I
I
I
I
I
I
DISPERSANT UTILIZATION REPORT
DREDGE BARGE PENNSYLVANIA OIL SPILL
ROCKAWAY JETTY, NEW YORK
7/31/78-8/14/78 tfk
V3>
\ '
EPA
902/
' 1978.1
BY
U.S. Environmental Protection Agency
Emergency Response & Inspection Branch
Region II
and
Industrial Environmental Research Laboratory - Ci
Oil & Hazardous Materials Spills Branch
Edison, New Jersey 08817
-------
DISCLAIMER
The mention of trade names or commercial products within this report
does not constitute endorsement or recommendation for use by the
U.S. Environmental Protection Agency.
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
CONTENTS
I
I
I
I
I
I Purpose of Report
II Conclusion
III Background
I 1. Spill Details
2. Cleanup Progress and Difficulties
3. Decision to Use nispersant
IV Dispersant Application
1. Choice of Dispersant
I 2. Initial Set Up
3. Dispersant Application
V Dispersant Efficiency
VI Environmental Impact
VII Figures and Tables
I VIII Photographs
IX References
I
I
DO
-------
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I. PURPOSE OF REPORT
The material contained herein documents the first Federally authorized
use of an "EPA-accepted" dispersant during an actual oil spill in the
United States.* This information is presented with the recognition of
the interest which others have in the events described. This document
also serves as the documentation which is required under Annex X,
Section 2003.2-3, of the National Contingency Plan (40 CFR 1510).
II. CONCLUSION
The ecological impact of dispersant application in this incident is
estimated to have been negligible due to the relatively small amounts of
oil treated, the large area of treatment, the small total amount and low
concentration of dispersant used, and the dynamic physical processes
present to dilute the dispersant and dispersed oil. Total organic
carbon measurements of water column samples supported this conclusion.
Attempts to accurately record environmental impact where small amounts
of dispersant are applied (in this case TOO gallons or less per day over
five days) are likely to be futile, and therefore may warrant exemption
from the requirement to document environmental impact.
The beach cleanup operation was hastened by preventing additional oil
slick from impacting beaches once they were cleaned. This promoted a
more prompt re-opening of the New York City beaches affected.
Mo standard technique for dispersant detection in estuarine waters was
available to provide the necessary data on residual dispersant concentra-
tions. It is recommended that when a product for treating oil has its
data accepted, a testing, proven analytical technique for use in
estuarine waters be made available.
Standard methods for measuring environmental impact of dispersant remain
to be defined. Once defined substantial difficulty may yet be encountered
in applying the method on a real-time basis.
The Coast Guard 32 ft. Port Security boat is an ideal platform for applying
dispersants, and is preferable to the Coast Guard's 41 foot vessel.
Accurate regulation of dispersant dilution, rate of flow, and area of
application was attainable. The eductor system was easily modified to
incorporate the dispersant application/regulation system, and was readily
accessible for maintenance and repairs.
*NOTE: Previous EPA authorized uses of dispersants for oil pollution
control occurred prior to EPA's data acceptance program: Polycomplex
A-ll authorized by US EPA Region IX for use in the Santa Barbara Marina
during the 1960's blowout; Corexit used on the open ocean in the early
stages of the Santa Barbara blowout (authorized by Region IX), and
several years ago Region VIII authorized the use of dispersants in a
lagoon where migratory waterfowl were threatened.
-------
Disadvantages to application of dispersant by vessel included a limited
range of vision, limited vessel speed and range of the flow system (as
compared to a helicopter application), periodic system leakage and
pressure loss, the logistical difficulty of hoisting full dispersant
drums on board, and the lack of a proper drum storage rack allowing
drums to move on deck during heavy seas.
III. BACKGROUND
At 2:10 p.m. on July 31, 1978, the Dredge Barge PENNSYLVANIA, under tow
from the Tug Grace Moran, ran aground and sank at 40-32N, 73-56W,
approximately 1/3 mile southwest of the Rockaway Jetty off the coast
of Brooklyn, New York (see attached map, Figure 1). The barge was
carrying approximately 6,000 gallons of No. 2 fuel oil and 37,000 gallons
of No. 6 fuel oil. Shortly after sinking, oil began leaking from the
barge.
The barge was owned by the American Dredging Co., Staten Island, New York,
and they initially assumed financial responsibility for the oil spill
cleanup operation. The U.S. Coast Guard Captain of the Port (COTP), New
York office, assumed the role of On Scene Coordinator (OSC) and began to
monitor the cleanup operation. The spill cleanup operation was directed
from the U.S. Coast Guard Station Rockaway. The U.S. Coast Guard Atlantic
Strike Force was called in to assist the COTP office.
The Coast Guard's High Seas boom and also 30-inch Bennet boom was deployed
around the barge in an attempt to contain the mixture of No. 2 and No. 6
fuel oils, which were continuously leaking from the barge. However,
because of high tidal currents, the booming operation was ineffective.
During the booming operation, attempts were made by Atlantic Strike
Force divers to plug the holes which were allowing the oil to escape.
Some of the discharge points were plugged, but oil continued to leak
from the barge from unknown points.
