The GHSL relies on the design and implementation of new spatial data mining technologies allowing to automatically process and extract analytics and knowledge from large amount of heterogeneous data including: global, fine-scale satellite image data streams, census data, and crowd sources or volunteered geographic information sources.

The GHS-SMOD is the rural-urban Settlement classification MODel adopted by the GHSL. It is the representation of the degree of urbanization (​​DEGURBA​​) concept into the GHSL data scenario. Each grid in the GHS-SMOD has been generated by integrating the GHSL built-up areas and GHSL population grids data for reference epochs: 1975, 1990, 2000, 2015.

The DEGURBA classification schema is a people-based definition of cities and settlements: it operates using as main input a 1 km² grid cell accounting for population at a given point in time. The DEGURBA discriminates the population grid cells in three main classes: '​​urban centers​​​' (cities), '​​urban clusters​​​' (towns and suburbs), and '​​rural grid cells​​'. (base). These class abstractions translate to 'high density clusters (HDC)', 'low density clusters (LDC)', and 'rural grid cells (RUR)', respectively, in the GHS-SMOD implementation.

The 'HDC' differ from the DEGURBA '​​urban centers​​' in that they account for the over-fragmentation of cities in regions with large low-density residential development by integrating the built-up layer. In the GHS-SMOD representation, the 'HDC' are the spatial generalization of contiguous population grid cells (4-connectivity, gap-filling) with a density of at least 1500 inhabitants per km² or a density of built-up surface > 50%, and a minimum total resident population of 50000. The 'LDC' are continuous grid cells with a density of at least 300 inhabitants per km² and a minimum total population of 5000. The 'RUR' are grid cells outside 'HDC' and 'LDC' with population > 0 and < 300. Everything else is classified as inhabited areas where population = 0.

This dataset was produced in the World Mollweide projection (EPSG:54009).

For more information visit: ​​Global Human Settlement - GHS SETTLEMENT GRID - European Commission​​.

The Global Human Settlement Layer (GHSL) project is supported by the European Commission, Joint Research Center, and Directorate-General for Regional and Urban Policy. The GHSL produces new global spatial information, evidence-based analytics, and knowledge describing the human presence in the planet.

GHSL 依赖于新的空间数据挖掘技术的设计和实施,允许从大量异构数据中自动处理和提取分析和知识,这些数据包括:全球、精细规模的卫星图像数据流、人口普查数据和人群来源或自愿地理信息来源。

GHS-SMOD 是 GHSL 采用的城乡聚落分类模型。它是城市化程度 ( ​​DEGURBA​​ ) 概念在 GHSL 数据场景中的表示。GHS-SMOD 中的每个网格都是通过整合 GHSL 建成区和 GHSL 人口网格数据生成的,用于参考时期:1975、1990、2000、2015。

DEGURBA 分类模式是基于人的城市和住区定义:它使用 1 平方公里的网格单元作为主要输入来计算给定时间点的人口。DEGURBA 将人口网格单元分为三个主要类别:“​​城市中心​​​”(城市)、“​​城市群​​​”(城镇和郊区)和“​​农村网格单元​​”。(根据)。在 GHS-SMOD 实现中,这些类抽象分别转化为“高密度集群 (HDC)”、“低密度集群 (LDC)”和“农村网格单元 (RUR)”。

“HDC”与 DEGURBA 的“​​城市中心​​”的不同之处在于,它们通过整合建成层来解释具有大型低密度住宅开发区域的城市的过度碎片化。在 GHS-SMOD 表示中,“HDC”是连续人口网格单元(4 连通性,间隙填充)的空间概括,其密度至少为每平方公里 1500 名居民或建筑表面密度 > 50% ,最小总居民人口为 50000。“LDC”是连续网格单元,密度至少为每平方公里 300 名居民,最小总人口为 5000。“RUR”是“HDC”和“LDC”之外的网格单元' 人口 > 0 且 < 300。其他一切都被归类为人口 = 0 的居住区。

该数据集是在 World Mollweide 投影 (EPSG:54009) 中生成的。

欲了解更多信息,请访问: http: ​​//ghsl.jrc.ec.europa.eu/ghs_smod.php​​。

全球人类住区层 (GHSL) 项目得到了欧盟委员会、联合研究中心以及区域和城市政策总局的支持。GHSL 产生新的全球空间信息、基于证据的分析和描述地球上人类存在的知识。

分辨率
1000米

波段


姓名

描述

​smod_code​

城市化程度


smod_code 类表


价值

颜色

描述

0

000000

居住区

1

448564

RUR(农村网格单元)

2

70daa4

LDC(低密度集群)

3

ffffff

HDC(高密度集群)


使用条款

GHSL 由 EC JRC 作为开放和免费数据制作。授权重复使用,前提是确认来源。如需更多信息,请阅读使用条件(​​欧盟委员会重复使用和版权声明​​)。

代码:

var dataset = ee.ImageCollection('JRC/GHSL/P2016/SMOD_POP_GLOBE_V1')
.filter(ee.Filter.date('2015-01-01', '2015-12-31'));
var degreeOfUrbanization = dataset.select('smod_code');
var visParams = {
min: 0.0,
max: 3.0,
palette: ['000000', '448564', '70daa4', 'ffffff'],
};
Map.setCenter(114.96, 31.13, 4);
Map.addLayer(degreeOfUrbanization, visParams, 'Degree of Urbanization');

Citations:

Google Earth Engine(GEE)——全球人类居住区网格数据 1975-1990-2000-2014 (P2016)_人类