Abstract
Supra-national thematic Land Use Cover (LUC) datasets are not very common. While there are several general datasets mapping all the land uses or covers in different supra-national areas across the world, LUC datasets with a similar extent that focus on the mapping of specific land covers in greater thematic detail are scarce. In this chapter, we review six different supra-national thematic LUC datasets. Three others were also found in the literature, but are not fully available for download, namely the TREES Vegetation Map of Tropical South America, the Central Africa—Vegetation map and FACET. The Circumpolar Arctic Region Vegetation dataset was also excluded from this review because of its specificity and coarse scale (1:7,500,000). Europe is the continent with the most relevant, most updated and most detailed LUC thematic datasets at supra-national scales. This is due to the work being done by the European Commission through its Joint Research Centre (JRC) and the Copernicus Land Monitoring Programme. The High-Resolution Layers (HRL) provide very detailed information, both thematically and spatially (from 10 m), for five different themes: imperviousness, tree cover, grasslands, water and wet covers, and small woody features. The European Settlement Map also provides information on built-up areas at very detailed scales (from 2.5 m). HRL and ESM are recently launched datasets which, therefore, do not provide a long series of historical data. In addition, ESM is an experimental dataset produced within the framework of a research project funded by the European Commission and no updates are expected. The datasets reviewed in this chapter for other parts of the world focus on vegetation covers of tropical forests and other relevant areas in terms of biodiversity and environmental studies. These datasets were produced within projects funded by the European Commission and the United States Agency for International Development. Unlike the previous datasets for Europe, they are already outdated and are usually produced at coarser spatial resolutions: Insular Southeast Asia—Forest Cover Map (1 km, 1998/00); Continental Southeast Asia—Forest Cover Map (1 km, 1998/02). For its part, the Congo Basin Monitoring dataset, although outdated, provides information at a higher resolution (57 m) for two different dates: 1990, 2000. The Joint Research Centre of the European Commission also produced an African cropland mask as a source of information for policy-makers. Of all the datasets reviewed in this chapter, it is the only one to focus on agricultural covers. It was obtained from data fusion at 250 m. Consequently, it does not show the cropland areas of Africa for a specific date across the whole continent.
You have full access to this open access chapter, Download chapter PDF
Similar content being viewed by others
Keywords
- Supra-National
- Forest
- Vegetation
- Built-up
- Insular Southeast Asia—Forest Cover Map
- Continental Southeast Asia—Forest Cover Map
- Congo Basin Monitoring Maps
- MARS Crop Mask Over Africa
- High Resolutions Layers
- European Settlement Map
1 Insular Southeast Asia—Forest Cover Map
| Product |
LULC thematic | |
Dates | |
1998 / 00 | |
Formats | |
Raster | |
Pixel size | |
1 km | |
Theme | |
4 forest classes out of 10 | |
Extent | |
Insular Southeast Asia | |
Updating | |
Not expected | |
Change detection | |
No (only one date) | |
Overall accuracy | |
Not specified | |
Website of reference | Website Language English |
https://forobs.jrc.ec.europa.eu/products/veget_map_insulare-sea/insularSEasia.php | |
Download site | |
https://forobs.jrc.ec.europa.eu/products/veget_map_insulare-sea/download_forest_cover_map_isea.php | |
Availability | Format(s) |
Open Access | .tiff |
Technical documentation | |
Other references of interest | |
– |
Project
The Joint Research Centre (JRC) of the European Commission produced a map for Insular Southeast Asia which sought to provide a more accurate characterization of the forest covers in this region. It aimed to overcome the limitations associated with the mapping of vegetation covers in tropical regions, due to the persistence of cloud covers.
The dataset covers Malaysia, Singapore, Indonesia, Brunei, East Timor, the Philippines and Papua New Guinea. It is especially useful for research into deforestation and biodiversity due to the significance of the insular Southeast Asia forest ecosystem for the world as a whole.
The dataset was produced within the context of the TRopical Ecosystem Environment observations by Satellite (TREES) project. The project aimed to produce regularly updated information to monitor forest covers in tropical regions at regional scales.
Production method
The forest map for Insular Southeast Asia was produced through the unsupervised classification (clustering and maximum likelihood classification) of a mosaic of imagery collected by the VEGETATION sensor of the SPOT satellite for the period 1998–2000.
The unsupervised classification identified 60 spectral clusters. They were manually interpreted and labelled on the basis of information provided by other satellite imagery, maps of reference and field data. In addition, the initial set of clusters was regrouped on the basis of information provided by two auxiliary datasets: GTOPO30 DEM and WCMC forest map. After this initial processing, the remaining clusters were finally grouped into 8 LUC categories and a No-Data category.
Product description
The forest map for Insular Southeast Asia can be downloaded as a single compressed file (.zip) containing the raster with the LUC information. No auxiliary information is provided.
Downloads
Insular Southeast Asia—Forest Cover Map | |
---|---|
– A raster with the LUC information (.tiff) |
Legend and codification
Code | Label | Code | Label |
---|---|---|---|
0 | No data | 5 | Cropland |
1 | Evergreen montane forest | 6 | Burnt/dry/sparse vegetation |
2 | Evergreen lowland forest | 7 | Non-forest vegetation |
3 | Mangrove forest | 8 | Water |
4 | Swamp forest |
Practical considerations
A full characterization of the dataset is provided in the technical report published by the European Commission and in the technical documentation cited above.
The map comes with several limitations: a few seasonal monsoon forests in Sulawesi, New Guinea and Philippines were not mapped as an individual category, while degraded forest cover and mature stages of forest regrowth were sometimes mapped as forest.
2 Continental Southeast Asia—Forest Cover Map
| Product |
LULC thematic | |
Dates | |
1998 / 00 | |
Formats | |
Raster | |
Pixel size | |
1 km | |
Theme | |
8 forest / wood classes out of 14 | |
Extent | |
Bangladesh, Myanmar, Thailand, Laos, Cambodia, the Himalayas mountain range, north-eastern India and southern China | |
Updating | |
Not expected | |
Change detection | |
No (only one date) | |
Overall accuracy | |
Not specified | |
Website of reference | Website Language English |
https://forobs.jrc.ec.europa.eu/products/veget_map_continental-sea/continentalSEasia.php | |
Download site | |
Availability | Format(s) |
Open Access | .tiff |
Technical documentation | |
Stibig et al. (2004) | |
Other references of interest | |
– |
Project
The forest map for Continental Southeast Asia was developed by the Joint Research Centre (JRC) of the European Commission within the context of the TRopical Ecosystem Environment observations by Satellite (TREES) and GLC2000 projects. Other LUC maps on forest covers for Insular Southeast Asia and Central Africa were also developed as part of the TREES project, following similar mapping workflows. They are all reviewed in this chapter.
