A simultaneous land surface temperature (LST)-emissivity retrieval algorithm has been developed and tested for the GOES-8 and Advanced Very High Resolution Radiometer (AVHRR) split-window channels. By assuming that the spectral emissivities are constant over a short time period (12-24 hours), two sets of split-window radiance measurements taken at two different times are used to retrieve the spectral emissivities and the two LSTs. The algorithm employs an optimization approach rather than a direct solver because of constraint requirements in the approach. The retrieved variables are those that minimize the sum of the squared differences between the measured satellite radiances and those predicted by a spectrally detailed radiative transfer model.
Validation of the algorithm has been done using GOES-8 measurements and in situ ground measurements for the Department of Energy's Atmospheric Radiation Measurement-Cloud and Radiation Testbed (ARM-CART) site. In addition, an AVHRR version of the algorithm has been validated using in situ ground measurements from the First ISLSCP Field Experiment (FIFE) site, the Hydrological-Atmospheric Pilot Experiment-Sahel (HAPEX-Sahel) site, and a LST validation site operated in Australia. The biases of the retrieved LSTs for the validation sites in the Australia, FIFE and ARM-CART study areas are less than 0.08 C, 1.7 C, and 1.9 C, respectively; The associated bias adjusted root mean square errors are less than 0.78 C, 4.8 C, and 4.5 C, respectively. The GOES-8 retrieved LSTs exhibit some noise and a daily cycle in their errors when compared to in situ measurements. The noise is attributed to random errors in the input satellite observations. The systematic differences between validation measurements and retrievals are near-zero during the nighttime but exhibit a slight sinusoidal oscillation during the daytime. The latter cycle can be attributed to too coarse sampling of the diurnal evolution of the thermodynamic structure of the atmospheric boundary layer, which determines changes in the GOES-8 split-window weighting functions.
The simultaneous retrieval algorithm is now being applied to GOES-8 measurements over the entire ARM-CART site in order to study the diurnal and seasonal changes in the spectral emissivities. Since these factors are a function of the type and amount of vegetation within the satellite's field of view, the retrieved emissivities can also be used to infer seasonal changes of land use within the agricultural zones situated within the ARM-CART domain