|
|||||
|
|
| 基于MonoRTM模型的微波辐射计反演方法研究 | |
| 引用本文： | 黄兴友,张曦,冷亮,李峰,樊雅文.基于MonoRTM模型的微波辐射计反演方法研究[J].气象科学,2013,33(2):138-145. |
| 作者姓名： | 黄兴友 张曦 冷亮 李峰 樊雅文 |
| 作者单位： | 1. 南京信息工程大学大气物理与大气环境重点开放实验室,南京,210044 2. 中国气象局武汉暴雨研究所暴雨监测预警湖北省重点实验室,武汉,430074 3. 中国气象局气象探测中心,北京,100081 |
| 基金项目： | 公益性行业(气象)科研专项(GYHY200806015;GYHY201106002) |
| 摘 要： | 基于辐射传输模型MonoRTM计算天空亮温度,使用多元线性统计回归方法和BP神经网络方法分别对大气温度和水汽密度廓线进行了反演,检验并分析了两种方法的反演精度.结果表明,多元线性回归方法反演的温度偏差总体不大于6K,反演的水汽密度偏差小于4 g/m3；神经网络方法反演的温度偏差小于2K,反演的水汽密度误差总体不大于2 g/m3.与探空数据的对比表明,对于大气温度和水汽密度反演,BP神经网络方法的反演结果都要比多元线性回归方法的反演结果更接近探空资料值. |
| 关 键 词： | 大气温湿廓线 MonoRTM辐射传输模型 多元线性回归 BP神经网络 |
| 收稿时间： | 2012/2/23 0:00:00 |
| 修稿时间： | 2012/6/14 0:00:00 |
Study on retrieval methods with MonoRTM for microwave radiometer measurements |
|
| HUANG Xingyou,ZHANG Xi,LENG Liang,LI Feng and FAN Yawen.Study on retrieval methods with MonoRTM for microwave radiometer measurements[J].Scientia Meteorologica Sinica,2013,33(2):138-145. | |
| Authors: | HUANG Xingyou ZHANG Xi LENG Liang LI Feng and FAN Yawen |
| Institution: | Key Laboratory of Atmospheric Physics & Environment, Nanjing University of Information Science & Technology, Nanjing 210044, China;Key Laboratory of Atmospheric Physics & Environment, Nanjing University of Information Science & Technology, Nanjing 210044, China;Hubei Key Laboratory for Heavy Rain Monitoring and Warning Research, Institute of Heavy Rain, China Meteorological Administration, Wuhan 430074, China;Meteorological Observation Centre, China Meteorological Administration, Beijing 100081, China;Key Laboratory of Atmospheric Physics & Environment, Nanjing University of Information Science & Technology, Nanjing 210044, China |
| Abstract: | This paper focuses on calculating brightness temperature based on monochromatic radiative transfer model MonoRTM, retrieving atmosphere temperature and water vapor density profiles by multiple linear regression and the BP neural network method, as well as comparing and analyzing retrieval accuracies of these two methods. The results show that the deviation of temperature profiles retrieved by multiple linear regression method is no more than 6 K and that of retrieved water vapor density profiles is less than 4 g/m3 on the whole, while the deviation of retrieved temperature using the BP neural network method is less than 2 K, and that of retrieved water vapor density is generally lower than 2 g/m3. Compared to the radiosonded data, the retrieval results of temperature and water vapor density from BP neural network method are better than those from multiple linear regression method. |
| Keywords: | Temperature and water vapor density profile The monochromatic radiative transfer model MonoRTM Multiple linear regression BP neural network |
| 本文献已被 万方数据 等数据库收录！ | |
| 点击此处可从《气象科学》浏览原始摘要信息 | |
| 点击此处可从《气象科学》下载免费的PDF全文 | |