基于土地利用类型的村级人口空间分布模拟——以湖北鹤峰县为例

发布时间：2018-04-20 17:00

本文选题：村级人口 + 空间分布　；参考：《地球信息科学学报》2014年03期

【摘要】：在人口分布及其相关研究中,常常会遇到小尺度人口数据部分缺失的问题。本文以湖北省鹤峰县为例,在分析土地利用与人口分布关系的基础上,从全局与局部、线性回归与非线性回归考虑,基于土地利用类型,分别利用地理加权回归(GWR)方法、格网方法、BP神经网络方法对缺失数据的行政村人口数据进行模拟,并进行了多角度精度对比验证。研究结果表明:(1)各种土地利用类型中,耕地、林地、城镇村及工矿用地、交通用地是影响研究区村级人口分布的主要因素;(2)30个调查村中,3种方法模拟的人口总数误差小于3%,通过每个村的模拟值与实际值相比,BP神经网络方法能更好地模拟研究区村级人口的分布,格网方法次之,GWR方法最差;(3)研究区各村人口分布呈现较高的空间正相关性,各乡镇的人口密度在空间上并不独立,而是呈现紧密的集聚特征。
[Abstract]:In the study of population distribution and its related research, the problem of partial absence of small scale population data is often encountered. Taking Hefeng County in Hubei Province as an example, based on the analysis of the relationship between land use and population distribution, considering the global and local factors, linear regression and nonlinear regression, and based on land use type, this paper respectively uses the method of geographical weighted regression (GWR) to analyze the relationship between land use and population distribution. The grid method and BP neural network method are used to simulate the missing data of administrative village population, and the multi-angle accuracy is compared and verified. The results show that among the various land use types, cultivated land, woodland, urban villages and industrial and mining land, Traffic land is the main factor affecting the population distribution at village level in the study area.) in 30 villages, the error of total population simulated by three methods is less than 3. The simulation value of each village is better than the BP neural network method compared with the actual value. To study the distribution of population at the village level, The population distribution of each village in the study area showed a high spatial positive correlation, and the population density of each township was not independent in space, but showed a close agglomeration characteristic.
【作者单位】：首都师范大学资源环境与地理信息系统北京市重点实验室;首都师范大学三维信息获取与应用教育部重点实验室;首都师范大学城市环境过程与数字模拟国家重点实验室培育基地;
【基金】：国家自然科学基金项目(41371375) 北京市自然科学基金项目(8132018) “十二五”国家科技支撑计划项目(2012BAH33B03、2012BAH33B05)
【分类号】：C924.24;P208

【参考文献】