By this time, several days had passed since the initial grounding and a
substantial quantity of oil was in Lower New York Bay, being washed back
and forth by the tidal cycle. Favorable winds had kept the oil from
impacting any beaches on New York or New Jersey. However, on August 2,
1978, the winds shifted to the south and oil began to be driven onto the
Coney Island and Rockaway beaches resulting in their being closed for
several days.
On August 2, 1978, Dr. Dewling, EPA Region II Deputy Regional Administrator,
spoke with Captain Fleishel (COTP) and expressed to him EPA's concern
about preventing further accumulations on the beaches. Dr. Dewling
suggested that dispersants be considered.
At the RRT meeting on August 3, 1978, CDR Mullen, USCG, the Acting
Regional Response Team (RRT) Chairman, requested that EPA further explore
the possibility of using dispersants.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Paul Elliot, the RRT representative, contacted Dr. Richard Dewling,
who in turn made the proper contacts and arrangements. Dr. Dewling
called Commissioner Berle's office (NYSDEC) and Commissioner O'Hearn's
office (NJDEP) and advised them of EPA's position. Since oil was not
impacting the New Jersey coast, and EPA aerial photographic overflights
indicated fairly light concentrations of oil off the New Jersey Coast,
Dr. Glenn Paulson, Deputy Commissioner, suggested we postpone any treatment
of the sheens along the New Jersey shore until actually needed. The
currents, which run parallel to the shore, prevented beach impact. Mr.
Russell Mt. Pleasant, Pure Water Division (Surveillance and Monitoring),
New York State Department of Environmental Conservation, did agree to
the use of dispersants to help protect the New York beaches from further
damage.
EPA then relayed to the U.S. Coast Guard their approval to utilize
a dispersing agent in deep water in the vicinity of the sunken barge
to attempt to disperse the oil which was continuing to leak from the
barge. This decision was approved by the RRT members present on
August 4, 1978 and dispersing operations began at 1:00 p.m. on August 4,
1978.
IV. DISPERSANT APPLICATION
1. Choice of Dispersant
Selection criteria for the dispersant was based on the following factors:
(a) It had to be one for which technical data had been accepted by
EPA. Background chemical and biological (bioassay) data had
to be available. The product selected had to be tested using
EPA procedures, and the data furnished to EPA prior to this
incident;
(b) Since a "high energy" delivery system was not available, the
product selected had to have a high surface active
agent content (40-50 percent). One of the new "self-energizing"
dispersants was desirable, particularly since the anticipated
use was for treating only "sheens" and "rainbows";
-------
(c) Last, but not least, and probably most important, the
product had to be available from the cleanup contractor,
Clean Venture, Inc. Since the revolving fund was not
being used, EPA or the U.S. Coast Guard had no authority to
purchase a chemical. Since the contractor had a
stockpile of COREXIT 9527 which met the conditions
outlined in (a) and (b) above, and since he could
make delivery by the next morning (August 4, 1978),
approval was given for this particular product.
Initial estimates indicated that between 5-10 barrels
of COREXIT 9527 would be needed and that amount was
delivered to the staging site during the early morning
hours of August 4;
2. Initial Set Up
It was initially decided to use the EPA Region II vessel, CLEAN WATERS,
as a platform to apply the dispersant. Personnel from the Oil and Hazardous
Materials Spills Branch, Industrial Environmental Research Laboratory-Ci,
and from Mason & Hanger, Environmental Emergency Response Unit (EERU)
operating contractor, located the necessary equipment to apply dispersants
from the CLEAN WATERS.
For the response, two systems were prepared. The primary unit consisted
of the modified Warren Springs Laboratory system which had been tested
at OHMSETT. This system consisted of a diesel (1% HP Lister model LT1)
driven Viking Rotary pump (model HL124) and 100 feet of IV fuel transport
hose fastened to two spray booms. These booms are each 15 feet of 2"
galvanized pipe with three Veejet spray nozzles (Spraying Systems Co.,
model No. H % V95100) spaced five feet apart. The dispersant concentrate
would be diluted by the use of IV eductor (Penberthy model LL 1^} and
an adjustable flowmeter (Dwyer 0-3.5 gpm water model RMC SSV) to vary
the concentration ratio. The total water/dispersant flow rate would be
measured by the installation of a flowmeter (RCM Industries model #2-71-
R-200-DF). This entire system would be capable of being mounted on the
EPA's CLEAN WATERS.
A second system was assembled for use with a hand-held fire hose. This
unit consisted of a 9 HP gasoline driven fire pump (Hale model #25FZZ-B23)
and a W eductor (Rockwood model #FWC) to dilute the concentrate. The
dispersant and water/dispersant flow rates would be measured by flow-
meter identical to ones used on the first system. This unit would
be used to apply dispersant to oil inaccessible to the spray booms
mounted on the boat. Hose and hand-held nozzle for this system would
be obtained from equipment available to the U.S. Coast Guard.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
A third pump (ITT Marlow model 2% - G15S) was carried in the event that
a second hand-held line was to be assembled. This pump also was to be
used in the event one of the others failed.
In all of the above systems, V' clear tygon tubing was to be used for
delivery of the dispersant to the eductor. This enabled the fluid to
be viewed for obstructions or air bubbles and to insure uninterrupted
flow.
