Land Use and Land Cover (LULC)Follow Us: Twitter Follow Us on Facebook YouTube Flickr | Share: Twitter Facebook
Global Mangrove Forests Distribution, v1 (2000)
Description of the Data
The extent of the world's mangrove forests circa 2000 are mapped in the Global Mangrove Forest Distribution, v1 (2000) data set. The mangrove data are compiled using more than 1,000 Landsat Thematic Mapper (TM) scenes from the Global Land Survey (GLS) and the USGS archives, and applying hybrid supervised and unsupervised digital image classification techniques. The data are available at ~30-m spatial resolution. The total area of mangroves in the year 2000 was estimated at 137,760 km2 in 118 countries and territories in the tropical and subtropical regions of the world, about 12% less than previous estimates.
This dataset was created by Chandra P. Giri et al. (2010) as described in the journal Global Ecology and Biogeography, and distributed by the NASA Socioeconomic Data and Applications Center (SEDAC).
The mangrove forest distribution data set was created by classifying more than 1,000 Landsat TM scenes using a hybrid unsupervised and supervised classification. The authors applied an unsupervised ISODATA clustering algorithm using ERDAS IMAGINE to generate 50–150 spectral clusters at the 99% convergence level. Through iterative labeling, mangrove classes were identified and labeled with reference to field data and high-resolution QuickBird and IKONOS imagery, then merged into a single mangrove category. Qualitative validation was performed with the help of local experts and high-resolution satellite data such as QuickBird and IKONOS by dividing the entire area into 500 by 500 grids and checking each visually in order to identify and correct gross errors found in the classified maps.
The spatial data are provided as 10° by 10° tiles of GeoTiff rasters. Raster cell sizes are 0.0003 degree decimal (one arc-second or about 30 m at the equator). The data set is in the geographic coordinate system. The downloaded compressed zipfiles contain raster data and a readme file.