Geospatial environmental and socioeconomic data

## Context

This dataset is a compilation of different vector and raster (gridded) data sources, presenting several relevant environmental and socioeconomic variables, such as the world population and GDP per 250 m grid, deforestation, air temperature, solar potential, among others. The data were divided into 12 folders:

1. Cities and towns (points and polygons)
2. Roads and railroads
3. Airports and ports
4. Power plants
5. Gridded Population 2015 (250 m)
6. Gridded Gross Domestic Product and Human Development Index over 1990-2015
7. Gridded Land cover 2015
8. Tree Cover Loss by Dominant Driver
9. Carbon accumulation potential from natural forest regrowth in forest and savanna biomes
10. Solar energy potential
11. Air temperature
12. Global cattle distribution in 2010 (5 minutes of arc)

Next, a description of the contents of each folder will be made. Additional metadata can be found in the respective folders and also in the original sources.

----------------------
1- Cities and towns (from Natural Earth)

**Description:** Shapefiles cointaining point symbols with name attributes and polygons with the cities areas (almost 7000 locations). Includes all admin-0 and many admin-1 capitals, major cities and towns, plus a sampling of smaller towns in sparsely inhabited regions. Also includes the name of cities in different languages, which makes it easier to compare with other data. Use the scale rankings to filter the number of towns that appear on your map. LandScan derived population estimates are provided for 90% of the cities. Those lacking population estimates are often in sparsely inhabited areas. Provide a range of population values that account for the total “metropolitan” population rather than it’s administrative boundary population. Use the PopMax column to size your town labels.

**Source:** https://www.naturalearthdata.com/downloads/10m-cultural-vectors/10m-populated-places/

**License:** Public Domain

----------------------
2- Roads and railroads (from Natural Earth)

**Description:** Shapefiles of roads and railroads of the North America, from the CEC North America Environmental Atlas. The database is at 1:1,000,000 scale (but holds up to 1:250,000 scale), covering all of the United States and Canada, Alaska, Hawaii, Puerto Rico, Mexico, and Belize and includes some attributes, such as Route number, Class, Type and State.

**Source:** https://www.naturalearthdata.com/downloads/10m-cultural-vectors/roads/ and https://www.naturalearthdata.com/downloads/10m-cultural-vectors/railroads/

**License:** Public Domain

----------------------
3 – Airports and ports (from Natural Earth)

**Description:** Shapefiles for the global airports and ports. Airports derive from Mile High Club, a detailed GIS compilation of world wide airports that is in the public domain. The “location” column is based on this Airport infrastructure diagram: https://en.wikipedia.org/wiki/File:Airport_infrastructure.png. Ports derives from High Seas, a detailed GIS compilation of world wide ports that is in the public domain.

**Source:** https://www.naturalearthdata.com/downloads/10m-cultural-vectors/airports/ and https://www.naturalearthdata.com/downloads/10m-cultural-vectors/ports/

**License:** Public Domain

----------------------
4- Global Power Plant Database (from WRI – World Resource Institute)

**Description:** The Global Power Plant Database is a comprehensive, open source database of power plants around the world (csv format). It centralizes power plant data to make it easier to navigate, compare, and draw insights for one’s own analysis. The database covers approximately 30,000 power plants from 164 countries and includes thermal plants (e.g., coal, gas, oil, nuclear, biomass, waste, geothermal) and renewables (e.g., hydro, wind, solar). Each power plant is geolocated, and entries contain information on plant capacity, generation, ownership, and fuel type.

**Source:** https://datasets.wri.org/dataset/globalpowerplantdatabase

**Citation:** Global Energy Observatory, Google, KTH Royal Institute of Technology in Stockholm, Enipedia, World Resources Institute. 2018. Global Power Plant Database. Published on Resource Watch and Google Earth Engine; http://resourcewatch.org/ https://earthengine.google.com/

**License:** Creative Commons Attribution 4.0 International License.

----------------------
5- Gridded Population 2015 (250 m) (from EC JRC and CIESIN)

**Description:** Distribution and density of population, expressed as the number of people per 250 m cell. The European Commission Joint Research Centre (EC JRC) and the Columbia University Earth Institute Center for International Earth Science Information Network (CIESIN) created a 250-meter resolution population grid. Residential population estimates for 2015 were generated by disaggregating census or administrative unit population data and re-mapping them to grid cells. The distribution and density of built-up area (from the Global Human Settlement Layer) informed the re-mapping of the population data. The model used raster-based density mapping relying on the Global Human Settlement – Built-Up Area data to define the distribution of people and inform the respective density. The Built-Up grid is the distribution of built-up areas displayed as the proportion of occupied area within each cell. Population estimates came from country-based census data and administrative polygons with estimated residential populations. Population grids were created using a volume-preserving density mapping approach.

