Annual Map of Temporary Agricultural Cultures

The Annual Map of Temporary Agricultural Cultures (Macat) is a product of the Land Occupation Monitoring System (SMOS), an innovative initiative, designed and developed by the Directorate-General of the Territory, with the aim of continuously producing cartographic information on land use and occupation. SMOS uses the latest developments in space technologies and Artificial Intelligence to create products with more detail, quality, speed and made available with open data policy. All products can be viewed on Vismos (https://smos.dgterritorio.gov.pt/vi-smos). Macat is an experimental product, resulting from DGT’s research and development activities, and users are encouraged to report a critical analysis on the use of the map to [e-mail]. Macat was created to provide complementary information to the Conjuntural Soil Occupation Charter (COSC), with an annual production frequency and update, and a 10 m pixel raster format. Two experimental editions were produced for the year 2020 and 2021. The Macat classifies annual agriculture within the temporary agricultural areas delimited in the 2018 Land Use and Occupation Charter (COS), specifically classes 2.1.1.1 and 2.1.1.2 (the SOC nomenclature is described at https://www.dgterritorio.gov.pt/sites/default/files/documentos-publicos/ET-COS-1995-2007-2010-2015-2018-v1.pdf). Within this perimeter, the Macat provides additional information for COSC areas classified as annual autumn/winter crops and annual spring/summer crops, specifically classes 211 and 212 (the COSC nomenclature is described at https://www.dgterritorio.gov.pt/sites/default/files/documentos-publicos/Nomenclatura_COSc.pdf). The Macat also includes the pixels classified in COSC as Matos, spontaneous herbaceous vegetation and vegetation-free areas (classes 410, 420 and 500) as long as they occur within COS2018 temporary agriculture areas. The Macat nomenclature consists of more than 30 classes identifying agricultural crops and other occupations (the Macat nomenclature is described at https://www.dgterritorio.gov.pt/sites/default/files/ficheiros-cartografia/nomenclatura_macat.pdf). Agricultural crops correspond to annual crops (e.g. wheat) selected on the basis of the Land Parcel Identification System (LPIS) of the Institute for Financing for Agriculture and Fisheries (IFAP). The selection of crops is made independently in the 14 landscape units that divide the mainland in COSC’s production methodology. In each landscape unit, all annual crops occupying 1 % or more of the LPIS area in the respective landscape unit are selected. Thus, Macat includes all major annual crops even if their distribution is representative only in a restricted area of the territory. The remaining non-agricultural classes included in the Macat nomenclature are the three classes mentioned above (Matos, spontaneous herbaceous vegetation and vegetation-free areas) that may occur in agricultural areas of COS2018, for example as a result of fallow or other management practices. Macat is produced through space technologies and Artificial Intelligence, which includes machine learning algorithms and expert knowledge rules to automatically classify multispectral and intra-annual series of Sentinel-2 satellite optical image data. To carry out this process is built a spectral database formed by monthly composites, spectral indices and intra-annual metrics of Sentinel-2 images for the agricultural year under analysis (i.e., from October 2020 to September 2021 for MACAT2021). The training database of each class for automatic classification is obtained by automatic processing of auxiliary information and by photointerpretation. If you are not familiar with spatial data being made available through viewing services (e.g. WMS) and downloading services, you can consult the Support Guides on the DGT Open Data Page (https://www.dgterritorio.gov.pt/dados-abertos).年度临时农业用地地图（Macat）是领土总局（Directorate-General of the Territory，DGT）设计开发的创新项目——土地占用监测系统（Land Occupation Monitoring System，SMOS）的产出成果，其核心目标是持续生成土地利用与占用的制图信息。SMOS依托航天技术与人工智能（Artificial Intelligence）的最新进展，打造出兼具更高精度、质量与时效性的产品，并以开放数据政策对外发布。所有产品均可通过Vismos平台（https://smos.dgterritorio.gov.pt/vi-smos）查看。 Macat属于DGT研发活动产出的试验性产品，鼓励用户针对该地图的使用情况提交批判性分析意见，反馈邮箱为[e-mail]。 Macat旨在为《临时土地占用宪章》（Conjuntural Soil Occupation Charter，COSC）提供补充信息，其生产与更新频率为年度，像素栅格格式为10米。目前已针对2020年与2021年推出两个试验版本。 Macat将2018年土地利用与占用宪章（Land Use and Occupation Charter，COS）划定的临时农业区域内的年度农业用地进行分类，具体对应SOC分类体系（SOC nomenclature）中的2.1.1.1与2.1.1.2类，该分类体系的说明文档可参见https://www.dgterritorio.gov.pt/sites/default/files/documentos-publicos/ET-COS-1995-2007-2010-2015-2018-v1.pdf。在此范围内，Macat针对COSC分类体系（Conjuntural Soil Occupation Charter，COSC）中划为年度秋/冬作物与年度春/夏作物的区域提供补充信息，具体对应COSC分类体系中的211与212类，其说明文档可参见https://www.dgterritorio.gov.pt/sites/default/files/documentos-publicos/Nomenclatura_COSc.pdf。此外，只要位于2018年COS划定的临时农业区域内，Macat也会纳入COSC分类体系中被划为灌丛（Matos）、自然草本植被与无植被区域（对应分类代码410、420与500）的像素。 Macat的分类体系包含30余个类别，涵盖农作物与其他用地类型，其分类体系说明可参见https://www.dgterritorio.gov.pt/sites/default/files/ficheiros-cartografia/nomenclatura_macat.pdf。农作物类别基于农业与渔业融资研究所（Institute for Financing for Agriculture and Fisheries，IFAP）的土地地块识别系统（Land Parcel Identification System，LPIS）选取，包括小麦等年度作物。作物选取工作在COSC生产方法论中划分的葡萄牙大陆14个景观单元内独立开展，每个景观单元内，凡占对应单元LPIS总面积1%及以上的年度作物均会被纳入。因此，即便某些主要年度作物仅在境内局部区域广泛分布，Macat也会将其涵盖。Macat分类体系中其余非农业类别即为前文提及的三类用地（灌丛、自然草本植被与无植被区域），这类用地可能出现在2018年COS划定的农业区域内，例如休耕或其他管理措施导致的区域。 Macat通过航天技术与人工智能生成，其中集成了机器学习算法与专家知识规则，以自动对Sentinel-2卫星（Sentinel-2）光学影像的多光谱数据与年度内时序序列进行分类。该生产流程构建了光谱数据库，该数据库由待分析农业年度（例如MACAT2021对应的2020年10月至2021年9月）内的Sentinel-2影像月度合成数据、光谱指数与年度内时序指标构成。自动分类所用的各类别训练数据库，通过辅助信息自动处理与目视解译获取。 若您不熟悉通过查看服务（如WMS服务，WMS）与下载服务获取空间数据的方式，可查阅DGT开放数据页面（https://www.dgterritorio.gov.pt/dados-abertos）上的支持指南。