Nutrient Movement Through Beach Media:
Problems and Field Results
Application to Enhance Cleanup of Oil Contaminated Shoreline
Steven C. McCutcheon
U.S. EPA
Environmental Research Laboratory
Athens, GA 30613
John R. Haines
U.S. EPA
Risk Reduction Engineering Laboratory
26 W. Martin Luther King Dr.
Cincinnati, Ohio 45268
John A. Glaser
U.S. EPA
Risk Reduction Engineering Laboratory
26 W. Martin Luther King Dr.
Cincinnati, Ohio 45268
-------
INTRODUCTION :
On March 24, 1989, the EXXON VALDEZ went aground in Prince William
Sound, Alaska, releasing approximately 11 million gallons of Prudhoe Bay crude
oil. After learning of the magnitude of the spill, the EPA Office of Research
and Development (ORD) convened a meeting of nationally and internationally
recognized scientists in the field of oil biodegradation to evaluate the
feasibility of using bioremediation to assist in cleanup operations.
Recommendations from the meeting urge ORD to plan and conduct a field
demonstration project to evaluate the use of fertilizers for accelerating
natural biodegradation of the spilled oil.1"6
Specific conclusions and recommendations were:
• Oil' biodegradation in Prince William Sound waters is probably limited by
the availability of nitrogen and phosphorus; therefore, fertilizing the
beaches with these nutrients will enhance natural degradation of the
oil.1'3-4-8
• Past studies have shown convincingly that the enhancement of oil
biodegradation by nutrient addition readily occurs. Further
verification of these studi.es by laboratory experiments are unnecessary.
• Successful bioremediation will require consideration of the engineering
requirements of long-term nutrient application and the physical
agitation of oil.
•. Bioremediation should be applied to residual quantities of oil, once the
bulk oil has been removed.
• Treatment of the beaches with fertilizer may not remove the black oil
residues due to the high molecular weight of the oil and consequent
lower rates of biodegradation but will reduce the ecological
availability of the oil.
A detailed oil spill bioremediation research plan was then developed by
EPA ORD scientists. The major objectives of this plan were to:
• Examine the occurrence and extent of natural biodegradation of oil on
the contaminated beaches,
• Determine the effect of nutrient addition on the rate of biodegradation
under field conditions.
• Develop application methodology for long-term application of nutrients
to contaminated beaches.
• Develop information on the movement of nutrients in beach substrata
(beach mechanics).
-------
The importance of nutrient movement in beach strata led to convening of a
workshop to develop study plans for hydrologic evaluation of PWS beaches. The
workshop considered the state of knowledge of water flow in beaches. The
results of the workshop were presented in a report that made several
conclusions and recommendations. First and foremost, it was noted that very
little is known about water movement in extremely porous, steep beaches
typical of PWS. Consequently, knowledge of nutrient movement in these beaches
is similarly limited. After limited observations on the original experimental
beaches, a research plan was prepared to study hydrology and nutrient movement
in a typical beach in PWS.
Items considered for study were the extent of wave action and tidal
influence on beach hydraulics, loss of applied nutrients to sea water, and
effect of rainfall on beach hydrology. The results of these studies were
designed to provide information for application to model development of
subsurface water flow and mixing of fresh and salt water layers in the beach
material.
BACKGROUND
The site of the EXXON Valdez oil spill is a harsh and diverse environment
with poor accessibility. The shoreline is geologically young. It is composed
largely of metamorphic rock, and ranges from vertical cliffs to cobble and
pebble beaches. In some sheltered bays, beaches were composed of sand and
gravel. High-energy beaches are common, with tides that vary from +4 to -1 m.
In some areas, glacier and snow melt introduce large amounts of fresh water to
nearshore water of the Prince William Sound.
'The spilled oil spread over an estimated 350 miles of shoreline in Prince
William Sound. Major contaminated shoreline areas include Knight Island,
Eleanor Island, Smith Island, Green Island, and Naked Island. Knight Island,
the largest and one of the most heavily polluted of these islands, has
restricted tidal flushing action in some bays and coves.
The oil settled into the beach gravel and on rock surfaces and the faces
of vertical cliffs.. Contamination occurred primarily in the intertidal zone.
