极区海冰密集度AMSR-E数据反演算法的试验与验证  

The experiment and validation of sea ice concentration AMSR-E retrieval algorithm in polar region

作　　者：苏洁[1] 郝光华[1] 叶鑫欣[1,2] 王维波[1] 

机构地区：[1]中国海洋大学教育部物理海洋重点实验室,山东青岛266100 [2]北京大学物理学院大气与海洋科学系,北京100871

出　　处：《遥感学报》2013年第3期495-513,共19页NATIONAL REMOTE SENSING BULLETIN

基　　金：国家高技术研究发展计划(863计划)(编号:2008AA121701)~~

摘　　要：海冰密集度是极区海冰监测的重要参数,目前分辨率最高的微波海冰密集度产品为德国Bremen大学发布的针对AMSR-E89GHz频段数据利用ASI算法反演的网格数据。为实现中国极区遥感产品从无到有的战略步骤,本文针对AMSR-E89GHz频段微波数据的ASI算法,进行了插值算法、系点值和天气滤波器一系列试验。针对北极海区,着重对影响反演结果的主要参数——纯冰和纯水的亮温极化差异阈值,即系点值(P1和P0)进行了2009年全年的统计分析。研究表明,2009年北极纯冰和纯水的代表区域P1和P0年平均值分别为10.0K和46.67K;2K以上的系点值差异引起的海冰密集度差别较为显著;同样的系点值差异在不同极化差异P取值范围对海冰密集度的影响也不同。通过统计确定的系点值推算并修正了海冰密集度反演公式,对2009年全年北极海冰密集度进行了反演,并与Bremen大学产品进行了比较。继而对白令海和楚科奇海12个晴空下MODIS可见光样本数据进行反演,以验证AMSR-E冰密集度反演结果,并对误差原因进行了分析。本研究反演结果与MODIS样本比对的误差略小于Bremen大学的反演产品,空间平均误差为3.84%,空间平均绝对误差10.83%。Sea ice concentration is an important parameter for polar sea ice monitoring. Currently, the highest resolution micro- wave-detected sea ice concentration gridded product is provided by University of Bremen, which is derived by the Arctic Radiation and Turbulence Interaction Study （ARTSIST） Sea Ice （ASI） algorithm based on 89 GHz Advanced Microwave Scanning Radiome- ter for Earth Observing System （AMSR-E） data. For the purpose of providing the fu＇st generation Chinese sea ice remote sensing product for polar regions, we implemented a series of experiments including the interpolation algorithm, tie-points, and the weather filter based on the ASI algorithm for AMSR-E 89 GHz data. The main parameters, tic-points threshold value of pure ice （P~） and pure water （P0） of ASI algorithm in the Arctic region were statisticaUy analyzed throughout year 2009. The results showed that the yearly average value of P1 was 10.0 K, and that of P0 was 46.67 K in the pure ice and pure water typical regions of Arctic. The retrieved values of ice concentration is sensitive to P1 and Po when the differences of tie-points value are larger than 2 K. The effects of P~ and P0 on the ice concentration also changed for different P values. The calculation fomula was amended based on the statistic tie points. The sea ice concentration inversion fields in Arctic region of whole 2009 was obtained and compared with the products of University of Bremen. Furthermore, 12 cloud-free samples were selected from the visible band of Moderate-resolution Imaging Spectroradiometer （MODIS） image in Bering Sea and Chukchi Sea, and the derived sea ice concentration of MODIS data is averaged in the AMSR-E grid to validate the AMSR-E sea ice concentration. The comparisons of these samples showed that the error of our results is slightly less than that of University of Bremen＇s product with the spatial average error and spatial average absolute error of 3.84% and 10.83 %, respectively.

关 键 词：AMSR—E 海冰密集度 反演算法 验证 极区 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] P731.15[自动化与计算机技术—计算机科学与技术]