中国公里网格人口分布数据集

传统的人口数据来源于全国人口普查数据，以行政区为基本单元的统计型数据，空间分辨率低，无法充分揭示人口数据的空间差异，同时，无法与以格网等基础地理单元数据的共享与整合。因此，统计型人口空间化，即将以行政区为基本单元的统计人口护甲展布到一定空间尺度（本数据集为1 公里）格网上，构建人口空间分布数据集，能够与土地利用数据、自然资源及生态环境等自然要素数据进行综合分析，实现人文社会经济数据与自然要素的整合，具有重要的理论与现实意义。中国公里格网人口分布数据集的生产，首先，综合分析我国人口分布的空间特征及区域差异进行人口区划。在区划后的各子区域中建立多元统计人口空间化模型进行人口空间化，并利用城市人口密度、交通状况、DEM 及总量控制四个因子对模型进行校正，以确保模型的合理性和准确性。为了对人口数据的精度进行校验，选择我国东部、西部及中部选择典型省份中具有完整乡镇数据人口的地区作为检验样本，以统计数据为真值，空间化人口数据的相对误差在4.5% 到13.6% 之间，而且大部分样本的相对误差小于10%。最终，形成一套具有统一空间坐标参数、统一数据格式、统一元数据标准的公里格网人口空间分布数据集。

Traditional population data originates from national census data, which is statistical data taking administrative divisions as basic units. It has low spatial resolution, failing to fully reveal the spatial variations of population data, and cannot be shared or integrated with basic geographic unit data such as grids. Therefore, the spatialization of statistical population, i.e., extending the administrative division-based statistical population distribution to spatial grids of a certain scale (1 kilometer in this dataset) to construct a population spatial distribution dataset, enables comprehensive analysis with natural factor data such as land use data, natural resources and ecological environment, and realizes the integration of humanistic, social and economic data and natural factors, which has important theoretical and practical significance.

The production of the China 1-kilometer grid population spatial distribution dataset begins with population regionalization, which is carried out by comprehensively analyzing the spatial characteristics and regional differences of China's population distribution. Then, multivariate statistical spatialization models for population are established in each sub-region after regionalization to conduct population spatialization, and four factors including urban population density, traffic conditions, DEM and total amount control are used to calibrate the models to ensure their rationality and accuracy.

To verify the accuracy of the population data, regions with complete township-level population data in typical provinces across eastern, western and central China are selected as test samples. Taking the statistical data as the true value, the relative error of the spatialized population data ranges from 4.5% to 13.6%, and the relative error of most samples is less than 10%.

Finally, a 1-kilometer grid population spatial distribution dataset with unified spatial coordinate parameters, unified data format and unified metadata standards is finalized.

搜集汇总

背景与挑战

背景概述

该数据集是中国2005-2010年期间的公里网格人口空间分布数据集，覆盖全国范围，分辨率为1公里，采用栅格格式。它通过人口空间化方法将行政区域统计人口数据展布到网格上，并利用城市密度、交通、DEM等因素校正模型，以提高空间分布的准确性，验证显示多数样本相对误差低于10%，适用于与自然因素数据的集成分析。