**Source:** https://jeodpp.jrc.ec.europa.eu/ftp/jrc-opendata/GHSL/GHS_POP_MT_GLOBE_R2019A/

**Suggested Citation:** European Commission, Joint Research Centre (JRC); and Columbia University, Center for International Earth Science Information Network (CIESIN). 2015. "GHS Population Grid, Derived from GPW4, Multitemporal (2015)." Dataset: https://jeodpp.jrc.ec.europa.eu/ftp/jrc-opendata/GHSL/GHS_POP_MT_GLOBE_R2019A/.

**License:** Creative Commons Attribution 4.0 International License.

----------------------
6- Gross Domestic Product and Human Development Index over 1990-2015

**Description:** Multiannual datasets in gridded form for Gross Domestic Product (GDP) and Human Development Index (HDI). To provide a consistent product over time and space, the sub-national data were only used indirectly, scaling the reported national value and thus, remaining representative of the official statistics. This resulted in annual gridded datasets for GDP per capita (PPP), total GDP (PPP), and HDI, for the whole world at 5 arc-min resolution for the 25-year period of 1990–2015. Additionally, total GDP (PPP) is provided with 30 arc-sec resolution for three time steps (1990, 2000, 2015). Gross domestic product (GDP) is a key economic indicator that measures the monetary value of goods and services provided within that area. This gridded GDP dataset is important because it can be used for research at the sub-national level.

Subfiles content:
**- Administrative units:** Represents the administrative units used for GDP per capita (PPP) and HDI data products. National administrative units have id 1-999, sub-national ones 1001-
admin_areas_GDP_HDI.nc

**- GDP_per_capita_PPP_1990_2015:** The GDP per capita (PPP) dataset represents average gross domestic production per capita in a given administrative area unit. GDP is given in 2011 international US dollars. Gap-filled sub-national data were used, supplemented by national data where necessary. Datagaps were filled by using national temporal pattern. Dataset has global extent at 5 arc-min resolution for the 26-year period of 1990-2015. Detail description is given in a linked article and metadata is provided as an attribute in the NetCDF file itself.

**- GDP_PPP_1990_2015_5arcmin:** This global dataset represents the gross domestic production (GDP) of each grid cell. GDP is given in 2011 international US dollars. The data is derived from GDP per capita (PPP) which is multiplied by gridded population data HYDE 3.2 (the years of population data not available (1991-1999) were linearly interpolated at grid scale based on data from years 1990 and 2000). Dataset has global extent at 5 arc-min resolution for the 26-year period of 1990-2015. Detail description is given in a linked article and metadata is provided as an attribute in the NetCDF file itself.

**-HDI_1990_2015:** HDI is a composite index of average achievement in key dimensions of human development (dimensionless indicator between 0 and 1). This index is based on method introduced 2010 and updated 2011. The subnational data for HDI were collected from multiple national-level datasets, and national-level HDI was collected from UNDP. Years with missing data were interpolated over time thin plate spines, assuming smooth trend over time. The dataset has a global extent at 5 arc-min resolution, and the annual data is available for each year over 1990-2015. HDI sub-national data covers 39 countries and 66% of global population in 2015.

**-pedigree_GDP_per_capita_PPP_1990_2015:** This is the source data for GDP per capita (PPP), published as an indication of accuracy and precision. Reports the scale (national, sub-national) and type (reported, interpolated, extrapolated) of each year of data. Detail description is given in a linked article and metadata is provided as an attribute in the NetCDF file itself.

**- pedigree_HDI_1990_2015:** This is the source data for Human Development Index (HDI), published as an indication of accuracy and precision. Reports the scale (national, sub-national) and type (reported, interpolated, extrapolated) of each year of data. Detail description is given in a linked article and metadata is provided as an attribute in the NetCDF file itself. Detail description is given in a linked article and metadata is provided as an attribute in the NetCDF file itself.

