Data

This section provides access to various datasets published by the Land Cover ECV project.

Interactive globe of Land cover class defined in LCCS (1992-01–2015-12) Version: 2.0.7

Products at moderate spatial resolution:

10-20m demonstration prototypes using Sentinel-2 over Africa and Mesoamerica:


MRLC maps series from 1992 onwards (v207 and v2.1.1)

Consistent global annual land cover maps at 300m spatial resolution from 1992 to 2020

Each pixel value corresponds to the label of a land cover class defined based on the UN Land Cover Classification System (LCCS). LCCS classifiers support the further conversion into Plant Functional Types distribution required by the Earth System Models. The typology counts 22 global land cover categories (see the complete CCI MRLC legend).

The MRLC maps series is delivered along with 4 quality flags documenting the full-time series (they are not year-specific):

These maps are derived from a unique baseline MRLC map which is generated thanks to a classification chain applied to the entire MERIS FR and RR archive from 2003 to 2012.

Independently from this baseline, MRLC changes are detected at 1 km based on a time series of annual global classifications generated from AVHRR HRPT (1992 - 1999), SPOT-Vegetation (1999 - 2012) and PROBA-V (2013 - 2015), and Sentinel-3 (recent years). A systematic analysis of the temporal trajectory of each pixel revealed significant land cover changes for a simplified land cover typology matching the IPCC classes: cropland, forest, grassland, wetlands, settlements and other lands. The last category is further split into shrubland, sparse vegetation, bare area and water.

From 2003 onwards, the changes detected at 1km are re-mapped at 300 m. The last step consists of back- and updating the 10-year baseline MRLC map to produce the annual MRLC maps over the entire period.

Some remarks:

Global Plant Functional Type (PFT) map series

Spatially explicit vegetation fractions for climate models

The PFT global dataset 1992-2020 has 14 layers per year, each describing the percentage cover (0-100%) of a plant functional types at a spatial resolution of 300 m: broadleaved evergreen trees, broadleaved deciduous trees, needleleaved evergreen trees, needleleaved deciduous trees, broadleaved evergreen shrubs, broadleaved deciduous shrubs, needleleaved evergreen, needleleaved deciduous shrubs, shrubs natural grasses, herbaceous cropland (i.e., managed grasses), built, water, bare areas, and snow and ice.

“Plant Functional Types” (PFTs) refer to globally representative and similarly behaving plant types. PFTs can be related to physiognomy and phenology, climate (which defines the geographical ranges in which a plant type can grow and reproduce under natural conditions, and physiological activity (e.g., C3/C4 photosynthetic pathways).

All terrestrial zones of the Earth between the parallels 90°N and 90°S are covered. The PFT dataset has a regular latitude-longitude grid with a grid spacing of 0.002777777777778°, corresponding to ~300 m at the equator and ~200 m in the mid-latitudes. The projection is a Plate-Carrée with a geographic latitude-longitude representation based on the WGS84 ellipsoid.

The plant functional type (PFT) distribution was created by combining auxiliary data products with the CCI MRLC map series. The LC classification provides the broad characteristics of the 300 m pixel, including the expected vegetation form(s) (tree, shrub, grass) and/or abiotic land type(s) (water, bare area, snow and ice, built-up) in the pixel. For some classes, the class legend specifies an expected range for the fractional covers of the contributing PFTs and broadly differentiates between natural and cultivated vegetation. We used a quantitative, globally consistent method that fuses the 300-meter MRLC product with a suite of existing high-resolution datasets to develop spatially explicit annual maps of PFT fractional composition at 300 m. The new PFT product exhibits intraclass spatial variability in PFT fractional cover at the 300-meter pixel level and is complementary to the MRLC maps since the derived PFT fractions maintain consistency with the original LC class legend.

This Data Set was generated to reduce the cross-walking component of uncertainty by adding spatial variability to the PFT composition within a land cover category. This work moved beyond fine-tuning the cross-walking approach for specific LC classes or regions and, instead, separately quantified the PFT fractional composition for each 300 m pixel globally. The result is a dataset representing the cover fractions of 14 PFTs at 300 m, consistent with the CCI MRLC LC maps for the corresponding year.

Citation Harper, K. L., Lamarche, C., Hartley, A., Peylin, P., Ottlé, C., Bastrikov, V., ... & Defourny, P. (2022). A 29-year time series of annual 300-metre resolution plant functional type maps for climate models. Earth System Science Data Discussions, 1-37. https://essd.copernicus.org/preprints/essd-2022-296/

Global Water Bodies 4.0

A global map of open permanent inland water bodies, ocean and land at 150 m spatial resolution.

