Environmental Engineering Reference
In-Depth Information
Recharge estimates produced by streamflow
duration and hydrograph analysis techniques
represent spatially extensive diffuse recharge
that is assumed to occur uniformly over some
area of the aquifer. Accurate delineation of that
contributing area can be problematic; com-
monly, it is assumed equal to the surface drain-
age area. Base-flow estimates are integrated over
time scales that are largely unknown. Base-flow
estimates by themselves provide no insight into
the travel time or the travel path that water
takes from the time it arrives at the water table
as recharge until the time it discharges to a
stream. Within any stream reach there can be a
wide range of ages in discharging groundwater.
Modica et al . ( 1997 , 1998 ), using a groundwater
flow model and chlorofluorocarbon analyses of
water, determined that groundwater dischar-
ging along a cross section of the Cohansey River
in New Jersey ranged in age from about 3 to 29
years. The oldest water was near the center of
the channel, and ages decreased with increas-
ing distance from the center.
Flow duration and hydrograph analysis
techniques are usually applied with mean
daily stream discharge for the entire period of
record, but the techniques also can be applied
over specific time intervals to examine trends.
Callow and Smettem ( 2007 ) found that replace-
ment of native vegetation with dry-land crops
produced a marked shift in the flow duration
curves for streams in southwestern Australia,
indicating increased recharge and base flow.
Gebert et al . ( 2007 ) determined annual base-
flow estimates (using hydrograph separation)
for streams in Wisconsin and found a trend of
increasing base flow with increasing agricul-
tural land use. Risser et al . ( 2005a ) applied the
recession-curve displacement method sepa-
rately to each month for the period of record for
197 stream gauge sites in Pennsylvania ( Figure
4.13 ) and determined that 80% of recharge to
these streams occurred between the months of
November and May.
Numerous studies have compared results
from different streamflow hydrograph ana-
lysis methods and found that the methods
described herein estimate base flow within a
range of about 25%, with the recession curve
displacement method providing a high esti-
mate and the local minima method within the
computer program HYSEP (Slotto and Crouse,
1996 ) providing a low estimate (Daniel and
Harned, 1998 ; Risser et al ., 2005a ). Kinzelbach
et al . ( 2002 ) determined that hydrograph sep-
aration in arid and semiarid regions estimates
recharge within a factor of 2 of actual ground-
water recharge. Stewart et al . ( 2007 ) determined
a similar accuracy for hydrograph analysis
methods for 10 sites in humid and subhumid
regions. Application of the Rorabaugh ( 1964 )
method requires more effort than application of
most other hydrograph analysis methods, and
the results may be no more accurate because
real aquifers are often quite different from the
hypothetical aquifer described by Rorabaugh
( 1964 ). Once the watersheds and stream gauges
of interest have been selected and the stream-
flow data assembled, it is a relatively easy matter
to apply several hydrograph analysis methods.
Comparisons of results from different methods
will provide insight on the uncertainty associ-
ated with the methods.
The wealth of available data on streamflow
and surface-water quality ( Table 2.1 ) allows
methods such as streamflow duration and
hydrograph analyses to be applied without the
need to collect new data. Streamflow statistics,
including duration curves and values of base-
flow index, are available for US Geological
Survey stream gauging sites (http://streamstats.
usgs.gov; accessed February 11, 2009). For meth-
ods that require data collection, streams and
lakes are much more accessible than aquifers
for measuring water levels and flow rates and
for collecting water samples.
Computer programs for performing tasks
such as surface-water hydrograph analysis
and flow-duration analysis provide easy-to-use
approaches that permit rapid analysis of many
years of streamflow data, but the programs
also make it easy to overlook poor data and
to ignore important assumptions. Sloto and
Crouse ( 1996 ) suggested plotting and visually
inspecting all data to avoid inclusion of invalid
data in base-flow calculations. Assumptions
for relating base-flow estimates to estimates
of recharge are discussed in Section 4.1.2 .
Search WWH ::




Custom Search