**- GDP_PPP_30arcsec:** The GDP (PPP) data represents average gross domestic production of each grid cell. GDP is given in 2011 international US dollars. The data is derived from GDP per capita (PPP), which is multiplied by gridded population data from Global Human Settlement (GHS). Dataset has a global extent at 30 arc-second resolution for three time steps: 1990, 2000, and 2015. Detail description is given in a linked article and metadata is provided as an attribute in the NetCDF file itself.
kummu_etal_scidata_code: This file contains the scripts for data handling and production

**Source:** https://datadryad.org/stash/dataset/doi:10.5061/dryad.dk1j0

**Citation:** Kummu, Matti; Taka, Maija; Guillaume, Joseph H. A. (2020), Data from: Gridded global datasets for Gross Domestic Product and Human Development Index over 1990-2015, Dryad, Dataset, https://doi.org/10.5061/dryad.dk1j0

**License:** CC0 1.0 Universal (CC0 1.0) Public Domain Dedication.

----------------------
7- Gridded Land cover 2015 (from ESA-CCI)

**Description:** Consistent global land cover map at 300 m spatial resolution for 2015 (GeoTIFF format). Each pixel value corresponds to the label of a land cover class defined based on the UN Land Cover Classification System (LCCS). LCCS classifiers support the further conversion into Plant Functional Types distribution required by the Earth System Models. The typology counts 22 classes.

**Source:** http://maps.elie.ucl.ac.be/CCI/viewer/download.php

**License terms:**
> The present products are made available to the public by ESA and the consortium. You may use one or several CCI-LC products land cover map for educational and/or scientific purposes, without any fee on the condition that you credit the ESA Climate Change Initiative and in particular its Land Cover project as the source of the CCI-LC database:
Copyright notice:
© ESA Climate Change Initiative - Land Cover led by UCLouvain (2017)
Should you write any scientific publication on the results of research activities that use one or several CCI-LC products as input, you shall acknowledge the ESA CCI Land Cover project in the text of the publication and provide the project with an electronic copy of the publication ([e-mail]).
If you wish to use one or several CCI-LC products in advertising or in any commercial promotion, you shall acknowledge the ESA CCI Land Cover project and you must submit the layout to the project for approval beforehand ([e-mail]).

----------------------
8- Tree Cover Loss by Dominant Driver (from GFW, Global Forest Watch)

**Description:** This dataset shows the dominant driver of tree cover loss from 2001-2019 within each 10 km grid cell, using the following five categories:
• Commodity-driven deforestation: long-term, permanent conversion of forest and shrubland to a non-forest land use such as agriculture (including oil palm), mining, or energy infrastructure
• Shifting agriculture: small to medium-scale forest and shrubland conversion for agriculture that is later abandoned and followed by subsequent forest regrowth
• Forestry: large-scale forestry operations occurring within managed forests and tree plantations
• Wildfire: large-scale forest loss resulting from the burning of forest vegetation with no visible human conversion or agricultural activity afterward
• Urbanization: forest and shrubland conversion for the expansion and intensification of existing urban centers.
The commodity-driven deforestation and urbanization categories represent permanent deforestation, while tree cover usually regrows in the other categories. The data were generated using decision tree models to separate each 10 km grid cell into one of the five categories. All model code, reference samples, decision trees, and the final model are available in the Supplementary Materials of the paper: Curtis, P.G., C.M. Slay, N.L. Harris, A. Tyukavina, and M.C. Hansen. 2018. “Classifying Drivers of Global Forest Loss.” Science. https://science.sciencemag.org/content/361/6407/1108

**Source:** https://data.globalforestwatch.org/datasets/tree-cover-loss-by-dominant-driver

**Citation:** Curtis, P.G., C.M. Slay, N.L. Harris, A. Tyukavina, and M.C. Hansen. 2018. “Classifying Drivers of Global Forest Loss.” Science. Accessed through Global Forest Watch on 04/11/2020. www.globalforestwatch.org.

**License:** Creative Commons Attribution 4.0 International License.

----------------------
9- Carbon accumulation potential from natural forest regrowth in forest and savanna biomes (from GFW, Global Forest Watch)

**Description:** This map (1 km resolution) shows the rate at which forests could capture carbon from the atmosphere and store it in aboveground live biomass over the first 30 years of natural forest regrowth in potentially reforestable areas (Mg carbon/ha/yr). It was created by combining ground-based measurements at thousands of locations around the world with 66 co-located environmental covariate layers in a machine learning model to produce a wall-to-wall map. Forest plot data used to train the model are sourced from published literature, which can be found in the Forest Carbon database (ForC, maintained by the Smithsonian Institute (https://github.com/forc-db)), as well as georeferenced data from publicly available national forest inventories. Although rates were estimated over all forest and savanna biomes globally, they are filtered here by “reforestable” area, as defined in Griscom et al. 2017 (PNAS). Reforestable areas exclude areas of native grasslands and croplands to safeguard the production of food and fiber and habitat for biological diversity.

