Environmental Engineering Reference
In-Depth Information
35
1800
Figure 4.14
Average daily
streamflow, conductivity, and
base flow as determined from
end-member mixing analysis
for Joshua Creek, Florida,
USGS gauging station 02297100
(Stewart
et al
.,
2007
, Figure 3,
Ground Water
, Wiley-Blackwell).
Conductivity
Discharge
Base flow
1600
30
1400
25
1200
20
1000
800
15
600
10
400
5
200
0
0
Dec-01
Mar-02
Jun-02
Sep-02
Nov-02
Feb-03
Date
States (Stewart
et al
.,
2007
). At six of the sites,
the methods produced comparable estimates
(differences were not greater than 12%). At the
other four sites, the HYSEP estimates of base flow
were from 37% to 197% greater than those pro-
duced by the EMMA model. These differences
were attributed to an inadequate definition of
N
, the number of days in which streamflow is
due solely to surface runoff. A revised version
of Equation (
4.6
) was proposed:
where
Q
pump
is water injection rate,
C
inj
is tracer
concentration in injected water,
Q
x
is stream
discharge at some point x-distance down-
stream,
C
x
is measured tracer concentration at
that same point, and
C
bg
is background tracer
concentration.
Q
x
is determined by rearranging
Equation (
4.19
):
Q Q C CC
x
=
/(
−
)
(4.20)
pump
inj
x
bg
Estimates of discharge may be obtained in this
manner for a number of sampling locations
along the course of a stream. A water budget
(generally a simplified form of Equation (
4.2
))
is then constructed for the reach between any
two sampling locations. Groundwater discharge
and surface-water inflow from tributaries to
the reach are estimated from the difference in
discharge between the two locations. If a reach
between measurement points at distances x1
and x2 downstream from the injection point
contains no surface-water inflow, then net
groundwater flow to the stream is equal to
Q
x2
-
Q
x1
, assuming other terms in Equation (
4.2
)
are negligible. The method is only valid for
gaining reaches of streams.
The tracer must be conservative, background
concentration in the stream must be uniform,
and there should be no significant sources or
sinks for the tracer in groundwater or surface
water contributing to streamflow. Rhodamine
dye, chloride, bromide, and lithium are com-
monly used tracers. Streamflow and tracer injec-
tion rate must be steady throughout the length
(4.18)
N
=
0.48
A
0.44
where
A
is drainage area in km
2
. Stewart
et al
.
(
2007
) suggested that the mixing model method
may be a convenient approach for calibrating
hydrograph separation programs such as HYSEP.
Given the simplifying assumptions in both the
mixing and hydrograph separation methods,
other factors could be contributing to the dis-
crepancy in results. The authors pointed out
that overestimation of
C
Roff
will lead to under-
estimation of base flow.
4.6.2 Tracer-injection method
Streamflow can be measured by injecting a con-
servative tracer into a stream at a known rate
and then measuring the concentration of the
tracer at downstream locations (Kilpatrick and
Cobb,
1985
). The approach is based on a mass-
balance equation for the injected tracer in the
stream:
Q
C
=
QC C
(
−
)
(4.19)
pump
inj
x
x
bg
