摘要
依据随机过程线性均方估计的正交原理，可由水文站获得的海洋要素估计工程点的海洋要索。利用1984～2003年的卫星海表温度（SST）数据将该方法用于估计海表温度，并对该方法产生的误差进行了分析，结果表明该方法比传统方法更能有效地减小误差；其意义在于可在陆地台站和工程点进行一段时间的同步观测，找出两者之间的相关系数，之后便可长期由台站气温估计工程点的海表温度。
From the orthogonality principle in liner mean square estimation of stochastic processes, the sea factor obtained at a project site can be estimated by that at some routine hydrographic station. This method was used to estimate the annual extreme wave heights at a project site before. By use of SST in 1985-2004, this paper applies the method in estimating the sea surface temperature and analyzes the error of the method. It is found that the method is able to effectively reduce the error. Its meaning is that we can acquire correlation coefficient of SST during the short-term synchronous observation at a project site and routine station（s） and then we can estimate SST at the project from SST at the routine station（s） forever, the error cannot be enlarged even if the correlativity is small between SST at the project and those at the routine station（s）.
出处
《中国海洋大学学报（自然科学版）》
CAS
CSCD
北大核心
2006年第1期127-130,共4页
Periodical of Ocean University of China
关键词
正交原理
均方误差
相关系数
估计
海表温度
orthogonality principle
mean square error
correlation coefficient
estimation
sea surface temperature