**Source:** https://www.arcgis.com/home/item.html?id=2b1e75c7d6274e448954178b3bc31bea

**Citation:** Cook-Patton, S.C., Leavitt, S.M., Gibbs, D. et al. Mapping carbon accumulation potential from global natural forest regrowth. Nature 585, 545–550 (2020).

**Credits:** Cook-Patton, S.C., S.M. Leavitt, D. Gibbs, N.L. Harris, K. Lister, K.J. Anderson-Teixeira, R.D. Briggs, R.L. Chazdon, T.W. Crowther, P.W. Ellis, H.P. Griscom, V. Herrmann, K.D. Holl, R.A. Houghton, C. Larrosa, G. Lomax, R. Lucas, P. Madsen, Y. Malhi, A. Paquette, J.D. Parker, K. Paul, D. Routh, S. Roxburgh, S. Saatchi, J.van den Hoogen, W.S. Walker, C.E. Wheeler, S.A. Wood, L. Xu, B.W. Griscom. 2020. Mapping carbon accumulation potential from natural forest regrowth. Nature, in press. https://www.nature.com/articles/s41586-020-2686-x. This work resulted from a collaboration between The Nature Conservancy, World Resources Institute, and 18 other institutions.

**License:** CC BY 4.0

----------------------
10- Solar energy potential (from World Bank Group/ESMAP/Solargis)

**Description:** The Solar Energy Potential dataset is made up of three GeoTIFF layers: yearly average daily global horizontal irradiation (GHI), global irradiation for optimally tilted surface (GTI) and photovoltaic power potential (PVOUT). Data layers are provided in a geographic spatial reference (EPSG:4326). The resolution (pixel size) of solar resource data (GHI and GTI,) is 9 arcsec (nominally 250 m), PVOUT is 30 arcsec (nominally 1 km).
GHI is the sum of direct and diffuse radiation received on a horizontal plane presented in the units kilowatt hour per square meter (kWh/m²). Radiation is energy that moves from one place to another and in this dataset it refers to energy released by the sun traveling to Earth. Irradiance refers to the average amount of radiation received in a given area on Earth. Photovoltaic power potential, or Potential photovoltaic electricity Production (PVOUT), is the estimated amount of energy converted by a photovoltaic system into electricity presented in kilowatt-hour per kilowatt peak of the photovoltaic system (kWh/kWp) according to the geographical conditions of a site and configuration of the photovoltaic system. GHI acts as an important base measurement to help determine regions that receive enough sunlight and is used to assess an area for flat-plate photovoltaic and solar heating technologies. PVOUT works to calculate the actual effectiveness of a solar energy project in the region based on local conditions. The datasets were created using geostationary satellite imagery and meteorological models, including SolarGIS’s global solar model (v2.1). Satellite imagery data comes from the National Oceanic and Atmospheric Administration (NOAA), North American Space Agency (NASA), European Organisation for the Exploitation of Meteorological Satellites (EUMESTAT), and Japan Meteorological Administration (JMA).
For more information and terms of use, please, read metadata, provided in PDF and XML format in the file. For other data formats, resolution or time aggregation, please, visit Solargis website.

**Source:** https://globalsolaratlas.info/download/world

**Suggested citation:** Global Solar Atlas. 2018. Retrieved from http://globalsolaratlas.info/. Accessed through Resource Watch, (date). https://www.resourcewatch.org.

**License:** Creative Commons Attribution 4.0 International

----------------------
11- Air temperature (from World Bank Group/ESMAP/Solargis)

**Description:** A GeoTIFF layer with the air Temperature at 2 m above ground level in °C (TEMP). Data layers are provided in a geographic spatial reference (EPSG:4326). The resolution (pixel size) of TEMP is 30 arcsec (nominally 1 km). For more information and terms of use, please, read metadata, provided in PDF and XML format in the file. For other data formats, resolution or time aggregation, please, visit Solargis website.