The project aimed to provide regularly updated LUC information on tropical forests to help monitor activities in these regions. The obtained dataset covers Bangladesh, Myanmar, Thailand, Laos, Cambodia, the Himalaya mountain range and tropical areas of north-eastern India and southern China.
Production method
The dataset was produced through unsupervised classification of a cloud free mosaic of VEGETATION imagery for the period 1998–2000. The classification identified 70 spectral clusters, which were manually labelled and interpreted on the basis of information provided by Landsat imagery, field-collected data and a DEM. For the labelling and interpretation of spectral classes, the mapped area was split into 11 geographic strata, covering the different types of climate, landscape and land cover in the region. Finally, the labelled clusters were grouped together in 12 land cover categories.
Product description
The forest map can be downloaded in a single compressed file (.zip). No additional information is provided.
Downloads
Continental Southeast Asia—Forest Cover Map | |
---|---|
– A raster file containing the LUC information (.tiff) |
Legend and codification
Code | Label | Code | Label |
---|---|---|---|
0 | No data | 7 | Evergreen wood and shrubland and regrowth mosaics |
1 | Evergreen Mountain forests | 8 | Deciduous wood and shrubland and regrowth mosaics |
2 | Evergreen Lowland forests | 9 | Mosaics of Cropping and Regrowth |
3 | Fragmented and degraded evergreen forest cover | 10 | Other lands |
4 | Deciduous forests | 11 | Other lands |
5 | Mangrove forests | 12 | Rocks |
6 | Swamp forests and inundated shrubland | 13 | Water bodies/Sea |
Practical considerations
Although a technical report describing the characteristics of the dataset was published, it is not currently available. The available information is therefore limited. In addition, the spatial resolution of the map (1 km) limits its capacity to map gradual local transitions in tree canopies, such as the degradation or fragmentation of forest canopies.
3 Congo Basin Monitoring Maps
| Product |
LULC thematic | |
Dates | |
1990 / 00 | |
Formats | |
Raster | |
Pixel size | |
57 m | |
Theme | |
Forest extent (2000) Forest probability (2000) Forest cover clearing (1990–2000) | |
Extent | |
Congo River Basin | |
Updating | |
Not expected | |
Change detection | |
Information on forest cover clearing for the period 1990–2000 | |
Overall accuracy | |
Not specified | |
Website of reference | Website Language English |
Download site | |
Availability | Format(s) |
Open Access | .tiff, .img |
Technical documentation | |
Hansen et al. (2008) | |
Other references of interest | |
Lindquist et al. (2008) |
Project
Maps of the Congo Basin Monitoring project were developed within the context of the Central African Regional Program for the Environment (CARPE), funded by the United States Agency for International Development (USAID). The program aims to promote sustainable resource management in the Congo Basin region, for which the provision of accurate monitoring data is vital.
The resulting LUC maps provide a useful resource for monitoring humid tropical deforestation at high spatial resolutions. Previous LUC datasets mapping humid tropical regions had insufficient spatial resolution. Central Africa forest covers are not subject to large-scale clearings and instead suffer smaller clearing processes taking place at a local level. This means that monitoring projects at coarse resolution miss many of the key landscape dynamics. Previous attempts to map the humid forests of Central Africa also faced important methodological limitations because of the lack of cloud-free imagery for the area. The Congo Basin Monitoring project aimed to overcome these limitations.
Two maps were produced for the Congo Basin as part of this project: a forest mask and a forest probability map that also offers information on forest clearing for the period 1990–2000. Forest clearing is defined as complete removal of the forest over story.
Production method
A forest mask was first created from a forest percent tree cover layer at 250 m generated after the classification of MODIS imagery (2000–2004) using the Vegetation Continuous Field (VCF) method. 34 metrics from MODIS imagery were extracted to carry out the classification. A threshold of 60% was applied to this layer to generate the forest mask: all pixels with a forest percentage of over 60% were considered forest. All the remaining pixels were considered non-forest. Two other categories were also classified from MODIS imagery based on a classification tree algorithm: water and rural complex. Water pixels were treated as non-land in the forest mask, and rural complex pixels were considered non-forest.
A forest probability layer was obtained from the classification of Landsat imagery at the scene level for two different epochs: pre-1996 (1986–1996) and post-1996 (>1996–2003). The classification was performed on the basis of tree models using the previously obtained forest mask as the dependent variable and the Landsat imagery as the independent variable. Forest cover changes between the two periods were mapped through a multi-date direct classification of change methodology, using training data at the same locations for the two available epochs.
Product description
The forest map can be downloaded as a single compressed file (.zip) in. tiff format. The forest probability layer is available in two different formats (.tiff and .img). In both cases, the download includes the raster file with the LUC information and a text file with a technical description of each dataset.
Downloads
Forest probability and forest cover clearing | |
---|---|
– Raster file with information on forest probability and forest cover clearing (.tiff) – A text file with a technical description of the dataset (.txt) |
MODIS-based evergreen tropical forest map (forest mask) | |
---|---|
− Raster file with information on the forest extent (.tiff) − A text file with a technical description of the dataset (.txt) |
Legend and codification
Forest probability and forest cover clearing | |
---|---|
Code | Label |
0–100 | Forest probability (0–100%) |
253 | Forest clearing between 1990s and 2000s |
250 | Water |
254, 255 | No data |
MODIS-based evergreen tropical forest map (forest mask) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non forest | 1 | Forest |
4 MARS Crop Mask Over Africa
| Product |
LULC thematic | |
Dates | |
One-date (varies from one product to the next) | |
Formats | |
Raster | |
Pixel size | |
250 m | |
Theme | |
Cropland extent | |
Extent | |
Africa | |
Updating | |
Not expected | |
Change detection | |
No (only one date) | |
Overall accuracy | |
Expected to be > 70% for most of the mapped countries | |
Website of reference | Website Language English |
Download site | |
Not available | |
Availability | Format(s) |
On request to authors | .tiff |
Technical documentation | |
Vancutsem et al (2013) | |
Other references of interest | |
Project
The Monitoring Agricultural Resources (MARS) unit of the Joint Research Centre (JRC) produced a cropland mask for Africa to assist the unit and Commission’s activities with crop and food security monitoring. The mask aimed to provide the most accurate information possible on cropland covers for Africa by merging the best available LUC cropland data sources.
