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.

2001

TABLE 5. Summary of statistics for two Ridge temperature partitions for seeded (s) and non~seeded (ns) cases.
44 s Control Zone I Zone 2 Zone 3
Temperature ~ -9OC 56 ns (3 gages) (12 gages) (9 gages) (7 gages)
Target-control correlation All 0.75 0.66 0.66
coefficients s 0.78 0,60 0.57
ns 0.81 0,73 0.75
Mean precipitation (mm) s 2.14 3.21 1.51 1.31
ns 3.01 2.90 1.54 1.28
Median precipitation (mm) s 1.27 1.98 0.65 0.49
ns 1.73 2.10 0.61 0.20
Days with zero precipitation s 6 2 9 7
at all gages ns 6 3 II 10
One-tailed probabilities Wilcoxon 0.008 0,13 0.09
MRPP 0.01 0.14 0.08
Double ratio Mean 1.56 1.38 1.43
Median 1.29 1.46 3.35
22 s Control Zone I Zone 2 Zone 3
Temperature ~ -13 oC 27 ns (3 gages) (12 gages) (9 gages) (7 gages)
Target-control correlation All 0.77 0.53 0.42
coefficients s 0.85 0.58 0.52
ns 0.78 0,67 0.54
Mean precipitation (mm) s 1.37 2.29 1.35 1.21
ns 2.81 2.46 0.94 0.76
Median precipitation (mm) s 0.51 1.86 0.62 0.40
ns 1.10 1.71 0.40 0.15
Days with zero precipitation s 3 0 4 2
at all gages ns 4 0 5 4
One-tailed probabilities Wilcoxon 0.009 0.008 0.005
MRPP 0.007 0.01 0.004
Double ratio Mean 1.90 2.93 3.24
Median 2.35 3.40 5.96

inclusion of temperature observations for times without
suitable clouds may dilute any "signal" in the data.
As discussed in Section 5, rawinsondes were
launched only when potential storm conditions existed.
Therefore, the use of T700 might be expected to
"sharpen" the temperature partition. The average T700
from all available rawins for each experimental day
was used to partition the day. Again the entire pop-
ulation was first subdivided into halves, thirds and
quarters and these were subjected to the Wilcoxon and
MRPP tests with the results given in Table 6.
Table 6 shows the coldest half, third and quarter of
all days to have low probability'values in Zone 1. Low
values are also shown for the two colder quarters of
the cases in Zone 2 and the second coldest in Zone
3. However, unlike all other low probability values
. shown in this table, the quarter with 39 cases between
-13.8 and -to.50C yielded mean and median double
ratios below unity, both in Zone 2 and Zone 3. These
double ratios ranged from 0.32 to 0.68 suggesting de-

creased snowfall on seeded days. The temperature range
is low enough so that artificial ice nuclei concentrations
should have been significant. It may be that snowfall
which naturally would have fallen on Zones 2 and 3
was redistributed onto Zone 1 for this temperature
range. This could have resulted if increased ice particle
concentrations caused by seeding resulted in more ag-
gregation and, thereby, increased snowflake fall ve-
locities as found by Holroyd and ]iusto (l971).
The colder cases were tested at 2 oC intervals in Table
7. Probabilities of 0.05 or less are seen for both tests
in Zone 1 for all but the warmest partition shown.
Both double ratios for Zone 1 are largest for T700 of
~ -13 oc. Probabilities less than 0.10 appear in Zones
2 and 3 only for T700 of ~ -150C.
The results of partitioning by T700 are similar to
those obtained with Ridge temperature. This is prob-
ably because Ridge temperature did not usually change
markedly over the course of the day. This resulted in
a correlation coefficient of 0.90 between 24 h mean