ENSO-tuna relations in the eastern Pacific Ocean and its prediction as a non-linear dynamic system

摘要

During the years between 1967 and 1994 the trimestral abundance of the yellowfin tuna fish (Thunnus albacares), expressed as the number of individuals per group age and cohort, for the Eastern Pacific Ocean (EPO), was used to calculate the biomass of this species per trimester. This information was obtained from the Inter-American Tropical Tuna Commission publications (ITTC). Graphic methods were applied to this data (crude data, phase space, subseries per trimester as well as annual averages) in order to identify the behavior of the dynamics of this variable; this data was correlated with information on the presence of El Nino Southern Oscillation (ENSO) for the same period. The analysis suggests that the presence of strong ENSO events correlate with the decline of tuna fish biomass, which is followed by a rapid increase of this variable in a 3: 1 time ratio; this means that it takes three times as long for the biomass to decrease than it does to recover and return to similar or higher values. After the ENSO-tuna fish biomass relations were established, two types of models were adjusted to the tuna fish biomass information: neuronal network and ARIMA. Both models described adequately the tuna fish biomass dynamics; however, the ARIMA model also permitted an adequate prediction of the behavior of ENSO variable, emphasizing that this model correctly predicted the presence of 1997 ENSO.