Research on Three-Dimensional Ocean Temperature Field Variation Model Based on Assimilation and Reconstruction of Regression Equation

For a long time, natural marine ecosystems have been subject to strong human intervention and global environmental changes. Therefore, it is a meaningful challenge to objectively evaluate the constituent elements of marine material transport, energy flow, and system functions. Besides, the calculation method of three-dimensional (3D) ocean temperature fields is improved in the paper by establishing regression model assimilation and reconstruction of 3D ocean temperature fields. Moreover, according to Guinethut's method, the model is reconstructed with the help of 3D monthly average temperatures and the salinity field of the Copernicus Marine Environment Monitoring Service. Then, monthly average sea surface data sea level anomaly and sea surface temperature from the satellite observations of the next year are used to derive 3D monthly average temperatures of the next year field. Additionally, the data sets of adjacent or similar years are applied to obtain linear regression relationships between sea surface and underwater features, which can improve the accuracy of reconstruction. The most optimal interpolation method is used to assimilate the measured data of Argo to increase accuracy. Finally, after data analysis and comparison, it is proved that compared with other methods, the observation value of the data model established in research content of the paper is further improved, which provides important reference for marine ecological protection work.