The methodology applied in the production of this dataset has also been used in the development of other cropland masks (ASAP Land Cover Masks) by the same team.
Production method
The MARS crop mask was obtained by merging the best available LUC data sources on cropland covers. To this end, all the input data sources were resampled or rasterized to a common spatial resolution (250 m) and projected with the same parameters. Cropland categories were extracted from each input dataset. LUC categories were considered as cropland when at least 50% of their surface was covered by cropland. LUC categories with a cropland proportion of between 20 and 50% were manually checked by experts, who decided whether to include them as cropland categories at a global level or for just one specific region.
The accuracy of each dataset was assessed against Google Earth imagery. When several datasets were available for the same area, the most accurate one was selected. If several datasets had similar levels of accuracy, the most detailed or recent was selected.
The input datasets were Globcover, SADC, Cropland Use Intensity datasets from USGS, Woody Biomass map of Ethiopia, AFRICOVER, JRC-MARS crop masks, LULC 2000 USGS datasets and national land cover maps of the Democratic Republic of Congo, Mozambique and Senegal.
Product description
The crop mask is available in Google Drive on request to the producers of the map. The download includes a document with a technical description of the product as well as the raster file with the LUC information. Another raster file is provided with information about the data source that was finally selected to create the crop mask in each case.
Downloads
MARS crop mask over Africa | |
---|---|
− Raster file with crop extent (.tiff) − Raster file with information on the data source used to map each area (.tiff) − Document with a technical description of the dataset (.doc) |
Legend and codification
Code | Label | Code | Label |
---|---|---|---|
0 | Cropland | 1 | No cropland |
Practical considerations
Users interested in accessing the dataset should apply to the map’s authors (Christelle.vancutsem@ec.europa.eu). This map was obtained by merging data from selected data sources. The dataset cannot provide LUC information for any specific reference year as each source had its own.
5 HRL—High Resolution Layers
| Product |
LULC thematic | |
Dates | |
2006, 2009, 2012, 2015, 2018 (Imperviousness) 2012, 2015, 2018 (Forests) 2015, 2018 (Grassland, Wetness and Water) 2015 (Small Woody Features) | |
Formats | |
Raster | |
Pixel size | |
5 m (Small Woody Features) 10 m (Products since 2018) 20 m (Products up to 2015) 100 m (Mosaics) | |
Themes | |
Extent and percentage of impervious areas Percentage of tree cover areas, leaf type and forest type Extent of grassland areas Wetness and water covers (5 water/wet classes out of 8) Extent of Small Woody Features | |
Extent | |
Europe (39 countries) | |
Updating | |
Planned every 3 years | |
Change detection | |
Through change layers | |
Overall accuracy | |
Imperviousness HRL, Forests: expected to be > 90% Grassland HRL, Wetness and Water HRL: expected to be > 80–80% Wetness and Water: HRL expected to be > 80% | |
Website of reference | Website Language English, German and French |
https://land.copernicus.eu/pan-european/high-resolution-layers | |
Download site | |
https://land.copernicus.eu/pan-european/high-resolution-layers | |
Availability | Format(s) |
Open Access after registration | .tiff |
Technical documentation | |
Copernicus Land Monitoring Service (2020a, b, c, d), D’amico et al. (2019), Faucqueur et al. (2018), Langangke (2015, 2016), Langangke et al. (2017, 2018a, b, 2019), Pennec et al. (2019a, b), Smith et al. (2019), Weirather et al. (2019a, b) | |
Other references of interest | |
Büttner et al. (2016), Manakos et al. (2018), Sannier et al. (2017) |
Project
The High-Resolution Layers are produced within the framework of the Copernicus Land Monitoring Programme. They were created as a means of overcoming some of the limitations associated with CORINE Land Cover (CLC), such as lack of detail, the presence of mixed classes and the difficulty of adapting the CLC legend to other common classification schemes, such as the FAO LCSS. Each High-Resolution Layer is associated with one of the CLC Level 1 classes: artificial surfaces (Imperviousness HRL), agricultural areas (Grassland HRL), forest and semi-natural areas (Forests HRL), wetlands and water bodies (Water & Wetness HRL).
The different High-Resolution Layers are separately produced using specific methods. Since 2018, they have been produced at enhanced spatial resolution (10 m) based on Sentinel imagery. This marks a change in the methodology applied in the production of HRL compared to the layers created for previous years of reference.
Some of the HRL layers have been produced for more years than the others, such as the Imperviousness HRL, available since 2006, and the Forests HRL, available since 2012. However, when available, the reference years are almost all the same for all the layers. The only exception is the recently created Small Woody Features HRL. In some cases, when more than one date is available, change layers have been developed.
Production method
Each HRL has its own specific production method, as each theme is characterized in a different way. Nevertheless, all the HRLs are obtained by automatic classification and interactive rule-based classification of high-resolution imagery, mostly from the Sentinel constellation. The Imperviousness HRL and Water and Wetness HRL are obtained from both optical and raster data, while the Forests, Grasslands and Small Woody Features HRLs are obtained exclusively from optical data.
Change layers are obtained by comparing the status layers for two different years of reference. For the changes between 2018 and the previous year of reference, some uncertainties may arise because of the change in the spatial resolution: 10 m vs 20 m. The production teams have implemented various different measures to prevent such uncertainties, including the development of supporting layers that inform about the changes that take place due to technical reasons and the level of confidence of the obtained change layer.