However, mechanical difficulties with the CLEAN WATERS forced a change
in plans, and it was decided to utilize the U.S. Coast Guard 32 ft. Port
Security boat as the dispersing platform. This boat has a number of
design features which made it ideal for use. The boat has:
(1) A variable speed fire fighting pump which can supply water
through either a monitor located on top of the wheelhouse
or a fire hose.
(2) A fixed eductor system which was designed to educt foam
into the water stream but which was easily adaptable to
allow educting of dispersant into either the monitor or
the fire hose.
(3) Controls which allow the pump flow rate to be varied and,
when using the monitor, the type of water stream can be
varied from solid stream to fog.
(4) The monitor itself can be swiveled from a control handle
located from the wheelhouse.
(5) The boat has a flat deck aft of the wheelhouse which
provided an ideal storage location for the 55 gallon
drums of dispersant and an excellent working platform
to stage other equipment as needed.
The other available vessels were the Coast Guard 41 footer and, once
repaired, the EPA CLEAN HATERS. The USCG 41 footer has a fire monitor
system but one that had several disadvantages as compared to the USCG
32 footer.
(1) The 41 footer normally is driven by two screws with each
screw being driven by its own engine. However, to run
the fire monitor pump, one of the drive engines has to be
disconnected from its drive screw. This seriously limits
the maneuverability of the 41 footer during use of its
fire monitor.
-------
(2) It was reported to be more difficult to vary the fire
monitor pump speed.
The CLEAN WATERS was too large (65 feet) to work immediately adjacent to
the wreck area and also had no fire monitor system. This meant that a
complete system of spray booms and pumps would have to be used. Conse-
quently, it was decided to request the USCG 32 footer.
The Coast Guard approved the use of their 32 footer as a platform, the
necessary modifications in the eductor were made, and the dispersant
loaded on board. A flow meter was fitted in the hose going from the
dispersant drum to the eductor to allow the dispersant flow rate to be
easily adjusted as conditions warranted. Also, a reducing bushing
was installed in the eductor system. For a schematic showing, the
eductor system and where our hose was installed, refer to Figure No. 2
and Photo No. 8.
3. Dispersant Application
August 4, 1978 - The initial shakedown run began at 1:00 p.m. Initially,
the dispersant was applied using a fire hose. A hose operator stationed
at the bow manually swept the hose over an arc of 120 while the boat
slowly cruised through the oil slick. This operation continued for
approximately 45 minutes. We then began to use the monitor to deliver
the dispersant. This ^as physically much easier and also allowed a
much wider coverage as the flow rate was increased from 100 gpm to 250
gpm. After approximately 45 minutes of use with the monitor, we returned
to the base to discuss the results and establish a plan of operation for
the remainder of the spill. During the subsequent discussion, it was
decided to utilize two 32 footers and deliver the dispersant/water mixture
with the monitor. It was also decided not to work 24 hrs/day as poor
visibility at night would not allow effective application of the dis-
persant. Following this meeting, another run was made until dark. For
complete details concerning application rates and amounts, see Table
#1 and Table #2.
August 5, 1978 - During the day's operation, two boats were utilized.
As indicated earlier, the entire dispersing operation was conducted in
the vicinity of the sunken barge. Calibration of this dispersant flow
meter was performed during the afternoon. A five gallon opaque plastic
drum was filled with dispersant. With all lines leading to the eductor
and monitor full, a timed drainage of the five gallons was performed.
A calculation of the actual rate of flow versus that indicated by the
dispersant flow meter indicated that the flow meter reading was 440
percent higher than the actual flow rate. Since flow meter readings
during the entire dispersing operation varied from .3 gpm to 3.0 gpm,
actual flow rates were between .06 gpm and .68 gpm. On this day, the
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
sea swell was 3% to 4 ft. and it rained periodically. EPA and USCG
helicopters assisted in spotting the oil slicks. The angle of vision
from the vessels made it difficult to locate the major concentrations
of oil. However, once in a slick, it was fairly easy to stay within
the slick.
August 6,1978 - Again, two boats operated. Very little oil was sighted
during the initial dispersant run so only one boat operated during this
period. However, during the afternoon, a large slick was sighted
coming from the Rockaways towards the barge. Both boats operated in
this slick for several hours. This slick contained black oil and
heavy rainbow sheen. The weather was very foggy and hazy with
occasional rain showers. The sea swell was down to 2-3 feet. Towards
the end of the day, little oil was sighted coining from the barge.
August 7. 1978 - Only one boat operated. The main job attempted was
to apply dispersant (using divers and 100 feet of fire hose) directly
into the oil pools trapped in between girders and in voids aboard the
sunken barge. The rate of dispersant application was increased because
of the thickness and size of the oil pockets located onboard the barge.
This operation involved maneuvering the 32 foot vessel inside the boom
and within 15-20 feet of the barge. The operation continued for about
1.75 hours. From reports from the Atlantic Strike Force divers, this
appeared to work well. The oil was dispersed and then forced out of
the barge using the venturi effect of the hose discharging through an
open port hole. Following this operation, a normal dispersing opera-
tion was conducted using the monitor. Again, very little oil was
coming off of the barge by the end of the afternoon. The weather
was warm and sunny with sea swell averaging between 2-3 feet.
August 8, 1978 - Again, only one boat operated. We repeated the dis-
persing operation inside the barge. However, there was little oil
inside the barge that would be found by the divers. Again, following
the operation inside the barge, we conducted a normal dispersing
operation on a small patch of black oil before going back in. The
weather was again warm and sunny.
