CAJ | 학술논문

人口居住空间分布不均，以及通勤和迁移等人口流动导致人口呈时空动态分布，这是人口承灾体的重要属性。人口承灾体的时空分布模拟可通过人口密度网格化方法和基于个体时空标记的模拟方法来实现。人口密度网格化方法主要有平均分配法、格点内插法、人口分布影响因子分析法、人口分布规律法和遥感估算法等五类。其中，人口分布影响因子分析法和遥感估算法是目前研究的热点，广泛应用于不同尺度人口密度网格化数据库的构建。大数据时代，基于个体时空标记的方法为监测与模拟人口承灾体时空动态分布提供了新的技术与途径，主要包括基于众源地理数据和基于移动基站数据的人口时空模拟方法。在此基础上，进一步分析了人口时空动态分布数据在自然灾害风险分析、风险管理与应急管理中的应用。基于手机通话数据和众源地理数据的模拟方法将与人口网格化方法互补，共同构建高时空分辨率、动态的人口空间化数据。另外，城市和脆弱性群体的时空分布将是人口承灾体研究的重要方向。
인구거주공간분포불균，이급통근화천이등인구류동도치인구정시공동태분포，저시인구승재체적중요속성。인구승재체적시공분포모의가통과인구밀도망격화방법화기우개체시공표기적모의방법래실현。인구밀도망격화방법주요유평균분배법、격점내삽법、인구분포영향인자분석법、인구분포규율법화요감고산법등오류。기중，인구분포영향인자분석법화요감고산법시목전연구적열점，엄범응용우불동척도인구밀도망격화수거고적구건。대수거시대，기우개체시공표기적방법위감측여모의인구승재체시공동태분포제공료신적기술여도경，주요포괄기우음원지리수거화기우이동기참수거적인구시공모의방법。재차기출상，진일보분석료인구시공동태분포수거재자연재해풍험분석、풍험관리여응급관리중적응용。기우수궤통화수거화음원지리수거적모의방법장여인구망격화방법호보，공동구건고시공분변솔、동태적인구공간화수거。령외，성시화취약성군체적시공분포장시인구승재체연구적중요방향。
The distribution of population is spatially uneven.Qeople’s daily commute and long-term migration further result in dynamic distribution of population.The spatial-temporal variation in population forms the basic characteristic of population exposure,which is of significant importance for disaster risk assessment and manage-ment.There are two ways to simulate the spatial-temporal distribution of population,namely,the dasymetric map-ping of population census data and individual geo-information mining.The dasymetric mapping of population census data is an important field in demographic research and population exposure assessment recently.It includes average allocation,grid-point interpolation,factor-weighted analysis,population distribution theory method and remote sensing-based estimation,among which factor-weighted analysis and remote sensing-based estimation are most pop-ular.In the era of big data,individual geo-information mining provides an emerging way to monitor and simulate population distribution and population exposure to natural disaster.Individual geo-information mining mainly focu-ses on two data sources,namely,crowd-sourcing geographic data and cell-phone-based spatial information.Based on the review of above methods,this paper further discusses their applications in natural disaster risk analysis,risk management and emergency management.Finally,it argues that future research should combine the dasymetric mapping of population census data with individual geo-information mining to construct a dynamic database of popu-lation exposure with high spatial-temporal resolution and multi-level vulnerability.Special attentions should be paid to spatio-temporal urban population and vulnerable groups in developing such a database,e.g.the elders, women,children,persons with disabilities and migrants,who are more vulnerable to hazards and diseases.