Initial production of the HRL is centralized. Then, each country reviews and verifies the results, so enhancing this initial product. For more detailed information about the production process of all the HRLs, readers are referred to the technical documentation cited above.
Product description
Imperviousness HRL
The Imperviousness HRL can be separately downloaded for each year of reference or for each period of changes. In the latter case, users can choose between an uncategorized file showing the change in the degree of imperviousness and a file that categorizes this change in a series of classes. For the reference year 2018, users can also download the Impervious built-up layer as a separate file. This is a binary map differentiating built-up areas from non-built-up areas.
The layers are disseminated at country level in 100 × 100 km tiles. Users download a single file with all the tiles covering the selected country. A mosaic of all the mapped countries is also available as a single file at two spatial resolutions: 10-20 m (the original resolution) and 100 m.
Different supporting layers are also available for download as part of the Imperviousness HRL. Unlike the previous layers, they are available in the “Expert Products” section as single files covering all of Europe. These supporting layers include (i) a layer indicating the change in the degree of imperviousness between 2015 and 2018 due to technical reasons (IMCS); (ii) a layer showing the confidence level of the Imperviousness density 2018 layer at 10 m (IMDCL); and (iii) an adaptation of the Imperviousness density 2015 layer to a spatial resolution of 10 m, to enable researchers to study changes in the impervious area between 2015 and 2018 (IMDR).
All downloads have the same contents: a raster file containing the LUC information, a file to symbolize it in any GIS software and a metadata file. Files for the pre-2018 editions of Imperviousness HRL also include an Excel file with technical information about the product.
Downloads
Imperviousness built-up 2018 (Status Map) | |
---|---|
Imperviousness density 2018 (Status Map) | |
Imperviousness Change 2015–2018 (Change Map) | |
Imperviousness Classified Change 2015–2018 (Change Map) | |
− Raster file with LUC information (.tiff) (DATA folder) − Text file to symbolize the raster in QGIS (.txt) (Symbology folder) − Metadata file (Metadata folder) |
Legend and codification
Imperviousness built-up (Status Map) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non built-up | 255 | Outside area |
1 | Built-up |
Imperviousness density (Status Map) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non-impervious areas | 254 | Unclassifiable |
1–100 | Degree of imperviousness (%) | 255 | Outside area |
Imperviousness Change (Change Map) | |
---|---|
Code | Label |
0–99 | Percentage of decreased imperviousness density |
100 | Unchanged areas with some degree of imperviousness |
101–200 | Percentage of increased imperviousness density |
201 | Unchanged areas with no degree of imperviousness |
254 | Unclassifiable (no satellite image available, or clouds, shadows, or snow) |
Imperviousness Classified Change (Change Map) | |
---|---|
Code | Label |
0 | Unchanged areas with Imperviousness Density = 0% |
1 | New cover (increasing imperviousness density, which was 0% at first reference date) |
2 | Loss of cover (decreasing imperviousness density, which was 0% at second reference date) |
10 | Unchanged areas with Imperviousness Density > 0% at both reference dates |
11 | Increased Imperviousness Density (>0% at both reference dates) |
12 | Decreased Imperviousness Density (>0% at both reference dates) |
254 | Unclassifiable |
255 | Outside area |
Forests HRL
For each available year of reference, three different types of layer can be downloaded as part of the Forests HRL: (i) a layer showing the forest density or the degree of tree cover (Tree Cover Density); (ii) a layer informing about the dominant leaf type, distinguishing mainly between broadleaf and coniferous trees (Dominant Leaf Type); and (iii) a layer informing about the dominant leaf type in treed areas covering more than 0.5 ha and with a tree cover density of over 10%, i.e. those areas considered as forest according to the FAO definition (Forest Type).
Change layers for Tree Cover and Dominant Leaf Type are also provided for each mapped period. A layer of tree cover density changes was initially created for the period 2012–2015. However, it has not been updated for the new mapping periods and is no longer distributed.
In all cases, the layers are distributed at a country level in 100 × 100 km tiles. A single file mosaic of each layer for all the mapped countries is also available at two spatial resolutions: 10–20 m (the original resolution) and 100 m.
Nine additional layers were also produced as supplementary information to the Forests HRL for the year 2018. These can be downloaded from the “Experts products” section. They provide information about the broadleaved and coniferous cover densities at 100 m (BCD, CCD) as well as other relevant technical information about the production of the Forests HRL: level of confidence, data sources, etc. The technical documentation of HRL Forests includes a detailed description of each of these supporting layers.
In all cases, the downloaded files include the raster with LUC information, a file to symbolize it in any GIS software and the product’s metadata. Files for the pre-2018 editions of Forests HRL also include an Excel file with technical information about the product.
Downloads
Tree Cover Density 2018 | |
---|---|
Tree Cover Change Mask 2015–2018 | |
Dominant Leaf Type 2015 | |
Dominant Leaf Type Change 2015–2018 | |
Forest Type 2018 | |
− Raster file with LUC information (.tiff) (DATA folder) − Text file to symbolize the raster in QGIS (.txt) (Symbology folder) − Metadata file (Metadata folder) |
Legend and codification
Tree Cover Density | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non–tree-covered areas | 254 | Unclassifiable |
1–100 | Percentage of tree cover density | 255 | Outside area |
Tree Cover Change Mask | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Unchanged areas with no tree cover | 10 | Unchanged areas with tree cover |
1 | New tree cover | 254 | Unclassifiable in any of parent status layers |
2 | Loss of tree cover | 255 | Outside area |
Dominant Leaf Type | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non–tree-covered areas | 254 | Unclassifiable |
1 | Broadleaved trees | 255 | Outside areas |
2 | Coniferous trees |
Dominant Leaf Type | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Unchanged areas with no tree cover | 10 | Unchanged areas with tree cover |
1 | New broadleaved cover | 12 | Potential change among dominant leaf types |
2 | New coniferous cover | 254 | Unclassifiable in any of parent status layers |
3 | Loss of broadleaved cover | 255 | Outside area |
4 | Loss of coniferous cover |
Forest Type | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non–tree-covered areas | 3 | Mixed forest (only for aggregated 100 m layer) |
1 | Broadleaved trees | 254 | Unclassifiable |
2 | Coniferous trees | 255 | Outside areas |
Grassland HRL
A status layer for each reference year and a layer of changes for each mapped period can be downloaded separately as part of the Grassland HRL. Moreover, three additional supporting layers are distributed as “Expert products”: (i) a layer showing the probability of each pixel being grassland (Grassland Vegetation Probability Index, GRAVPI); (ii) a layer informing about the number of years since the last ploughing (Ploughing Indicator, PLOGH); and (iii) a confidence layer for the Grassland 2018 status map (GRACL).
