creeks). These relatively rapid accumulation rates
Bird, S.T., C.H. Racine, M.E. Walsh, C.M. Collins,
may bury the WP sufficiently so that, over time,
D.J. Calkins, B.D. Roebuck, L.W. Metker, and R.
exposure of feeding waterfowl will be reduced.
Newsome (1991) Waterfowl mortality in Eagle
River Flats, Alaska: The role of munitions com-
pounds and human health risk assessment. In Pro-
Bigl, S.R., D.E. Lawson, L.E. Hunter, B.M.
Nadeau, P.B. Weyrick, and J.H. Bodette (1995)
ceedings, Caribbean Haztech International Conference
Tidal flooding dynamics, sedimentation, and im-
and Exhibition, 1315 November, San Juan, Puerto
plications to site remediation, Eagle River Flats,
Rico, p. 1A1.
Ft. Richardson, Alaska. In Proceedings, Geological
Since August 1982, an estimated 10002000
Society of America, Fairbanks, Alaska, p. 6.
waterfowl deaths have been observed each year at
Eagle River Flats is an 865-ha subarctic estua-
Eagle River Flats, a 2500-acre estuarine salt marsh
rine salt marsh and impact area used by the U.S.
complex on Cook Inlet near Anchorage, Alaska.
Army-Alaska. High rates of mortality in migrat-
This salt marsh has been used since 1949 as the pri-
ing waterfowl during the 1980s and 1990s had led
mary impact area for artillery training by the mili-
to extensive environmental study. Investigations
tary at Fort Richardson. Eagle River Flats is inhab-
by CRREL identified that small particles (~1 mm)
ited by waterfowl primarily during spring and fall
migrations when the deaths have been observed.
of white phosphorus (WP), from artillery illumi-
In the spring of 1990, the Army investigated the
nation rounds, ingested by waterfowl feeding
possibility that munitions compounds fired into
from pond bottoms was the cause of observed
high mortality rates. High levels of WP contami-
the salt marsh were the cause of the mortality.
nation remain despite cessation of firing of WP-
Because no chemicals analyzed for showed ele-
vated levels, the Army initiated a physical gross
bearing munitions in February 1990.
analysis of several selected sediment samples. The
Investigations of physical system dynamics as
physical gross analysis was designed to investi-
part of the site assessment are to develop physical
gate and distinguish between naturally occurring
models of the depositional system and WP fate in
and foreign particles within the glacial sediments.
the system, and to evaluate system response to
In the summer of 1991, a sediment sample being
remediation. Sedimentary processes and system
prepared for visual inspection by the Army released
dynamics are important because they affect the
timing and distribution of tidal flooding and asso-
a white puff of smoke upon stirring. White smoke is
ciated pondmudflat sedimentation. Instrumen-
characteristic of white phosphorus (WP) munitions
tation located in gullies linking pondsmudflats
and immediately led the Army to consider WP as a
and the Eagle River document duration of inun-
possible cause of the waterfowl mortality and con-
dation (depth), river versus tidal domination
duct a search of available literature on the toxicity of
(salinity/temperature), and sediment transport
WP to waterfowl. At this point, WP became the
(optical backscatter and total suspended solids).
prime suspect of the waterfowl mortality. The WP
Sedimentation rates were also measured with var-
does not exist in nature since it spontaneously oxi-
dizes in the presence of oxygen to form phosphorus
Semidiurnal flooding during high tidal events
oxides. When applied to well-drained upland soils,
causes pondmudflat inundation and drives sur-
WP has been shown to oxidize and even provide a
face hydrology. Ponds receive sediment influx
source of phosphate to growing plants. However,
from riverine and tidal sources at accumulation
when smoke-producing munitions introduce WP
rates of 1030 mm/yr. These rapid rates may pro-
particles into a salt marsh environment with anaer-
vide a means for natural remediation of WP con-
obic sediments and standing water, incomplete
tamination while enhancing the efficiency of arti-
ble. The Army conducted both field and labora-
ods. Amplification of tidal height by fjord geome-
tory studies of waterfowl and sediments from Ea-
try results in a higher frequency of flooding
gle River Flats and made an assessment of the tox-
events than predicted from Anchorage tides. The
icity of WP in the laboratory. (This is a truncated
result is repeated flooding and retardation of nat-
version of the Introduction. The conference paper
ural WP decomposition that is favored by warm,
did not have an abstract.)
dry conditions. Further understanding of physical
system dynamics is required to evaluate the best
Collins, C.M. (1997) Pond draining to treat white
means for site remediation.
phosphorus-contaminated sediments at Eagle
River Flats, Alaska. In Proceedings, 5th International