Open AccessSea surface temperature prediction model based on long and short-term memory neural network
Author(s) -
Xiaojing Li
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/658/1/012040
Subject(s) - la niña , sea surface temperature , anomaly (physics) , artificial neural network , climatology , term (time) , environmental science , meteorology , el niño southern oscillation , geology , computer science , geography , machine learning , physics , quantum mechanics , condensed matter physics
In response to the monitoring and forecasting of El Nino/La Nina phenomenon, This paper proposes a sea surface temperature prediction method based on long and short-term memory neural network for the average sea surface temperature in the NINO comprehensive area. This method uses the monthly anomaly sea surface temperature sequence for the mean sea surface temperature in the NINO Comprehensive area as the input of the long- and short-term memory neural network to establish a forecast model. The average sea surface temperature of the NINO comprehensive area is forecasted for the next 1 to 3 months. The results show that the method can better predict the average sea surface temperature of the NINO comprehensive area, which is useful for the monitoring and forecasting of El Nino/La Nina phenomenon. Provides a new approach.