Description: The Landsat-7 scan-line error affects the classification in certain areas, resulting in striping artefacts in the data.
Classification accuracy was assessed with over 53,800 randomly sampled points across 20 randomly selected regions. Overall accuracy was 95.25 %, while User’s and Producer’s accuracies for the mangrove class were estimated at 97.5% and 94.0%, respectively. Users should be aware that it is a global-scale dataset, generated with a single methodology applied over all regions, and as such, the accuracy of the map may vary between locations. Factors such as satellite data availability (due to clouds, cloud shadows and Landsat-7 scan-line error), mangrove species composition and level of degradation all influence the local accuracy. The mangrove seaward border is generally also more accurately defined than the landward side, where distinction between mangrove and certain terrestrial vegetation species can be unclear.
Areas known to be missing in this version (v2.0) of the dataset: Bermuda (U.K.); Europa Island (France); Fiji, east of longitude 180°E; Guam and Saipan (U.S.); Kiribati; Maldives; Peru, south of latitude S4°; and Wallis and Futuna Islands (France).