Environmental Engineering Reference
In-Depth Information
depth, a sample can be collected from the entire water column using a cylinder
inserted vertically into the wetland water (Swanson
1978
). If
C
P
P
is a significant
component of tracer input, then a specialized sampling device must be used to
collect atmospheric deposition (see Allan (
2004
) for specification). If a specialized
device is not available, it may be better to use published data as a proxy rather than
using erroneous data collected via inappropriate methods. Precipitation chemistry
data are available from national and international atmospheric deposition monitor-
ing networks, such as Global Atmospheric Watch (
http://gaw.empa.ch/gawsis/
)
.
Temporal and spatial variability in
C
Si
and
C
Gi
need to be properly represented
by collecting multiple samples over time. If there are multiple inflow streams,
samples should be collected at all streams and volume-weighted averages of
concentrations should be used in the mass-balance equation. Due to geologic
heterogeneity, solute concentrations in groundwater may have large spatial
variability even within a relatively small area. Therefore, it is necessary to collect
groundwater samples from several locations (and depths) in the areas of anticipated
groundwater discharge. Water samples usually are collected directly from monitor-
ing wells and piezometers, or springs; they also can be extracted from sediment core
samples (see Adams (
1994
) for methods). Wells should be bailed or pumped prior
to sample collection to ensure that the sample represents the composition of
groundwater in the aquifer surrounding the well screen. A common purging proto-
col is to pump 3-5 times the volume of water in the well to ensure that the stagnant
water in the casing has been completely removed prior to sample collection.
Another option is to slowly pump water through an intake tube that is placed in
the screened interval of the well (low-flow sampling method). The pumping rate
needs to be slow enough that virtually no drawdown occurs in the well, in which
case nearly all of the water supplied during pumping originates from the aquifer.
Detailed procedures for groundwater sampling are found in manuals and handbooks
on this subject (e.g., Yeskis and Zavala
2002
; Wilde
2006
). The number of
groundwater samples commonly used is insufficient to determine the precise
value of an average tracer concentration,
. Therefore, a recommended
practice is to use one standard deviation from the arithmetic mean of all ground-
water samples to represent the uncertainty in the mass-balance calculation.
<
C
Gi
>
3.10.3 Use of Multiple Tracers
If more than one conservative tracer is available for water-budget determinations, and
their concentrations are not correlated, then Eq.
3.51
can be written for each individ-
ual tracer. This increases the number of equations and thus, the number of unknowns
that can be determined. For example, using naturally occurring chloride and artifi-
cially introduced bromide, a set of three equations can be solved for three unknowns.
Alternatively, the water-budget equation is solved separately with each mass-balance
equation to provide separate estimates of the same unknowns. If resulting values
using separate tracers are greatly different, then possible errors in estimation or
measurement of known terms or missing terms in the equations are indicated.
