WMOD00315 - Laserfiche WebLink

Home Browse Search

2. INSTRUMENTATION, DATA REDUCTION AND DATA PROCESSING Although results in this Final Report are based mainly on data collected from the two mobile radiometers, nearly all data platforms were used in describing the 1991 storms and evaluating the characteristics of SL W in these storms. The primary instrumentation was positioned as noted in Fig. 2.1, a map of the 1991 Utah/NOAA project area. The DR! had primary responsibility for data from the DR! mobile radiometer, project soundings taken from the Mt. Pleasant Airport, and GOES-7 satellite data and synoptic data archived on the DR! MclDAS system. The DR! received radiometer, precipitation, and meteorological data from the U. S. Bureau of Reclamation (USBR) and K.-band radar data from the University of Utah (UU). Data editing and archival procedures were discussed in some detail in Huggins (1991). As necessary, additional new information will be presented here in order to describe the different forms of data that were used in the analyses presented in Sections 3 and 4. A. Microwave Radiometer Data The processing of microwave radiometer data has been discussed at length in previous reports; e.g. Long (1984), Long (1986), and Huggins (1990a,b). The method of "front-end" calibration using tipping curves is described in detail by Decker and Schroeder (1991). This calibration accounts for the internal variations of components, and ensures that the radiometer provides data in absolute terms. Several tipping curve calibrations were performed during clear conditions for both the DR! and the USBR radiometers. However, because of the erratic behavior of certain components, or because component variations occurred during lengthy periods between tipping calibrations, the final processed data of both radiometers still often required correction to eliminate a nonzero offset. For the USBR data set, the correction was applied to daily data files. Only the liquid data were corrected to eliminate a nonzero offset. Typically a value of .01-.03 mm was added to or subtracted from the entire set of 2-min depth averages in the daily file. The actual correction was determined from the average offset from zero noted during known clear conditions. The vapor data were not adjusted because no comparison to an absolute reference was available. The radiometer water vapor depths correlated very well with precipitable water estimates (integrated upward from the height of the Plateau) from the Mt. Pleasant soundings, but some natural difference could be expected between these independent measurements of vapor because of errors in both instruments, horizontal variations in water vapor between the measurement locations, or because some unknown amount of vapor measured by the sounding at levels below the top of the Plateau could be transported over the Plateau and subsequently 2-1 I I I I I I I I I I I I I I I I I I I