One of the most important physical parameters that can be retrieved from passive microwave radiometers is emissivity, in particular, over land. Emissivity can provide information on surface parameters such as soil condition, vegetation, and land cover. To advance the current state of empirically driven land-based precipitation algorithms, precipitation estimates can be improved over land if an accurate emissivity can be retrieved. Therefore, surface characterization is highly needed to advance the GPM-era precipitation over land algorithms.

An intercomparison study under the auspices of NASA's Precipitation Measurement Missions (PMM) Science Team was organized to assess the current status of established and emerging emissivity data sets and techniques. Currently, the PMM Land Surface Characterization Working Group (LSWG) is comparing seven algorithms that use different techniques to estimate emissivity (e.g., by direct observational, by using a land surface model, and by physical retrieval). These algorithms are compared during a three year period (July 2004 to June 2007), initially at three different sites (continental US (central, and southern) and in southern Canada). Amongst the data set being compared are estimates coming from passive microwave radiometer (e.g. SSMI/I, SSMI/S, AMSR-E, TMI, AMSU/MHS, WindSat), or in combination with ancillary data (e.g. land surface model data, ISSCP cloud mask, satellite-based precipitation data).

The comparison study tries to explore similarities and differences between the different techniques. These techniques are being compared by time series, monthly means, and percent of difference between them during periods of land surface change, such as snow cover, vegetation, presence of clouds, and precipitation. The results to date demonstrate that emissivity is affected by these conditions, more or less depending on the target site, but some of the discrepancies between these techniques are still unknown, and are under current evaluation.