Remote sensing of global wetland dynamics with multiple satellite data sets

This study is the first global effort to quantify seasonality and extent of inundation with a suite of satellite observations, including passive and active microwave along with visible and infrared measurements. A clustering technique which merges the satellite observations is used to detect inundation. Monthly flooded areas are then calculated by estimating pixel fractional coverage of flooding using the passive microwave signal and a linear mixture model with end-members calibrated with radar observations to account for vegetation cover. The global results, comprising natural wetlands, irrigated rice, and lakes/rivers, indicate a minimum inundated area for the July 1992-June 1993 period of 2.16 x 10(6) km(2), about 38% of the maximum 5.75 x 10(6) km(2), to be compared to maximum areas of 5.83 x 10(6) km(2) and 5.70 x 10(6) km(2) from independent data sets. Comprehensive evaluation requires substantial additions to the sparse observational record now available.