The fraction of land cover classes; derived variables from ESA CCI Land Cover time-series (1992 - 2018)
收藏资源简介：
GEOEssential project (http://www.geoessential.eu/) is built on an end-user-driven approach to first identify environmental policy indicators, then their associated Essential Variables (EVs), and finally the appropriate Earth Observation data sources. In the context of the GEOEssential project, we developed a workflow to aggregate the ESA CCI Land Cover time-series (https://www.esa-landcover-cci.org/) in a more end-user friendly format and structure. The ESA CCI Land Cover time-series describe the land surface with 22 classes at a spatial resolution of 0.002778° (~ 300 m) annually from 1992 to 2018. We calculated the fraction of each class at two windows of 10 x 10 pixels and 100 x 100 pixels for the entire time-series. As a result, we produced 22 layers corresponding to each of the classes, per year, presenting the fraction of each class in ~3 km and ~30 km globally from 1992 to 2018. This provides the end-users with the possibility to explore the changes in a fraction of a single land cover class, e.g. urban areas, and easily employ the ESA CCI Land Cover time-series in their works with a significantly less computational cost at the global scale. Furthermore, we merged the three classes of cultivated areas, forest areas, and urban areas, and calculated the fraction of each class following the classification used by the IPCC Assessment Report 5 and the IPBES Global Assessment. These derived variables could be used to monitor ecosystem structure and changes of specified land cover classes extent in time and space. In this contribution, we aim to inform the community about the availability of such derived variables and to present some example applications. "Niamir_POSTER_The Fraction of Land Cover Classes.pdf": a poster presented at the GEO BON #OSC2020 "frac100_allyears.7z": : All years, all classes, fraction in 100 x 100 pixels in NetCDF format "frac10_allyears.7z": : All years, all classes, fraction in 10 x 10 pixels in NetCDF format "fraccover_classbased_100.7z" : All years, all classes, class by class, fraction in 100 x 100 pixels in GeoTIFF format "fraccover_classbased_10.7z" : All years, all classes, class by class, fraction in 10 x 10 pixels in GeoTIFF format
GEOEssential项目（http://www.geoessential.eu/）以终端用户需求为导向的研究方法构建，首先明确环境政策指标，继而确定其关联的核心变量（Essential Variables, EVs），最终筛选适配的地球观测（Earth Observation, EO）数据源。本项目中，我们开发了一套工作流，将欧洲空间局气候变化倡议陆地覆盖时间序列数据集（ESA CCI Land Cover time-series，https://www.esa-landcover-cci.org/）整合为更贴合终端用户使用习惯的格式与结构。 ESA CCI陆地覆盖时间序列数据集以0.002778°（约300米）的空间分辨率，每年生成22类地表覆盖分类结果，时间跨度为1992年至2018年。我们针对全时间序列，分别在10×10像素与100×100像素两个窗口下计算每一类别的占比。最终得到22层数据，对应每一年的每一类覆盖类型，在全球范围内以约3千米与约30千米的尺度呈现各类别的占比，时间跨度同样为1992年至2018年。该数据集可为终端用户提供探索单一陆地覆盖类别（例如城市区域）占比变化的可能，并帮助用户以极低的全局计算成本，便捷地将ESA CCI陆地覆盖时间序列数据集应用于相关研究工作中。 此外，我们将耕地、森林与城市三类覆盖类型合并，并依据政府间气候变化专门委员会第五次评估报告（IPCC Assessment Report 5）及生物多样性和生态系统服务政府间科学政策平台全球评估（IPBES Global Assessment）的分类标准，重新计算各类别的占比。这些衍生变量可用于监测生态系统结构，以及特定陆地覆盖类别的时空范围变化。 本研究旨在向学界通报此类衍生数据集的可用性，并展示部分典型应用案例。 "Niamir_POSTER_The Fraction of Land Cover Classes.pdf"：于GEO BON #OSC2020会议展示的海报 "frac100_allyears.7z"：全年度、全类别100×100像素占比数据，格式为NetCDF "frac10_allyears.7z"：全年度、全类别10×10像素占比数据，格式为NetCDF "fraccover_classbased_100.7z"：全年度、全类别逐类100×100像素占比数据，格式为GeoTIFF "fraccover_classbased_10.7z"：全年度、全类别逐类10×10像素占比数据，格式为GeoTIFF
