Abstract

One of the most important fisheries in the Pacific is the Halibut fishery (Hippoglossus stenolepis, Schmidt 1904). The main objective of its management is to guarantee its sustainability while maintaining the rate of replacement of the population stable. To this end, this study analyzed the general characteristics of the halibut CPUE historical series (1998-2016) and evaluated the ability to model its behavior through univariate linear techniques, ARIMA models. Following the application of the models, it has been possible to detect historical changes in the series due to the efficiency of the fishery and its legislative framework. There has also been a steady decline in halibut catches from 1998 to the present, a clear indication of over-exploitation of the fishery. On the other hand, with 18 years of CPUE data available, the variability explained (r2 and E2) by the models improved, absolute (RMSE and MAE) and relative (% SEP and ARV) errors and time lag (PI) were reduced in the validation phase. Following the above results, the ARIMA models have been presented as a useful tool in the analysis of the management of the fishery resource, indicating the establishment of protection and conservation measures on the Pacific halibut stocks to avoid the collapse of this fishery.