协同克里格方法在东海表面温度场数据插值中的应用

Application of Co-Kriging Method to Interpolation of Surface Temperature Data in East China Sea

首次将协同克里格方法引入到海洋水文插值计算中处理东海实验区2008-01—2008-03三个月抽稀后的月平均海表面温场数据。将温场数据分为趋势分量和剩余分量,最终采用两分量的估计值之和实现海表面温场数据的格网插值估计。结果表明:协同克里格方法比传统方法(趋势分析、移动平均、普通克里格插值)效果好。提出变差函数局部拟合的方法,解决了在最大搜寻范围内变差函数无法收敛,各方向很难取得统一基台值的问题。

The Co-Kriging method is introduced into the interpolation of marine hydrological data for the first time and used to process the rarefied monthly average temperature data from the experimental area of East China Sea in the months of January,February and March.The temperature data are decomposed into physical trend component and residual component.Finally the two estimated components are used to realize the gridding interpolation of the sea surface temperature data.The results show that the effect of interpolation by using the method of Co-Kriging is better than the tradition method by using the trend analysis,the moving average or the ordinary Kriging method.The suggested method i.e.the local variogram fitting can solve the problems that the maximum variation within the search function is divergent and it is hard to obtain the uniform value of sill in all directions.