Prediction of seawater salinity using truncated spline regression method

Salinity is the level of dissolved salt in water, which is one of the factors that affect salt production, the higher the salinity dissolved in seawater, the better the resulting salt production. The factors that affect seawater salinity include air humidity, wind speed and seawater temperature. In this study, the Spline Truncated Regression method was applied to predict seawater salinity based on the variables that influence it. Data collection was obtained using satellite images obtained from Landsat 8. The data was taken within 1 year from January to December 2019. From the analysis results, it was found that the linear spline model with 1 knot point is the best model with a minimum GCV value.0.3648021178, with an R-Sq of 0.580443194, the MSE value is 0.136237833, and the MAPE is 0.98349512. Based on the MAPE value, the prediction model is said to be very accurate forecasting.