A Modis Sea Surface Temperature Composite Product

Short term forecasts of meteorological parameters (including precipitation) over the oceans is often difficult because of the lack of observations used to initialize the forecast models in these regions. While sea surface temperature (SST) fields derived from satellite and in situ data currently provided to the models provide some large scale information on surface forcing, the small scale gradients in SST that are important for regional forecast models, particularly in coastal regions, are not. The mission of the NASA's Short-term Prediction and Research Transition (SPoRT) program (Goodman et al. 2004) is to accelerate the infusion of NASA Earth Science Enterprise observations, data assimilation and modeling research into NWS forecast operations and decision-making at the regional and local level. In support of this, a 1 km MODIS SST composite product has been developed to use in numerical weather prediction modeling to study the possible forecast improvement the high-resolution SST data provides. In order to provide a high-resolution, spatially continuous SST product for assimilation into regional forecast models, daily MODIS SST data from the Aqua satellite were temporally composited and interpolated to a 1000 x 1000 km grid for the coastal region surrounding Florida. For the case study period of May 2004, the daily SST composites were assimilated into the SPoRT version of the Weather Research and Forecasting (WRF) predication system to study the improvement in short-term regional forecasts. This paper presents a description of the SPoRT Aqua MODIS SST compositing method, a comparison of the composite to independently produced GOES SST composites and the real-time global SST analysis field, and validated against GOES SST data and in situ buoy data. The results of the WRF forecasts are presented in the companion paper by LaCasse et al. (2006). Although only Aqua MODIS composites are discussed and used in this paper, the same method can be applied to Terra data with similar results.