Summary 1992
collected in ERF but in none of five wild
In Eagle River Flats there are a large number
ducks collected from Susitna Flats, a similar
salt marsh over 30 km distant from ERF.
(75100) of small to large open-water ponds
The behavior of laboratory ducks dosed
where waterfowl feed; these cover a total area of
with white phosphorus was consistent with
about 70 ha (175 acres), or 8% of ERF. Based on
that of sick ducks in ERF, while the behavior
our sampling and white phosphorus (WP) analy-
of laboratory ducks dosed with 2,4-DNT
sis of over 1000 sediment samples from these
was not similar to that observed in the field.
ponds, the major contamination and hypothe-
The concentrations and distribution of
sized source of most of the waterfowl poisonings
white phosphorus in duck tissues was simi-
in ERF is in two ponded areas (Area C and the
lar in both laboratory ducks dosed with
Bread Truck Pond), covering areas of about 18 ha
white phosphorus and in dead ducks col-
(45 acres) and 6.5 ha (16 acres), respectively. Sedi-
lected in ERF.
ment cores 2030 cm deep from the bottom of the
High methemoglobin blood levels, a symp-
contaminated ponds showed that WP can be bur-
tom of 2,4-DNT poisoning known and seen
ied in the sediments to depths of 30 cm (1 ft) and
in the laboratory toxicity studies and indi-
cated by a very brown coloration of the
sediments adjacent to the contaminated Area C
blood, was not obvious in any of the ERF
and Bread Truck Ponds was found in 6 out of 56
ducks autopsied in the field. All blood of
samples.
field-autopsied ducks was bright red.
A simple and repeatable index of waterfowl
In ERF, 2,4-DNT was found only in a tall
mortality by which to evaluate future remedia-
sedge marsh, a habitat not used by feeding
tion efforts was developed and applied over the
ducks, whereas limited sampling for white
past three migration periods using permanent
phosphorus showed occurrence in a shal-
transects through the ponds and in the adjacent
low pond used intensively by waterfowl.
woodlands. No decline in mortality levels were
Since white phosphorus is extremely reactive
detected from August 1991 to August 1992.
and quickly oxidized or burned in the presence of
oxygen, various scenarios describing a mechan-
particles of a wide range of sizes from less than
ism by which unburned white phosphorus entered
0.15 mm up to 3.5 mm. Other abundant items of
similar sizes on which waterfowl feed in the sedi-
and was stored in the salt marsh sediments are
presented. Although firing into ERF ceased in
ments include insect larvae and seeds. Moreover,
February 1990, waterfowl continued to die dur-
each sediment sample obtained from different
ing the spring and fall, indicating that white
pond locations appears to have a unique abun-
phosphorus is persistent. Frequent flooding,
dance and range of WP particle sizes. In the labora-
combined with heavy suspended clay sediments,
tory, disturbance of contaminated sediments con-
results in burial of white phosphorus under
taining large numbers of these smaller particles
anaerobic conditions conducive to long-term
(<0.15 mm) can result in their suspension in the
storage. Measured redox potentials in salt marsh
water column for up to 2 hours. Surrogate sam-
pling suggests that such suspended WP particles
sediments were particularly low (< 200 mV) in
may be transported out of ponds into distributary
the bottoms of ponds where the dabbling ducks
channels connecting with the Eagle River.
feed, confirming the anaerobic nature of the sedi-
ments. Based on this information, we feel that the
Sedimentation in the summer of 1992 (28 May
munition white phosphorus is the major cause of
22 Sept) ranged from 4 to 30 mm, with the highest
mortality in ERF.
in the ponds (2530 mm) and the lowest on the
levees adjacent to Eagle River. Sediments origi-
nate from the Eagle River and from tidal influx
Racine, C.H., M.E. Walsh, C.M. Collins, D. Law-
from Knik Arm. Resuspension and redeposition
son, K. Henry, L. Reitsma, B. Steele, R. Harris,
and S.T. Bird (1993) White phosphorus contami-
of pond bottom sediments occurred at a rate of
nation of salt marsh sediments at Eagle River Flats,
about 15 mm over the summer season. Processes
Alaska. Part II. Remedial investigation report.
included wind waves and sediment-feeding birds.
CRREL Contract Report to the U.S. Army Envi-
Headwalls of gullies are actively eroding by
ronmental Center, Aberdeen Proving Ground,
undermining and collapse; recession rates ranged
Maryland AEC Report No. ENAEC-IR-CR-93063.
from about 0.5 to 2.0 m.
57
to contents
to index
Previous Page
