phosphorus particles (1.8-mm diameter, 5.6 mg)
worth detailing did occur, however. The first con-
that were made in the laboratory (Walsh et al.
cerned the switch logic. As configured on all sys-
1996). Each particle was first inserted into a plug
tems except the original, the pump can restart
of saturated sediment, then the plug of sediment
during the cool-down cycle if the genset warm-
was placed in a nylon stocking, which was then
up timer is set lower than the cool-down timer.
placed within the top 5 cm of saturated sediment
The immediate fix was to extend the warm-up
at each monitoring station. At the end of the sea-
time and decrease the cool-down time to work
son, we recovered five of the plugs to determine if
around the timing problem. This didn't resolve
white phosphorus mass had decreased. The
the underlying logic problem, so the circuitry was
remaining plugs were left for multiyear monitor-
redesigned to eliminate the problem altogether.
ing. To determine if the white phosphorus parti-
The fix actually simplified the circuitry and elimi-
cles had changed, we placed the sediment samples
nated a relay. All five systems were modified,
containing particles into isooctane to extract white
tested, and made operable within three days.
phosphorus residue prior to analysis by gas
The second problem involved a substantial
chromatography.
fuel spill that occurred in late June. A fuel filter on
one of the units burst over a weekend, resulting
in a spill of approximately 750 L. Fortunately, this
Feasibility study of large-scale
occurred on the one shore-based genset (pump-
deployment (1998)
Results obtained from the initial deployment of
ing system 3), so although the gravel pad beneath
the 126-L/s pumping system in Pond 183 indicat-
the unit was contaminated, no fuel entered ERF.
ed that the enhancement of natural attenuation of
The probable cause of the failure is a manufactur-
white phosphorus from ponded areas using a
ing defect in the case of the filter, although the
pump was highly effective and feasible (see
manufacturer disputes this.
Results). Prior to the cessation of field activities in
Deployment of all systems went smoothly. All
September of 1996, two more sumps were blasted,
gensets and pumps, with the exception of system
one in Area A (Pond 290), and an additional one in
3, were sling loaded into place by an Alaska Na-
Area C (Pond 155). These ponds were chosen to
tional Guard (AKANG) UH-60L Blackhawk heli-
further the feasibility study initiated in 1997. The
copter. The pipe for these systems was carried by
Pond 155 deployment was planned to study the
a commercial UH-1H Huey helicopter. Pipe for
effects of pumping in adjacent, partially connected
pumping system 3 was deployed along the edge
pond systems. Pond 290 was targeted to work out
of the EOD pad, emptying into Eagle River near
the logistics of a system deployment in a remote
the southern end of the pad. In addition to the
area of ERF, across the river from the support area
gensets and pumps, two 1,900-L double-walled
on the EOD pad.
fuel tanks were used to ferry fuel from a refueling
In the spring of 1998, the decision was made to
point on the EOD pad to the remotely deployed
attempt to deploy the two pumping systems on
gensets. The tanks are equipped with 12-V elec-
hand and the four systems on order. The areas the
tric transfer pumps that can be powered from the
pumps were to be deployed were Ponds 146, 155,
battery of the genset. Table 2 denotes the location
and 183 in Area C and Ponds 256, 258, and 290 in
of each system and some pertinent information
Area A. Because of the magnitude of the deploy-
regarding its deployment. Systems 1,3, 5, and 6
ment, the lack of site preparation for the three
additional systems, the need to troubleshoot the
Table 2. Pump system deployments (1998).
four new systems, and the lack of experience with
Discharge
remote deployments, the systems were to be fielded
System
Capacity
Operational Drawdown line length
in two stages. This allowed the gaining of experi-
number Pond
(L/s)*
date
date
(m)
ence with a remote system (Pond 290) and the
1
183
125
1 June
1 June
336
blasting of the two additional sumps in Area A
2
258
125
25 June
Unknown
258
(Ponds 256 and 258). The pump in Pond 146
1 June†
3
146
190
27 May
490
4
256
125
23 June
26 June
398
would be deployed in the existing dredge channel
5
290
125
28 May
29 May
185
at the edge of the EOD pad. No sump was formed
6
155
65
1 June
2 June
313
for the pump in Pond 146 due to time constraints.
* Theoretical capacity. Pump 3 can operate at 63-, 126-, or
Some problems developed with the systems
190-L/s levels.
over the course of testing and deployment, but
† Drawdown date determined by level where pumping in
most were quickly resolved. Two major problems
Ponds 183 and 155 is minimal.
13
Previous Page
