Neural network approach for detecting spatial changes in catch probability of Engraulis ringens during El Niño-Southern Oscillation events in northern Chile

Armas, Elier; Arancibia, Hugo; Neira, Sergio; Marin, Maria Carla

Abstract

Engraulis ringens (anchovy) is a small pelagic fish of the Family Engraulidae that inhabits the neretic-coastal zone from northern Peru to south-central Chile. It is the main resource species of industrial fishing of northern Chile, representing 80% of the annual landings of the purse seine fleet. The history of this fishery (1985-2023) shows a strong decrease in annual industrial landings, especially during extreme El Nino-Southern Oscillation (ENSO) events. The greatest decrease in landings in two decades occurred in 2020, coinciding with a cold La Nina event, which had not been observed in previous La Nina events. We evaluated whether the current decrease in annual landings of E. ringens is associated with oceanographic changes in northern Chile during El Nino or La Nina events. We applied a neuronal network model to identify the spatial and temporal distribution of E. ringens using the catch probability of each boat of the industrial purse seine fleet. The selected oceanographic variables (sea surface temperature, salinity, depth of the mixed layer, sea height and currents) for the 2003-2020 period were obtained from the Copernicus Marine Environment Monitoring Service (CMEMS program) and used as predictor variables of the monthly landings of E. ringens. The neural network model explained 97% of the monthly variability of catch probability of E. ringens by the industrial purse seine fleet. The spatial distribution of catch probability of E. ringens was analyzed independently for El Nino (2015), La Nina (2007, 2013, and 2020), and Neutral (2004) years. We found that catch probability extended further west during La Nina events (except for 2020), occupying a greater area, but were limited to a 10 nautical mile coastal strip during the El Nino event. The spatial distribution of catch probability in the Neutral condition was near the coast, although not as restricted as during the El Nino event. The higher catch probabilities in the La Nina event of 2020 were near the coast, in contrast to the previous La Nina events of 2007 and 2013, due to the restriction of the optimal habitat of E. ringens in response to changes in oceanographic conditions. The application of the results of this study will allow understanding and probably anticipating the consequences that extreme ENSO events could have on the catch probability of the industrial anchovy purse seine fleet in northern Chile.

Más información

Título según WOS: ID WOS:001179690500001 Not found in local WOS DB
Título de la Revista: FISHERIES OCEANOGRAPHY
Volumen: 33
Número: 4
Editorial: Wiley
Fecha de publicación: 2024
DOI:

10.1111/fog.12672

Notas: ISI