WMOD00473

• Title: Evaluation of Bridger Range Winter Cloud Seeding Experiment Using Control Gages
• Date: 12/12/1983
• Weather Modification - Doc Type: Report

DECEMBER 1983

ARLIN B. SUPER AND JAMES A. HEIMBACH, JR.

2003

TABLE 7. 700 mb temperature partitions: colder days by 20 increments. Probabilities and double ratios are for 3 zones of Fig. 1.

Zone I Zone 2 Zone 3
Experimental
days Probability Double ratio Probability Double ratio Probability Double ratio
Range
(0C) All sins. W MRPP Mean/Median W MRPP Mean/Median W MRPP Mean/Median
~ -7.0 127 62/65 0.14 0.09 1.17/1.35 0.33 0.24 1.03/1.55 0.26 0.21 1.09/3.89
~ -9.0 104 52/52 0.02 0.02 1.37/1.00 0.24 0.24 1.14/1.12 0.17 0.15 1.23/2.94
~ -11.0 70 35/35 0.02 0.02 1.52/1.30 0.22 0.20 1.29/1.29 0.36 0.34 1.17/1.47
~ -13.0 50 22/28 0.02 0.02 1.74/1.76 0.10 0.13 1. 77 /2.07 0.12 0.11 1.57/2.06
~ -15.0 26 11/15 0.03 0.03 1. 72/1.12 0.01 0.02 3.81/1.73 0.06 0.09 3.41/3.00
· s-seeded/ns-nonseeded.

....

from about -lOoC to about - 24 oC within which seed-
ing can lead to increased precipitation. Grant and El-
liott appear to agree that for the particular circum-
stances of the BRE, temperatures near mountaintop
levels should define seeding effectiveness better than
cloud-top temperature. However, they suggest that the
latter should be crucial in establishing the requirement
for seeding. '
Cloud top estimates are known to be crude when
based on radiosonde units manufactured prior to de-
velopment of a new humidity duct introduced into
general use in about 1972 (Hill, 1980b). Montana State
University researchers were aware of the humidity
problems and obtained the new ducts for retrofitting
to their store of radiosondes as soon as they became
available. Records of the first use of the new duct no
longer exist, but examination oftemperature-dewpoint
differences suggests it was near the beginning of the
final winter season. This implies that cloud tops for
19'70- 71 were largely estimated from less reliable hu-
midity data than the 1971-72 cloud tops.

The method used to estimate cloud top temperature
from radiosonde data in the BRE is discussed in detail
in Part II. Basically, the temperature-dewpoint dif-
ference was used and the difference was linearly in-
creased with decreasing pressure such that a 30C dif-
ference was used at 700 mb, a 40C difference at 600
mb and 50C difference at 500 mb, etc. The manner
of handling such problems of multi-layered clouds is
also discussed in Part II. While no claim is made for
the accuracy of the cloud-top estimates due to lack of
independent supporting data, all estimates were cal-
culated in an objective and consistent fashion.
All available sounding data were averaged to estimate
the mean cloud-top temperature for each 24 h exper-
imental unit. The median value of the entire population
of 149 days was -25.50C, and 80% of all values were
between -24.0 and-55.0oC. The whole population,
and halves, thirds and quarters of the population, were
subjected to the Wilcoxon test for each of the three
zones of Fig. 1. None of the resulting 30 probabilities
was <0.05. This suggests that cloud-top temperature,

TABLE 8. Cloud-top temperature partitions for fractions of all days with Ridge temperature ~ -9.00C:
... Wilcoxon (W) and MRPP one-tailed probabilities are for the 3 zones of Fig. I.
Experimental days Zone I Zone 2 Zone 3
Range (0C) All sins. W MRPP 'W MRPP W MRPP
-55.0/-8.0 84 39/45 0.02 0.03 0.19 0.18. 0.17 0.15
-29.5/-8.0 42 19/23 0.17 0.27 0.40 0.13 0.33 0.28
-55.0/-29.6 42 20/22 0.04 0.04 0.04 0.06 0.13 0.21
-25.5/-8.0 27 11/16 0.20 0.27 0.24 0.18 0.29 0.17
-34.5/-25.6 28 16/12 0.05 0.Q7 0.14 0.15 0.40 0.32
-55.0/- 34.6 29 12/17 0.19 0.24 0.10 0.13 0.40 0.49
-22.7/-8.0 21 9/12 0.26 0.33 0.21 0.23 0.35 0.29
-29.5/-22.8 21 10/11 0.34 0.46 0.32 0.14 0.46 0.50
-37.9/-29.6 21 10/11 0.01 0.01 0.06 0.08 0.12 0.17
-55.0/- 38.0 21 10/1,1 0.34 0.31 0.15 0.19 0.18 0.29
· s-seeded/ns-nonseeded.