| dc.description.abstract | Artificial neural networks are adjusted to predict monthly landings of anchovy
(Engraulis ringens) and sardine (Sardinops sagax) in northern Chile (18°21´S-24°S).
Fishing effort (fe), landings and twelve environmental variables are considered from 1980
to 2012. External validation for the best models using all variables showed an R2 of 95%
for anchovy and 99% for sardine, with an efficiency of 0.94 and 0.96, respectively. To
simplify the models, only fe and the sea surface temperature from NOAA satellites (SSTNOAA)
were considered, with the following lags for anchovy: fe(t-0), fe(t-12), fe(t-26), SSTNOAA(
t-2), SST-NOAA(t-14) and SST-NOAA(t-26); and for sardine: fe(t-0), fe(t-12), SSTNOAA(
t-5), SST-NO0AA(t-17), SST-NOAA(t-28) and SST-NOAA(t-40). When comparing both
the models with all variables and reduced variable models, fitness and predictive capacity is
maintained. With these simplified models and local SST projections for the A2 IPCC
scenario, which predicts an approximate 2°C increase in the local SST, estimated anchovy
landings decrease 6% and sardine landings also decrease in lesser measure due to an
increased SST of approximately 2°C from climate change.
Keywords: forecast, landings, pelagic, northern Chile, climate change. | |
