sea level heights and meteorological factor measurements were obtained from a tide gauge in Antalya Province. Then, three different scenarios were established to explore most feasible input combinations for sea level prediction. These scenarios use lagged sea level observations (scenario-1), lagged meteorological factors observation (scenario-2) and both lagged sea level and meteorological factors observations (scenario-3) as input for predictive modeling. Auto correlation and cross correlation analysis were conducted to determine the optimum input combination for each scenario. Then, several predictive models were developed using multiple linear regressions (MLR) and adaptive neuro-fuzzy inference system (ANFIS) techniques. The performance of the developed models was evaluated in terms of root mean squared error (RMSE) and Nash Sutcliffe Efficiency (NSE) indices. The results showed that adding meteorological factors as input parameters increases the performance accuracy of the MLR models up to 33% for sea level predictions. Moreover, the results contributed a more precise understanding that ANFIS is superior to MLR for sea level prediction.