P. Gong et al. land-cover classification system as well as the International Geosphere-Biosphere Programme (IGBP) system. Using the four classification algorithms, we obtained the initial set of global land-cover maps. The SVM produced the highest overall classification accuracy (OCA) of 64.9% assessed with our test samples, with RF (59.8%), J4.8 (57.9%), and MLC (53.9%) ranked from the second to the fourth. We also estimated the OCAs using a subset of our test samples (8629) each of which represented a homogeneous area greater than 500 m × 500 m. Using this subset, we found the OCA for the SVM to be 71.5%. As a consistent source for estimating the coverage of global land-cover types in the world, estimation from the test samples shows that only 6.90% of the world is planted for agricultural production. The total area of cropland is 11.51% if unplanted croplands are included. The forests, grasslands, and shrublands cover 28.35%, 13.37%, and 11.49% of the world, respectively. The impervious surface covers only 0.66% of the world. Inland waterbodies, barren lands, and snow and ice cover 3.56%, 16.51%, and 12.81% of the world, respectively.

No abstract

In this paper, we report the first wetland mapping of the entire China using Landsat enhanced thematic mapper plus (ETM+) data. These data were obtained from the Global Land Cover Facility at the University of Maryland spanning from 1999 to 2002. A total of 597 scenes of Landsat images were georeferenced and mosaiced. Manual image interpretation of satellite images was aided with elevation data, soil data, land cover/land use data and Google Earth. The minimum mapping unit is 10 pixel × 10 pixel, equivalent to 9 ha. The aim of our first round of mapping was only targeted at the boundary delineation of any type of wetland except those wetlands that are under agricultural use (i.e., paddy fields), which has already been well mapped by others. Our interpretation results indicate that a total of 359478 km 2 of wetlands are of non-agricultural use. Among our preliminarily mapped wetland, 339353 km 2 are inland wetland, 2786 km 2 are non-agricultural artificial wetland, and 17609 km 2 are coastal wetland. Because low-tide is rarely captured in satellite images, an under-estimation of coastal wetland is inevitable. We conducted some statistics based on our mapped wetlands and compared them with those previously obtained from a number of sources including a land cover/land use map made with satellite images during the late 1990s and early 2000s, a marshland map developed in approximately the same period, survey data of coastal wetland in early 1980s, and area data for approximately 400 larger patches of marshland in China compiled in 1996. Because some inconsistencies exist in the guidelines of those different wetland surveys, difference in area is expected. Some further comparison indicates that the wetland distributions derived from the preliminary wetland map are reasonable and more objective than other sources. The mapping process also indicated that the method adopted by us was efficient and cost-effective. We also found that in order to ensure comparability of the wetland maps developed at different times, a set of standard guidelines on the wetland categories to be mapped, and the mapping methods to be used must be well conceived, developed and effectively employed. We carried out some initial geographical analysis on the distribution of wetlands.China wetlands, remote sensing, wetland mapping Wetlands host a biodiversity far beyond any other ecosystem. Along with oceans and forests, wetlands form the three major ecosystems on earth [1] . In the past two hundred years, human activity and settlement has drastically reduced wetlands [2] . From the 1780s to 1980s, America's wetlands were reduced by 53% [3] . In England, since the Rome times, 23% of the river deltas, 50% of marsh and 40% of the bog have disappeared (http:// www.ramsar.org/about/about_wetland_loss.htm, 1996) [4] .