Modelling and Forecasting the Catch of the Scads (Decapterus macrosoma, Decapterus russellii) in the Javanese Purse Seine Fishery Using ARIMA Time Series Models

Abstract

The standard of living of the purse seine fishers operating in the northern coast of the Java Island is closely related to scad resources. Scads landings of the semi-industrial purseseine fishery were analyzed using Seasonal Auto Regressive Integrated Moving Average (SARIMA) techniques. Several models were found to be suitable for describing the temporal fishery pattern and for forecasting the landings two years ahead. Forecasts were matched to actual data. It appears that such models can describe and forecast the dynamics of the scads fishery in the Java Sea. Until now, such results are difficult to predict owing to the strong influence of the year to year changes in the monsoon regime. Forecasting is more reliable with neritic species such as Decapterus russellii, since their spatial distribution in the Java Sea is relatively independent of the monsoon variability. On the contrary, oceanic species such as Decapterus macrosoma are more difficult to forecast as their distribution is closely related to the environmental changes. In that case, transfer function models including environmental parameters would be more appropriate.