The status layer and the change layers are distributed at country level in 100 × 100 km tiles. A single file European mosaic is also available at two spatial resolutions: 10-20 m (the original resolution) and 100 m. The three supporting layers can be downloaded as single files covering the whole of the mapped area.
All downloads include the raster with LUC information, a file to symbolize it in GIS and a metadata file. Downloads for the pre-2018 editions of the layers also include an Excel file with technical information about the product.
Downloads
Grassland 2018 (Status Map) | |
---|---|
Grassland Change 2015–2018 (Change maps) | |
− Raster file with LUC information (.tiff) (DATA folder) − Text file to symbolize the raster in QGIS (.txt) (Symbology folder) − Metadata file (Metadata folder) |
Legend and codification
Grassland (Status Map) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non–grass areas | 254 | Unclassifiable |
1 | Grassy and non–woody vegetation | 255 | Outside area |
Grassland Change (Change maps) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | All non-grassland areas | 11 | Unverified grassland gain |
1 | Grassland gain | 22 | Unverified grassland loss |
2 | Grassland loss | 254 | Unclassifiable in any of parent status layers |
10 | Unchanged grassland in both years | 255 | Outside area |
Water and Wetness HRL
The Water and Wetness HRL is made up of a main product mapping the different types of water and wetness covers in Europe. Users can also download an additional layer (Expert products) showing the probability of each pixel being water or wetness. Two extra technical layers are also available as expert products: one informs about the confidence of the 2018 status map (WACL) while the other studies the differences in the mapping of water and wetness covers between 2015 and 2018 (WAWCSL).
Different files can be downloaded for each available layer and year. The main layer is distributed at country level in 100 × 100 km tiles. However, a single file mosaic is also available at the original resolution of the product (10–20 m) and at 100 m. The supporting layers are only available at the original resolution as single files covering the whole of Europe.
All downloads include the raster with LUC information, a file to symbolize it in any GIS software and a metadata file. The available layer for 2015 also includes an Excel file with technical information about the product.
Downloads
Water and Wetness 2018–WAW (Status Map) | |
---|---|
− Raster file with LUC information (.tiff) (DATA folder) − Text file to symbolize the raster in QGIS (.txt) (Symbology folder) − Metadata file (Metadata folder) |
Legend and codification
Water and Wetness (Status Map) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Dry | 4 | Temporary wet |
1 | Permanent water | 253 | Sea water |
2 | Temporary water | 254 | Unclassifiable |
3 | Permanent wet | 255 | Outside areas |
Small Woody Features HRL
The Small Woody Features HRL is available in either vector or raster files. Vector files can be downloaded in two different formats: ESRI Geodatabase and GeoPackage. Raster files can be downloaded at two different spatial resolutions: 5 and 100 m.
The vector and raster files at 5 m are distributed in tiles obtained after splitting each European country into a series of large regions. To find out which tile corresponds to their particular area of interest, users should consult the viewer on the dataset’s website.Footnote 1 The rasters at 100 m are distributed as single files covering the whole of Europe, without splits into regions.
The raster at 5 m only differentiates between Small Woody Features (SWF) and Additional Woody Features (AWF). The vector file also differentiates between SWF and AWF, although it splits the first category into linear and patchy structures. Three different layers are available at 100 m: (i) the density of small woody features (SWF); (ii) the density of Additional Woody Features (AWF); and (iii) the density of both small and additional woody features (SWFAWF).
Downloads
Small Woody Features 2018 (Geodatabase) | |
---|---|
− Vector file with LUC information (DATA folder) − Raster file with information about the accuracy of the product (.tiff) − PDF with a guide about how to use the ESRI Geodatabase in QGIS (Documents folder) − PDF with information about the product (Documents folder) − Metadata about the product (Metadata folder) |
Small Woody Features 2018 (Raster 5 m) | |
---|---|
− Raster file with LUC information (.tiff) (Data folder) − File to symbolize the raster in GIS (.clr) (Data folder) − PDF with information about the product (Documents folder) − Metadata about the product (Metadata folder) |
SWF density (Raster 100 m) | |
---|---|
AWF density (Raster 100 m) | |
SWF + AWF density (Raster 100 m) | |
− Raster file with LUC information (.tiff) (Documents folder) − PDF with information about the product (Documents folder) − Metadata about the product (Metadata folder) |
Database
Small Woody Features 2018 (Geodatabase) |
---|
|
− Gid: Unique identifier for each polygon − Code: Thematic code for each polygon − Area: Area of the polygon, in square meters − Class_name: Category assigned to each polygon |
Legend and codification
Small Woody Features (Geodatabase and GeoPackage) | |||
---|---|---|---|
Code | Label | Code | Label |
1 | Linear structures of trees, hedges, bushes and scrub | 3 | Additional woody features |
2 | Patchy structures of trees, hedges, bushes and scrub |
Small Woody Features (Raster 5 m) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non-SWF area | 254 | Unclassifiable |
1 | Patchy structures of trees, hSWF area (Linear or patchy structures of trees, hedges, bushes and scrub) | 255 | Outside areas |
3 | Additional woody features |
SWF density (Raster 100 m) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non-SWF area | 254 | Unclassifiable |
0–100 | Small Woody Features density | 255 | Outside areas |
AWF density (Raster 100 m) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non-SWF area | 254 | Unclassifiable |
0–100 | Additional Woody Features density | 255 | Outside areas |
SWF+ AWF density (Raster 100 m) | |||
---|---|---|---|
Code | Label | Code | Label |
0 | Non-SWF area | 254 | Unclassifiable |
0–100 | Small+ additional Woody Features density | 255 | Outside areas |
Practical considerations
Users can consult the layers via the online viewers available at the product’s download website. The technical documents provide useful descriptions of the characteristics of the products and all the layers available for each year of reference, including the expert products, which we have not been reviewed in detail.
