Global Broadband Infrared Surface Emissivity Computed from Combined ASTER and MODIS Emissivity over Land (CAMEL)

Infrared surface emissivity is an important variable in the estimation of longwave net radiation, a critical parameter in weather and climate models and the earth radiation budget. Earth's surface emits radiation to the atmosphere and space as a non-ideal blackbody with spectral variations due to the mineral composition of the soil and seasonally varying vegetation cover. Due to a prior lack of spatially and temporally variant global broadband emissivity (BBE) measurements, it is common practice in land surface models to use BBE set as a single constant over the globe. This can lead to systematic biases in the estimated net radiation for any particular location and time of year. Several efforts have recently been put forth to develop BBE datasets using satellite measurements (e.g. Wang et al. 2005; Tang et al. 2010; Ogawa et al. 2002, 2008; Huazhong et al. 2013; Cheng et al. 2012, 2015). Such methods involve regressions of BBE to the measured narrowband emissivities.

Under the NASA MEASUREs project a new global, high spectral resolution land surface emissivity database is being made available as part of the Unified and Coherent Land Surface Temperature and Emissivity Earth System Data Record. This CAMEL emissivity database, is created by the merging of the MODIS baseline-fit emissivity database (UWIREMIS) developed at the University of Wisconsin-Madison and the ASTER Global Emissivity Database produced at the Jet Propulsion Laboratory. Available for 2000 through 2017 at monthly mean, 5km resolution for 13 bands within the 3.6-14.3 micron region, this dataset is extended to 417 infrared spectral channels using a principal component regression approach.

This work presents BBE calculated using this new database with MODIS surface temperature as input. BBE over two wavelength ranges is computed—the full available MEASUREs spectrum from 3.6-14.3 microns and 8.0-13.5 microns, which has been determined to be an optimal range for computing the most representative all wavelength, longwave net radiation (Ogawa and Schmugge, 2000, 2004; Cheng et al., 2013). This dataset provides the advantages of being consistent with the MEASUREs HSR emissivity and not requiring regression schemes—BBE can be calculated by simple numerical integration over the MEASURES high spectral resolution emissivity product. Variations of monthly BBE since 2003 are investigated and are separated by land cover classification schemes.