PREDICTING CASPIAN SEA-SURFACE WATER-LEVEL BY ANN AND ARIMA MODELS

Fluctuations of the Caspian Sea's mean monthly surface water level for the period of January 1986 to December 1993 were studied. The time se ries data showed an increasing trend and seasonal variations. Artifici al neural network (ANN) and multiplicative autoregressive integrated m oving average (ARIMA) modeling were used to predict the time series da ta. The ANN's input and output consisted of the last 12 months and the current month surface water levels, respectively. The selected ARIMA model required one-month regular differencing, 12-month seasonal diffe rencing, and had a moving average component of lag 12. The ANN and ARI MA predictions for the period of January to December 1993 were very re asonable when compared with the recorded levels. On average, the ANN m odel underestimated the sea level by three cm, whereas the ARIMA model overestimated it by three cm. The monthly predictions for January to December 1994 presented a continuation of the Caspian Sea water surfac e level rise that would have various adverse effects or its neighborin g countries.