基于多约束变分融合的海表温度数据重构OA

By introducing multiple constraint conditions into the analysis of Argo buoy temperature anomaly data,a sea surface temperature(SST)data reconstruction method based on multi-constraint variational fusion is proposed to enhance the reconstruction accuracy of the SST field.First,temperature data are extracted at Argo buoy observation points and processed to obtain discrete temperature anomaly data.Then,a multi-constraint 2-dimensional variational(MC2D-Var)method is employed to generate a continuous anomaly field.During assimilation,three different schemes are implemented:sea surface height(SSH)constraint,advection constraint induced by eastward and northward sea water velocity(UV),and a combined advection+SSH constraint to optimize the generation of the continuous anomaly field.Finally,the WOA18 climatological SST field is superimposed onto the continuous temperature anomaly field to reconstruct the SST field.Using global SST data from June 2023 as an example,the Argo observation data were randomly divided into a training set(70%)and a validation set(30%).Based on the three constraint schemes,SST fields were generated with a resolution of 0.25°×0.25°.Validation results show that compared to OISST data(RMSE of 0.902℃),all schemes achieved improvements.Specifically,Scheme 1(SSH constraint)and Scheme 2(advection constraint)reduced RMSE to 0.866℃and 0.891℃,respectively,while Scheme 3(combined constraint)performed best,lowering RMSE to 0.864℃.Additionally,reconstruction experiments from January to May further validated the stability of Scheme 3.Compared to OISST,its RMSE was reduced by 5.55%,2.97%,7.26%,4.91%and 11.46%,respectively,while its accuracy improvement over WOA18 data reached 49.15%,42.09%,44.56%,47.30%and 47.21%,respectively.These results indicate that incorporating multiple physical constraints can effectively enhance SST reconstruction accuracy,providing a new technical approach for high-precision sea temperature field construction.