面向观测信息量不足时地表反照率反演算法研究

地表反照率是影响地球能量收支平衡的决定性参数之一，被广泛应用于全球变化研究。地表的各向异性反射特征和遥感观测信息量的不足制约了地表反照率的反演精度。针对这一问题，本项目期望借助于先验信息改善地表反照率的反演精度。具体研究包括：(1)耦合物理模型和核驱动模型，分析不同模型的一致性，根据核系数的变化提取对地表各向异性反射敏感的地表结构参数，建立敏感参数与多角度数据关系查找表；(2)建立基于历史多角度数据和BRDF产品提取BRDF先验知识的方法，建立适用于不同尺度的全球BRDF先验知识库，并对BRDF先验知识的精度及尺度效应进行评价；(3)分析影响地表反照率反演精度的因素，探索将BRDF先验知识用于观测信息量不足时改善地表反照率反演精度的方法。本项目为高分辨率地表反照率产品生产提供了一种新的思路，对30m分辨率地表反照率的估算及多尺度生态和陆表模型的发展有着重要意义。

Land surface albedo is one of the crucial parameters affecting the land surface energy balance and has been wildly used in global change studies. However, the inversion accuracy of albedo is restricted by the land surface reflectance anisotropy and limited observation information acquired by remote sensing. In this project, we plan to: (1) Couple the physical and kernel driven BRDF model to analysis the consistency of the models. Extract the reflectance anisotropy reflectance sensitive surface structure parameters based on the change of kernel parameters, and build the lookup table for sensitive sturcture parameters and multi-angular observations; (2) Develop a method to extract prior BRDF knowledge from historical multi-angular observations and BRDF product, and evaluate the accuracy and the scale effect of the prior knowledge; (3) Analysis the facts that affect the inversion accuracy of albedo, explore a method to improve the accuracy of albedo retrieved from insufficient observations based on prior BRDF knowledge. This project provides a new idea for high resolution albedo retrieval, and can help improve the accuracy in estimating albedo with 30m spatial resolution and facilitate the development of multi-scale ecological and land surface models.