积雪算法探究与分析.docVIP

山东科技大学本科毕业设计（论文） 摘要 积雪是地球表面变化最为频繁的自然因素之一，近四分之三的淡水资源都由积雪所存储。同时，在气象学和水文学的领域中，积雪也是十分重要的参数，即使是在全球气候和全球水循环的系统中也发挥着巨大的作用。前向模型是获取遥感参数反演的基础，要想得到最为实用的遥感反演算法就必须充分理解遥感器成像过程中的物理机制。雪水当量（Snow Water Equivalent）是描述全球季节性积雪覆盖情况的关键变量。但因为传统方法已经无法准确计算得到雪水当量变量信息了，例如地面测站数据插值或是适用单一航天测量数据的独立算法对积雪密度、分布及其演变有极大的不确定性。 本文章借助AMSR-E传感器的36GHz和18GHz水平极化被动微波遥感数据和地面基站实测数据来进行回归分析得到具有高度回归拟合的雪水当量的计算模型，并借用DMRT-QMS模型进行比较分析结果模型精度。最后分析植被覆盖类型、雪粒粒径、干湿雪等影响因子对雪水当量反演的影响。 总之，经过分析总结，我们发现地形因素、时间序列以及土地覆盖类型等三个变量对雪水当量分别有着不同程度的影响，例如在陡峭的山地区域的被动微波遥感数据和地面基站实测数据拟合效果差，在每一年的一月到三月的拟合效果好，在人口密集的城市回归拟合效果不明显。 关键字：被动微波、雪水当量、DMRT-QMS PAGE 1 ABSTRACT Snow is the most active natural component on the earth surface.3/4 of the total fresh water resource on land is stored in the form of snow.Moreover,snow is an important parameter for hydrology,meterology,climate and water cycle researches.The forward model is crucial for the retrieval of snow parameters by remote sensing.The development of inversion algorithms depend on the understanding of imaging mechanism. In the previous active/passive microwave and optical models for snow electromagnetic scattering assumed spherical,ellipsoild or other single shaped shaped particles in snow.But the microstructure of snow is very complex and random,and this can be verified by tomographic imaging or section images.This thesis focuses on the development and analysis of the most advanced electromagnetic models for active&passive microwave remote sensing and optical remote sensing considering the interaction of electromagnetic waves with the complex microstructure of real snow. Snow Water(Equivalent)is a key variable to describe the seasonal snow cover in the world.But because the traditional method has been unable to accurately calculate the snow water equivalent variable information,such as the ground station data interpolation or independent algorithm for single space measurement data has great uncertainty on snow density,distribution and evolution.With the help of AMSR-E sensor 36GHz and 18GHz leve