6 ESM—European Settlement Map
| Product |
LULC Thematic | |
Dates | |
2012, 2015 | |
Formats | |
Raster | |
Pixel size | |
2 m, 10 m (2015) 2.5 m, 10 m, 100 m (2012) | |
Theme | |
Extent of Built-up areas (2015) Extent of Residential areas (2012) 13 built-up categories (2012) Percentage of built-up areas (2012) | |
Extent | |
Europe | |
Updating | |
Not planned | |
Change detection | |
No | |
Overall accuracy | |
Expected to be > 80% (ESM 2015 - 2 m) Expected to be > 70% (ESM 2015 - 10 m) | |
Website of reference | Website Language English |
https://land.copernicus.eu/pan-european/GHSL/european-settlement-map | |
Download site | |
https://land.copernicus.eu/pan-european/GHSL/european-settlement-map | |
Availability | Format(s) |
Open access after registration | .tiff |
Technical documentation | |
Ferri et al. (2014, 2016a, 2017), Florczyk et al. (2016), Pafi et al. (2016a), Pesaresi et al. (2013), Sabo et al. (2019), Smith and Sannier (2017) | |
Other references of interest | |
Project
The European Settlement Map (ESM) is part of the Global Human Settlement Layer (GHSL) project, supported by the European Commission through the Joint Research Centre (JRC) and the Directorate General for Regional and Urban Policy (DG REGIO). ESM complements the GHSL global products by providing an urban settlement map for Europe at a very detailed spatial resolution: 2–2.5 m versus 30 m for the GHSL. Both products share similar automatic methods for extracting LUC information from satellite imagery.
ESM was initially released in 2014, with successive updates in 2016, 2017 and 2019. In 2014, a dataset was created for the reference year 2012, showing the percentage of the surface area that was built up. This was revised with a new production methodology in 2016 and again in 2017. The first update improved the accuracy of the product and its consistency with population data. The spatial resolution was also improved: from 100 to 10 m. The second update increased the spatial and thematic detail of the product, at 2.5 m and differentiating between 12 classes. A new dataset at 2 m for the year 2015 was released in 2019, using a different production methodology. Unlike previous editions, this map only shows the extent of built-up areas, without providing further information about the built-up fraction per pixel.
In addition to the base layer delineating built-up areas, the latest edition of the product (2019) includes a classification differentiating residential from non-residential areas at a spatial resolution of 10 m.
Production method
The ESM production method has changed over time, although it has always been fully automatic. The latest edition (2019) was produced at 2 m on the basis of the Copernicus VHR_IMAGE_2015 imagery dataset, made up of images captured by the satellites Pleiades, Deimos-02, WorldView-2, WorldView-3, GeoEye-01 and Spot 6/7. The imagery was classified through a scene-based classification algorithm: Symbolic Machine Learning (SML).
The first three editions of ESM were obtained at 100, 10 and 2.5 m through a textural and morphological technique of unsupervised built-up area detection. Spot 6/7 imagery was used as an input. In the third edition (2017), auxiliary data sources (Open Street Map, Urban Atlas…) were also used to provide more thematic detail, distinguishing between 13 LUC categories, instead of just between built-up and non-built-up areas.
Product description
The ESM for each of the available editions can be downloaded separately as a single file. If more than one spatial resolution is available, users must separately download the specific product for the spatial resolution they require.
The ESM layers at 100 m for the 2014 and 2016 editions are distributed as a single European file. For the 2017 edition of ESM at 100 m, users must download a different file covering the entire mapped area for each of the categories (13 in total). The 2016 edition at 10 m is also distributed in 400 × 400 km tiles. Finally, the ESM layers at 2–2.5 and 10 m are distributed in 100 × 100 km tiles for the 2017 and 2019 editions of the product. In all cases, users can find out which tile or tiles fall within their area of interest by consulting the viewer available on the ESM website.
Downloads
Due to the complexity of this product, with different editions available for the same years of reference at different spatial resolutions, in the following table we present an overview of all the available maps, classified according to the year they were released, their spatial resolution and the year of reference, i.e. the year for which they map the LUC covers. The different files available for download are described below the table.
Available products for download | ||
---|---|---|
Product | Edition | Pixel size |
ESM 2012 | 2014 | 100 m |
2016 | 100 m | |
2016 | 10 m | |
2017 | 100 m | |
2017 | 10 m | |
2017 | 2.5 m | |
ESM 2015 | 2019 | 10 m |
2019 | 2 m |
ESM 2012 (2014)—100 m | |
---|---|
− Raster file with built-up percentage (EU_GHSL100m folder) − Raster files with technical information about the product (\EU_GHSL100m_Data_Mask and EU_GHSL100m_Data_Processed_Ref_Year folders) |
ESM 2012 (2016)—100 m, 10 m | |
---|---|
– Raster file with built-up percentage – Text file with a description of the product (.txt) |
ESM 2012 (2017)—100 m | |
---|---|
− Raster file with class percentage per pixel for one of the classes mapped in 2nd edition of ESM |
ESM 2012 (2017)—10 m | |
---|---|
− Raster files with class percentage per pixel for each of the classes mapped in 2nd edition of ESM |
ESM 2012 (2017)—2.5 m | |
---|---|
− Raster file with LC information − Layer style file for ArcGIS (.lyr) and QGIS (.qml) − PDF with technical information about the product |
ESM 2015 (2019)—10 m, 2 m | |
---|---|
− Raster file with LC information − TXT files with map legend and copyright information − File for symbolizing the raster in GIS(.clr) |
Legend and codification
ESM 2012 (2014)—100 m | |||
---|---|---|---|
Code | Label | Code | Label |
0–1 | Built-up percentage (0–100%) | -2 | No data |
ESM 2012 (2016)—100 m | |
---|---|
Code | Label |
0–1 | Built-up percentage (0–100%) |
ESM 2012 (2016)—10 m | |
---|---|
Code | Label |
0–100 | Built-up percentage (0–100%) |
ESM 2012 (2017)—100 m-Class 50 (Buildings) | |
---|---|
Code | Label |
0–1 | Percentage (0–100%) of the selected class (50) |
ESM 2012 (2017)—10 m-Class 50 (Buildings) | |
---|---|
Code | Label |
0–100 | Percentage (0–100%) of the selected class (50) |
ESM 2012 (2017)—2.5 m | |||
---|---|---|---|
Code | Label | Code | Label |
50 | BU Buildings | 20 | NBU Area-Green NDVI |
45 | BU Area-Street Green NDVI | 15 | NBU Area-Streets |
41 | BU Area-Green UA | 10 | NBU Area-Open Space |
40 | BU Area-Green NDVI | 2 | Railways |
35 | BU Area-Streets | 1 | Water |
30 | BU Area-Open Space | 0 | No Data |
25 | NBU Area-Street Green NDVI |
ESM 2015 (2019)—10 m | |||
---|---|---|---|
Code | Label | Code | Label |
0 | No data | 250 | Non-residential built-up area |
1 | Land | 255 | Residential built-up area |
ESM 2015 (2019)—2 m | |||
---|---|---|---|
Code | Label | Code | Label |
0 | No data | 2 | Water |
1 | Land | 255 | Built-up area |
Practical considerations
All editions of ESM are available for download at the Copernicus Land programme website.Footnote 2 The ESM 2015 can be consulted through an online viewer as part of the GHSL framework.Footnote 3 It can also be downloaded from the same website in tiles.Footnote 4
The 2016 ESM edition at 10 m is distributed in 237 400 × 400 km tiles. However, of the 237 tiles available for download, only 86 fall within areas with impervious surfaces. Therefore, only 86 out of the 237 tiles include LUC information.
