Keywords

1 UMD LC Classification—University of Maryland Land Cover Classification

Product

LULC general

Dates

1992/93 (1 km)

1984 (8 km)

1987 (1°)

Formats

Raster

Pixel size

1 km, 8 km, 1°

Thematic resolution

15 Classes – 1 km products

1 (a), 1 (ag), 10 (v), 1 (m), 1 (na)Footnote 1

Compatible legends

UMD, IGBP

Extent

Global

Updating

No

Change detection

No (only one date)

Overall accuracy

Expected to be >65%

Website of reference

Website Language English, Spanish, French, Arabic, Russian

https://daac.ornl.gov/ISLSCP_II/guides/umd_landcover_xdeg.html

Download site

http://iridl.ldeo.columbia.edu/SOURCES/.UMD/.GLCF/.GLCDS/.lc/datafiles.html

Availability

Format(s)

Open Access

.lan, .img

Technical documentation

Hansen et al. (2000)

Other references of interest

DeFries and Townshend (1994), DeFries et al. (1995), Hansen and Reed (2000), McCallum et al. (2006)

Project

The Department of Geography of the University of Maryland hosted one of the first research groups to use the classification of satellite imagery for global LUC mapping. They initially produced an LUC map at a spatial resolution of 1 degree for the year of reference 1987. This was followed sometime later by the production of a finer map at 8 km for 1984. Finally, the project delivered a map at 1 km, which at that time was the finest resolution at which global LUC mapping had ever been carried out.

The Global Land Cover Facility that hosted all this data recently went offline. This means that there is currently no official website that supports the datasets and provides information about their particular specifications. The map at 1 km can however be downloaded from external sites. The earlier maps at coarser resolutions are no longer available.

Production method

The UMD LC was obtained through supervised classification with a decision tree algorithm of imagery captured by the AVHRR sensor. Urban and built-up areas were not mapped, nor were water covers. Instead, they were extracted from auxiliary sources. The classification obtained in this way was then improved in a post-classification stage by expert regional labelling, based on inconsistencies that were identified by the experts.

Product description

Users can download the UMD LC Classification in two formats (.lan, .img), which are available in the section “GIS-Compatible Formats”. The download is not easy and does only include the raster file with LUC information.

Downloads

LAN file

– Raster file with LUC map

Legend and codification

Code

Label

Code

Label

0

Water

8

Closed Shrubland

1

Evergreen Needleleaf Forest

9

Open Shrubland

2

Evergreen Broadleaf Forest

10

Grassland

3

Deciduous Needleleaf Forest

11

Cropland

4

Deciduous Broadleaf Forest

12

Bare Ground

5

Mixed Forest

13

Urban and Built-up

6

Woodland

14

Unclassified

7

Wooded Grassland

  

Practical considerations

There is no official website hosting this dataset, which makes it more difficult to access and understand. Users must bear in mind that this was one of the first global LUC datasets ever developed and it can therefore be considered outdated in technical terms.

Coarser versions of the 1 km map, resampled at 0.25, 0.5 and 1 degree of spatial resolution, are also available.Footnote 2

2 GLCC 2.0 Global—Global Land Cover Characterization 2.0

Product

LULC general

Dates

1992 / 93

Formats

Raster

Pixel size

1 km

Thematic resolution

100 classes (Global ecosystems legend)

19 classes (IGBP legend):

1 (a), 1 (ag), 10 (v), 2 (m), 2 (na)

Compatible legends

Global Ecosystems, IGBP, USGS LULC system, SiB, SiB 2, BATS, Vegetation lifeforms

Extent

Global

Updating

No

Change detection

No (only one date)

Overall accuracy

Expected to be > 66%

Website of reference

Website Language English

https://www.usgs.gov/centers/eros/science/usgs-eros-archive-land-cover-products-global-land-cover-characterization-glcc?qt-science_center_objects=0#qt-science_center_objects

Download site

https://earthexplorer.usgs.gov/

Availability

Format(s)

Open Access after registration

.tiff, .bil

Technical documentation

Belward et al. (1999), Brown et al. (1999), Loveland and Belward (1997), Loveland et al. (2000), Reed et al. (2000)

Other references of interest

Hansen and Reed (2000)

Project

The GLCC dataset was the result of collaboration between several international institutions: the U.S. Geological Survey (USGS), the Earth Resources Observation and Science (EROS) Center, the University of Nebraska-Lincoln (UNL) and the Joint Research Centre (JRC) of the European Commission. The project aimed to create a dataset of reference for global land monitoring. One of the LUC maps obtained from the project is usually referred to as the DISCover LUC map and follows the IGBP classification scheme.