The stranded oil was weathered with a loss of approximately 15% to 20% of
the oil by volatilization. Volatilized components included normal aliphatic
hydrocarbons less than 12 carbon atoms and low molecular weight aromatic
hydrocarbons (benzene, toluene, xylene, and naphthalenes). The resulting
residue consisted of approximately 40% to 50% high-molecular-weight waxes and
asphaltenes. On most beaches in Prince William Sound the weathered oil was
black and viscid rather than brown and mousse-like(emulsified oil).
Beaches were physically cleaned by EXXON with a combination of flooding
and the application of water under high and low pressure and/or high
temperature. Gross quantities of oil were removed from the Passage Cove
Beaches. Vacuum extraction and physical 'skimming were used to remove the
released oil from the water surface. The cleaning process partially removed
-------
oil from the surface of rocks and beaches, particularly the pools of oil but
did not effectively remove the oil trapped in and below the matrix of gravel
and cobble. However, the washing process spread a thin layer of oil over a
much greater surface area of rock and gravel. The extent of physical washing
was dependent upon the degree of contamination.
To investigate bioremediation, two test sites were chosen. Criteria for
the selection of the test sites were based on the following:
Typical shoreline of Prince William Sound; i.e., mixed sand- gravel, and
cobblestone beaches
• Sufficient area with fairly uniform distribution of sand, gravel, and
cobble for the test plots
• Protected embayment with low energy wave motion, adequate staging areas
and sufficient size to support several test and control plots
• Uniform oil contamination
• Shoreline with a gradual vertical rise
PASSAGE COVE PROJECT SITE
Passage Cove is located on the northwestern side of Knight Island. This
site was originally heavily contaminated with oil and was subjected to
physical washing by Exxon. Even after physical washing, considerable amounts
of oil remained at this site, mostly spread uniformly over the surface of
rocks and in the beach material below the rocks. Pools of oil and mousse-
like(emulsified oil) material were minimal on the surface. Contamination was
apparent to about 50 cm below the beach surface. The shoreline area and the
designated beaches in Passage Cove are shown in Figure 1. The test site chosen
had a mixture of sand, gravel, and cobble material. Except for storm surges
the wave action was generally low energy. The calm conditions made possible
daily accessibility, safety, and durability of the apparatus over the period
of the study. The tidal fluctuation over the period of the study varied from
a high of 4.3 m. to'a low of -0.7 m. This tidal fluctuation was expected to
dramatically affect hydraulic behavior of water in the beach face. As was
typical of PWS beaches, the ground level rose rapidly behind the beach area,
with the exception of a small pond directly behind the primary berm of the
beach. The pond was expected to contribute freshwater flow to the beach area
and moderate rapid changes in water level due to rainfall events.
METHODS AND ANALYSIS
Hydrological evaluation of Kittiwake Beach at Passage Cove, Knight Island
was implemented through installation of sample wells, instrument packages, a
tide gauge and a weather station. Concurrently, this beach was used to test
the efficacy of nutrient application via a sprayer using water-soluble
fertilizer. The orientation of wells installed on the beach and a diagram of
the instrument packages installed in the wells are shown in Figures 2 and 3,
respectively . Samples of surface and subsurface sediments were collected for
-------
oil composition analysis about two weeks before fertilizer application, one,
four, and six weeks after fertilizer began. Routine interstitial water
samples for nutrient analysis were collected on the same schedule.
Nutrient Collection
Nutrient samples were collected every two weeks between August 6 and
September 12, 1989. Water samples were withdrawn from each of three small
tubes placed alongside the major well casing. At each sample location these
sample tubes extended to.specific depths: two feet below the beach surface,
one foot above the bottom of the well, and the bottom of the well, respec-
tively.
Due to the vertical changes in sea level over a tidal cycle, often a
complete series of nutrient samples could not be obtained. Survey samples
were collected in groups. Groups 1 through 7 were collected August 6 and 7,
groups 8 through 17 were collected August 20 and 21, and 18 through 25 were
collected September 10 and 11, 1989. Sample sets were collected at about
three hour intervals, unless otherwise indicated.
Water Samples . _ ' .
Clean 250 ml polyethylene bottles were filled with water and frozen as
soon as possible after collection. For each sampling period, samples were
collected every three hours over two tidal cycles. It was not feasible to
collect samples over 24 hour cycles due to weather conditions and the hours of
darkness. Salinity and temperature data were collected in the field concur-
rently with nutrient sample collection. Samples were analyzed for ammonium,
nitrite, nitrate, and phosphorous.
