31 Jan 94
(2) Locations and descriptions of river cross
(1) A record of the daily average water tempera-
sections on the mainstem and tributaries. Generally,
ture at the upstream and downstream ends of the
cross section descriptions, referenced to known eleva-
mainstem, at the upstream end of the modeled reach
tions, must be provided every 5 to 20 river widths.
of each tributary, and at selected intermediate points.
In reaches where the river geometry varies signifi-
cantly along the river, the cross sections must be
(2) A record of the daily average air temperature
spaced closer together. All geometric features known
during the same period. A record of the high and
to influence the flow should be included.
low temperature recorded each day is usually suffi-
cient because the average of these measurements is a
(3) Location and operation rules of hydraulic
good indication of the daily average air temperature.
control structures such as locks and dams.
The air temperature should be measured at a suffi-
cient number of locations so that the actual air tem-
perature variations along the river system are well
represented. If the air temperature normally has
brate an unsteady flow model, the channel roughness
strong gradients -- for example, if the river rises
or resistance to flow under a variety of flow condi-
rapidly in elevation, or if the river travels directly
tions must be determined. Energy loss coefficients
north and south -- more air temperature measurement
may also have to be determined. The minimum data
stations are required.
required to calibrate an unsteady flow model are
records of discharges and their associated stages
during at least three flow events, which should com-
prise low, medium, and high flows. For each flow
greatest uncertainties in river ice forecasting is apply-
event the following is needed:
ing river ice mechanics to specific situations. This is
because there is no broad, underlying, physically
(1) A record of the designated boundary condi-
based theory, but rather a series of rules determined
tions at the upstream and downstream ends of the
through observation in the field and the laboratory,
mainstem, and at the upstream end of each tributary.
plus theory based on analogies that apply in certain
Typically, the upstream boundary conditions are
limited situations. As a result it is important that data
known discharges, and the downstream condition is a
on the ice conditions be as thorough as possible.
known stage or rating curve.
(2) Records of the discharges at each lateral
(1) A record of the location and extent of ice
along the river. It is important to note the reaches
where the ice is moving, and estimate the surface
(3) Records of stages at several locations, includ-
concentration of ice; the locations where the ice has
ing the downstream end of the mainstem. The chan-
bridged across the channel and formed a stable cover;
nel roughness can be estimated only for reaches
the upstream progression rate of the leading edge of
upstream of each measured stage location.
the ice cover; and the locations where the ice cover
cannot progress, such as regions of high velocity.
Frequent aerial surveys using video cameras can
collect this information.
mal model determines the river water temperature at
each location of interest, and calculates the volume of
(2) An estimate of the ice thickness at as many
ice produced once the water temperature has declined
locations as possible. It is not possible to estimate
to 0C (32F). The heat transfer between the river
the ice thickness by just looking at it; direct measure-
water and the atmosphere is usually the most impor-
ment is required. This is a labor-intensive process
tant factor in the change of water temperature. This
that must be conducted with due regard to safety.
heat transfer rate is calculated simply as a linear
However, these measurements are very important for
calibrated function of the temperature difference
model calibration. Generally, it is better to have
between the air and the water. This approach is
measurements over a wide area at a few points in
sufficiently accurate and relies only on air temper-
time rather than to have many measurements at one
ature, which is separately available from meteoro-
location. Measurements of thickness should indicate
logical forecasts. The heat transfer rate must be
calibrated during the time when no ice exists in the
river. This requires: