Environmental Engineering Reference
In-Depth Information
to be 510 gal/day (1,927 L/day), researchers concluded that
up to 445 gal/day (1,682 L/day) of groundwater was used by
the trees.
Much like the method described in the first part of this
chapter, the total discharge of groundwater through a cross-
sectional area of the aquifer was determined by the
researchers using Darcy's Law to be 44 gal/day/ft of aquifer
thickness (166 L/day/m). To account for the estimated
445 gal/day (1,682 L/day) of groundwater uptake by the
trees, water from a 10-ft (3 m) thick section of the aquifer
would need to be tapped to supply the estimated demand.
The authors concluded that this demand of groundwater by
the planted trees at the site should have resulted in a measur-
able water-table depression. None, however, was observed
in the five monitoring wells. Ferro et al. (2001) suggest that
the lack of water-table depression may be due to the fact that
the groundwater flow rate was similar to the rate of ground-
water uptake by the trees. Hence there would be no change in
the water-table elevation, even though groundwater was
being taken up.
Trenches were dug at least 3 ft below grade to near the top of
the seasonal high water table, filled with peat and sand, and
nearly 1,000 hybrid poplar trees (
P. deltoides x P. nigra
cuttings) were planted on 6-ft (1.8 m) centers across
0.8 acre of the 1.2-acre area of interest. By May 1998, only
60% of the planted trees had survived (Ferro et al. 2000); no
explanation for the high mortality was provided. In early
1999, about 400 white willow trees (
S. alba
) were planted
in the trenched area but in separate boreholes created
with an auger, and only 5% mortality was observed. Other
tree species were subsequently added to replace dead
hybrid poplars. In the spring of 2002, the remaining
hybrid poplars had to be removed due to a canker infes-
tation by
Cryptodiaporthe populea
. Other native facultative
phreatophytes were used, including pin oak, sweet gum,
silver maple, river birch, tulip poplar, and eastern red
bud.
Monitoring of the pilot phytoremediation system by Ferro
et al. (2000) focused on the flow of groundwater through the
stand of trees and sap-flow rates. Sap flow was determined
between 5 and 7 times during May and September 2000 and
2003 using the heat-balance method. Sap-flow rates were
normalized by the cross-sectional area of the sapwood of
each tree measured and ranged from 6 to 16 gal/day/tree
(26.6-71.9 L/day/tree). The highest sap-flow rate was
measured in 2001 when precipitation was the lowest during
the sampling period. The fact that
ET
P
exceeded precipita-
tion at that time suggests that much of the sap flow consisted
of groundwater. However, in 2003, the highest precipitation
was recorded during the sampling period at the site, near
31 in. (80 cm), and exceeded
ET
P
such that recharge
occurred, and sap-flow measurements probably reflect a
mix of water from all sources. When these values were
used by Ferro et al. (2000) to extend individual sap-flow
rates to the entire planted area, the mean water use ranged
from 2.1 to 8.0 gal/min (7.9 to 30 L/min).
Estimates were made using Eq.
8.4
to determine how
much groundwater would be used by the trees. Estimates
ranged from less than 1 gal/min (3.78 L/min) for the second
year to almost 10 gal/min (37.8 gal/min) by the fifth year. If
this rate is superimposed upon the average mechanical pump
rate of 19 gal/min (71 L/min) attained by the pump-and-treat
system, the pump-and-treat could be cut back to near 10 gal/
min (37.8 gal/min) during the summer, as the removal by
trees would account for the difference. Of course, Ferro et al.
(2000) state that the pump rate would have to be increased
during the winter to offset the reduction in transpiration by
the dormant plants (Figs.
8.7
and
8.8
).
Because groundwater-level data were not provided in
Ferro et al. (2000), it is not clear if the water-table level
decreased due to the mechanical or the planted system.
This type of information would be useful, because it is
unclear
8.4.1.2 Case Study: Superfund Site, Connecticut
The approach taken at this site was the same as described in
8.4.1.1
. The site is located near Southington, Connecticut.
The facility was used between 1955 and 1991 to recycle used
industrial solvents. Depth to groundwater is between 4 and
5 ft (1.2-1.5 m) below land surface. Groundwater at this site
was affected by volatile organic compounds (VOCs), such as
chlorinated solvents, present in both the dissolved phase and
DNAPL phase. In 1983, the site was listed as a Superfund
site by the USEPA. The contamination is found throughout
the aquifer thickness to bedrock at 30 ft (9.1 m) (Ferro et al.
2000).
To contain the DNAPL source and prevent additional
downgradient flow of the dissolved-phase contaminants to
offsite areas that include the Quinnipiac River, two tradi-
tional groundwater containment control structures had been
used: a sheet-pile cutoff wall and a pump-and-treat system.
The sheet-pile wall was constructed in a 700-ft (213 m) long
section roughly perpendicular to groundwater flow, and
placed up to 30 ft (9.1 m) deep through glacial overburden
to bedrock. The pump-and-treat system consisted of 12
recovery wells, placed on the upgradient side of the sheet-
pile wall. The average pumping rate for all 12 wells com-
bined was less than 20 gal/min (75 L/min). This rate reflects
the low hydraulic conductivity and low specific yield of the
glacial overburden at the site. The pumped groundwater that
contained VOCs was treated onsite by ultraviolet oxidation.
In order to determine if the amount of groundwater
pumped by the wells could be enhanced or replaced by
groundwater removed by the trees, a series of pilot plantings
was performed in a 1.2-acre (4,856 m
2
) part of the site. This
initial phytoremediation system was installed in early 1998.
if well-and-plant
interference would present a
Search WWH ::
Custom Search