August 14, 1978 - The Coast Guard requested that a final dispersing
operation be conducted on the area inside the boom prior to removing
the boom. After inspecting the area by boat and by USCG helicopter,
it was decided that there was not sufficient oil to warrant the applica-
tion of dispersants.
V. DISPERSANT EFFICIENCY
During the dispersant application, an attempt was made to estimate the
efficiency of the dispersing agent in dispersing the oil. The slick
-------
was a mixture of No. 2 fuel and No. 6 oil and, where it coalesced in
black oil, it had the consistency of a No. 4 oil. It spread out into
a thin black slick and did not form the heavy, tarry patches normally
associated with a No. 6 fuel oil spill. Consequently, the oil was
ideal for dispersing in terms of thickness and consistency.
Because the slick to be dispersed consisted mainly of sheen inter-
spersed with black oi", and covered a large area, we needed to dilute
the dispersant a great deal to allow efficient coverage of the thin
slicks. COREXIT 9527 was visually effective at dilutions which averaged
approximately one thousand-fold.
The best estimate is that we treated between 2,000 and 3,000 gallons of
oil in the course of our surface operations. No estimate was available
for the amount of oil treated within the barge during the underwater
operations. A total of 288 gallons of COREXIT 9527 was used during
both surface and subsurface operations.
VI. ENVIRONMENTAL IMPACT
It was quickly recognized that the chances of recording the environmental
impact of this dispersant application were minimal due to the circumstances
involved, and the absence of a standard method for measuring the
environmental impact of dispersant application. Consider the following
theoretical calculations.
The appearance of the oil in the water at the time that dispersant was
applied ranged from a silvery sheen (0.08 gallons/acre) to bright bands
of color (0.32 gallons/acre) to thin black"slicks (2 gallons/acre)(l).
The primary area of dispersant application was roughly a circle with its
center at the barge and a diameter of about 0.75 miles (an area of 283
acres). Estimating 10 percent coverage of the primary area of dispersant
application with silvery sheen, 35 percent with bright bands of color,
10 percent coverage with black slicks, and 45 percent having no
sheen as the conditions during which dispersant was applied on the
surface, we estimate the amount of oil present at a point in time in the
primary application area to have been 90 gallons. This serves as a
frame of reference for further calculations. No actual measurements are
available regarding the rate at which oil was being released or the
amount of oil treated with dispersant. The dispersant application
occurred four days afi:er the barge sank, and observations indicate that
by that time the bulk of the 37,000 gallons of No. 6 oil and 6,000 gallons
of the No. 2 fuel oil had been released.
Acute toxicity data for COREXIT 9527 is reported as follows:
Species LC (ppm)
Fathead minnow
Mummichog
Brine shrimp
201 - 96 hr.
92 - 96 hr.
40 - 48 hr.
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
Acute toxicity data for COREXIT 9527 and No. 2 oil in a one to ten ratio
are reported as follows:
Species LC5Q (ppm)
Fathead minnow 58 - 96 hr.
flummichog 22 - 48 hr.
Brine shrimp 38 - 48 hr.
Note that no acute toxicity data were available for crustacean and mollusk
larvae, or icthyoplankton, which together may be considered to be the
key populations to monitor during dispersant applications.
I
I
I
I
I
I
If we assume that in any one day involved a maximum of 400 gallons of
I oil was treated in a 100 percent effective manner with 100 gallons of
dispersant in the primary area of dispersant application, and that
homogenous mixing occurred only in the top meter of water (sea chop
_ varied from 2 to 4 feet), then the concentration of additional oil
placed into the water column by dispersant is calculated to have been
1.4 ppm (400 gal. * 849 acre feet) and the concentration of dispersant
would have been 0.35 ppm (100 gal. ? 849 acre feet). Let us be more
I conservative in calculating an average concentration in water, and assume
that 400 gallons of oil was treated but that only one-fourth of the dilution
volume was actually available (because only a part of the primary dispersant
I application area was involved and/or there was a smaller depth of
homogeneous mixing). Concentrations are then calculated to have been
5.6 ppm for oil and 1.4 ppm for dispersant. In fact, however, dispersants
are not 100 percent effective in dissolving oil which further indicates
that a concentration of 5.6 ppm of oil may be conservatively large. The
9 calculated concentration of dispersant in the water column is 29 times
less than COREXIT 9527's TL50 for the most sensitive specie tested
I (brine shrimp, 48 hrs.). The calculated concentration of dispersant
in the water column is 16 times less than COREXIT 9527's TL5Q for the most
sensitive specie tested with a one to ten ratio of dispersant to oil
(Mummichog, 48 hours).
of 285U true north, and 1.4 knots at the ebb at a direction of 140" true
Currents at the site are typically 1.2 knots on the flood, at a direction
_.C Or*l~Ğ J. ._ J_l_ I 1 t\ I, Ğ .^ ^ Ğ . J. J. U n A U k -s-t *\ s>J^uĞn.n^-!nv-i **.£ T Af\ ^ V*l Jrğ ^H
I north(2). Depths in the area range from 9 to 31 feet(3). Tidal range
from August 4 through August 8 was predicted at as much as 5.2 feet(4).