The global LUC map was created by joining various continental LUC maps together, and the final product consisted of a generalized global map and a set of more detailed continental maps.

Two versions of the dataset have been produced so far, with the first being released in 1997. The second version (2.0) improved on the first by applying both the lessons learnt and user feedback. Version 1.2 of the product included the IGBP classification (DISCover LUC map).

Production method

The dataset was obtained through unsupervised classification (CLUSTER classifier) of AVHRR imagery at a spatial resolution of 1 km. The classification obtained was further refined with the help of auxiliary data from the Digital Elevation Model (DEM), Ecoregions data and other thematic maps specific for each region. Label-assignment for the spectral classes was based on expert interpretation.

The dataset production was split into different continents, according to their specific characteristics. A detailed LUC map was produced for each continent and these were then joined together to create the global LUC product.

Product description

Two GLCC maps are available for download: the global product and the specific LUC product for each continent. The continental LUC maps show more detail than the global one and have specific legends that disaggregate the complexity of the land uses and covers for each continent.

The data can be downloaded in two different formats (.bil, .tiff). The download for each format includes the LUC maps with all the various classification schemes, together with technical documentation about the product. The continental product also includes a specific binary raster which maps the built-up land cover.

The product is distributed in two different projections: the Goode projection and a geographic projection.

Downloads

Global land cover product—Goode projection (“glccgbg20_tif”)

– Raster files with LUC maps for each of the 7 classification schemes included in the product

– PDF document with technical information about the product

European land cover product—Goode projection (“glcceag20_tif”)

– Raster files with LUC maps for each of the classification schemes included in the product

– Raster file with urban land cover information (built-up/non built-up)

– PDF document with technical information about the product

Legend and codification

LUC maps for each continent include a specific regional classification scheme, which is not shown here. The global dataset also supports seven different classification schemes. The most detailed of these is the Global Ecosystems (GLCC) scheme. In this case, however, we will only display the IGBP Land Cover classification scheme (IGBP), because it is the most commonly used of all the schemes provided by the dataset.

Information about the codification and the meaning of all the other classification schemes can be found in the technical documentation included in the downloaded product, as well as in the documentation available on the project’s website.Footnote 3

IGBP Land Cover (IGBP) Legend

Code

Label

Code

Label

1

Evergreen Needleleaf Forest

11

Permanent Wetlands

2

Evergreen Broadleaf Forest

12

Croplands

3

Deciduous Needleleaf Forest

13

Urban and Built-Up

4

Deciduous Broadleaf Forest

14

Cropland/Natural Vegetation Mosaic

5

Mixed Forest

15

Snow and Ice

6

Closed Shrublands

16

Barren or Sparsely Vegetated

7

Open Shrublands

17

Water Bodies

8

Woody Savannas

99

Interrupted Areas (Goode’s Homolosine Projection)

9

Savannas

100

Missing Data

10

Grasslands

  

Practical considerations

For more information about the product, users are referred to its readme file,Footnote 4 which explains the project history, the dataset production workflow and all the characteristics of the product.

3 GLC2000—Global Land Cover 2000

Product

LULC general

Dates

2000

Formats

Raster

Pixel size

1 km

Thematic resolution

23 classes:

1 (a), 1 (ag), 15 (v), 3 (m), 1 (na)

Compatible legends

FAO LCCS, IGBP

Extent

Global

Updating

No

Change detection

No (only one date)

Overall accuracy

Expected to be >68%

Website of reference

Website Language English

https://forobs.jrc.ec.europa.eu/products/glc2000/glc2000.php

Download site

https://forobs.jrc.ec.europa.eu/products/glc2000/products.php

Availability

Format(s)

Open Access

.tiff, ESRI GRID, .img and Binary

Technical documentation

Hua et al. (2018), McCallum et al. (2006), Neumann et al. (2007), Pérez-Hoyos et al. (2012), Tchuenté et al. (2011)

Other references of interest

Bartholomé et al. (2002), Bartholomé and Belward (2005), Eva et al. (2004), Fritz et al. (2003)

Project

GLC2000 was a project run by the Joint Research Centre (JRC) of the European Commission in collaboration with regional teams across the globe. The objective of the project was to create a homogeneous, coherent global LUC map that was suitable for environmental monitoring. The reference year 2000 was chosen because of its particular significance for that purpose.

