| 用蒙特卡洛方法研究海洋盐度遥感误差 |
| |
| 引用本文： | 马文韬,李紫薇,刘玉光,李轶斐,吕学珠.用蒙特卡洛方法研究海洋盐度遥感误差[J].中国海洋大学学报(自然科学版),2012(10):10-18. |
| |
| 作者姓名： | 马文韬 李紫薇 刘玉光 李轶斐 吕学珠 |
| |
| 作者单位： | 中国海洋大学物理海洋实验室;中科院遥感应用研究所 |
| |
| 基金项目： | 高等学校博士学科点专项科研基金项目(20100132110002);海洋公益性行业科研专项(201005027-4)资助 |
| |
| 摘 要： | 海洋的盐度观测对于气候和海洋科学的研究有重要的意义,盐度的卫星遥感观测需要估计各种因素带来的误差影响。本文基于海面微波辐射理论和海水相对电容率等模型,采用蒙特卡洛模拟方法研究了在盐度遥感中温度误差、仪器误差以及风速误差对于后续的盐度反演的影响。通过计算温度误差产生的盐度误差,并与敏感性方法的对比发现,在低温低盐时温度误差对盐度反演误差的影响较大,2种方法的偏差较大;而在高温高盐时温度误差对盐度反演误差的影响较小,2种方法的偏差较小。辐射计仪器噪声对盐度误差的影响普遍在0.1psu以上,在低温低盐时可达0.5psu以上。风速误差对盐度反演误差的影响在水平极化状态下随入射角增大,在温度低于20℃时普遍超过1psu;在垂直极化状态下随入射角先减小后增大,在温度低于20℃以及较小的入射角下误差也会超过1psu。对误差的综合分析发现,采用垂直极化状态在高温时这2种误差的影响较小。研究发现,当入射角是45.6°和垂直极化状态下,对于3种典型海面状态(35℃和35psu,20℃和35psu,5℃和30psu),反演的盐度反演误差可达到0.162,0.153和0.444psu,达到了卫星单次扫描对盐度反演的误差要求。
|
| 关 键 词： | 盐度遥感 温度 误差 蒙特卡洛模拟 |
Study on the Error of Ocean Salinity Remote Sensing by Using Monte Carlo Simulation |
| |
| MA Wen-Tao,LI Zi-Wei,LIU Yu-Guang,LI Yi-Fei,LV Xue-Zhu.Study on the Error of Ocean Salinity Remote Sensing by Using Monte Carlo Simulation[J].Periodical of Ocean University of China,2012(10):10-18. |
| |
| Authors: | MA Wen-Tao LI Zi-Wei LIU Yu-Guang LI Yi-Fei LV Xue-Zhu |
| |
| Institution: | 1.Physical Oceanography Laboratory,Ocean University of China,Qingdao 266100,China;2.Institute of Remote Sensing Applications,Chinese Academy of Sciences,Qingdao 100101,China) |
| |
| Abstract: | The sea surface salinity(SSS) observation has important significance for studying climate change and ocean dynamics.For salinity remote sensing,we need to estimate the errors caused by various factors.Based on the microwave radiation theory of sea surface,the relative permittivity model,and emprical model on wind-derived sea surface brighness temperature,by using the Monte Carlo simulation,we investigate the errors of retrieved SSS generated by the observed errors of sea surface temperature(SST) and wind,as well as NEDT.It is found that when considering the errors influenced by only a single factor SST,the root mean square(rms) errors obtained by Monte-Carlo simulation equal aproximatedly to those obtained by the differential method.By calculating the rms error of retrieved SSS,it is found that,the same SST error has greater impact on retrieved SSS accuracy for the sea with low temperature and low salinity,and less impact on that for the sea with high temperature and high salinity.At the areas with a temperature higher than 25 ℃ and a salinity higher than 35psu,the salinity retrival errors caused by SST error of 0.5 ℃ is less than 0.2psu,however,the errors increase even up to 0.5~0.9psu for the sea with lower temperature and lower salinity.In the vertical polarization,the rms errors of retrieved SSS increase with the incidence angle,while it is opposite in horizontal polarization.In the same incidence angle,the errors in vertical polarization are larger than those in horizontal polarization,especially in the case of large incidence angle.This study reveals that in the incidence of 45.6 with vertical polarization for the three sea states of 35 ℃ and 35psu,20 ℃ and 35psu,as wel as 5 ℃ and 30psu,the total rms errors of retrieved SSS,caused by SST,NEDT and wind,are 0.162psu,0.153psu and 0.444psu,respectively,which reach the precision requirement for satellite remote sensing of SSS in a single observation. |
| |
| Keywords: | salinity remote sensing sea surface temperature error Monte-Carlo simulation |
| 本文献已被 CNKI 等数据库收录！ |
|