Nutrient Application
During the experimental period from August 2 through September 11, 1989
fertilizer solution was applied to Kittiwake beach daily. Seven pounds of
triple super phosphate{0-45-0) and 17 pounds of NH4N03(34-0-0) were dissolved
in about 35 gallons of seawater. The fertilizer solution was metered into a
sprayer system and distributed over the beach plot over a period of about an
hour.
Meteorological Monitoring
Automatic data recorders were used to record meteorological data, water
conditions in the sample wells, and tidal conditions in the bay. The
meteorological station was damaged in a storm after three weeks operation.
The well data recorders recorded water depth, salinity, and temperature. The
tide gauge recorded water depth, temperature, and salinity off the beach.
Oil Chemistry
Beach samples consisted of either mixed sand and.gravel frozen in 400 ml
I-Chem jars or cobblestones wrapped in aluminum foil and frozen. The mixed
sand and gravel was thawed immediately prior to the initiation of oil
-------
analysis, and the contents were mixed thoroughly. A weighted 100 gm subsample
was removed and mixed thoroughly with 300 mis of methanol in a separatory
funnel. The slurry was shaken for five minutes, and the methanol was decanted
into a 2 L separatory funnel. The samples were similarly re-extracted two
times with 300 ml of pesticide - or HPLC grade methylene chloride. The three
organic fractions were combined and back-extracted with 100 ml of 3% aqueous
sodium chloride. The phases were separated and the aqueous portion was
extracted with 50 ml of fresh methylene chloride. This aqueous extraction in
methylene chloride was added to the combined organic fraction.
The combined organic fractions and 3 or 4 clean boiling chips were placed
into a 1 L round bottom flask fitted with a three-ball Snyder column. The
volume of solvent was reduced until the color was approximately the color of
dilute weathered oil (ca 15 mg/2 ml methylene chloride). The final volume of
the extract was measured with a syringe having an appropriate graduated
cylinder, and an aliquot was transferred to a GC autosampler vial.
All of the cobblestones were extracted using the same procedure
(methanol, followed by methylene chloride), except that shaking was replaced
by gentle swirling to remove oil from the rock surfaces.
Gas chromatographic (GC) analysis was accomplished with an instrument
capable of reproducible temperature programming with a flame ionization
detector and a reliable autosampler. The GC conditions were:
Column: DB-5, 30 m X 0.25 mm, film thickness 0.25 urn
Initial Temperature: 45»C, 5 min. hold
Temperature Rate: 3.5«C/min
. Final Temperature: 280»C, 60 min. analysis
Injector: splitless, 1 min valve closure
Injector Temperature: 285*C
Injection: 2.0 01
Detector: FID, 350»C
Those samples that demonstrated significant evidence of biodegradation
were fractionated to allow separate determination of aliphatics and aromatics.
Aliquots of the sediment or oil extracts selected for fractionation were
solvent exchanged to hexane under a stream of dry nitrogen. A volume of 50 @1
of hexamethyl benzene (80 ng/01) and 25 31 of n-decyclohexane (1 @g/(?l) was
added to each sample extract prior to fractionation. The fractionation was
accomplished using a 10 mm X 23 'cm glass column that was slurry packed (with
hexane) with 60/200 mesh silica gel activated at 210 C for 24 hours. The
aliphatic fraction was eluted with 30 ml of hexane and the aromatic fraction
was eluted with 45 ml of hexane/benzene (1:1). Aliphatic and aromatic
fractions were analyzed using the GC methods described above.
Subsamples of the final concentrated extract were subjected to mass
spectral analysis using GC/MS. The analytical procedure is given in Fucus oil
analysis protocols.
Subsamples (5-15 ml) of the final concentrated extract were also removed,
filtered through sodium sulfate, and placed in tared watch glasses. After
-------
passive evaporation of the solvent, the oil residue weight was determined.
Changes in oil composition were determined using three procedures:
a. The branched hydrocarbons pristane and phytane were used as
internal standards, under the assumption that they were slow to
degrade, and weight ratios of C17:Pristane and C18:Phytane were
calculated as indicators of biodegradation.
b. The total weight of all alkanes, appearing on the chromatograph,
normalized to the total residue weight of oil, were compared on a
sample by sample basis.
c. Assuming that hopane and norhopane were not biodegraded, weight
ratios with other identifiable hydrocarbons were calculated.
