North Pacific Ocean squid central fishery predicting method

Abstract

A method for forecasting the central fishery of squid in the North Pacific Ocean, including time-space scale setting, environmental factor setting, and establishment of a central fishery prediction model. The factor setting uses surface temperature (SST) as the main environmental factor, supplemented by two environmental factors, sea surface height (SSH) and chlorophyll a (Chl-a), and the environmental factors are divided into four situations when establishing the central fishery prediction model: According to the setting of time-space scale and environmental factors, a sample plan set of 24 situations is established; the central fishery prediction model adopts the classic error backpropagation BP neural network model, and the BP neural network model has a three-layer structure, namely the input layer and the hidden layer And the output layer, the input layer inputs the spatio-temporal factors and environmental factors of the fishery, and the output layer outputs CPUE or the fishery level index converted from CPUE for forecasting.

Description

technical field [0001] The invention relates to a forecasting method for a central fishery, in particular to a forecasting method for a squid central fishery in the North Pacific Ocean. Background technique [0002] The central fishery forecast is a kind of quick fishery status report. Accurate central fishery forecast can increase the catch yield and reduce fuel costs for fishing production. The fishery status quick report is the position of the central fishery and the movement of fish in the next 24 hours or a few days. And the possibility of prosperous development is predicted, and the fishery news command unit transmits the forecast content to the production ships quickly and accurately through the telecommunication system every day, so as to achieve the purpose of commanding on-site production. [0003] At present, there are many methods to predict the central fishing grounds of oceanic economic soft fishes. These methods are based on the relationship and law between fi...

AI Technical Summary

Problems solved by technology

[0003] At present, there are many methods to predict the central fishing grounds of oceanic economic soft fishes. These methods are based on the relationship and law between fish behavior, biological status and environmental conditions, and are essentially based on production statistical data samples. "Experience knowledge" is used for forecasting, but in the past, there was no in-depth study on the selection of sample time-space scales and environmental factors, which were basically set based on experience (such as large fishing areas and small fishing areas, etc.), without considering the impact of different time-space scales and environmental factors. The influence of the central fishery forecast model; in the selection of the model, the real-time problem of marine environmental factors is rarely considered

Images