项目名称: 红外高光谱遥感资料直接四维变分同化技术研究
项目编号: No.41305101
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 天文学、地球科学
项目作者: 赵延来
作者单位: 中国人民解放军国防科学技术大学
项目金额: 25万元
中文摘要: 针对现有红外高光谱遥感资料同化方法中的缺陷和不足,研究基于主成分分数的红外高光谱遥感资料直接四维变分同化技术。首先面向四维变分同化,研究一种新的红外高光谱资料云检测方法,寻找对云雨不敏感性的观测通道,提高云雨区红外高光谱资料的同化利用率;其次,研究适用于风云四号卫星干涉式红外高光谱仪的主成分快速辐射传输模式及其切线性和伴随模式;然后研究基于主成分分数的红外高光谱资料的四维变分同化方法,突破针对主成分分数的非线性观测算子、多分辨率增量算法、变分偏差订正和变分质量控制等关键技术。在此基础上,实现基于主成分辐射传输模式的全球四维变分同化试验系统,评估新方法对改进红外高光谱卫星遥感资料同化的有效性和优越性。
中文关键词: 红外高光谱资料;主成分快速辐射传输模式;云检测;四维变分资料同化;
英文摘要: To overcome the deficiency of current assimilation methods developed for infrared hyperspectral data, a new four-dimensional variational data assimilation technique is designed with the principal component scores of infrared hyperspectral radiances. Firstly, a new cloud detection scheme is investigate for infrared hyperspectral data to search observation channels which is not sensitive to cloud and precipitation, and to improve the utilization of infrared hyperspectral data in the area of cloud and precipitation. Secondly, the PC radiance transform model with its tangent and adjoint version are revised to fit the FY-4 satellite's infrared hyperspectral sensor. Then activities are focus on the new four-dimensional variational data assimilation technique using the principal component scores of infrared hyperspectral radiances, including the study of nonlinear observation operator, multiple resolution increment algorithm, variational bias correction and variational quality control. Finally, a global four-dimensional data assimilation experiment system is constructed with the PC radiance transform model, to evaluate the new method's validity and superiority for assimilation of infrared hyperspectral data.
英文关键词: infrared heperspectral data;PC_RTTOV;cloud detection;4DVAR;