Nutrient Analysis
Water samples taken for nutrient analysis were filtered (Whatman glass
fiber filter) and then placed in 150 ml plastic screw capped bottles. The
bottles were immediately frozen with a dry ice-antifreeze solution. Water
samples taken offshore were collected with a clean bucket and subsamples were
taken for nutrient analysis. Beach water samples were collected behind or in
front of an ebbing or flooding tide, a vacuum device capable of withdrawing
intersticial water. The sampler was inserted approximately 20 cm into the
mixed sand and gravel and flushed for one minute prior to sampling.
Nutrient concentrations were determined using the following standard
methods:
Nitrate-- . , '
Nitrate was determined by reduction to nitrite followed by a colorimetric
assay for nitrite. The procedure for nitrate was derived from the non-
automated technique. Detection limits for nitrate and nitrite were
expected to be about 0.05 and 0.01 @M, respectively. Estimates of the
precision for the nitrate measurements were made at the 20 @H level in the
samples was calculated as the mean of n determinations ±0.5 (mean/n2) in @M.
Nitrite--
Nitrite was determined by the Geiss reaction in an aqueous acidic
solution to form an intensely pink diazo dye with an adsorption maximum at
540-543 nm.
Ammonium--
Ammonium was determined by the Berthelot reaction by forming indophenol
blue, an intensely blue chromophore with an absorption maximum at
approximately 637-640 nm. The detection limit for ammonium was expected to be
approximately 0.1 @M.12 An estimate of precision at the 1 @M level was
calculated as the mean of n determinations ± 0.1 (mean/n2) in @M.
Phosphate--
Orthophosphate was determined as phosphomolybdic acid, which has an
-------
absorption maximum at 880-885 nm in its reduced form in the presence of
antimony.12 The detection limit for phosphate was expected to be about 0.03
@M. An estimate of the precision at the 3 (?M level was calculated as the mean
of n determinations ± 0.03 (mean/n ) in units of @M.
Total Kjeldahl Nitrogen (TKN)--
TKN was measured by heating the sample in a sulfuric acid solution
containing K2SO, and HgS04 and comparing colorimetrically with standards and
blanks using a Technicon AutoAnalyzer (EPA method 365.4).
RESULTS
The results of the field investigation of nutrient and water movement in
a typical PWS beach were used to develop conceptual models and determine
potential effects on bioremediation efforts in PWS . Models of saturated and
unsaturated zones of the beach were tested to determine gross circulation
patterns of the beach. Oil composition changes were used to evaluate the
effectiveness of nutrient addition for stimulation of biodegradation.
Nutrient survey data were used to follow nutrient movement through the body of
.the beach and provide estimate of subsurface conditions and nutrient movement.
The first survey conducted two days after application of fertilizer began
showed penetration of NH4 at 1.7 to 2.3 meters below the surface of the
beach. No nitrate data have been developed as yet for this sample series.
The NH4 data varied between 3.8 and 179 uM depending on sample time and depth.
Offshore NH4 did not exceed lOuH. The pond behind the beach yielded NH4
concentrations between 0.8 and 35 uM. The salinities of interstitial water
varied between 0 and 18 ppt. Open water salinities were typically 21ppt or
higher. The pattern of salinity data indicates that the sample wells
penetrated the zone of saline and fresh water mixing. The data present a
complex picture of water movement in the beach subsurface. Both vertical and
horizontal variations in salinity were evident^across the tidal cycle.
The second nutrient survey was conducted August 20 and 21, 1989. Both
NH4 and N03 data were available from this survey. In most cases, the N03 was
greater in concentration than NH4. In two cases NH4 exceeded N03
concentration. The highest nutrient concentrations were associated with the
lower salinity values. The lowest nutrient concentrations were associated
with the highest salinities. The offshore samples had nutrient concentra-
tions less than 20 uM. NH4 concentrations varied between 0.4 and 400 uM and
N03 concentrations varied between 0.4 and 290 uM. Salinity values varied
between 2 and 21 ppt in the well'samples. The highest salinities were
recorded at high tide and flood tide. The lowest values were recorded on ebb
tide and at low tide. With the exception of high tide, most salinities were
represented as contributions of about half seawater and half freshwater
indicating a complex vertical and horizontal flow of water under the beach.
