WMOD00063

• Applicant: Steven M. Hunter
• Sponsor Name: California Energy Commission
• Project Name: Optimizing Cloud Seeding for Water and Energy in California
• Title: Optimizing Cloud Seeding for Water and Energy in California
• Prepared For: California Energy Commission
• Prepared By: Steven M. Hunter
• Date: 3/31/2006
• State: CA
• Country: United States
• Weather Modification - Doc Type: Report

..J
...
<Il
~
,
o
o
<

~
-
o
....
-
o
'"

Merced River at Pohono Bridge

100
90
80
70
60 -
50
40
30
20
10
o
1920

~. .- ,./'--. .-.,..~__~. -.-..~_ .~__ _a._
-,JJ -.. . - - _~.... .-. .. .....
. I . __ a, .... _ ._ _ _._ .
. .. . - ..-
. . .. .
~ ~ ~ --- ._._a_
April-September

.

.
c

~October.March_c;l_

- __o---r- 00.
00 cP 0 00 []
[] [] n 0.' '[J l"Ooo-nc_c or;;!........o B [']
"fI g l"I.... D-o 0 GJ [] o~o (h orP-'
OU rP~oo i:tb .(J o-lhDC/1:l 0 D--o~-ODQJ-,." [] 0 rr- 00

1930

1940

1950

1960

1970

1980

1990

2000

Water Year

Figure 5. The percentage of water year streamflow at the Merced River near Pohono Bridge
control site, MDP, during the October-March and April-September periods as a function of water
year.

8. Conclusions
^n independent evaluation of 7 ongoing, operational (non-randomized) cloud seeding
programs in the major watersheds of the Sierra Nevada Mountains of Cali fomi a has been earried
out. Pending further study and analyses, the principal findings of these evaluations are as
follows:
a. The historical regression method of evaluating opcrational cloud seeding programs was
not valid for use 011 streamflow data involved in these evaluations. 'Jne data docs not satisfy the
assumptions upon which the historical regression is based. Consequently. the evaluations
obtained using the historical regression method yields results that arc inconsistent with the
seeding conceptual model upon which the operational seeding programs arc based or any other
known seedmg conceptual model.
b. The usc of ratio statistics and, in particular, the bias-adjusted rcgression ratio is a morc
valid evaluation methodology for this seeding project than the traditional historical regression
methodology, Although randomization is the only surc way of salcguarding against bias and its
intluencc on the cvaluation results. thc adjustmcnt to the regression ratio is deemed adcquate to
provide reasonably unbiased results. Thc ratio statistics methodology is very robust to dcpartures
from its inherent assumptions and it was shown that ratio statistics results approximate those from
re-randomil'.ation analyses extremcly well.
c. Thc evaluation of the Sierra Nevada i\.'1ountains opcrational seeding programs is an a
posteriori evaluation of non-randomized secding operations and their results must. therefore, be
vicwed with caution. Nevertheless. the size and statistical strength of the conservatively
estimated seeding enects using the ratio statistics method support a conclusion that the cloud
seeding operations to enhance streamflO\..., in thc major watersheds of the Sierra Nevada

46