Environmental Engineering Reference
In-Depth Information
Hybrid poplar trees (
Populus deltoides
; poplars are native
to this part of Texas) were installed in early 1996 in two
areas using different approaches. Each planted area was
rectangular in shape, 49 by 246 ft (15 by 75 m), and located
perpendicular to the generalized direction of groundwater
flow. The upgradient planting used 440 whips, or vegetative
cuttings from mature trees of eastern cottonwood, and
consisted of clones from local trees. The initial mean diam-
eter of these whips at 3.9 in. (10 cm) above ground surface
was 1.1 in. (2.8 cm) (Vose et al. 2000). The downgradient
planting used 224, 1-year-old seedlings of eastern cotton-
wood grown at a local nursery. The initial mean diameter
was 1.8 in. (4.6 cm). Rather than installing separate holes for
each tree, linear trenches were dug to the depth of 3.2 ft
(1 m), 7.8 ft (2.4 m) apart. This design facilitated the instal-
lation of a drip-irrigation system in the trenches. The drip
irrigation system was used on alternating days following
planting, due to drought conditions. It was determined that
tree roots reached the water table during the end of the
second growing season (1997).
The amount of water transpired by the trees was
estimated using sap-flow measurements. Between May and
October 1997, up to 14 trees were measured, including
recently planted whips as well as the 1-year-old seedlings.
Transpiration ranged from 0.42 to 2.4 gal/day/tree (1.6 to
9.2 L/day/tree) for the whips, and from 0.2 to 3.9 gal/day/
tree (0.92 to 15 L/day/tree) in the 1-year-old trees (Vose
et al. 2000).
Groundwater levels were monitored in all the wells
installed at the site. The maximum decrease in groundwater
level measured was 3.9 in. (10 cm), and this was in a well
located between the two plantings. The change in ground-
water flux due to the removal of groundwater by trees was
estimated for each year after planting from 1996 to 1999.
The change ranged from 2% to 12% less than when trees
were just planted and no uptake of groundwater was
assumed. As can be imagined from the reported decrease
in groundwater flux of no greater than 12% from preplanted
conditions, groundwater continued to discharge to the creek.
Even in such a thin aquifer, not all groundwater flowlines
were being diverted upward as a result of groundwater
uptake by the trees.
8.4.3.1 Case Study: Areas 317 and 319, Argonne
National Laboratory, Illinois
A phytoremediation project was installed at a contaminated
site located at the Argonne National Laboratory, near
Chicago, Illinois, in mid-1999. The site consists of two
contaminated areas, the East Argonne Areas 317 and 319,
totaling about 3.2 acres (12,950 m
2
). The contamination
resulted from the disposal of VOCs and tritiated water,
respectively, in french drains used to dispose of the wastes.
The french drains are no longer used, but the sites remain
active facilities for waste processing and storage. The
phytoremediation system designed for the sites was to (1)
degrade the contaminants in groundwater (discussed in
Chap. 13) and (2) provide hydrologic containment of the
contaminant plumes such that groundwater flow across
the downgradient property boundary could be decreased. The
phytoremediation project was funded by the Department of
Energy Accelerated Site Technology Development (ASTD)
Program. After 1999, the U.S. Environmental Protection
Agency SITE Program became involved. More information
on the history of this project is given in Quinn et al. (2001).
The hydrogeologic setting of the disposal areas consists
of multiple aquifers and confining units, reflecting the gla-
cial history at the site. These layers of sediments are of
widely different permeability and hydraulic conductivity,
from low-permeability tills consisting of silts and clays to
high-permeability sands and gravels. The aquifers systems
are not of uniform thickness or of great lateral extent, and
interconnection between separate sand layers occurs.
Groundwater from these areas flows offsite to discharge in
ravines in a downgradient forest preserve and ultimately to
the Des Plaines River. Recharge to the aquifer system is by
leakage from overlying units of high permeability or directly
by precipitation. The hydraulic conductivity at the sites
was determined from pump and slug tests, and averages
8.8 ft/day (2.6 m/day), but with considerable variation over
relatively short distances.
Unlike the other case study sites discussed so far, because
the contaminants were introduced to the subsurface through
french drains dug into the overlying shallower aquifer and
confining unit, the groundwater contamination was present
in a confined aquifer at depths between 25 and 30 ft
(7.6-9.1 m) below land surface. The contaminated confined
aquifer of interest ranges from 3 to 10 ft (0.9-3 m) thick.
Conventional planting approaches as have been previously
described could not be used. Rather, a deep-rooting method
called TreeWell
8.4.3 Numerical Model Framework
This approach is based on numerical groundwater-flow
models that simulate the various parts of the water budget
at a contaminated site. For the model to function, a mass
balance of water is calculated for each time step of the
model. The simulation of plants and groundwater
interactions in such groundwater-flow models is at best an
approximation, however, as is discussed in Chap. 14.
,
Applied Natural Sciences, Inc.) was employed to get the
roots into the contaminated confined aquifer. In brief, a
deep borehole is created to the appropriate depth through
overlying aquifer and confining unit materials. A caisson is
added to keep the hole from collapsing, as well as to limit or
constrict lateral root formation, and materials having greater
Treatment System (or TreeMediation
™
®
Search WWH ::
Custom Search