One of the most successful aspects of the project was the coordination of different teams across the globe to produce a global LUC map. To this end, GLC2000 provides a global dataset, together with a set of more detailed regional datasets adapted to the specificities of each territory.

Production method

GLC2000 was produced by different work teams across the globe. To this end, the world was split into 18 different regions, with each team mapping either a specific region or an area of special interest within a region.

A LUC map for each region was obtained through unsupervised classification of imagery captured by the VEGETATION sensor. The classifications obtained were then labelled by each regional team according to their local expertise in the area. Input for the classification varied in line with the particular characteristics of each region.

Regional LUC maps were merged into the global product, which is a coherent and homogeneous generalized mosaic of the set of regional maps. However, these regional maps provide more detail than the global one.

Product description

GLC2000 consists of two main products: the harmonized global LUC dataset covering the whole earth and the set of detailed regional LUC datasets. The Global LUC map can be downloaded in four different formats (ESRI, Binary, Tiff, Img), whereas the regional maps are only available in two (ESRI, Binary). The product for download includes a file to symbolize the raster LUC map as well as auxiliary information to interpret the legend.

Downloads

GLC2000 (Global)

– Raster file with LUC map

– Colormap file to symbolize the raster in ArcGIS (.clr)

– Excel spreadsheet with the map legend

GLurope)

– Folder with raster file of the regional LUC map (glc_eu_v2)

– Colormap file to symbolize the raster in ArcGIS (.clr)

– DBF file with the map legend

Legend and codification

Code

Label

1

Tree Cover, broadleaved, evergreen

2

Tree Cover, broadleaved, deciduous, closed

3

Tree Cover, broadleaved, deciduous, open

4

Tree Cover, needle-leaved, evergreen

5

Tree Cover, needle-leaved, deciduous

6

Tree Cover, mixed leaf type

7

Tree Cover, regularly flooded, fresh

8

Tree Cover, regularly flooded, saline, (daily variation)

9

Mosaic: Tree cover/Other natural vegetation

10

Tree Cover, burnt

11

Shrub Cover, closed-open, evergreen (with or without sparse tree layer)

12

Shrub Cover, closed-open, deciduous (with or without sparse tree layer)

13

Herbaceous Cover, closed-open

14

Sparse Herbaceous or sparse shrub cover

15

Regularly flooded shrub and/or herbaceous cover

16

Cultivated and managed areas

17

Mosaic: Cropland/Tree Cover/Other Natural Vegetation

18

Mosaic: Cropland/Shrub and/or Herbaceous cover

19

Bare Areas

20

Water Bodies (natural and artificial)

21

Snow and Ice (natural and artificial)

22

Artificial surfaces and associated area

23

No data

Practical considerations

Information about map metadata is easily available on the project’s website together with technical documents describing the products. This information can help users gain a better understanding of the maps and all their specific characteristics, advantages and disadvantages. GLC2000 has also been widely analysed in the scientific literature. Users can find out more about the particular characteristics and the accuracy of the database by consulting some of the references of interest cited above.

4 Geo-Wiki Hybrid

Product

LULC general

Dates

2000 / 05

Formats

Raster

Pixel size

300 m

Thematic resolution

10 classes:

1 (a), 1 (ag), 3 (v), 1 (m), 0 (na)

Compatible legends

FAO LCCS

Extent

Global

Updating

Not planned

Change detection

No (only one date)

Overall accuracy

Expected to be > 82% (87.9% for Hybrid Map 1 and 82.8% for Hybrid Map 2)

Website of reference

Website Language English

https://www.geo-wiki.org/

Download site

https://application.geo-wiki.org/Application/index.php

Availability

Format(s)

Open Access after registration

.img

Technical documentation

See et al. (2015)

Other references of interest

Fritz et al. (2012)

Project

This project aimed to merge available global LUC maps to create a new, more accurate dataset, in a bid to enable more accurate global LUC mapping. Reference LUC data collected by the Geo-Wiki platform via crowdsourcing was employed in the fusion process, so pioneering a practice that has become more common in recent years. The dataset obtained in this way was one of the first, best-known examples of data fusion for global LUC mapping.

Production method

The hybrid map of the Geo-Wiki project was produced by merging three global LUC datasets: GLC2000, GlobCover and MODIS LC. Whereas GLC2000 shows the LUC state of the world for the reference year 2000, the other two sources provide LUC information for the reference year 2005. The spatial resolution of the hybrid map is the same as applied in the dataset with the highest resolution: GlobCover (300 m). The other two datasets, which had a spatial resolution of 1 km, were resampled to fit this resolution.