**Source:** https://globalsolaratlas.info/download/world

**License:** Creative Commons Attribution 4.0 International

----------------------

12- Global cattle distribution in 2010 (5 minutes of arc)

**Description:** This dataset contains the global distribution of cattle in 2010 expressed in total number of cattle per pixel (5 min of arc) according to the Gridded Livestock of the World database (GLW 3). Please go through the 1_Ct_2010_Metadata.html file for more information about this dataset and the set of included files.

**Source:** https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GIVQ75

**Citation:** Gilbert, Marius; Nicolas, Gaëlle; Cinardi, Giusepina; Van Boeckel, Thomas P.; Vanwambeke, Sophie; Wint, William G. R.; Robinson, Timothy P., 2018, "Global cattle distribution in 2010 (5 minutes of arc)", https://doi.org/10.7910/DVN/GIVQ75, Harvard Dataverse, V3

**More information:** https://www.nature.com/articles/sdata2018227#Sec9

**License:** Public Domain.

本数据集汇集了多种矢量数据和栅格（网格）数据源，涵盖了多个与环境和社会经济相关的变量，例如全球人口、每250米网格的国内生产总值（GDP）、森林砍伐、空气温度、太阳能潜力等。数据被分为12个文件夹：

1. 城市和城镇（点与多边形）
2. 公路和铁路
3. 机场和港口
4. 发电站
5. 2015年网格化人口（250米）
6. 1990-2015年网格化国内生产总值（GDP）和人类发展指数
7. 2015年网格化土地覆盖
8. 树冠损失的主要驱动因素
9. 森林和草原生物群落中自然森林恢复的碳积累潜力
10. 太阳能潜力
11. 空气温度
12. 2010年全球牛群分布（5分弧度）

接下来，将对每个文件夹的内容进行描述。相关元数据可在各自文件夹和原始来源中找到。

----------------------
1- 城市和城镇（来自自然地球）

**描述：** 包含具有名称属性的点符号形状文件和表示城市面积的 polygon 形状文件（近7000个地点）。包括所有 admin-0 和许多 admin-1 首都、主要城市和城镇，以及稀疏居住区域中的小型城镇样本。还包括不同语言中城市的名称，这使得与其他数据比较更加容易。使用规模排名来过滤地图上出现的城镇数量。为90%的城市提供了来自 LandScan 的推导人口估计。那些缺乏人口估计的通常位于稀疏居住区域。提供一系列人口值，以反映“大都市”总人口，而不是其行政边界总人口。使用 PopMax 列来调整城镇标签的大小。

**来源：** https://www.naturalearthdata.com/downloads/10m-cultural-vectors/10m-populated-places/

**许可证：** 公共领域

----------------------
2- 公路和铁路（来自自然地球）

**描述：** 来自北美环境图集的北美公路和铁路的 shapefiles。数据库以1:1,000,000比例尺（但实际上可以达到1:250,000比例尺），覆盖了美国和加拿大、阿拉斯加、夏威夷、波多黎各、墨西哥和伯利兹，并包括一些属性，如路线号、类别、类型和州。

**来源：** https://www.naturalearthdata.com/downloads/10m-cultural-vectors/roads/ 和 https://www.naturalearthdata.com/downloads/10m-cultural-vectors/railroads/

**许可证：** 公共领域

----------------------
3 – 机场和港口（来自自然地球）

**描述：** 全球机场和港口的 shapefiles。机场数据来自 Mile High Club，这是一个详尽的全球机场 GIS 编纂，属于公共领域。'位置'列基于以下机场基础设施图：https://en.wikipedia.org/wiki/File:Airport_infrastructure.png。港口数据来自 High Seas，这是一个详尽的全球港口 GIS 编纂，属于公共领域。

**来源：** https://www.naturalearthdata.com/downloads/10m-cultural-vectors/airports/ 和 https://www.naturalearthdata.com/downloads/10m-cultural-vectors/ports/

**许可证：** 公共领域

----------------------
4- 全球发电厂数据库（来自世界资源研究所）

**描述：** 全球发电厂数据库是一个综合性的开源数据库，收集了全球范围内的发电厂数据（csv格式）。它集中了发电厂数据，以便于导航、比较和为个人分析提供见解。该数据库涵盖了来自164个国家的约30,000个发电厂，包括热电厂（例如，煤炭、天然气、石油、核能、生物质、废物、地热）和可再生能源（例如，水力、风能、太阳能）。每个发电厂都进行了地理定位，条目包含有关发电厂容量、发电、所有权和燃料类型的信息。