The receiving waters involved here can readily be characterized as dynamic,
I involving as much as 479 million gallons of water added to the primary
area of dispersant application by tidal flux alone. Currents could be
expected to transport waters more than 1 nautical mile out of the primary
- dispersant application area in one hour.
I
9
I
I
-------
A total of 288 gallons of COREXIT 9527 was applied over a period of five
days in connection with this incident. The maximum used in any one day
involved was approximately 100 gallons (on August 5, 1978). The mean
concentration of dispersant as applied at the eductor (weighted by
volume of various concentrations applied - see Tables No. 3 and No. 4)
is calculated to have been 800 ppm.
It was thus estimated that due to the small total amount of dispersant
used, the extended period and dilute concentrations in which it was
applied, the relatively small amount of oil treated, and the dynamic
coastal water movements involved, efforts to evaluate the impact of the
dispersant application would not likely be fruitful. Nevertheless,
limited efforts were mace to measure the environmental impact of the
dispersant application in an attempt to comply with the National Contingency
Plan. What follows should not be misconstrued to be the result of
planned research into the impact of dispersants. It should further be
recognized that the procedures used here are not necessarily recommended
for use in future attempts to document the environmental impact of
dispersants. Standard methods for measuring environmental impact of
dispersant remain to be defined.
Chemical Analyses
Environmental sampling was performed using a Huey helicopter on loan
from the U.S. Army. It was equipped with pontoons for ocean landings,
if necessary, and had an opening in the cabin floor through which sampling
could be accomplished by U.S. EPA Region II Surveillance and Monitoring
Branch technical staff. The following grab samples for dispersant analysis
were obtained near the surface of the water using a Kemmerer sampler
(see Figure 3).
SAMPLE
52881
52882
52883
52884
52885
52886
52887
52888
52889
52890
52891
52892
52893
DATE
8/5/78
8/5/78
8/5/78
8/5/78
8/5/78
8/5/78
3/5/78
8/5/78
8/5/78
8/5/78
8/5/78
8/5/78
8/5/78
SITE
1
2
3
4
6
7
8
9
10
11
12
13
SAMPLE SOURCE
Black & White Buoy, Ambrose Channel
Flashing Red Buoy, Ambrose Channel
R N 2, Red Buoy, Ambrose Channel
1/2 Mile South of Station LIC02
1 1/2 Mile South of Station LIC02
3 Miles South of Station LIC02
1/2 Mile South of the Sunken Barge
1 Mile South of the Sunken Barge
1/2 Mile East of the Sunken Barge
400 Feet North of the Sunken Barge
400 Feet East of the Sunken Barge
400 Feet West of the Sunken Barge
100 Yards South of the Sunken Barqe
10
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
In lieu of an available, accepted, standardized procedure for detecting
dispersants in estuarine waters, a modified methyl - green dye test was
used in an attempt to detect the presence of COREXIT - 9527. Since
only one component of the dispersant was amenable to detection by this
method and its amount in the product (only 25 percent), chances of
detection, because of the dilution factor, were questionable. Despite
the known limitations of the test procedure, an attempt was made to
detect the dispersant.
Modified Methyl - Green Dye Test Procedure
A 250 ml water sample in a 500 ml separatory funnel was treated with
125 ml buffer solution (pH 2.5) and 2 ml methyl - green solution and
shaken with 20.0 ml benzene for one minute. The aqueous layer was
discarded and the benzene extract washed with a mixture of 166 ml dis-
tilled water and 65 ml buffer solution, and allowed to stand for 20 -
30 minutes for clarification. Pipette into a 1 cm cell (4 cm if nec-
essary) without filtration and scan absorbance in the range of 510 mm
to 750 mm. (Absorbance at 640 mm is used for calculating the concen-
tration).
Reagent
Methyl - green solution is prepared by dissolving 0.5g of dye in 100 ml
distilled water. The reagent solution is then extracted with chloro-
form until the latter takes-up no further color.
Buffer
pH 2.5; 7.5g glycol (glycine) and 5.8g NaCl are dissolved in one liter
of distilled water and adjusted to pH 2.5 with 0.1N HC1.
Standard Solution
Weigh 0.0542g COREXIT - 9527 and diluted to 200 ml in volumetric flask.
ml Std. Solution ml - H^O Absorbance (1 cm cell) Absorbance (4 cm cell)
3 (0.81 mg) 250 0.02 0.05
5 (1.36 mg) 250 0.05 0.23
7 (1.90 mg) 250 0.14 0.523
EPA samples 52887, 52891, 52893 were selected for residual COREXIT - 9527
analysis. The above locations were in close proximity to the COREXIT -
9527 application area. It was assumed that if the dispersant was not
detected in these samples, there would be little chance in finding it in
the remaining samples more distant from the application site. If this
was the case, the remaining samples would not be analyzed.
Dispersant was not detected in samples 52887, 52891 and 52893. With a
4 cm cell the minimum detection limit was 1 mg/1. Increasing the sample
volume from 250 ml to one liter would not increase the absorbance greater
than 0.03.
11
-------
IR and GC Approach
Attempts were made to obtain additional information using infrared and
gas chromatography. Infrared spectra of COREXIT - 9527 indicate strong CH
stretching, strong carbonyl stretching and a strong alkyl ether band.
