An analysis and accurate prediction of the sea level uctuations in Caspian Sea is always important because it potentially aects the natural processes occurring in the basin and inuences the infrastructure built along the coastlines. In this article, dierent approaches in analysis and forecasting of Caspian Sea level anomalies derived satellite altimetry are presented. Compared with the conventional linear regression methods, such as the routine Autoregressive Moving Average (ARMA) models, neural network methodologies and arti�cial intelligence approaches are the more powerful tools inproviding reliable results of the short-term Caspian Sea level anomaly prediction according to our study. Based on the analysis of the minimum Root Mean Square Error and the maximum coeƒcient of determination, the Support Vector Machine (SVM) shows the best performance in predicting the Caspian Sea level anomalies. …e excellent methods and models for the Caspian Sea level analysis included in this article can be employed to monitor water level changes in other water bodies whose time series of water levels have the stochastic behavior in the future.
All Science Journal Classification (ASJC) codes
- Agricultural and Biological Sciences(all)
- Environmental Science(all)
- Earth and Planetary Sciences(all)