**来源：** https://datasets.wri.org/dataset/globalpowerplantdatabase

**引用：** 全球能源观测站、谷歌、瑞典皇家理工学院、Enipedia、世界资源研究所。2018年。全球发电厂数据库。在资源守望和谷歌地球引擎上发布；http://resourcewatch.org/ https://earthengine.google.com/

**许可证：** 创作共用Attribution 4.0国际许可证。

----------------------
5- 2015年网格化人口（250米）（来自欧洲委员会联合研究中心和哥伦比亚大学地球研究所国际地球科学信息网络）

**描述：** 人口分布和密度的分布，以每250米单元格中的人数表示。欧洲委员会联合研究中心（EC JRC）和哥伦比亚大学地球研究所国际地球科学信息网络中心（CIESIN）创建了一个250米分辨率的人口网格。2015年的住宅人口估计是通过将人口普查或行政单元人口数据分解并重新映射到网格单元格中产生的。建成区面积的分布和密度（来自全球人类居住层）为人口数据的重新映射提供了信息。所使用的模型是基于栅格的密度映射，依赖于全球人类居住-建成区数据来定义人口分布和相应的密度。建成区网格是每个单元格内占用面积的分布，以占用面积的比例表示。人口估计来自基于国家的人口普查数据和估计的住宅人口行政多边形。

**来源：** https://jeodpp.jrc.ec.europa.eu/ftp/jrc-opendata/GHSL/GHS_POP_MT_GLOBE_R2019A/

**建议引用：** 欧洲委员会，联合研究中心（JRC）；哥伦比亚大学，国际地球科学信息网络中心（CIESIN）。2015年。“GHS人口网格，源自GPW4，多时相（2015年）。”数据集：https://jeodpp.jrc.ec.europa.eu/ftp/jrc-opendata/GHSL/GHS_POP_MT_GLOBE_R2019A/。

**许可证：** 创作共用Attribution 4.0国际许可证。

----------------------
6- 1990-2015年国内生产总值（GDP）和人类发展指数

**描述：** 以网格形式提供的多年国内生产总值（GDP）和人类发展指数（HDI）数据集。为了在时间和空间上提供一致的产品，仅间接使用次国家数据，通过缩放报告的国家值，从而保持官方统计数据的代表性。这导致了1990-2015年整个世界的年度网格化GDP人均（PPP）、总GDP（PPP）和HDI数据集，分辨率为5弧分。此外，总GDP（PPP）以30弧秒分辨率提供了三个时间步长（1990年、2000年、2015年）。国内生产总值（GDP）是衡量该区域内提供商品和服务货币价值的宏观经济指标。这个网格化GDP数据集的重要性在于它可以用于次国家层面的研究。

子文件内容：

**- 行政单元：** 表示用于GDP人均（PPP）和HDI数据产品的行政单元。国家行政单元的ID为1-999，次国家行政单元的ID为1001-
admin_areas_GDP_HDI.nc

**- GDP_per_capita_PPP_1990_2015：** GDP人均（PPP）数据集表示给定行政单元的平均国内生产总值。GDP以2011年国际美元表示。在必要时补充了国家数据，以填补次国家数据中的数据空缺。数据空缺是通过使用国家时间序列模式填补的。数据集具有全球范围，分辨率为5弧分，涵盖了1990-2015年26年的时期。详细信息在链接的文章中给出，NetCDF文件本身提供了元数据。

**- GDP_PPP_1990_2015_5arcmin：** 这是全球数据集，表示每个网格单元的国内生产总值（GDP）。GDP以2011年国际美元表示。数据源自GDP人均（PPP），乘以HYDE 3.2网格化人口数据（人口数据不可用的年份（1991-1999年）在网格尺度上基于1990年和2000年的数据线性插值）。数据集具有全球范围，分辨率为5弧分，涵盖了1990-2015年26年的时期。详细信息在链接的文章中给出，NetCDF文件本身提供了元数据。

**- HDI_1990_2015：** HDI是一个复合指数，衡量人类发展关键维度的平均成就（介于0和1的无量纲指标）。该指数基于2010年引入并更新于2011年的方法。HDI的次国家数据来自多个国家层面的数据集，国家层面的HDI来自联合国开发计划署（UNDP）。数据缺失的年份通过时间薄板样条插值，假设时间趋势平滑。