However, dilution of the standard to one milligram (25 ml CCl^ extractant)
produced a very weak and small peak at the CH stretching area and the
the carbonyl band at 1730 cm using 1 cm NaCl cells.
disappearance of
COREXIT - 9527 does not seem to dissolve completely in CCl,.
based on the cloudiness in the COREXIT - 9527/CCl4 standard.
This is
An attempt was made to resolve the different organic compounds of COREXIT -
9527 by FID/Gas chromatography. Ultra - Bond PEGS by RFR Corporation
appeared to be an excellent column material. One micro!Her of COREXIT -
9527 (neat) injected produced two large peaks (off scale), four medium
peaks (at least 10 percent of scale) and many small peaks (less than 10
percent of scale). However, dilution of the standard with hexane produced
a different chromatogram: one large peak missing, two medium peaks
mission and the appearance of a new medium peak. The main obstacle
appears not to be gas chromatography and/or infrared spectrometry, but
the proper solvent (or solvent mixture) to use.
Nonionic Surfactant Procedure
It appeared that a nonionic surfactant procedure would be the best way
to analyze for COREXIT - 9527 (5). ASTM lists one method with a
sensitivity of one gram - (D 2024-65 Standard Method of Test For Cloud
Point of Nonionic Surfactant). This method covers the determination
of the solubility inversion temperature or "cloud point" of nonionic
surfactants which are characteristically less soluble in water at higher
temperatures than at lower temperatures. It is limited to thoseQsurfactants
for which the visible solubility change occurs over a range of 1 C or
less at concentrations of 0.5 to 1.0 percent in water between 30 and
95 C. Ionic surfactants in concentrations down to 1 percent or less of
the nonionic surfactant drastically raise the characteristic cloud point
of the latter. The presence of salts and bases (non-surface active
materials) will lower the characteristic cloud point. Acids tend to
raise the cloud point.
A list of several nonionic procedures can be found in the text by W. Lei the -
"The Analysis of Organic Pollutants in Water and Waste Water". The methods
require various reagents which were not on hand in the laboratory. Also,
an all glass apparatus for defoaming the nonionic surfactant would have
to be purchased. However, there was one brief method of extracting
nonionic surfactants from an alkaline bicarbonate solution with butanol.
All references, however, were in German.
12
I
I
I
I
I
I
I
-------
1
1
1
1
1
1
1
I
w
1
w
1
I
1
V
1
1
1
Total Organic Carbon Analysis
An indication of gross amounts of dispersant and/or oil was sought
through the analysis for total organic carbon (TOC) utilizing a Beckman
TOC Analyzer.
Results
Sample
52826
52827
52828
52876
52877
52878
52879
52887
52888
52889
52890
52891
52892
52893
were as
Date
8/1/78
8/1/78
8/1/78
8/4/78
8/4/78
8/4/78
8/4/78
8/5/78
8/5/78
8/5/78
8/5/78
8/5/73
8/5/78
8/5/78
fol 1 ows :
TOC
mg/1
8
6
4
5
5
4
4
5
5
3
3
4
4
4
Site
14
15
7
16
17
18
8
7
7
9
10
11
12
13
13
Sample Source
Collected 2 ft. off bottom,
Rockaway Inlet, 300 ft.
Southwest of Barge
Collected 2 ft. off bottom,
Rockaway Inlet, 300 ft.
West of Barge
Collected 2 ft. off bottom,
% mile South of Barge
Collected (Surface) 500 ft. East
of Barge - Rockaway Inlet
Collected (Surface) North of
Barge - Rockaway Inlet
Collected (Surface) 500 ft. West
of Barge - Rockaway Inlet, Red
Buoy #2
Collected (Surface) 1 mile South
of Barge
Collected (Surface) ^ mile South
of Barge
Collected (Surface) h mile South
of Barge
Collected (Surface) % mile East
of Barge
Collected (Surface) 400 ft. North
of Barge
Collected (Surface) 400 ft. East
of Barge
Collected (Surface) 400 ft. West
of Barge
Collected (Surface) 100 yards South
of Barge
-------
Note that there was no difference in the TOC values of samples taken
before and after dispersant application, even for August 5th when the
largest amount of dispersant was used. Bottom and surface water samples
did not differ substantially. Values obtained also did not differ
substantially from samples collected in the area by EPA in past years.
This suggests that the concentrations of oil and dispersant in the water
column were quite minimal, as expected.
Biological Analyse!
The following grab samples obtained near the surface of the water were
analyzed for phytoplankton:
Sample Date^
52876 8/4/78
52377 8/4/78
52878 8/4/78
52879 8/4/78
52887 8/5/73
52890 8/5/78
52891 8/5/78
Site
52892
52893
8/5/78
8/5/78
Sample Source
Collected 500 ft. East of Barge
Rockaway Inlet
Collected North of Barge -
Rockaway Inlet
Collected 500 ft. West of Barge
Rockaway Inlet, Red Buoy #2
Collected 1 mile South of Barge
\ mile South of the Barge
400 ft. North of the Barge
400 ft. East of the Barge
400 ft. West of the Barge
100 yards South of the Barge
The density and species present were determined by allowing the samples
to settle for at least 48 hours, then concentrating the samples to 100 mis
by siphoning off the sjpernatent. An aliquot from each sample was then
taken to fill a Palmer-Maloney Counting Cell. Fifty Whipple fields were
counted to give a raw count which was then converted to number of cells/ml.