For each dataset, a probability layer was produced indicating the probability of that source representing the correct LUC class on the ground. These layers were obtained by regressing the datasets with validation points created through Geo-Wiki campaigns. A Geographically Weighted Regression (GWR) algorithm was employed to this end.

The probability layers were later merged in two different ways, delivering two LUC maps. For Hybrid Map 1, the LUC category from the dataset with the highest probability in the probability layers was selected. For Hybrid Map 2, when two LUC datasets agreed on a LUC category, this was selected. When the LUC datasets disagreed, the LUC category from the dataset with the highest probability in the probability layers was chosen.

Product description

Users can download the hybrid map in a compressed folder (.rar) which also contains the raster layers that store the LUC information. No other auxiliary information is provided.

Downloads

Geo-Wiki Hybrid (folder)

– A raster file with LUC information (.img)

Legend and codification

Code

Label

Code

Label

1

Tree cover

6

Flooded/wetland

2

Shrub cover

7

Urban

3

Herbaceous vegetation/Grassland

8

Snow and ice

4

Cultivated and managed

9

Barren

5

Mosaic of cultivated and managed/natural vegetation

10

Open water

Practical considerations

The Hybrid map is available online through the Geo-Wiki platform.Footnote 5 Although two hybrid maps were produced, only one was finally distributed. No information is provided as to which of these two maps is the one available online and for download.

5 LADA LUC Map—Land Degradation Assessment in Drylands

Product

LULC general

Dates

2007

Formats

Raster

Pixel size

5 arc minutes

Thematic resolution

40 classes

1 (a), 7 (ag), 23 (v), 0 (m), 0 (na)

Compatible legends

–

Extent

Global

Updating

No

Change detection

No (only one date)

Overall accuracy

Not specified

Website of reference

Website Language English

http://www.fao.org/land-water/land/land-governance/land-resources-planning-toolbox/category/details/en/c/1036360/

Download site

https://data.apps.fao.org/map/catalog/srv/eng/catalog.search?currTab=simple&id=37139#/metadata/fc32c5de-440c-46aa-9cad-81f4c8b84c6a

Availability

Format(s)

Open Access

ESRI GRID, .tiff

Technical documentation

Nachtergaele and Petri (2013)

Other references of interest

–

Project

Land Degradation Assessment in Dryland (LADA) is a project led by the Food and Agriculture Organization (FAO) of the United Nations that aims to assess and map land degradation at different scales and levels, so as to understand its impact on land use. As part of the datasets created in the project, a map of the world’s Land Use Systems (LUS) was developed. Many other datasets were also created within the framework of this project, which may be of interest to users.

Production method

The dataset was obtained after the interpretation of LUC units over a spatial dataset generated by the overlay of different spatial thematic layers: the GLC2000 LUC map, cropland LUC maps, livestock distribution data, ecosystem and ecological indicators and socioeconomic factors such as population density.

Product description

The LADA LUC map can be downloaded in two different formats (ESRI GRID or TIF). In each case, users download the raster files containing the LUC information, together with a layer style file to symbolize the dataset in a GIS.

Downloads

ESRI GRID folder

– Folder with raster files including LUC information (“lus”)

– Folder with product metadata (“info”)

– Layer style file for ArcGIS (.lyr)

TIF folder

– Raster file with LUC map (.tiff)

– Layer style file for ArcGIS (.lyr)

Legend and codification

Label

Code

Label

1

Forest—Virgin

2

Forest—Protected

3

Forest—With agricultural activities

4

Forest—With moderate or high livestock density

5

Forest—Agroforestry

6

Forest—Plantations

7

Grasslands—Unmanaged

8

Grasslands—Protected

9

Grasslands—Low livestock density

10

Grasslands—Moderate livestock density

11

Grasslands—High livestock density

12

Grasslands—Stable fed

13

Shrubs—Unmanaged

14

Shrubs—Protected

15

Shrubs—Low livestock density

16

Shrubs—Moderate livestock density

17

Shrubs—High livestock density

18

Shrubs—Stable fed

19

Agricultural land—Rainfed crops (subsistence/commercial)

20

Agricultural land—Crops and mod. Intensive livestock density

21

Agricultural land—Crops and intensive livestock density

22

Agricultural land—Crops with large scale irrigation and mod. Intensive or higher livestock density

23

Agricultural land—Large-scale irrigation (>25% pixel size)

