中国人口空间分布形态模拟与预测——基于“五普”和“六普”的分县尺度人口密度研究

Simulation and Projection on the Pattern of Chinese Population Spatial Distribution:Research with Population Density of County Level Based on the 5^(th) and 6^(th) Population Census of China

人口空间分布不平衡是一个普遍现象,但其"是否有分布和统计规律"的研究议题非常鲜见。通过中国第五次和第六次人口普查的分县尺度数据模拟出中国分县尺度人口密度的概率分布函数,并进行概率分布预测,分析了未来中国人口分布统计特征。主要结论有:模拟显示,中国分县尺度人口密度符合对数正态分布形态;以2000年分县尺度数据的前向(未来)预测检验和2010年数据的后向(过去)预测检验发现,"人均产出"和"地均产出"两个条件能很好地模拟未来人口密度的分布规律;以2010年分县尺度为基础预测了2020-2100年的中国人口密度概率分布特征和趋势,结果显示中国人口分布极化特征将持续加大,例如2050年,无人区(小于1人/平方公里)的县域比例将从当前的1.02%(29个)增加到4.49%(128个),超高密集区(大于30000人/平方公里)的区县比例将从0.11%(3个)增加到3.93%(112个)。

Population spatial distribution unevenness is an universal phenomenon,but the issue that wether the population distribution is in accordance with some statistic law is worth to be studied.This paper will simulate and project the probability distribution function of population density based on the county level data of the 5th and 6th China Population Census,and then discuss the distribution characteristics of future population density in China. The main results show that 1） Chinese county level population density is in accordance with Log Normal Distribution; 2） output per capita and output per land can in a large extent fit the future probability distribution of population density in China based on the 5th and 6th China Population Census data; 3） the gap of population distribution in China from the year 2020 to 2100 will be larger based on the projection with the 6th China Population Census data; For instance,from 2010 to 2050,the ratio of Non-people Area countries（ where density 1 per/km2） will increase from 1. 02%（ No. 29） to 4. 49%（ No. 128） and the ratio of Dense-people Area countries（ where density 30000 per/km2） will increase from 0. 11%（ No.3） to3. 93%（ No. 112）.