Estimating, Validating and Conveying Measurement Differences in the Land Surface Temperature and Emissivity Products from ASTER, MODIS and AIRS

One of the key Earth System Data Records (ESDRs) identified by NASA and many national and international organizations is Land Surface Temperature and Emissivity (LST&E). LST&E data are critical for many studies including climate and ecosystem modeling, land use land cover change, and atmospheric retrieval schemes. Understanding the uncertainties associated with LST&E data across different sensors is essential if these data are to be used as effectively as possible in modeling studies and as long-term ESDRs. Currently there are significant discrepancies among the LST&E products provided by NASA that arise due to the different retrieval methods used to derive them, and the different spatial and spectral scales associated with each sensor. Sensors such as ASTER, MODIS and AIRS on NASA's Terra and/or Aqua satellites produce LST&E products at a variety of spatial, spectral and temporal scales. While validation studies have been undertaken for a particular product from a given sensor, there have been relatively few studies that have examined and conveyed the sources of uncertainty between the products with a view to producing a single unified LST&E data record. The lack of such studies limits the usefulness of satellite-derived LST&E data as an ESDR, and as inputs to ecosystem and climate models. For example, an error of 0.1 in emissivity will result in current climate models having errors of up to 7 Wm-2 in their upward longwave radiation estimates - a much larger term than the surface radiative forcing due to an increase in greenhouse gases (~2-3 Wm-2), making accurate knowledge of the LST&E a critical component for climate change studies. An in-depth study on error and uncertainty analysis between current LST&E products produced by NASA is essential to provide an uncertainty metric by which LST&E climate data sets and trends from futures sensors can be measured. In meeting this need our goal will be to identify and resolve the largest sources of uncertainty for each of the LST&E products from ASTER, MODIS and AIRS using a global set of validation sites representing a broad range of different conditions; produce a comprehensive, land-cover dependent error budget for the standard LST&E products from ASTER, MODIS and AIRS; and evaluate the implications of these uncertainties when the products are used as inputs to ecosystem and climate models for determining important variables such as evapotranspiration and net longwave radiation.