References
Büttner G, Maucha G, Kosztra B (2016) High-resolution layers. In: Feranec J, Soukup T, Hazeu G, Jaffrain G (eds) European landscape dynamics. CORINE land cover data. CRC Press, Boca Raton, pp 61–69
Copernicus Land Monitoring Service (2020a) Copernicus land monitoring service high resolution land cover characteristics. Lot1: imperviousness 2018, imperviousness change 2015–2018 and built-up 2018. https://land.copernicus.eu/user-corner/technical-library/hrl-imperviousness-2018-user-manual. Accessed 30 Sept 2020
Copernicus Land Monitoring Service (2020b) Copernicus land monitoring service high resolution land cover characteristics. Tree-cover/forest and change 2015–2018. https://land.copernicus.eu/user-corner/technical-library/forest-2018-user-manual-v1-0.pdf. Accessed 30 Sept 2020
Copernicus Land Monitoring Service (2020c) Copernicus land monitoring service high resolution land cover characteristics. Grassland 2018 and grassland change 2015–2018. https://land.copernicus.eu/user-corner/technical-library/hrl-grassland-2018-user-manual. Accessed 30 Sept 2020
Copernicus Land Monitoring Service (2020d) Copernicus land monitoring service high resolution land cover characteristics. Lot4: water & wetness 2018. https://land.copernicus.eu/user-corner/technical-library/hrl-water-and-wetness-2018-user-manual. Accessed 30 Sept 2020
D’amico Q, Corsini N, Vandeputte R, Pennec A (2019) HRL water and wetness 2015 validation report. https://land.copernicus.eu/user-corner/technical-library/hrl-water-and-wetness-2015-validation-report. Accessed 30 Sept 2020
Faucqueur L, Desclée B, Pennec A et al (2018) HRL small woody features validation concept. https://land.copernicus.eu/user-corner/technical-library/clms_hrl_swf_validation_concept_sc03_1_1-2.pdf. Accessed 30 Sept 2020
Ferri S, Syrris V, Florczyk A et al (2014) A new map of the European settlements by automatic classification of 2.5m resolution SPOT data. Int Geosci Remote Sens Symp 1160–1163. https://doi.org/10.1109/IGARSS.2014.6946636
Ferri S, Siragusa A, Halkia M (2016a) The ESM green components. A dedicated focus on the production of the green in the European settlement map’s workflow. https://publications.jrc.ec.europa.eu/repository/handle/JRC102521. Accessed 28 Sept 2020
Ferri S, Siragusa A, Pafi M, Halkia S (2016b) How green are the European Cities? Exploring the green european settlement map 2016. https://ec.europa.eu/jrc/en/publication/how-green-are-european-cities-exploring-green-european-settlement-map-2016-systematic-comparison. Accessed 28 Sept 2020
Ferri S, Siragusa A, Sabo F et al (2017) The European settlement map 2017 release. https://publications.jrc.ec.europa.eu/repository/bitstream/JRC105679/kjna28644enn.pdf. Accessed 28 Sept 2020
Florczyk AJ, Ferri S, Syrris V et al (2016) A new European settlement map from optical remotely sensed data. IEEE J Sel Top Appl Earth Obs Remote Sens 9:1978–1992. https://doi.org/10.1109/JSTARS.2015.2485662
Hansen MC, Roy DP, Lindquist E et al (2008) A method for integrating MODIS and landsat data for systematic monitoring of forest cover and change in the Congo Basin. Remote Sens Environ 112:2495–2513. https://doi.org/10.1016/j.rse.2007.11.012
Langangke T (2015) GIO land (GMES/Copernicus initial operations land) High Resolution Layers (HRLs)–summary of product specifications
Langangke T (2016) Copernicus land monitoring service–high resolution layer imperviousness. https://land.copernicus.eu/user-corner/technical-library/hrl-imperviousness-technical-document-prod-2015. Accessed 30 Sept 2020
Langangke T, Desclee B, Faucqueur L et al (2019) Copernicus Land monitoring service–high resolution layer small woody features—2015 reference year. Product specifications & user guidelines. https://land.copernicus.eu/user-corner/technical-library/hrl_lot5_d5-1_product-specification-document_i3-4_public-1.pdf. Accessed 30 Sept 2020
Langangke T, Hermann D, Ramminger G et al (2017) Copernicus land monitoring service–high resolution layer forest: product specifications document. https://land.copernicus.eu/user-corner/technical-library/hrl-forest. Accessed 30 Sept 2020
Langangke T, Richter R, Sandow C et al (2018a) Copernicus land monitoring service–high resolution layer grassland: product specifications document. https://land.copernicus.eu/user-corner/technical-library/hrl-grassland-technical-document-prod-2015. Accessed 30 Sept 2020
Langangke T, Moran A, Dulleck B, Schleicher C (2018b) Copernicus land monitoring service–high resolution layer water and wetness: product specifications document. https://land.copernicus.eu/user-corner/technical-library/hrl-water-wetness-technical-document-prod-2015. Accessed 30 Sept 2020
Lindquist EJ, Hansen MC, Roy DP, Justice CO (2008) The suitability of decadal image data sets for mapping tropical forest cover change in the Democratic Republic of Congo: implications for the global land survey. Int J Remote Sens 29:7269–7275. https://doi.org/10.1080/01431160802275890