The CCI WB v4.0 is composed of two layers:

  1. A static map of open water bodies at 150 m spatial resolution resulting from a compilation and editions of land/water classifications: the Envisat ASAR water bodies indicator, a sub-dataset from the Global Forest Change 2000 - 2012 and the Global Inland Water product. Legend: 1-Land, 2-Water.
    This map is delivered at 150 m as a stand-alone product but it is consistent with the class "Water Bodies" of the baseline map used as input to generate the annual LC Maps after resampling to 300 m using an average algorithm.
  2. A static map with the distinction between ocean and inland water is available at 150 m spatial resolution. Legend: 0-Ocean, 1-Land. It is fully consistent with the CCI WB-Map v4.0 at 150 m.

Citation Lamarche, C., Santoro, M., Bontemps, S., d’Andrimont, R., Radoux, J., Giustarini, L., ... & Arino, O. (2017). Compilation and validation of SAR and optical data products for a complete and global map of inland/ocean water tailored to the climate modelling community. Remote Sensing, 9(1), 36. https://www.mdpi.com/2072-4292/9/1/36

CCI-LC User Tool

A user tool dedicated to climate modellers

This tool fits land cover products to climate modellers' needs by sub-setting, resampling, re-projecting and converting land cover classes into Plant Functional Types according to default or user-defined cross-walking tables.

Functionalities:

Note that the current version of the user tool (v3.3) is not working with the CCI-LC Water Bodies product yet. This will be included in the next user tool version and this is the reason why it can still be fine-tuned.

More info here: http://maps.elie.ucl.ac.be/CCI/viewer.

For any further questions or concerns, email us at contact@esa-landcover-cci.org.

Land Surface Seasonality Products

Three global climatological 7-day time series describing the natural variability of the vegetation, the snow cover and the burned areas

On a per-pixel basis, these LC seasonality products reflect, along the year, the average dynamics and the inter-annual variability of the land surface over the 1998-2012 period. They are expressed as 7-day time profiles of the average and standard deviation for the vegetation greenness (NDVI) or as temporal series of occurrence probabilities for the snow and the burned areas.

Although they are built from existing and independent datasets, they were found to be quite consistent among themselves and with the land cover classes. These products are complementary to the three global CCI-LC maps products characterizing the same period.

Each climatology product is delivered in 52 files (1 file per 7-day time interval) and each file is made of measurements and quality flags.

More info here: http://maps.elie.ucl.ac.be/CCI/viewer.

For any further questions or concerns, contact us at contact@esa-landcover-cci.org.

MERIS surface reflectance time series

The surface reflectance (SR) products consist of MERIS global time series covering the 2003-2012 period. The spectral content encompasses the 13 surface reflectance channels – the atmospheric bands 11 and 15 being removed – and the spatial resolution is of 300 m for the Full Resolution (FR) and 1000 m for the Reduced Resolution (RR). The time series are made of temporal syntheses obtained over a 7-day compositing period. To simplify the handling and analysis of global datasets, the MERIS SR time series are delivered in 5°x5° tiles.

Pre-processing: The pre-processing chain generates global SR time series by a series of pre-processing steps, including radiometric corrections, geometric correction, pixel identification, atmospheric correction with aerosol retrieval, BRDF corrections as well as compositing and mosaicking.

Quality control of input products: The MERIS dataset is very valuable and the use of the full mission dataset in the CCI-LC project in a consistent manner is a major effort. It requires advanced techniques for the development of specific quality checks related to the input data

The quality of each global multispectral SR composite is described, on a per-pixel basis, by a set of flags and values: uncertainties for each spectral band, the current status of the surface, uncertainties for each spectral band, the number of observations with clear sky land coverage, water coverage, clear sky snow and ice coverage, cloudy coverage and cloud shadow coverage for each pixel. The uncertainties of the surface directional reflectance value are calculated from the contributions of each error source, assuming a negligible correlation between the different error sources.

The obtained values are compared with in-situ data from CEOS LandNet sites and with reflectance products available from other sensors and other projects. Besides assessing the quality of individual composites, the quality of the global SR time series is also documented, to quantify their discrimination potential.

More info here: http://maps.elie.ucl.ac.be/CCI/viewer.

CCI-LC Prototypes using Sentinel 2

These were generated within the CCI MRLC initiative using Sentinel-2 data at 20m over the whole of Africa and at 10m over Mesoamerica.