Laserfiche WebLink
<br />212 JOURNAL <br /> 2.50 <br /> 2.25 <br /> 2.00 <br /> 1.75 <br /> E <br /> .s 1.50 <br /> Cl <br /> 5 1.25 <br /> :J <br /> <l <br /> <l1.00 <br /> 0 <br /> z <br /> 0.75 <br /> 0.50 <br /> 0.25 <br /> <br />OF A TMOSPHERIC AND OCEANIC TECHNOLOGY <br /> <br />VOLUME 4 <br /> <br />....' <br />.,..".s;" <br /> <br />~ <br />I <br /> <br />. .;.;~ <br />Irl':':"~' <br />...... 0" <br />.:=-,::' ... <br />'" . . <br />~-'. . <br />..~j~f:.' <br /> <br />~.,~~~~::! <br />. . t1,:<~~:t.. <br />':'f..d f;: : <br />!~."...: <br />. ;..s:,.:.:...; <br />..~I.:... <br />~. ."*\7'.,' <br />. .. <br />0" '''~ 00. <br />. ...-. l' <br />f.. <br />o. 00 <br /> <br />L~..,. <br />. - <br /> <br />.. <br />~ .~::i; <br /> <br />0.00 <br />0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 <br />USSR LIQUID (mml <br /> <br />FIG. 8. As in Fig. 7. but for liquid water. <br /> <br />on these occasions was approximately 8 km upwind <br />of the radiometer site. This separation ensured that the <br />rawinsonde would not pass to the lee of the mountain <br />during the ascent. Two tipping curves were performed <br />during each rawinsonde release. . <br />The results of the tipping curve/rawinsonde calIbra- <br />tion measurements of integrated vapor are presented <br />in Table 3. During all measurements, the maximum <br />departure in measured vapor was 0.05 cm. The la~gest <br />discrepancy differences between rawinsonde and either <br />radiometer occurred during the 1800 UTC December <br />1984 comparison. The radiometer measurements in <br />this case were identical. Half of the measurements by <br />the radiometers. departed from the rawinsonde mea- <br />surements by 0.02 cm or less. In general, agreement <br />between the different instrumentation systems.was very <br /> <br />good. <br /> <br /> TABLE 3. Tipping curve calibrations. <br /> NOAA USBR Rawinsonde <br /> Time vapor vapor vapor <br />Date (UTC) (cm) (cm) (cm) <br />4 Dec 1984 0600 0.33 0.35 0.36 <br /> 0.33 0.32 <br />4 Dec 1984 1800 0.14 0.14 0.19 <br />4 Dec 19.84 2100 0.17 0.16 0.16 <br /> 0.18 0.18 <br /> <br />4. Summary and discussion <br /> <br />Two studies have been performed to check the va- <br />lidity of radiometer measurements. One was designed <br />to determine independently the accuracy of the vapor <br />channel of the radiometer. A second study, an inter- <br />comparison of radiometers, was not a test of accuracy, <br />but rather of stability, reliability, and general system <br /> <br />integrity. <br />The study comparing vapor measurements from ra- <br />winson des and the radiometer showed excellent agree- <br />ment. The rms difference between the two instruments <br />was 0.07 cm, with a slight bias (0.04 cm) towards the <br />rawinsonde measurement. These differences were less <br />than 15% of the means, which was less than the theo- <br />retical estimation of error for radiometer-retrieved val- <br />ues calculated by Westwater (1978). <br />The collocated radiometer study examined the sys- <br />tem stability. The vapor and liquid channels were found <br />to exhibit similar trends between radiometers, though <br />a small bias often existed. Comparisons of paired data <br />from the vapor channel yielded a correlation coefficient <br />of 0.95 with a rms difference of 0.05 cm, while the <br />liquid water channel data yielded a correlation coeffi- <br />cient of 0.99 with a rms difference of 0.02 mm. <br />For determining integrated amounts of vapor and <br />liquid, the radiometer appears to be an appropriate <br />device capable of operation over a wide range of severe <br />wintertime conditions. Comparison with rawinsonde <br />data indicates that radiometric measurements of in- <br /> <br />" <br />