太平洋长时间序列遥感数据的参数化模型研究

The Pacific is not only the response region to global changing, but also the global change driving area， and the spaceborne remote sesing has advantages of macrography, speedy and synthesized. Quantitative conjoint analysis of multiple ocean elements is one of the major ways in ocean remote sensing. For the reason that the ocean phenomena is affected by more than one ocean factors, increasing the efficiency of processing method and the number of processing variables by researching on principles of correlation analysis has great significance. Researching the time-space relationship of remote sensing inversion parameters is significant in air-sea interaction and global climate change. The project will improve existing retrieval method and develop remote sensing model of biological-optical parameters and ocean dynamic parameters that base on satelite remote sensing data during 30 years. Spatial teleconnection, temporal and spatial autocorrelation, grey correlation analysis, canonical correlation analysis, empirical orthogonal function and singular value decomposition are used to establish biological-optical parameters chlorophyll-a, sea surface wind, sea surface temparature,sea surface height anomaly and rainfall, Spatio-temporal correlation model in the Pacific， which applied to reveal the spatial relevance among marine parameters and clarify the laws of dynamic change and time response. The project will provid for parameters-reanalysis theory and method for observing ocean dynamic and ecology environment in the Pacific.

海洋遥感具有高空间、高时间分辨率等优势；太平洋是全球海洋和气候系统中最为重要的海区之一，如何在太平洋海洋研究中发挥长时间积累的遥感数据的作用，必须解决数据的可用性这一关键问题，开展太平洋区域卫星遥感数据的参数化研究，对获得准确的遥感产品具有现实意义。本项目旨在基于30多年的卫星遥感数据，将海洋参数的区域时空特性融入到时空插值和时空聚合算法中；利用卫星遥感、观测等多源数据的融合技术，把海气相互作用的遥感模型参数化，从而建立统一时空标准的太平洋海洋参数化模型，为太平洋海洋遥感产品的参数化提供理论和方法，提高海洋遥感数据的可用性。

本项目研究了海洋遥感数据的参数化方法。利用重构的海洋遥感数据，研究了太平洋海洋过程的遥感监测方法。首先建立一套遥感数据重构方法，并对国产FY-3 海表温度（SST）数据进行了重构、误差分析；其次利用重构的数据构建了一种新的基于遥感数据的ENSO指数；分析了太平洋海洋过程的时空变化规律。为太平洋海洋遥感产品的参数化提供理论和方法，提高海洋遥感数据的可用性。