Environmental Engineering Reference
In-Depth Information
The two-member EMMA described in this sec-
tion is widely used in hydrologic studies. More
complex three-member EMMA applications
have been used to study watershed dynamics
(Dewalle
et al
.,
1988
). Concentrations of end
members must be sufficiently different for suc-
cessful application.
The EMMA method can provide detailed
information on watershed hydrology. As with
stream water-budget methods, this method
is easily adaptable to different hydrologic
systems. However, application of the EMMA
method may require a large amount of data
collection, including continuous streamflow
gauging upstream and downstream from the
reach of interest (when the method is applied
to a small watershed, there may be no surface-
water inflow to measure). Tracer concentrations
in the stream also must be continuously moni-
tored. Tracer concentration in groundwater (or
in base flow) is required, as is concentration in
precipitation or in surface runoff. For these rea-
sons, the method is usually applied over small
watersheds (1-10 km
2
drainage area), and even
then the method can be expensive and time
consuming to apply.
EMMA applications are many and diverse;
only a small number are cited in this section.
Hooper and Shoemaker (
1986
) used stable iso-
topes and major ions to track snowmelt in the
Hubbard Brook watershed in New Hampshire.
McDonnell (
1990
) used stable isotopes to iden-
tify preferential flow paths through soils.
Buttle and Peters (
1997
) used oxygen-18 and
silica to track the pathways of snowmelt in
a small basin. Uhlenbrook
et al
. (
2002
) used
oxygen and hydrogen isotopes and dissolved
silica to determine groundwater contribution
to streamflow in a 40 km
2
watershed in the
Black Forest, Germany. Becker
et al
. (
2004
) used
stream and groundwater temperatures and an
energy-budget equation to estimate ground-
water discharge to a stream in western New
York. Jones
et al
. (
2006
) used a combined sur-
face/subsurface-water flow and solute transport
model to analyze assumptions inherent in the
EMMA method and found that due to hydro-
dynamic mixing of waters, EMMA methods
may underestimate actual base flow. Baillie
et al
. (
2007
) used major ions as tracers to dis-
tinguish local, monsoon-derived groundwater
from waters recharged in distant mountains in
base flow in the San Pedro River in southern
Arizona and northern Sonora State. Gu
et al
.
(
2008
) used sulfur and oxygen isotopes to deter-
mine the contributions of different aquifers to
streamflow in southern Arizona. Young
et al
.
(
2008
) used radium isotopes to identify sources
of groundwater discharge to a tropical lagoon.
Genereux (
1998
) demonstrated how uncer-
tainty in hydrograph separation can be esti-
mated on the basis of analytical uncertainty in
tracer concentrations.
Example: specific conductance as a tracer
Stewart
et al
. (
2007
) used continuous measure-
ments of streamflow and specific conductance
of stream water (
Fig ure 4.14
) to estimate base
flow. A two-member mixing model was used:
Q QC C
bf
=
sw
[(
−
) /(
C C
−
)]
(4.17)
sw
Roff
gw
Roff
where
Q
bf
is base flow,
Q
sw
is stream discharge,
C
sw
is specific conductance of stream water,
C
gw
is specific conductance of groundwater,
and
C
Roff
is specific conductance of surface-wa-
ter runoff.
C
gw
and
C
Roff
were not measured dir-
ectly in this study.
C
gw
was assumed equal to
the stream conductance when streamflow was
at its lowest value of the year, i.e. when stream-
flow is assumed to be entirely derived from
groundwater discharge.
C
Roff
was set equal to
stream conductance at the highest stream-
flows of the year; groundwater contribution to
streamflow was considered negligible at those
times. These assumptions, which were found
to be reasonable at an intensively monitored
field site (Stewart
et al
.,
2007
), greatly facili-
tated application of the two-member EMMA
model. Within the United States, several hun-
dred US Geological Survey streamflow moni-
toring sites are equipped with continuous
specific-conductance meters (http://waterdata.
usgs.gov/nwis; accessed October 29, 2009).
Conceivably, this model could be applied at
many of these sites.
Base flow was estimated by the EMMA model
and by hydrograph separation using HYSEP
at 10 locations across the southeastern United
