Japanese squid winter-spawning group resource abundance prediction method

Abstract

The invention provides a Japanese squid winter-spawning group resource abundance prediction method. The method comprises steps: 1, the sea surface temperature (SST) of a marine environmental factor ina Japanese squid winter-spawning group spawning ground sea area is acquired; 2, SST time sequence values of sample points in the spawning ground in a spawning month and corresponding Japanese squid CPUE values are calculated for correlation analysis, and sea areas with high correlation are selected; 3, six sea areas S1 to S6 with a high correlation coefficient in continuous three months are selected; 4, a multivariate linear prediction model is built for the SSTs from S1 to S6 and the CPUE; 5, the correlation coefficients for six SSTs and the CPUE are ranked from large to small, and input factors are sequentially added to build four BP neural network forecasting models; and 6, the multivariate linear prediction model and the four BP neural network forecasting models are compared, and a BPneural network forecasting model with a 6-4-1 structure is selected as the Japanese squid winter-spawning group resource abundance prediction model.

Description

technical field [0001] The invention relates to a method for forecasting fishery conditions, in particular to a method for forecasting the resource abundance of winter squid populations in Japan. Background technique [0002] Fishery forecast is a key link in fishery production. Forecasting the abundance of winter squid resources in Japan will help to forecast and grasp the production of winter squid in Japan, the location of central fishing grounds and the fish season. The temperature factor has a great influence on the growth and distribution of Japanese squid, and the study found that the regression of sea surface temperature and the fish catch per unit fishing boat (CPUE, unit: thousand tails / day) is extremely significant. Therefore, using It is feasible to use sea surface temperature as a predictor of resource abundance of winter squid species in Japan. [0003] Fishery forecast is also the focus of fishery research. Accurate fishery forecast can guide enterprises to ...

AI Technical Summary

Problems solved by technology

[0005] Due to the endless variety of forecasting methods and models, the scope of application of the interpretation is also different, which will actually cause confusion for us to use these methods