Environmental Engineering Reference
In-Depth Information
on the year-to-year variability in recharge over
that period.
Individual environmental and historical
tracers are listed in
Tables 9.1
and
9.2
as opposed
to the specific tracer methods (profile, mass bal-
ance, peak displacement, and age dating) dis-
cussed in
Chapter 7
. This was done for purposes
of clarity; multiple methods can be applied
with data for a single tracer, and the time scales
listed for each tracer generally are valid for
any of the methods. An exception to this rule
occurs when using the unsaturated zone peak-
displacement method for a tracer, such as chlor-
ide, under conditions of changing land use. In
this case, times scales given for applied tracers
are more appropriate. Individual applied trac-
ers are not listed in
Table 9.1
or
9.2
because, in
general, time and space scales are similar for all
applied tracers. The tritium/helium-3 method
generally is applicable only when the recharge
rate is at least 30 mm/yr because of diffu-
sion of helium in groundwater (
Section 7.3.3
).
Historical groundwater and unsaturated-zone
tracers, such as tritium, CFCs, and sulfur hex-
afluoride (SF
6
), are applicable only for specific
ranges in groundwater ages and sampling dates
(
Chapter 7
) depending on the history of atmos-
pheric concentrations (e.g. SF
6
is only good for
dating water recharged after about 1970).
A few additional points require clarification
with regard to
Table 9.1
. The data collection
frequency column indicates which methods
require no data collection (these methods rely
solely on the use of existing data), which meth-
ods (mostly tracer methods) require only a
single field trip to collect data, and which
methods require data collection at multiple
times (this could mean continuous data record-
ing or multiple field trips). Methods based on
streamflow data, such as streamflow duration
and hydrograph analysis, are usually based
on mean daily streamflows. However, the esti-
mates generated by these methods are averages
over multiyear or decadal periods. Hydrograph-
separation and recession-curve displacement
methods are applicable over identical space and
time scales. The Darcy unsaturated-zone unit-
gradient method is based on the assumption
that recharge is a steady process; the method
cannot be used to estimate nonsteady recharge.
Other methods (e.g. water-table fluctuation,
zero-flux plane, step-response function, and
streamflow recession-curve displacement
method) are based on the assumption that all
recharge occurs as discrete episodes in time;
therefore, these methods do not account for
recharge that occurs as steady flow.
Models deserve particular attention in any
discussion of time scales. They can be applied
over a multitude of time intervals, including
periods for which no data may exist (e.g. when
making predictions for future periods). Model
inputs can be based on actual data or synthetic
data (such as weather or climate trends) gen-
erated independently outside of the model. In
addition to providing recharge estimates, mod-
els have many other uses (
Chapter 3
). They can
be used to integrate diverse types of data and to
explore various hypotheses on recharge proc-
esses; in so doing, models can help to shape
conceptual models. Models can provide error or
uncertainty bounds on recharge estimates that
they generate. Sensitivity analysis can be used
to identify model parameters that most influ-
ence calculated recharge rates, and, therefore,
help guide data-collection efforts. The effects
of temporal and spatial variability of model
parameters on timing and locations of recharge
can also be investigated.
Spatial scales over which methods are
applicable are listed in
Table 9.2
. As with time
scales, estimates made at small scales can be
used to calculate estimates at larger scales, but
unlike the case for temporal scales, upscaling is
not simply a matter of summing up the smaller
scale estimates. Upscaling, as described in
Chapter 3
, usually takes into account the vari-
ability in factors described in
Section 9.2
, such
as climate and land use, that affect recharge
rates. Downscaling is problematic; if a recharge
estimate represents an integrated value over
some area (e.g. from application of a method
such as streamflow hydrograph separation), it
is generally not possible to infer information
on recharge processes at specific points within
that area.
