THE USE OF POLAR-ORBITING SATELLITE SOUNDING DATA TO ESTIMATE RURAL MAXIMUM AND MINIMUM TEMPERATURES

Atmospheric sounding products from NOAA's polar-orbiting satellites we re used to derive and test predictive equations of rural shelter-level maximum and minimum temperatures. Sounding data from both winter and summer months were combined with surface data from over 5300 cooperati ve weather stations in the continental United States to develop multip le linear regression equations. Separate equations were developed for both maximum and minimum temperature, using the three types of soundin g retrievals (clear, partly cloudy, and cloudy). Clear retrieval model s outperformed others, and maximum temperatures were more accurately p redicted than minimums. Average standard deviations of observed rural shelter temperatures within sounding search areas were of similar magn itude to root-mean-square errors from satellite estimates for most cle ar and partly cloudy cases, but were significantly less for cloudy ret rieval cases. Model validation for surrogate polar and tropical climat ic regions showed success in application of the four clear retrieval m odels (maximum and minimum temperature, for both winter and summer). T his indicates the potential adaptability of these models to estimates of rural shelter temperature in areas outside of the United States.