24

Agricultural land—Protected

25

Urban land

26

Wetlands—Not used/not managed

27

Wetlands—Protected

28

Wetlands—Mangrove

29

Wetlands—With agricultural activities

30

Sparsely vegetated areas—Unmanaged

31

Sparsely vegetated areas—Protected

32

Sparsely vegetated areas—Low livestock density

33

Sparsely vegetated areas—With mod or higher livestock density

34

Barren areas—Unmanaged

35

Barren areas—Protected

36

Barren areas—Low livestock density

37

Barren areas—With mod. livestock density

38

Open water—Unmanaged

39

Open water—Protected

40

Open water—Inland fisheries

Practical considerations

The LADA LUC dataset is not a standard LUC map. It is a map of land use systems that was specifically created for the purposes of the LADA project, i.e. to study land degradation.

6 GLC-SHARE—Global Land Cover-SHARE

Product

LULC general

Dates

Only one date, different for each part of the Earth

Formats

Raster

Pixel size

1 km

Thematic resolution

11 classes:

1 (a), 1 (ag), 6 (v), 0 (m), 0 (na)

Compatible legends

FAO LCCS

Extent

Global

Updating

None planned

Change detection

No (only one date)

Overall accuracy

Expected to be >80%

Website of reference

Website Language English

http://www.fao.org/land-water/land/land-governance/land-resources-planning-toolbox/category/details/en/c/1036355/

Download site

https://data.apps.fao.org/map/catalog/srv/eng/catalog.search?uuid=ba4526fd-cdbf-4028-a1bd-5a559c4bff38&currTab=distribution#/metadata/ba4526fd-cdbf-4028-a1bd-5a559c4bff38

Availability

Format(s)

Open Access

.tiff, .kml, WMS

Technical documentation

Latham et al. (2014)

Other references of interest

–

Project

GLC-SHARE was a project led by the Land and Water Division of the Food and Agriculture Organization (FAO), in collaboration with other institutions across the world. It aimed to create a global LUC map by mixing different sources of LUC information available at detailed scales. The objective was to improve the accuracy and quality of LUC information, so as to have a reliable source of global LUC information for policymaking.

Unlike other global LUC mapping projects, GLC-SHARE provides detailed LUC information in a single global product. Usually, this is only available in national, regional and local datasets.

Although the GLC-SHARE was produced in 2014, it was conceived as a living database that could integrate new LUC datasets as they were released or updated. Its production method has been made public, so enabling product replication.

As GLC-SHARE was produced by merging data from multiple databases, it has no specific date of reference. There are different dates for each part of the world, according to the main product that was used to map them.

Production method

GLC-SHARE was produced by merging and integrating high-quality LUC data for different areas of the world. LUC data at all scales (global, national, sub-national, regional) was used to produce the map.

In order to merge the various LUC datasets into a single product, their legends had to be harmonized. When different products were available for the same area, the one with the most detailed, most accurate data was chosen. If no products were available at detailed or national scales, global LUC datasets (Globcover 2009, MODIS VCF 2010 and Cropland database 2012) were used instead. The main areas not covered by high-resolution datasets included Latin America, West Africa, Indonesia and important parts of Asia, such as Thailand and the Arabian Peninsula.

An initial map for each of the 11 land cover classes that make up the classification legend of the GLC-SHARE was obtained. Each map shows the proportion that each land cover occupies in each pixel of the GLC-SHARE grid. Finally, from the 11 thematic rasters created, a general raster was obtained indicating the dominant land cover type in each pixel.

Product description

GLC-SHARE products can be downloaded in raster format or as a kml file to upload in Google Earth or any other GIS software. GLC-SHARE maps are also available through a WMS web service.

Users can download the global GLC-SHARE LUC map, which indicates the dominant land cover type in each pixel, or individual LUC rasters showing the proportions of each LUC type in each pixel. In these rasters, the pixel value refers to the proportion (0–100) at which each category is represented in the pixel. A pixel covered exclusively by artificial surfaces would have a value of 100 in the “GLC-Share – Artificial surfaces” raster.

Users can also download auxiliary information about the dataset from the website. This includes a technical report about the product (GLC-Share report) as well as a raster and an excel spreadsheet explaining which dataset was used to map each area of the world (GLC-Share—Sources).