**- pedigree_GDP_per_capita_PPP_1990_2015：** 这是GDP人均（PPP）的源数据，作为准确性和精度的指示。报告了每年数据的规模（国家、次国家）和类型（报告的、插值的、外推的）。详细信息在链接的文章中给出，NetCDF文件本身提供了元数据。

**- pedigree_HDI_1990_2015：** 这是人类发展指数（HDI）的源数据，作为准确性和精度的指示。报告了每年数据的规模（国家、次国家）和类型（报告的、插值的、外推的）。详细信息在链接的文章中给出，NetCDF文件本身提供了元数据。

**- GDP_PPP_30arcsec：** GDP（PPP）数据表示每个网格单元的平均国内生产总值。GDP以2011年国际美元表示。数据源自GDP人均（PPP），乘以来自全球人类居住（GHS）的网格化人口数据。数据集具有全球范围，分辨率为30弧秒，涵盖了三个时间步长：1990年、2000年和2015年。详细信息在链接的文章中给出，NetCDF文件本身提供了元数据。

kummu_etal_scidata_code：此文件包含数据处理和生产脚本。

**来源：** https://datadryad.org/stash/dataset/doi:10.5061/dryad.dk1j0

**引用：** Kummu，Matti；Taka，Maija；Guillaume，Joseph H. A.（2020），数据来自：1990-2015年全球国内生产总值和人类发展指数的网格化数据集，Dryad，数据集，https://doi.org/10.5061/dryad.dk1j0

**许可证：** CC0 1.0通用（CC0 1.0）公共领域奉献。

----------------------
7- 2015年网格化土地覆盖（来自欧洲航天局-CCI）

**描述：** 2015年一致性的全球土地覆盖地图，空间分辨率为300米（GeoTIFF格式）。每个像素值对应于基于联合国土地覆盖分类系统（LCCS）定义的土地覆盖类别标签。LCCS分类器支持进一步转换为地球系统模型所需的植物功能类型分布。典型学数有22个类别。

**来源：** http://maps.elie.ucl.ac.be/CCI/viewer/download.php

**许可条款：**
> 现有产品由ESA和联盟向公众提供。您可以在不收取任何费用的条件下使用一个或多个CCI-LC产品土地覆盖地图进行教育和/或科学研究，前提是您将ESA气候变化倡议及其特别项目土地覆盖作为CCI-LC数据库的来源进行信用。版权声明：
© ESA气候变化倡议 - 土地覆盖由UCLouvain领导（2017）
如果您撰写任何基于使用一个或多个CCI-LC产品作为输入的研究活动的科学出版物，您应在出版物的文本中承认ESA CCI土地覆盖项目，并向该项目提供出版物的电子副本（[e-mail]）。
如果您希望将一个或多个CCI-LC产品用于广告或任何商业推广，您应承认ESA CCI土地覆盖项目，并且必须在使用之前将布局提交给项目进行批准（[e-mail]）。

----------------------
8- 树冠损失的主要驱动因素（来自全球森林监视）

**描述：** 此数据集显示了2001-2019年每个10公里网格单元中树冠损失的主要驱动因素，使用以下五个类别：
• 商品驱动的森林砍伐：长期、永久地将森林和灌木丛转换为非森林土地利用，如农业（包括油棕）、采矿或能源基础设施
• 转移农业：小到中等规模的森林和灌木丛转换为农业，随后被放弃并随后森林恢复
• 林业：在管理森林和树木种植园内进行的大规模林业作业
• 山火：由于森林植被的燃烧而导致的大规模森林损失，之后没有明显的人类转换或农业活动
• 城市化：森林和灌木丛转换为现有城市中心的扩张和集约化。
商品驱动的森林砍伐和城市化类别代表永久性森林砍伐，而树冠通常在其他类别中恢复。数据使用决策树模型生成，将每个10公里网格单元分类为五个类别之一。所有模型代码、参考样本、决策树和最终模型都可在论文的补充材料中找到：Curtis，P.G.，C.M. Slay，N.L. Harris，A. Tyukavina，和 M.C. Hansen。2018年。“全球森林损失驱动因素的分类。”科学。https://science.sciencemag.org/content/361/6407/1108

