1000
800
600
400
200
Figure 15. AFDD at Columbus vs. date
0
for winter 1993-94.
Nov '93
Dec
Jan '94
Feb
Mar
3
40 x 10
20
Provisional
Minimum
Ice Event
Figure 16. Minimum discharge re-
Adjusted
quired for ice jam for 1 February to
0
8 March, 1994.
1 Feb
6
16
21
26
3 Mar
8
11
17 January 1994 and remained above 400 through
bilities that may be enhanced by the addition of
the end of March (Fig. 15). The peak AFDD (924)
missing ice data.
occurred on 1 March 1994. Forecasting discharge,
Using the model to predict an ice jam requires
or even simply monitoring discharge, however, is
the calculation and updating of AFDD using ob-
more complicated. The discharge criteria for the
served minimum and maximum daily air tem-
prediction model are based on the reported dis-
peratures at Columbus; also, some estimate of dis-
charges at North Bend, which are actually adjust-
charge at North Bend is necessary. Once the
ed from the real-time, provisional discharges. The
minimum AFDD criterion of 400 has been met,
provisional discharges are calculated from the
forecast discharges can be compared to the mini-
stages reported at the gage, which include the
mum discharge associated with an ice jam for a
effect of ice and are therefore too high. At the end
given date, which can be estimated using Figure
of each winter season, the provisional discharges
14. If the AFDD criterion has been exceeded, and
are adjusted to reflect the presence of ice and are
the forecast discharge is greater than the discharge
revised downward. The difference between the
for that date shown in Figure 14, there is a high
provisional and adjusted discharges can be sig-
probability that an ice jam will occur in at least
nificant (Fig. 16).
one location in the study area.
If the minimum ice event discharge criteria
Calculating or forecasting AFDD is relatively
were applied to the provisional discharges, sever-
simple. For example, the AFDD at Columbus
al ice events would have been predicted begin-
during the winter of 1993-94 had exceeded 400 by
29
Previous Page
