Laserfiche WebLink
<br />1774 <br /> <br />JOURNAL OF APPLIED METEOROLOGY <br /> <br />VOLUME 17 <br /> <br />TABLE 5. Stratification over ranges of LCTTI, the lifted cloud top temperature with respect to ice; (a) warmer than <br />-100C, (b) between (10 and -300C and (c) colder than -300C. <br /> <br />Projects All <br />Precipitation group Crest <br /> <br />Stratifications <br /> <br />LCTTI> -lOoC <br /> <br />M +N = 93 <br /> <br />Seeded Nonseeded <br />M=36 N=57 <br />Zeros = 10 Zeros = 19 <br />Mean= 2.94 Mean= 2.33 <br /> <br />Ratio (seeded/nonseeded) = 1.26 <br />Test statistic = 1.230 <br />P value =0,22 <br /> <br />(a) <br /> <br />Projects All <br /> <br />Precipitation group Crest <br />-100C< <br />LCTTI< -300C <br /> <br />Stratifications <br /> <br />M+N=861 <br /> <br />Seeded . Nonseeded <br />M=353 N=508 <br />Zeros = 35 Zeros = 108 <br />Mean= 2.72 Mean= 1.74 <br /> <br />Ratio (seeded/nonseeded) = 1.57 <br />Test statistic = 5.444 <br />P value<O.OOl <br /> <br />~ <br /> <br />, <br />1 <br /> <br />(b) <br /> <br />Projects All <br />Precipitation group Crest <br />Stratifications LCTTI< -30oC <br />M+N=294 <br /> <br />Seeded Nonseeded <br />M=110 N=184 <br />Zeros= 6 Zeros= 18 <br />Mean= 3.20 Mean= 3.09 <br /> <br />Ratio (seeded/nonseeded) = 1.03 <br />Test statistic=O,413 <br />P value =0,64 <br /> <br />(c) <br /> <br />CENSARE was found to have the largest effect when <br />eliminated. Table 6 shows the statistics with no <br />stratification and without CENSARE. The' seed/no- <br />seed ratio has been reduced from 1.34 to 1.13 (see <br />Table 4) while the P value has changed very little. <br />The effect of CENSARE can also be seen clearly in <br />Fig. 2. The change in ratio is large when it is considered <br />that CENSARE only had 118 cases. Since none of the <br />other projects had such a large effect on the seed/no- <br />seed ratio, it was decided to exclude CENSARE from <br /> <br />TABLE 6. Tabulated statistics for all projects except CENSARE <br />with no stratifications. <br /> <br />Projects All less CENSARE <br />Precipitation group Crest <br />Stratifications None <br /> <br />M+N=1130 <br /> <br />Seeded <br /> <br />Nonseeded <br /> <br />M";'428 N=702 <br />Zeros = 48 Zeros = 137 <br />Mean = 2.26 Mean = 2.00 <br /> <br />Ratio (seeded/nonseeded) = 1.13 <br />Test statistic =3.051 <br />P value <0.001 <br /> <br />the analysis at this point until the question of bias <br />is resolved, The next step then was to conduct analyses <br />based on multiple-variable stratifications using the 1130 <br />cases left in the sample. <br />Tables 7-11 display the crest group statistics for the <br />five-project sample stratified into five combinations of <br />variables which define expected positive and negative <br /> <br />TABLE 7. Statistics for a stable cloud, with crest trajectory, <br />moderate water content and LCTTI .between -10 and -300C. <br /> <br />J <br /> <br />Projects All less CENSARE <br /> <br />Precipitation group Crest <br /> <br />Stratifications <br />EPOS<O,OO5J g-l <br />-3600 < BTI < 3600 s <br />3<CBWS<8 g kg-1 <br />-30 < LCTII < -10oC <br /> <br />M+N=178 <br /> <br />Seeded <br /> <br />Nonseeded <br /> <br />M=57 N=121 <br />Zeros = 2 Zeros = 22 <br />Mean= 2.90 Mean= 2,40 <br /> <br />Ratio (seeded/nonseeded) = 1.18 <br />Test statistic = 2,357 <br />P value =0,018 <br />