**来源：** https://data.globalforestwatch.org/datasets/tree-cover-loss-by-dominant-driver

**引用：** Curtis，P.G.，C.M. Slay，N.L. Harris，A. Tyukavina，和 M.C. Hansen。2018年。“全球森林损失驱动因素的分类。”科学。通过全球森林监视于2020年4月11日访问。www.globalforestwatch.org。

**许可证：** 创作共用Attribution 4.0国际许可证。

----------------------
9- 森林和草原生物群落中自然森林恢复的碳积累潜力（来自全球森林监视）

**描述：** 此地图（1公里分辨率）显示了森林在自然森林恢复的前30年中从大气中捕获并储存在地上活生物量中的碳的速率，这些地区具有潜在的再造林能力（每公顷/年碳/兆克）。它通过将成千上万的地点的地面测量与66个共定位的环境协变量层结合在一个机器学习模型中创建。用于训练模型的森林样地数据来自已发表的文献，可在森林碳数据库（ForC，由史密森尼研究所维护的 https://github.com/forc-db）中找到，以及来自公开可用的国家森林清查的地理参考数据。尽管估计了全球所有森林和草原生物群落中的碳积累速率，但在此处通过“可再造林”区域进行筛选，如Griscom等人2017年（PNAS）中定义。可再造林区域排除了原生草地和农田，以保护食物和纤维的生产以及生物多样性的栖息地。

**来源：** https://www.arcgis.com/home/item.html?id=2b1e75c7d6274e448954178b3bc31bea

**引用：** Cook-Patton，S.C.，Leavitt，S.M.，Gibbs，D.等人。绘制全球自然森林恢复的碳积累潜力。自然 585，545–550（2020）。

**信用：** Cook-Patton，S.C.，S.M. Leavitt，D. Gibbs，N.L. Harris，K. Lister，K.J. Anderson-Teixeira，R.D. Briggs，R.L. Chazdon，T.W. Crowther，P.W. Ellis，H.P. Griscom，V. Herrmann，K.D. Holl，R.A. Houghton，C. Larrosa，G. Lomax，R. Lucas，P. Madsen，Y. Malhi，A. Paquette，J.D. Parker，K. Paul，D. Routh，S. Roxburgh，S. Saatchi，J.van den Hoogen，W.S. Walker，C.E. Wheeler，S.A. Wood，L. Xu，B.W. Griscom。2020年。绘制自然森林恢复的碳积累潜力。自然，待发表。https://www.nature.com/articles/s41586-020-2686-x。这项工作是由世界自然保护组织、世界资源研究所和其他18个机构之间的合作产生的。

**许可证：** CC BY 4.0

----------------------
10- 太阳能潜力（来自世界银行集团/ESMAP/Solargis）

**描述：** 太阳能潜力数据集由三个GeoTIFF层组成：年平均每日全球水平辐射（GHI）、优化倾斜表面的全球辐射（GTI）和光伏发电潜力（PVOUT）。数据层以地理空间参考（EPSG:4326）提供。太阳能资源数据的分辨率（像素大小）为9弧秒（相当于250米），PVOUT为30弧秒（相当于1公里）。GHI是直接辐射和散射辐射在水平平面上接收的总和，以每平方米千瓦时（kWh/m²）的单位表示。辐射是能量从一个地方移动到另一个地方，在此数据集中，它指的是太阳释放到地球的能量。辐照度是指地球上某个区域接收到的平均辐射量。光伏发电潜力或潜在光伏电力生产（PVOUT）是根据地理位置和光伏系统的配置估计的光伏系统转换成电力的能量量，以每千瓦峰值光伏系统千瓦时（kWh/kWp）表示。GHI作为一项重要的基础测量，有助于确定接收足够阳光的地区，并用于评估平板光伏和太阳能加热技术。PVOUT旨在根据当地条件计算地区太阳能项目的实际有效性。数据集使用地球静止卫星图像和气象模型创建，包括SolarGIS的全球太阳能模型（v2.1）。卫星图像数据来自国家海洋和大气管理局（NOAA）、北美航天局（NASA）、欧洲气象卫星组织（EUMESTAT）和日本气象厅（JMA）。

**来源：** https://globalsolaratlas.info/download/world

**建议引用：** 全球太阳能图集。2018年。从http://globalsolaratlas.info/检索。通过资源守望访问，（日期）。https://www.resourcewatch.org。

**许可证：** 创作共用Attribution 4.0国际

----------------------
11- 空气温度（来自世界银行集团/ESMAP/Solargis）

**描述：** 包含地面2米以上