Downloads

GLC-Share—Dominant land cover type

– Raster file with LUC map displaying the dominant land cover type

– Layer style file for ArcGIS (.lyr)

– Text document showing the classification legend for the dataset

GLC-Share—Sources

– Raster file with information about which LUC dataset was used to map each area of the world

– Layer style file for ArcGIS (.lyr)

– Excel spreadsheet with information about which LUC dataset was used to map each area of the world

– Text document explaining the downloaded product

GLC-Share—Artificial surfaces

– Raster file with information about the proportion of artificial surfaces in each pixel

Legend and codification

Code

Label

Code

Label

1

Artificial Surfaces

7

Mangroves

2

Cropland

8

Sparse vegetation

3

Grassland

9

Bare soil

4

Tree covered areas

10

Snow and glaciers

5

Shrubs covered areas

11

Water bodies

6

Herbaceous vegetation, aquatic or regularly flooded

  

Practical considerations

GLC-SHARE is a single product with no information about changes in LUC over time. It was created in 2014, which may therefore be considered the reference year for the dataset. However, this date may vary a great deal between the different parts of the world. GLC-SHARE is therefore not recommended for studies or analyses of LUC change.

Although the dataset was conceived as a live map, it has not been further updated with the inclusion of new LULC datasets since 2014.

7 OSM Landuse/Landcover

Product

LULC general

Dates

Only one date, which cannot be specified

Formats

Raster

Pixel size

10 m

Thematic resolution

14 classes:

4 (a), 3 (ag), 2 (v), 3 (m), 1 (na)

Compatible legends

CLC

Extent

Global (with gaps) / Europe (full coverage)

Updating

Completion of the map is ongoing, although new editions of the map for different years of reference are not expected

Change detection

No (only one date)

Overall accuracy

Not specified

Website of reference

Website Language English

https://data.osmlanduse.org

Download site

https://data.osmlanduse.org

Availability

Format(s)

Under request (email to producers)

.tiff

Technical documentation

Schultz et al. (2017)

Other references of interest

Fonte and Martinho (2017), Fonte et al. (2017a, b), Viana et al. (2019)

Project

OSM Landuse/Landcover (LULC) is a LUC dataset created as part of the H2020 project “LandSense”, which aims to engage citizens in the production of LUC information. The OSM LULC has been developed above all by the GIScience research group from Heidelberg University.

OSM LULC is an attempt to exploit the LUC information contained in the OpenStreetMaps (OSM) database. It is a test project and therefore cannot be regarded as a final product with full global coverage. Nevertheless, the project has developed a workflow to obtain LUC information from the OSM database as well as a methodology for obtaining an LUC map with full coverage over a specific test area (Heidelberg), filling the gaps in the OSM via classification of satellite imagery.

Production method

OSM LULC was produced using a very simple method. Authors downloaded the OSM database and translated the tags that define the features stored in the database into LUC terms (the legend for the Corine Land Cover (CLC) survey was used as a reference). An equivalence table between the OSM tags and the CLC level 2 legend was created.

The OSM LUC information, in vector, was generalized in a 30 m pixel side grid. In the event of feature overlap when aggregating information, preference was given to the smaller features.

Gap areas not covered by the OSM database were filled with the LUC information obtained by a supervised classification of Landsat imagery with the random forest classifier. This process was only carried out for a European test area, leaving important information gaps in the rest of the global map.

Due to the particular characteristics of the OSM database, LUC information is not provided for a single date. Each feature of the database has a different date. This makes it difficult to determine the date of reference for each pixel in the dataset.

Product description

The product was initially distributed in tiles. However, users can also request a specific file for their area of interest by email. These files contain the LUC map and an Excel spreadsheet with the pixel count for each category. They do not include the qualitative meaning of the category codes.

Downloads

OSM Landuse

– Raster file with LUC map (.tiff)

– Excel file with class codes and pixel count per class

Legend and codification

Code

Label

Code

Label

5

Water bodies

23

Pastures

11

Urban fabric

31

Forests

12

Industrial, commercial and transport units

32

Shrub and/or herbaceous vegetation associations

13

Mine, dump and construction sites

33

Open spaces with little or no vegetation

14

Artificial, non-agricultural vegetated areas

41

Inland wetlands

21

Arable land

42

Coastal wetlands

22

Permanent crops

NA

No data

Practical considerations

The website for this database includes a form for those who want to download the map. However, interested users are recommended to contact the map producers directly, as the first approach does not always work. Contact details for the map producers are available at the project’s website.Footnote 6

Users should be aware of the limitations of this dataset. As there is no single reference year for all the mapped areas, it may be difficult to use the map as a reference when analysing changes over time.