WMOD00527

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

00
Ll)
N

24 ..a 72 96 120 '''4 lS8 192 216 2~O 2a4 288 312 336 3600
Lf)
N

~ U!llnQ Model Precip.
- . U!lng Observed Precip.
6 -0- (\ OM~rved Inflow .

o
Lt)

, I
, I
(J) \
: "
: I~ ~
, I ~ \
: T ~ : ~
, I I,
I , \
9 : l~\
-/ \~
, f ''''
I I r
0'
~

o
o
N

"
i
I

..........0
CJ) 0
LL N
()

..-
o
(/)
"0
c 0
ro Lt)
(/)
:J
o
.c
I-
----

o
L.{')

~

?;
o
:+:: 0
c 0

,
I
,
,

o
o

~

~

E
o
(I)
o
LL

o
Lf)

01
o

72 96 120 144

'10
192 216 248 264 288 312 336 360

Hours After Start of inflow Prediction

Figure 4. - Example of simulated hydrographs of inflow to Folsom Reservoir (data are for the February 1986 storm)
(C.F.S '" fe/s).

l
I

~

To summarize the results statistically, values of computed and observed inflow peaks were
tabulated. Some leeway in timing was permitted between computed and observed peaks.
Some mismatches of 12 h occurred, but most were within about 6 to 8 h (recall that 6 h is
the time interval for which HED71 requires inputs).

The linear correlation between predicted and observed peaks was determined. Also, an
arbitrary high inflow value of 50,000 ft3fs (cubic feet per second) was chosen and the
probability of detection (forecasting) of observed inflows above this value was determined as
was the false alarm rate, i.e., the rate of predicting more than 50,000 fefs when the observed
inflow remained less than 50,000 fefs.

13