All of the above technique is described in the EPA Biological Methods
manual and is standard practice for performing cell density determinations
on phytoplankton samples.
14
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Sample
52876
52877
52878
52879
52887
52890
52892
52891
52893
Species
Skeletonema costatum
Guinardia flaccida
Skeletonema costatum
Guinardia flaccida
Leptocylindrus danicus
Skeletonema costatum
Guinardia flaccida
Leptocylindrus danicus
Skeletonema costatum
Skeletonema costatum
Dinophysis sp.
Leptocylindrus danicus
Skeletonema costatum
Leptocylindrus danicus
Thalassiosira sp.
Ceratium minutum
Skeletonema costatum
Leptocylindrus danicus
Skeletonema costatum
Leptocylindrus danicus
Guinardia flaccidia
Dinophysis sp.
Ceratium minutum
Olisthodiscus luteus
Skeletonema costatum
Leptocylindrus danicus
Chaetoceros sp.
Raw
Count
10
1
18
7
16
19
9
1
12
19
1
4
34
7
6
1
14
2
29
6
3
2
1
1
7
3
2
Cells/ml
4591
62
8265
3214
7346
8724
4132
459
5510
8724
459
1837
15611
3214
2755
459
6428
918
13315
2755
1377
918
459
459
3214
1377
918
Percent
Concentration
90.9
9.08
43.92
17.06
39.02
65.51
31.04
3.43
100.0
79.81
4.20
16.64
70.81
14.60
12.50
2.10
87.46
12.54
69.02
14.30
7.13
4.78
2.39
2.39
58.32
25.04
16.64
15
-------
The above effort to analyze phytoplankton did not in fact result in data
which were useful in assessment of the environmental impact of the dispersant.
Phytoplankton populations require intense monitoring to determine statistically
significant changes in soecies diversity and abundance. Other plankton
populations such as ichtiyoplankton and mollusc larvae would be more
important to analyze. Many of these plankton are surface dwellers by
reason of special floating mechanisms or phototaxis and would be more
likely to exhibit any effects of the oil dispersant. The low concentrations
of dispersant and dispersed oil predicted may have obviated the need for
detailed biological analysis.
16
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
VII. FIGURES AND TABLES
-------
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
VERRAZANO
NARROWS
BRIDGE
- IV - - -/' "IA" - AMBROSE
^ - - CHANNEL
// SANDY
" HOOK
FIGURE #1
LOCATION OF D/B PENNSYLVANIA
-------
Turret
Nozzle
Metering
Valve M y ğ
Hand
Nozzle
CD X*
Sea Chest
Turret
Nozzle
Metering
Valve M
Ball
Valve
8
Hand
Nozzle
1V2"
CH 04-
Sea Chest
Tygon
Tubing
Foam n Meter
Pickup U
Dispersanl
FIGURE #2
U. S. COAST GUARD - FOAM EDUCTOR SYSTEM
18
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1
14
ATLANTIC OCEAN
8
N
R
N"2"
fi
"2A"
AMBROSE
CHANNEL
BW
Mo(A)
19
FIGURE U3
SAMPLE COLLECTION SITES
-------
TABLE #1
DISPERSANT APPLICATION
Average Dispersant Usage Rate - Surface Operations
Date Hours Applied
8/4
8/5
8/6
8/7
8/8
TOTAL
4.0
12.5
6.0
.8
.2
23.5
Dispersant Applied
(Gallons)
18
101
74
8
1
202
Average Dispersant Usage Rate - Underwater Operations
Date Hours Applied
8/7
8/8
TOTAL
1 .75
.83
2.58
Dispersant Applied
(Gallons)
55
31
86
Usage Rate (gal/hr)
4.5
8.1
12.3
10.0
5.0
8.6
Usage Rate (gal/hr)
31.4
37.3
33
20
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
TABLE #2
DISPERSANT APPLICATION
PWB32318
DATE
Fri
8/4
Sat
8/5
Sun
8/6
PWB-32328
DATE
Sat
8/5
Sun
8/6
Mon
8/7
Tue
8/8
Legend:
- DISPERSANT
APPLICATION
METHOD
H-H/M
M
M
M
BOAT #1
TIME
OPERATING
(HOURS)
1.5
2.5
7.25
4.5
GALLONS PUMPING
OF RATE
DISPERSANT (GPM)
10 100
8 250
48 250
54 250
COMMENTS
Initial Shake-
down
Large Area of
Heavy Sheen
Encountered
Heaviest Oil
Concentrations
Found
- DISPERSANT BOAT #2
APPLICATION
METHOD
M
M
HH/UW
M
HH/UW
M
H-H -
TIME
OPERATING
(HOURS)
5.25
1.5
1.75
0.8
0.8
0.2
Hand Held Hose
21
GALLONS PUMPING
OF RATE
