WMOD00477 - Laserfiche WebLink

Home Browse Search

210 '-'" SCAN I 2... 2.00 1.70 1-'" 1.2, "'0 070 0.00 E 020 U 000 0 '0 eo 0 :; 2-'" SCAN 4 0.... ~ 2.00 JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY VOLUME 4 '-'" SCAN 2 2.20 2DO '-" 1-'" ,... IDO ..70 0.00 .... 2-'0 SCAN 3 .20 rl I 2.00 .1 I. '-" ..70 0.00 0... ODD ODD 120 160 200 240 280 320 J60 0 040 80 120 160 200 240 280 320 360 0 40 eo 120 160 200 240 280 320 360 2-'0 SCAN 6 2... ..0 SCAN 5 .... 2.00 070 0>0 .... .... QOO 000 QOO o 40 80 120 160 200 240 280 320 J60 0 40 80 120 160 200 240 280 320 360 0 40 80 120 160 200 240 280 320 360 ..20 1.70 ,-'" AZIMUTH (deg) Fla. 6. As in Fig. 5 except for liquid water. hemisphere. The second scan was the least coordinated and consequently may have been affected by temporal variations in the cloud system. Values in the vapor channel were similar in the 3200 to 1200 quadrant but differed significantly at other azimuths. An exception was in the region of liquid water peaks near 2300 where vapor values approached each other. During this scan the liquid water was above the noise level, but not as high as in later scans. The Bureau of Reclamation liquid values were higher in the eastern quadrant and lower in the western quadrant. A double peak in liquid water was observed near 2300 azimuth by USBR, but only one peak was seen by NOAA. Liquid water values during the remaining scans in- creased significantly. Close coordination was achieved on these scans. In the vapor channel, a trend similar to scan 1 was again observed in the northern quadrant (USBR higher). Such a bias was observed in all re- maining scans in the vapor channel. Both radiometers duplicated peaks and valleys associated with the liquid water distribution. However, with few exceptions, the USBR radiometer measured consistently higher values. The complete statistics for each scan are presented in Table 2. During all six scans, the vapor throughout the viewing region underwent only minor variations. The mean value of vapor measured by either radi- ometer in any scan varied from 2.55 to 2.89 cm. Little variation with azimuth was noted as evidenced by the small standard deviations. In general, the rms and mean absolute differences between radiometers were small over all scans; the correlation coefficient for all scans was 0.71. A scatter plot of all vapor points determined during the scanning portion of this study is shown on Fig. 7. Values of liquid water throughout the hemisphere varied considerably in space and time. Despite these large variations, measured values ofliquid water by the radiometers correlated extremely well (0.99). However, a consistent offset was present throughout the dataset with NOAA values exceeding those measured by the USBR. Root-mean-square differences between the two radiometers for the complete scan data varied between 17 and 19% of the mean liquid values measured. The mean absolute difference between radiometer mea- surements varied between 14 and 16% of the means. The scatterplot for all liquid water values determined during the scanning portion of this study is shown on Fig. 8. 3) RADIOMETER CALIBRATIONS Calibration of the instrument, including the antenna system, is accomplished using the clear, stable atmo- sphere as a signal source by means of elevation scan observations, commonly called "tipping curves." In the calibration procedure, the absolute absorption of the atmosphere at each operating frequency and in the zenith direction is determined from the slope of the relative values of absorption measured by the radi- ometer as a function of antenna elevation angle. Ab- solute absorption, T (in nepers) is related to the zenith brightness temperature, Tb, by the expression T=ln[(Tm-2.9)/(Tm- Tb)] (1)