基于RS和GIS的建筑物空间分布格网化方法研究  

Multi-source and Multi-factor Gridding Method of Building Distribution Based on RS and GIS

作　　者：杨海霞[1] 王晓青[1] 窦爱霞[1] 李振敏[1] 

机构地区：[1]中国地震局地震预测研究所,北京100036

出　　处：《地震》2015年第3期136-146,共11页Earthquake

基　　金："十二五"科技支撑课题(2012BAK15B02)

摘　　要：本文在概述目前地震风险评估以及震后快速评估中建筑物空间分布格网化处理方法基础上,提出了基于遥感和GIS的多源数据多因子建筑物空间分布格网化方法。以云南省东川区为例,利用DEM数据、土地利用数据、基础地理数据等,提取地形、地貌、河流、道路、居民地和其他土地利用类型等各种影响因子,以300m格网为单元,研究了建筑物空间分布与各类影响因子之间的相关性,确定了各类子因子的建筑面积密度,实现了建筑物空间分布格网化预测。综合分析表明,考虑多影响因子的建筑物空间非均匀性分布,较传统的均匀性分布方法更接近真实空间分布,因而有助于提高地震风险评估和震后快速损失评估的准确性。In this paper, based on a review of method of building spatial distribution for seismic risk assessment and post-earthquake fast estimation, a gridding method of building distribution method is proposed considering multi-source data and multi-factors based on RS and GIS. We take Dongchuan, Yunnan as the study area and DEM, land use, basic geographic data to study the impact of factors like topography, landform, river, road, resi- dent and land use and so on. Then considering a gridding unit of 300 m, we study the in- fluence of different factors on building distribution and determine the distributing weights of factors and of sub-factors, then the building distribution is estimated in the study area. The results indicate that the building distribution estimated by the gridding method considering multiple factors is much closer to reality than traditional method. The method is expected to improve the accuracy of seismic risk assessment and post-earthquake fast esti- mation.

关 键 词：建筑物空间分布 格网化方法 遥感 地理信息系统 

分 类 号：P315.7[天文地球—地震学]