Use of inverse methods in atmospheric sounding problems
The inverse problem of the retrieval atmospheric profiles (including temperature, humidity, O3, etc.) using remote sensing data is a nonlinear ill-posed problem. For this problem, a method of the iteratively Gauss-Newton method using the Bayesian approaches is provided in the paper. To overcome the ill-posedness of the retrieval, the method utilizes all the information of the background knowledge of the profiles and me observation, and the error of those information are also be considered in the process respectively. The experimental result using the observation of ATOVS indicates that this method can give a stable result which has notable improvements on the prior states. It also gives an approach to other inverse problems.
Hua-dong Du Si-xun Huang
College of Meteorology, PLA University of Science and Technology, Nanj ing 211101 ,China
国际会议
The 5th International Conference on Nonlinear Mechanics(第五届国际非线性力学会议)
上海
英文
1494-1498
2007-06-11（万方平台首次上网日期，不代表论文的发表时间）