Manakos I, Tomaszewska M, Gkinis I, et al (2018) Comparison of global and continental land cover products for selected study areas in South Central and Eastern European Region. Remote Sens 10. https://doi.org/10.3390/rs10121967
Pafi M, Ferri S, Siragusa A, Halkia M (2016a) Visual assessment of the green European settlement map. https://ec.europa.eu/jrc/en/publication/visual-assessment-green-european-settlement-map-exploration-and-quality-performance-measures-green. Accessed 28 Sept 2020
Pafi M, Siragusa A, Ferri S, Halkia S (2016b) Measuring the accessibility of urban green areas. https://ec.europa.eu/jrc/en/publication/measuring-accessibility-urban-green-areas-comparison-green-esm-other-datasets-four-european-cities. Accessed 28 Sept 2020
Pennec A, Deutscher J, Pennec A, Dufourmont H (2019a) HRL forest 2015 validation report. https://land.copernicus.eu/user-corner/technical-library/hrl-forest-2015-final-validation-report. Accessed 30 Sept 2020
Pennec A, Vandeputte R, Sannier C et al (2019b) Analysis of the mixed forest in the 20 m HRL forest type validation report. https://land.copernicus.eu/user-corner/technical-library/analysis-of-the-mixed-forest-in-the-20-m-hrl-forest-type-validation-report. Accessed 30 Sept 2020
Pérez-Hoyos A, Rembold F, Gallego J et al (2017a) Development of a new harmonized land cover/land use dataset for agricultural monitoring in Africa. In: ESA world cover conference, 14–17 Mar 2017. Frascati, Rome (Italy)
Pérez-Hoyos A, Rembold F, Kerdiles H, Gallego J (2017b) Comparison of global land cover datasets for cropland monitoring. Remote Sens 9. https://doi.org/10.3390/rs9111118
Pesaresi M, Huadong G, Blaes X et al (2013) A global human settlement layer from optical HR/VHR RS data: concept and first results. IEEE J Sel Top Appl Earth Obs Remote Sens 6:2102–2131. https://doi.org/10.1109/JSTARS.2013.2271445
Sabo F, Corbane C, Politis P, Kemper T (2019) The European settlement map 2019 release. https://ec.europa.eu/jrc/en/publication/european-settlement-map-2019-release. Accessed 28 Sept 2020
Sannier C, Pennec A, Dufourmont H (2017) Comparative validation of HRL tree cover density and University of Maryland global forest change products. https://land.copernicus.eu/user-corner/technical-library/comparative-validation. Accessed 30 Sept 2020
Smith G, Pennec A, Sannier C, Dufourmont H (2019) HRL Imperviousness degree 2015 validation report. https://land.copernicus.eu/user-corner/technical-library/hrl-imperviousness-2015-validation-report. Accessed 30 Sept 2020
Smith G, Sannier C (2017) European settlement map 2016, validation report. https://land.copernicus.eu/user-corner/technical-library/the-european-settlement-map-validation-report/view. Accessed 30 Sept 2020
Stibig H-J, Achard F, Fritz S (2004) A new forest cover map of continental Southeast Asia derived from satellite imagery of coarse spatial resolution. Appl Veg Sci 7:153–162
Stibig HJ, Beuchle R, Janvier P (2002) Forest cover map of insular Southeast Asia at 1:5 500 000 derived from SPOT-vegetation satellite images. https://land.copernicus.eu/user-corner/technical-library/analysis-of-the-mixed-forest-in-the-20-m-hrl-forest-type-validation-report. Accessed 20 Apr 2021
Stibig HJ, Beuchle R, Achard F (2003a) Mapping of the tropical forest cover of insular Southeast Asia from SPOT4-vegetation images. Int J Remote Sens 24:3651–3662. https://doi.org/10.1080/0143116021000024113
Stibig HJ, Malingreau JP (2003b) Forest cover of insular Southeast Asia mapped from recent satellite images of coarse spatial resolution. AMBIO A J Hum Environ 32:469. https://doi.org/10.1639/0044-7447(2003)032[0469:fcoisa]2.0.co;2
Vancutsem C, Marinho E, Kayitakire F et al (2013) Harmonizing and combining existing land cover/land use datasets for cropland area monitoring at the African continental scale. Remote Sens 5:19–41. https://doi.org/10.3390/rs5010019
Weirather M, Zeug G, Pennec A et al (2019a) HRL Grassland 2015 validation report. https://land.copernicus.eu/user-corner/technical-library/analysis-of-the-mixed-forest-in-the-20-m-hrl-forest-type-validation-report. Accessed 30 Sept 2020
Weirather M, Zeug G, Pennec A et al (2019b) HRL Grassland 2015 validation report. https://land.copernicus.eu/user-corner/technical-library/hrl-grassland-2015-validation-report. Accessed 30 Sept 2020
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Copyright information
© 2022 The Author(s)
About this chapter
Cite this chapter
García-Álvarez, D., Jurado Pérez, F.J., Lara Hinojosa, J. (2022). Supra-National Thematic Land Use Cover Datasets. In: García-Álvarez, D., Camacho Olmedo, M.T., Paegelow, M., Mas, J.F. (eds) Land Use Cover Datasets and Validation Tools. Springer, Cham. https://doi.org/10.1007/978-3-030-90998-7_22
Download citation
DOI: https://doi.org/10.1007/978-3-030-90998-7_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-90997-0
Online ISBN: 978-3-030-90998-7
eBook Packages: Earth and Environmental ScienceEarth and Environmental Science (R0)