本文利用AMSR/E被动微波扫描辐射计的亮度温度数据(L3产品)及对应的地表观测台站的地表温度数据研究了青藏高原地区2004年一月份至2004年五月份土壤冻融的变化规律以及空间分布。所采用的土壤冻融监测算法主要是基于Zuerndorfer等人提出的经验算法，由于该算法是针对于SMMR和SMM/I数据的。本研究首先通过分析AMSR/E被动微波扫描辐射计不同频率波段的光谱特性，针对AMSR/E数据寻找出了适合冻融分类的亮度温度频率波段Tb36.5Ghz以及Tb18.7Ghz，V极化频率波段。采用统计回归的方法，通过选取典型区域的亮度温度观测数据和对应的地面实测温度数据(样本总数为236个)来进行回归分析，根据回归分析的结果，建立了适合我国青藏高原地区的土壤冻融的判别算即Tb36.5GhzV极化亮度温度数据小于262K和18.7Ghz至36.5Ghz V极化谱梯度值小于0K。然后运用所确定的土壤冻融判别算法，针对青藏高原地区冻融交替的几个月份分别进行了土壤冻融监测制图，并根据相应日期的实测地表温度数据对监测结果进行了验证。通过对我国青藏高原地区2004年一月至2004年五月份的土壤冻融监测结果我们可以看出,运用该方法监测我国青藏高原地区的土壤冻融界限具有很大的潜力，进一步说明了本文所建立的算法能够较好的反映研究区域的土壤冻融状况。由于该算法目前还不能区分冻土和积雪以及植被等因素的影响，加之青藏高原地区是一个特殊的高海拔地区，最终导致局部地区土壤的冻融监测误差较大。本文进一步讨论了影响土壤冻融判别结果的各种因素。最后，本研究对今后应该进一步研究的问题提出了一些新的思考。

Based on the brightness temperature data by AMSR/E(the advanced microwave scanning radiometer) L3 daily product and the earth surface temperature data(0cm depth)by the observation sites. This thesis has researched the freeze/thaw boundary of soil and its space distributing in Tibet-Qinghai Plateau from January to May in 2004. In virtue of the freeze/thaw algorithm, which is based on the empirical algorithm established by Zuerndorfer et.al.,it mainly deals with SMMR and SMM/I data. We have got the AMSRE frequency band Tb36.5Ghz and Tb 18.7Ghz V polarization that are suitable for the soil freeze/thaw classification through frequency and polarization mode of AMSR/E L3 daily brightness temperature data. Using the linear regression method with 236 soil ter nperature data (0 cm depth) and the corresponding brightness temperature data, we established the soil freeze/thaw algorithm, Tb36.5Ghz V polarization less than 262K and negative spectral gradient of Tb18.7Ghz and 36.5Ghz, which is suitable for the Tibet-Qinghai Plateau.Then we do the soil freeze/thaw mapping for Tibet-Qinghai Plateau with the new soil freeze/thaw algorithm January to May in 2004.We can find that this method has great potential for surveying the soil freeze/thaw status from the mapping result. As a result of the fact that we cannot distinguish frozen soil and snow now by using this method.In addition, due to the special high altitude terrain of Tibet-Qinghai Plateau, it results in a little bigger error in some part of the research area.We also research more about the factors, which affect the soil freeze/thaw mapping result. In the end, we provide some new thoughts for the furtherstudy.