基于遥感数据的三维温度场参数化分析方法研究

Research on parameterized analysis method of 3D temperature field based on remote sensing data

为了满足海洋研究以及海洋调查的需求,本文基于Argo剖面和海表面温度数据开发了一个新的拟合三维温度场的算法。选取西北太平洋区域作为验证算法有效的实验海区。该水域的经纬度范围设定为:30°~40°N, 140°~155°E,水平分辨率为0.25°。深度方向为从海表到2 000 m水深,水域划分为29层。拟合算法首先将Argo温度剖面以5个深度划分为6层,分别为混合层、夹层、温跃层、过渡层、第一深层、第二深层,然后以第一猜想值和线性回归得到的海表面温度作为初始条件重构三维温度场。重构的三维温度场的剖面与原观测剖面的均方根误差较小,相关性较好,表明该算法是合理有效的。

To meet the needs of marine research and marine surveys, a new algorithm for fitting three-dimensional temperature fields was developed based on the Argo profile and sea surface temperature data. The Northwest Pacific region was selected as the experimental sea area to validate the algorithm. The latitude and longitude range of the water area was set as: 30°–40°N, 140°–155°E, and the horizontal resolution was 0.25°. The depth direction was from the sea surface to 2 000 m underwater, and the water area was divided into 29 layers. First the fitting algorithm divided the Argo temperature profile into six layers with five depths. The layers were mixed layer, inter layer, thermocline, transition layer, the first deep layer, and the second deep layer. The first guess and the sea surface temperature obtained by linear regression were used as the initial conditions to reconstruct the three-dimensional temperature field. The RMSE of the reconstructed three-dimensional temperature field was small and well correlated with the original observation profile, indicating that the algorithm is reasonable and effective.