WMOD00527

• Title: Quantitative Precipitation Forecasting for Improving Reservoir Operations
• Date: 4/1/1995
• Weather Modification - Doc Type: Report

............

CJ)
LL

00
-0
ON
en
"0
c
as
en
:J
o
..cO
CI.O
~
as
Q)
0..
:;:
o
;;:::0
c 0

E
0
en
0
LL
"0
Q) 0
~ 1.0 '*
Q)
en
.0
0
0
0

00
1.0
N

100

150

200

25Q:,
If)
N

50

POD = 14/14 = 1.00
FAR = 5/19 = .26

'*

o
o
N

,r = .92
Std. Error = 16.5 k-cfs
Y = .97X - 1.6

..

'*

o
If)

o
o

'*
'* '*
*
'*
'* '*
'*

o
If)

o
250

50 100 1 50 200
Predicted Folsom Inflow Peak (Thousands of C.F.S.)

Figure 6. - Scatterplot of observed inflow peaks to Folsom Reservoir with predicted inflow peaks based upon observed
22-site mean precipitation (C.F.S = fe/s).

of 0.92 was found. Letting the predictor variable be the set of predicted inflow peaks from
the hydro graphs generated with orographic model precipitation as HED71 input, the slope
of the resulting regression equation (for predicting the inflow peaks computed using observed
precipitation) turned out to be 0.87, or once again, the same as the overall ratio of observed
to model precipitation. A quite small intercept of -3000 refs was found (fig. 7).

It is interesting that the orographic precipitation model seems to be even more useful as
input to the HED71 hydrologic model for predicting peaks in inflow to Folsom Reservoir
(correlation coefficient, r = 0.87) than would be implied by its direct correlation to observed
12-h totals of precipitation (r = 0.73). This higher correlation is caused at least in part by the
leeway permitted in time when matching peaks of inflow.

These encouraging results for the ARB demonstrate the potential operational applicability
of the method, viz., using Oakland rawinsonde observations for the first several hours and
then using predicted upper air conditions for orographic precipitation model input for future
time periods. The use of predicted conditions is discussed in the next section.

16