ed site investigation: Evaluation of white phosphorus
prevent poisoning of waterfowl in Eagle River
contamination and potential treatability at Eagle River
Flats, Alaska. In Interagency expanded site investi-
Flats, Alaska (C.H. Racine and D. Cate, Ed.).
gation: Evaluation of white phosphorus contamination
and potential treatability at Eagle River Flats, Alaska
CRREL Contract Report to U.S. Army Garrison,
(C.H. Racine and D. Cate, Ed.). CRREL Contract
Alaska, Directorate of Public Works, FY93 Final
Report to U.S. Army, Alaska, Directorate of Public
Report, p. 2584.
Works, FY94 Final Report, p. 445470.
Eagle River Flats is an estuarine salt marsh that
The objective of the study was to evaluate the
is actively undergoing progressive and signifi-
potential of four geosynthetic barriers to prevent
cant changes in the physical environment. These
the movement of WP from beneath the barrier to
changes result from the interaction and response
above it. Emphasis was on the retention of parti-
of multiple physical processes to various factors,
cles larger than 0.15 or 0.25 mm in diameter. A
including external forces such as a high tidal
field test in Area C Pond involved the establish-
range, a glacial river, two relatively large sedi-
ment of five test circles (A, B, C, D, and a control
ment sources, and a cold climate. The location of
without any geosynthetic), 2.4 m in diameter,
ERF within an active earthquake zone adds fur-
constructed with a clear plastic Lexan barricade
ther to the complexity and the potential for future
(to prevent mixing of water on the inside and out-
rapid, probably unpredictable, physical changes.
side of the barrier). The bottoms of three circular
In addition, ground explosions and cratering dur-
ing the use of ERF as a military firing range since
plots were covered by a geocomposite (consisting
the early 1940s have caused physical changes to the
of a needle-punched polyester geotextile approx-
imately 540 g/m2, with an apparent opening size
terrain, hydrology, and surface drainage. The
of 0.149 mm) overlain by a drainage matrix mate-
inherent complexity of this dynamic environ-
rial (geomesh) consisting of an nylon entangled
ment makes it extremely difficult to predict if and
mesh. The geotextile had 8-cm-diameter holes cut
how WP particles in the sediments might be
into it, 0.3 m on center, to allow for the venting of
transported and what effects potential remedial
gas. In one of these three test cells, an 8-cm-thick
measures will have on the physical system and
bentonite layer, provided by the Denver Wildlife
conversely what short- and long-term effects the
Research Center, was placed on top of the geo-
physical system will have on proposed remedial
composite. A 10-cm gravel fill layer, provided by
measures. Understanding both the system's
Ft. Richardson, was placed on top of the geosyn-
response and the effects of such remedial meas-
thetic in a second circle. No gravel or bentonite
ures are critical to deciding on cleanup actions.
layer was placed over the third test cell geocom-
Sedimentation rates in ponds and mudflats are
posite. A biodegradable woven coir (a fiber made
relatively high, ranging from l0 to 15 mm on
from coconut husks) mat was used to cover the
mudflats and 20 to 40 mm in ponds. The primary
ground surface of the fourth test cell.
source of this sediment is tidal inundation, which
We disturbed the barriers by vigorously stir-
carries sediments from Knik Arm (Cook Inlet) or
ring the water with a canoe paddle and by drop-
Eagle River or both. The timing and height of
ping a mass onto it to simulate a moose walk. The
responses measured were amounts of sediment
urements in the Bread Truck and C Pond areas) is
resuspended during the test and percentage of
controlled mainly by tides but enhanced by the
the resuspended sediment particles that were
discharge of the Eagle River (whose discharge is
larger than 0.15 mm and that were larger than
controlled in turn by glacial melt and precipita-
0.25 mm. The results showed that the barriers
tion). Preliminary analyses suggests that sedi-
reduced the amount of sediment resuspended by
mentation in the coastal two-thirds of ERF is tid-
about 30% or more. The sediment that moved
ally dominated, whereas the inland one-third is
across the barriers contained less than 3% by
apparently river-dominated.
weight of particles larger than 0.15 mm in diame-
While sedimentation rates are high, erosion
ter. The barrier consisting of geocomposite cov-
and recession of the headwalls and lateral walls
ered with 10 cm of gravel was most effective.
of gullies are also relatively high. Gullies are
therefore progressively extending into the mud-
None of the barriers tested were damaged by
flats toward the Bread-TruckC-Pond complex.
loading tests that simulated a moose walking.
Headwalls and adjacent lateral walls receded at
Lawson, D.E., S.R. Bigl, and J.H. Bodette (1994)
rates ranging from 0.1 to 4.9 m during the sum-
Physical system dynamics. In Interagency expand-
mer of 1992, 0.4 to 6.3 m during the winter of
46
to contents
Previous Page
