面向震害评估的人口数据空间化—以银川市为例

Spatialization of Population Data for Earthquake Damage Assessment—Taking Yinchuan as an Example

摘要: 本文以银川市为例，基于街道(乡镇)级人口统计数据和公里格网土地利用数据，建立多元线性回归模型，进行人口数据空间化，生成公里格网银川市人口地图，并基于此，计算了银川市地震烈度为7度、8度、9度时白天和夜晚的人员伤亡地图。结果表明，本文的人口空间化结果精度为26%，较为可靠；基于公里格网进行地震人员伤亡评估，打破了行政区划界限，能从整体上反应地震人员伤亡情况，同时也能反映各行政单元内部地震人员伤亡情况，从而为地震应急避难场所设置及震后救援提供更精准的数据支持。

Abstract: Taking Yinchuan as an example, based on the statistical population data of the street (township) level and the land use data of the kilometer grid, the multiple linear regression model is set up, the population data is spatialized, and the population map of the kilometer grid of Yinchuan is gener-ated. Based on this, distribution of casualty in the day and night is calculated when the seismic in-tensity in Yinchuan is at 7, 8 and 9 degrees. The results show that the precision of the population spatial results is 26% and more reliable, and the assessment of earthquake casualties based on the kilometer grid breaks the boundary of the administrative division, and can reflect the casualty of the earthquake personnel as a whole, and can also reflect the casualty of the internal seismic personnel in the administrative units, thus for the earthquake emergency refuge. Site settings and post earthquake rescue provide more accurate data support.