DISPERSANT (GPM)
53 250
20 250
55 100
8 250
31 100
1 250
M - Monitor UW -
COMMENTS
Large Area of
Heavy Sheen
Encountered
Heaviest Oil
Concentrations
Found
Operation to
Wash Out Sunken
Barge
Underwater
-------
BOAT #1
Date
8/4
8/4
8/5
8/6
BOAT #2
8/5
8/6
8/7
8/7
8/8
8/8
TABLE #3
Rate Time Total Volume Volume Average
(GPM) Operating Water Dispersant Dispersant
(Hours) (Gallons) (Gallons) Concentration
Leaving the
Eductor (ppm)
100 1.5 9,000 10 1,100
250 2.5 37,500 8 210
250 7.25 108,750 48 440
250 4.5 67,500 54 800
250 5.25 78,750 53 673
250 1.5 22,500 20 888
100 1.75 10,500 55 5,238
250 0.8 12,000 8 666
100 0.8 4,800 31 6,458
250 0.2 3,000 1 333
22
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
-------
1
1
P
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
TABLE #4
Calculation of Volume Weighted Average Concentration of
Total Dispersant/
Water Mixture X Average ppm
Boat Applied Dispersant
Date Number (in gallons) Concentration
8/4 1 9,010 x 1,110
8/4 1 37,508 x 210
8/5 1 108,798 x 440
8/5 2 78,803 x 673
8/6 1 67,554 x 300
8/6 2 22,520 x 888
8/7 2 10,555 x 5,238
8/7 2 12,008 x 666
8/8 2 4,831 x 6,458
8/3 2 3,001 x 333
TOTAL 354,583
To determine volume weighted average concentration:
288,306,628 4 354,588 =813 ppm
23
Dispersant Applied
= 10,001,100
7,876,680
= 47,871,120
= 53,034,419
= 54,043,200
= 19,997,760
= 55,287,090
7,997,328
= 31,198,598
999,333
288,306,628
-------
-------
VIII. PHOTOGRAPHS
-------
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
OJ
o
: v-ğ id i
>-
oo
r- Z
CO
uj cn
D- i
o>-P c
e£ S_ O v
Z J3 CL-i-
Ğ£ - (O -P t.
to s_
z: o o> a)
2; to -C
LU (U O) -M
O. C7>-£3
S- 4J M-
Ğ ĞJ O
-i E -(->
=*=
O
o
> fO
-OS 1
Oğ O> >- T3
S- -M OO Ğ
T3 O
COO)
3 xa -c o
<1J *->
r O
-Q XT
-1- Q.
o o
nğ 4-> o
JD C -r-
(U E O
o c
i. O (/>
M o a>
= -u
en in
c 01 o
O.-M -a
to c
O OğT-
10 C
o <""--=
I Ğ/> Q.
o i ia
3: -i-
CM
=*:
\. ' ğ .
;.f "
. ğ'5f ,
25
-------
SI
0)
I
O
5
o
in
0)
O)
S_
T3 4J
dl Jii
i- -c
JD I/) 4->
=*te l/l 5- -M
S- Oğ C
O (O
H- Q.T- W
3C -i- OJ O>
G_ O T3 O.
3 -> -O 3
c a>
(O > 01
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
uo
O
>
o
ai
Q.
r- O)
TJ T3 S^
c +j c at
rO C fO *~
3i 5
O> */> O 0)
C 3 r S-
-i*: '"-J^- <13
O "O
r~ FO If- O)
-C O 4J
+-> i- (O T3
O O> O O)
f M- -ğ-> -i- -
*^~ fQ ^3 ^^
O T3 5- C CL
d) ^^ CL
4- 3 > to
O O c w
O T3 >,
r- 03 C
+-" >. S- OJ 01
-o o -3
QJ E
D -(-> -t-> (D 4->
i- C OJ C
f^ TO i^ ^ fO
3 V) V) +-> l/)
I S_ S- 01 S-
Q) OJ 0> 01
01 CL CL 2 CL
O +- -O
3
O
C
(0
W
O)
a.
Ğf- 01
O (->
S
O
-------
C3
o
D-
Ol O
I/)
l/l
^i£ -Q
O O
r- T3 -d
4-J !-> O)
Ğ3 i
CU i- T3
-C -> "O
fl> O
> CO ItJ
Q> IB M
- 3
-C S- O
M CO C
Oj -i-
o c:
(O O -i£
OJ O
w -p
o c
JT -r- E
CD
O O> +->
r O) (/>
M S- >,
-
- o s-
T3 O
O T3
>-
C/)
CDM- Ol
C *i l/>
- TJ Irt
t-J Q 3
CU CD
00
to O - O
I O.
O V) T3
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
IX. REFERENCES
1. Hornstein, Bernard, 1972, The Appearance and Visibility of Thin
Oil Films on Water. Aug. 72, EPA-R2-72-039, National Environmental
Research Center, Office of Research and Monitoring, U.S. EPA,
Cincinnati, Ohio 45268.
2. Tidal Current Tables 1978 - Atlantic Coast of North America, NOAA,
National Ocean Survey, p. 147.
3. National Ocean Survey C&GS Chart 1215, 27th Ed., Oct. 6/73
4. Tide Tables 1978 - East Coast of North America and South America,
NOAA, National Ocean Survey, pp. 66 and 215.
5. A method for measuring anionic surfactant is listed in ASTM, but
the detection limit is at the percent level - ASTM Test D 1681-74
Test for Synthetic Anionic Active Ingredient in Detergents by
Cationic Titration Procedure, in Part 30 of Book of ASTM standards,
29
-------
-------
|