【作者】 邱洁；

【导师】 过仲阳；

【作者基本信息】 华东师范大学 ， 地图学与地理信息系统， 2006， 硕士

【摘要】 灾害天气，如暴雨、洪涝、龙卷风等，会给人们的生命财产和国民经济建设带来重大的损失。而暴雨等强对流天气与中尺度对流系统(Mesoscale Convective Systems，简称MCS)的发生发展密切相关。青藏高原的高海拔和特定地表的动力和热力作用能明显影响到MCS的形成和发展，从而造成长江流域的强降雨天气。然而，由于影响青藏高原上MCS移动的因素十分复杂，因此对中尺度对流系统及其所伴随的强对流天气的研究和预报具有重要的理论意义。 本文利用日本GMS红外辐射亮温(Tbb)资料，运用面积比法以及点模式匹配法对青藏高原上的MCS进行了自动追踪，并对这两种方法的追踪结果进行了比较分析。在此基础上，提取了MCS周边的环境物理量场特征值，进而运用空间数据挖掘中的关联规则建立了高原上MCS东移传播与其环境物理量场之间的数学模型，确立了长江流域暴雨中心周边的环境场特征与暴雨中心暴雨量之间的关系。研究成果对于提高灾害天气预报的准确性，减少因灾害天气所造成的损失具有重要的实践价值。 论文共分六章。第一章对近年来国内外对MCS的研究进展进行了详细的论述，在此基础上，提出了论文的研究思路。第二章阐述了空间数据挖掘中关联规则的主要原理。第三章主要介绍了基于点模式匹配的MCS的自动追踪方法，在此基础上，得到了1999-2000年6-8月各月MCS的移动路径、大小、强度、活动区域及生命史等信息，并将点模式匹配法与传统面积比法的追踪结果进行了比较分析。第四章运用空间关联规则研究了高原上东移的MCS与其环境物理量场之间的关系，揭示了影响高原上MCS移动传播的环境场条件，进而以地理信息系统软件为平台，构建了1999-2000年夏季高原上MCS及其周边环境物理量场的时空数据库。第五章运用空间关联规则提取了长江流域暴雨中心周边的环境场特征，揭示了暴雨发生的环境场条件。第六章对论文的研究成果进行了概括和总结。更多还原

【Abstract】 Disaster weather, for instance, heavy rainfall, flood, tornado, will usually bring about grave loss to the life and property of people and the construction of national economy. Strong convective weathers have a great deal with the origin and development of Mesoscale Convective System (MCS). The height and the exceptional dynamic field and thermal field of the Tibetan Plateau can have an effect on the origin and development of MCS, and beget heavy rainfall in Yangtze River Basin. However, the factors that affect the moving of MCS on the Tibetan Plateau are very complex, so it is very important to study and forecast strong convective weathers caused by mesoscale convective systems.In this paper, the trajectories of MCS over the Tibetan Plateau are automatically tracked by area-overlapped method and point-pattern matching method using GMS (Geostationary Meteorological Satellite) brightness temperature (Tbb), and then the results of the two methods are compared. Based on these, the environmental physical field values of MCS are abstracted and spatial association rule mining is used to study the relationships between the trajectories of MCS and their environmental physical field values and the relationship between the rainfall of rainstorm center in Yangtze River and the feature values of environmental physical field as well. These are of important value to improve the predicted accuracy of disaster weather and reduce the loss due to it.Six chapters are included in this paper. Chapter one discuss recent study resultsof MCS and heavy rainfall, based on this, methodology is provided. The fundamental principles of association rule of spatial data mining are represented in chapter two. Automatic tracking method based on point-pattern matching of MCS is discussed in chapter three, on the basis of these, some information about MCS, including trajectories, size, intensity, active regions, life characteristics etc, are obtained from June to August in 1999 and 2000. Besides, these results have been compared with that of area-overlapped method in this section. In chapter four, the relationship between the moving out of the Tibetan Plateau and to the east of MCS and their environmental physical field values is analyzed by spatial association rule. It reveals the condition of更多还原