Table 3. Mean, standard deviation for Qsub, and net subbasin yield per km2 or flow loss per m2 of channel.
Net yield 104
Net unit loss 107
Net yield 104
Net unit loss 107
Qsub
Qsub
(m3/s)
(m3/s-km2)
(m3/s-m2)
(m3/s)
(m3/s-km2)
(m3/s-m2)
Subbasin
January 1977
January 1979
Cr + WCCOg
0.48, 0.06
4.09, 0.50
0.57, 0.06
4.87, 0.55
Og + 2WCCKa
0.13, 0.15
0.18, 0.21
0.21, 0.13
0.28, 0.18
Ka + WROa
0.20, 0.34
0.30, 0.50
0.55, 0.31
0.81, 0.45
February 1976
November 1977
Cr + WCCOg
0.49, 0.12
1.01, 0.25
0.56, 0.05
4.84, 0.45
Og + 2WCCKa
9.05, 0.86
12.4, 1.18
1.94, 0.18
2.65, 0.25
Ka + WROa
0.53, 0.70
0.57, 0.75
1.81, 0.20
1.94, 0.21
would reduce the uncertainty in the winter hy-
of the study period. Even relatively mild winters
drologic balance and allow reliable estimates of
did not produce inflows from these subbasins,
the annual flow exchanges.
unless a runoff event occurred. In contrast, the
OglalaKadoka subbasin, situated between the
others, consistently contributed flow to the river.
CONCLUSIONS
The flow to this reach from three perennial creeks
is the probable cause of this anomalous behavior.
The semiarid White River basin is heteroge-
Very consistent monthly flow losses from the river
neous, with highly variable annual and winter
at a sandsilt seepage velocity provide evidence
average subbasin yields to the river caused by
of a predominantly perched river between
differences in soils and underlying strata. Winter
Crawford and Oglala. Small, variable flow yields
is the season of minimum flows throughout the
and losses suggest coupled hydrologic systems
basin. The winter water balance is simplified be-
downstream, with the alluvial water table near
cause of the absence of quantities, such as evapo-
(OglalaKadoka) or below (KadokaOacoma) the
transpiration and water withdrawals, that are large
in other seasons and have large uncertainties. We
hydrologic systems, based on the results of this
have developed a methodology for quantifying
study, are consistent with the field investigations
inflow to the river from a subbasin and the river
of Rothrock (1942).
alluvial aquifer flow exchange by month through
The mean, variance, and extremes obtained with
the winter. Important aspects of the method are a
the PEM for dependent variables such as water
winter water balance equation with a river ice
storage as ice and subbasin inflow allow defini-
growthmelt term and a point estimate method
tive conclusions to be developed or identify the
that uses deterministic models with variable or
independent variables responsible for uncertainty
uncertain parameters. The yield to the river from
in the results. Computation of air temperatures
by subbasin instead of over the complete basin,
and consumption. Trends in this parameter with
and additional river width data to characterize a
time or between subbasins have direct water man-
reach, would reduce the uncertainty in the present
agement implications.
water balance. Improved estimates of the exchange
The variable severity of the winters in our seven-
year study period did not significantly affect the
in a subbasin can be obtained by gaging all peren-
water balance. Water storage as ice is generally a
nial creeks. A well-defined water balance that
dominant component of the water balance on the
quantifies the winter river exchange with the allu-
main-stem White River below Oglala, where the
vial aquifer in semiarid regions, together with mea-
channel becomes wide. The large CrawfordOglala
surements of the relative riveralluvial aquifer
and KadokaOacoma subbasins on the main stem
levels throughout the year, can provide reliable
did not contribute flow to the river in most months
estimates of the annual flow exchange.
14