Microwave Surface Emissivity Database

The microwave land surface emissivity is derived from a combination of AMSR-E brightness temperature and MODIS LST and NCEP GFS data. Contents include global gridded monthly emissivity products at the AMSR-E 10.7 GHz, 18.7 GHz, 23.8 GHz, 36.5 GHz, and 89.0 GHz V-pol and H-pol channels.

Spatial coverage is global except for high latitude regions where LST is not retrieved by the MODIS day/night LST algorithm. Global sinusoid grid resolution is 27.79 km with 1440 columns and 720 rows.

Database brief description

The database is composed of two datasets: a multi-product database containing multiple AMSR-E channel emissivity products retrieved by so-called "1a", "1b" and "classification" algorithms, and a merged database derived from the multi-product database to provide the "best" available estimates of the average effective emissivities.

The 1a dataset is retrieved directly from AMSR-E brightness temperatures, using MODIS LST to represent the emitting temperature of the surface and ancillary data to define the atmospheric contribution.

The 1b algorithm estimates the microwave emission depth and effective emissivity by fitting the solution of a thermal diffusion equation to a one-month time series of clear-sky measurements, assuming sinusoidal diurnal surface forcing. It is produced as an alternative to the 1a product over arid or semi-arid regions where the microwave penetration depth effect is substantial and our time-series model solution provides credible estimates.

Classification-based dataset emissivity is produced where persistent cloud coverage precludes production of the ordinary 1a product. The emissivity and its variance are obtained from a combination of data from all clear grid points that have a compatible MODIS-retrieved land class and surface stability.

Note that the 1a dataset may include emissivities larger than 1.0 and may have large day/night differences over dry regions due to inconsistency between MODIS surface skin temperatures and effective microwave emitting temperatures that include subsurface emission.

Nighttime 1a emissivities are closer to the 1b emissivities than daytime 1a emissivities are because the largest subsurface thermal gradients occur during the day for an early afternoon overpass.

Data for January 2003 to December 2003 are available for download.  Instructions are below.

Acknowledgement

This material is based upon work supported by the National Aeronautics and Space Administration under Contracts No. NNH04CC43C and NNH08CC96C, issued through the Science Mission Directorate.

For more information, please contact Dr. Alan Lipton.