in most years the discharges before breakup on
where FDD is the accumulated degree days of
the Aroostook are well below 10,000 ft3/s, and
freezing (degrees Fahrenheit), and c is an em-
then in a matter of days increase to levels well
pirical constant to account for wind exposure
above 15,000 ft3/s. On this basis, we have used
and snow cover. For this analysis, and for h ex-
discharges surpassing 10,000 ft3/s as an indicator
pressed in inches, the value of the constant was
of probable ice cover breakup. The required
taken as c = 0.60. While eq 1 was developed to
breakup discharge varies, however, with the
predict ice growth on still bodies of water rather
actual freezeup discharge for a given year as well
than flowing rivers that generate and accumu-
as variations in the other terms listed in Table 2.
late frazil, it provides a useful, if approximate,
The index dates of DQmax and DQ10, when com-
estimate of the relative quantities of ice present
pared to the date of maximum freezing degree
from year to year.
The fourth column in Table 2 (Dmax) is the
days Dmax, can be used to reflect the arrival of
Julian day (days since 1 October of the year in
significant spring runoff attributable to warm
question) when the FDD term began to decrease.
weather. Columns 9 and 10 present the differ-
Low values for this term indicate relatively early
ence, in days, between the onset of negative
warming and, therefore, significant ice deterior-
freezing degree days and increased runoff.
ation and melting was likely before a significant
Finally, the last column lists the total snowfall
increase in runoff and resulting breakup. The
prior to the jam. Ideally, the effects of snow cover
next two columns, Qmax W and Qmax FF, list the
would be accounted for through the depth of
maximum mean daily discharge for Washburn
snow remaining on the ice prior to breakup, but
and Fort Fairfield during the estimated time of
such information is not generally available. In-
ice cover breakup. As described previously, the
stead, we have used total snowfall as an indica-
values for Washburn are derived directly from
tor. Thick snow covers prior to breakup can
the USGS gage records, while the Fort Fairfield
insulate the ice from deterioration by warm
values are transposed from the Washburn gage
weather and solar radiation. In the absence of
based on drainage area.
snow, even a relatively thick ice cover can be
Columns 7 and 8 indicate the Julian dates
weakened by solar radiation to reduce ice jam
when the maximum discharge occurred (DQmax)
flooding potential. In addition to insulating the
and when the discharge first exceeded 10,000
ice cover, the melting of a thick snow cover can
ft3/s (DQ10) respectively. We know that it takes
significantly increase the rate of rise of a flood
a certain magnitude of discharge and stage in-
hydrograph, further ensuring that a thick, strong
crease to release an ice cover and to allow it to
ice cover is present at breakup
As previously mentioned, the Fort Fairfield
move downstream. For typical flood hydro-
area regularly experiences ice jams. In some
graphs, the required increase in stage is on the
years, however, the ice-related flooding is more
order of three or four ice thicknesses above
severe. In Table 3, the same factors examined in
freezeup levels. If the increase in discharge is
Table 2 were evaluated for years with severe ice
rapid or the ice deteriorated, the required in-
jam flooding. The 1973 flood happened in mid-
crease in stage may be less. For the 2.5 ft of ice
winter, and as a result the absolute magnitudes
measured on the lower Aroostook River on 25
March 1992, this rule of thumb would have re-
of some of the terms in Table 3 are quite different
quired a stage increase in excess of 7.5 ft. How-
from those for other floods. Also, there are two
ever, there was significant deterioration of ice
sets of discharge values in Table 3 for each runoff
strength and thickness prior to breakup. Under
event as determined from the Washburn gage
the assumption that the ice had thinned to about
records. The first line represents the maximum
1.5 ft, the required stage rise would have been
discharge during the runoff event. These values
only about 4.5 ft.
are comparable to those presented in Table 2.
Lacking direct field observations of ice break-
The second line of discharge data was deter-
up, there is some uncertainty as to the actual
mined for the dates of peak stages rather than
date of breakup and peak flooding. For the typi-
peak discharge. These discharge values are typi-
cal freezeup and midwinter discharge of about
cally much lower than the peak discharges, indi-
1000 ft3/s in the study area, however, a spring
cating that the ice cover or jam became unstable
runoff event of between 10,000 and 15,000 ft3/s
and washed downstream before the event peak
should be sufficient to initiate breakup. Further,
was reached. Since these values are derived from
a review of the Washburn gage data shows that
tables of mean daily discharge, they do not nec-
5
Previous Page
