土壤含水量遥感反演及其在泥石流预报中的应用
文献类型：学位论文
| 作者 | 赵岩 |
| 学位类别 | 博士 |
| 答辩日期 | 2012-05-22 |
| 授予单位 | 中国科学院研究生院 |
| 授予地点 | 北京 |
| 导师 | 韦方强 |
| 关键词 | 泥石流 土壤含水量 遥感反演 泥石流预报 前期有效降水量 |
| 其他题名 | Soil Moisture Retrieval by Remote Sensing and Its Application in Debris Flow Forecasting |
| 学位专业 | 自然地理学 |
| 中文摘要 | 前期有效降水量EAR是泥石流预报中的关键因子之一，一般通过气象站点的观测数据内插分析得到，然而，受到地形的影响，山区降水分布很不均匀，现有的插值方法无法很好的把点雨量转成面雨量，从而影响了泥石流预报的准确率。本研究针对这一问题，分析前期有效降水量与土壤含水量的关系，采用热红外遥感的方法反演土壤含水量，利用土壤含水量替代前期有效降水量作为泥石流预报因子，改进基于泥石流成因的可拓预报模型，并以浙江省为例对改进前后的泥石流预报模型进行了对比研究，检验土壤含水量获取方法和用于泥石流预报的可靠性，以期能够为泥石流预报提供更可靠的前期降水数据支持，进一步提高泥石流预报的准确性。通过本项研究，取得了如下研究成果： 1. 在无云条件下，利用作物缺水指数CWSI法和温度植被指数TVDI法反演高植被覆盖地区土壤含水量。作物缺水指数法不需要对干边和湿边进行模拟，反演精度较高，可以更好地反演土壤含水量。 2. 通过建立裸地地表温度与云的光学厚度、气温的相关关系，实现了裸地地表温度的插值，同时建立了陆地表面温度与裸地地表温度和植被指数之间的相关关系，这样就可通过对气温的插值得到有云时的陆地表面温度。 3. 建立了晴朗度函数与云的光学厚度的回归方程，从而能够实现有云时坡地散射辐射的计算。 4. 利用MODIS产品中的MOD06云产品数据、MOD13植被指数和气象站气温的观测数据建立有云条件下的地表温度回归方程，并且讨论反照率获取方法，进而可以解决目前有云条件下，CWSI无法直接应用于土壤含水量的反演问题。 5. 分析了前期有效降水量与土壤含水量的关系，将土壤含水量作为泥石流预报因子替代前期有效降水量，对基于泥石流成因的可拓预报模型进行了改进，并将其应用到浙江省泥石流预报系统，对比改进前后两个系统的预报结果，改进后的系统对泥石流预报的准确性更高。论文的创新点如下： 1. 通过建立裸地地表温度与云的光学厚度、气温的相关关系，实现了裸地地表温度的插值，同时建立了陆地表面温度与裸地地表温度和植被指数之间的相关关系，这样就通过对气温的插值而得到了有云时的陆地表面温度；建立了晴朗度函数与云的光学厚度的回归方程，实现了有云时坡地散射辐射的计算。从而解决了在有云条件下无法利用现有反演方法获取土壤含水量的难题。 2. 通过分析前期降水量与土壤含水量的关系，利用土壤含水量替代了泥石流预报中的前期有效降水量因子，克服了因前期降水量内插结果误差大的问题，显著提高了泥石流预报的准确率。 |
| 英文摘要 | As a key factor to the debris flow forecasting, the effective antecedent rainfall is generally obtained by the way of interpolated analysis of the observation station data. However, due to the factors of very inhomogeneous rainfall distribution under the influence of topography and inadequate rainfall stations located in the valley mostly, the existing interpolation method can not satisfy this work that the area rainfall is induced from the point rainfall. Consequently, the inaccuracy of the effective antecedent rainfall interpolation caused by the above situation will further affect the accuracy of the debris flow forecasting. In order to solve this problem, the main purpose of this study is mainly to improve the mechanism-based extension debris flow forecasting model by the way of firstly establishing the relationship between the effective antecedent rainfall and the soil water content, then retrieving the soil water content using the remote sensing method, and finally using soil water content in place of the effective antecedent rainfall as the debris flow forecasting factor. The comparison study on the debris flow forecasting effect of the pre-and post forecasting models taking Zhejiang province for instance has been carried out for testing the reliability of the acquirement method of soil water content and its application into the debris flow forecast. In this study, this method of improving debris flow forecasting model is supposed to provide more reliable antecedent rainfall data for debris flow forecast and improve the forecasting precision further. The research results of this study are shown as follows: 1. In the condition of cloudless case, the soil water content of vegetation cover region is retrieved using the methods of crop water shortage index (CWSI) and temperature vegetation index (TVDI), respectively. By the comparative research of the two retrieval methods, the results show that the CWSI method need not simulate the dry edge and wet edge and retrieve the soil water content better with high retrieval accuracy. 2. The temperature interpolation of the bare soil has been carried out by determining the correlation of bare surface temperature and the variables including optical thickness, atmosphere temperature. Meanwhile, the correlation between land surface temperature and the variables with the bare surface temperature, the vegetation index is established. This way, the land surface temperature with the cloudy condition can be obtained by the way of atmosphere temperature interpolation. 3. The regression equation about the relationship of clearness index and cloudy optical thickness is established to server the requirement of the calculation of the scattering and radiation of the slopes in the cloudy condition. 4. The equation of the land surface temperature with cloudy condition based on the cloud product data of MOD06, vegetation index of MOD13 and the atmosphere temperature of the meteorological observing station is determined and the acquisition method of the albedo discussed. With the above methods, the problem that CSWI can not be used to retrieve the soil water content directly in the condition of the cloud covering has been solved. 5. Based on analyzing the relationship between the effective antecedent rainfall and the soil water content, the formation-mechanism-based extension debris flow forecasting model has been improved through replacing the effective antecedent rainfall with the soil water content considered as the debris flow forecasting factor, which is applied into the debris flow forecasting system of Zhejiang province. By contrasting the forecasting results of the pre-and post model, the forecasting accuracy of the latter model namely the improved model in this study is higher. The innovative points in this paper are shown as follows: 1. The temperature interpolation of the bare soil has been performed by establishing the correlation of bare surface temperature and the variables including the optical thickness, atmosphere temperature. Meanwhile, the correlation between land surface temperature and the variables with the bare surface temperature, the vegetation index is established. This way, the land surface temperature with the cloudy condition can be obtained by the way of atmosphere temperature interpolation. 2. The method that the effective antecedent rainfall considered as the debris flow forecasting factor is replaced by the soil water content by analyzing their relation will overcome the poor precision problem caused by the effective antecedent rainfall interpolation. And the accuracy of debris flow forecast is improved finally. |
| 语种 | 中文 |
| 公开日期 | 2013-01-16 |
| 源URL | [http://192.168.143.20:8080/handle/131551/4819]  |
| 专题 | 成都山地灾害与环境研究所_山地灾害与地表过程重点实验室 |
| 推荐引用方式 GB/T 7714 | 赵岩. 土壤含水量遥感反演及其在泥石流预报中的应用[D]. 北京. 中国科学院研究生院. 2012. |
入库方式： OAI收割
来源：成都山地灾害与环境研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。