The third nutrient survey was carried out on September 10 and 11, 1989.
Nutrient analysis results were incomplete at this writing. The salinity data
recorded in the field during this survey were much lower in general than the
earlier surveys. This observation reflects an extensive period of rainfall in
the two weeks preceding the sample collection. Salinities varied from 800
uMho to 16ppt in the sample wells. The field data for this survey do not
8 x- . ' •'
-------
permit determination of the movement of the interface of fresh and salt water.
Data collected by the automatic recorders in place on KIttiwake beach
were also used to evaluate the mixing transport of nutrients. Later these data
will be used to evaluated the efficacy of commercially available models of
groundwater flow. The plan was to use the models to evaluate engineering
design of nutrient application to beaches. At this time the two best
groundwater models have been tested. Neither model by itself was adequate to
describe the complete movement of groundwater under the surface of the beach.
The FEMWATER model13 was used to simulate the response of the beach
groundwater surface elevations to tidal changes. Ignoring fluid density
stratification as a first approximation, simulations based on the point
iterative solution technique indicate that tidal inflows can overwhelm the
freshwater flow near the surface of the beach and saltwater is pushed into the
beach surface from the surface of the bay during late summer conditions. The
depth of saltwater penetration can not be determined with the present models
but preliminary calculations indicate that penetration of one to two meters
(horizontal penetration into the beach) may be possible during the average
tidal and freshwater conditions occurring in August of 1989. Penetration of
seawater at the beach surface indicates that freshwater fluxes are not enough
to continually push water through the beach at all times. However, during ebb
tides and at other times fresh and brackish water is pushed out of the beach
continuously. These preliminary findings influence our understanding of how
the beaches were oiled originally and how oil on the surface of the beach may
continue to move into the beach subsurface.
Where effects of fluid density stratification are expected is in the
magnitude of the penetration and the slope of the of the water surface under
the beach surface during flood tide. The data on hand do not permit resolution
of uncertainty with regard to these parameters. For example, less penetration
than the data indicated may actually be occurring (on the order of 0.5 to 1.0
m). The representation of fluid mechanics is substantiated sufficiently that
surface penetration is occurring regularly (once a tidal cycle except in the
case of large freshwater flows). Numerous observations of fairly uniform
oiling to a depth of several feet tend to bear this out. As a result, this
offers the possibility to predict how deeply oil penetrated upon initial
oiling but additional work will be necessary to make precise estimates.
Another model, SUTRA14, was used to determine if it was possible to gain
limited information about the density underflow of saltwater into the beach
below the water surface. For these calculations, an idealized domain was
selected and the penetration of the seawater simulated. These calculations
are subject to interpretation, but at this time it seems clear that saltwater
does not completely penetrate and underlie the beach. This is important
because it indicates the dimensions of the freshwater lens under the beach.
The volume of this lens will influence the residence time of nutrients in the
beach.
. The fact that saltwater penetrates.the beach as a density underflow
indicates that dispersed oil may be pushed into the beach several meters below
the water surface and may collect on the substrate porous media or if
\
-------
sufficient sand is present to filter the water. As of yet there are no
observations of dispersed oil accumulating below the intertidal zone. Seawater
pushed into the beach mixes with freshwater, flows upward and moves seaward
above the intrusion at the toe of the beach. There is no back-flushing action
of the same seawater pushing backout the same area that it entered. In this
regard, some hydraulic pumping different from wave driven pumping is evident
in these simulations. Un fortunately, if there is a filtering effect on
dispersed oil by this pumping mechanism, there will be less opportunity for
nutrients to reach the oil. Freshwater is not expected to reach these low
levels. Any nutrients that reach the seawater will be diluted by mixing over
the surface of the embayment (Passage Cove) and will be diluted to near
background concentrations.
At this time, the models have ignored the slope of the beach face which
will modify the results significantly. The advective nature of tides pushing
seawater into the beach on a flood tide has also not been fully incorporated.
These alterations will change the upper part of the flow but will prevent the
penetration of the mixing cell into the beach. Density difference appears to
drive saltwater into the toe of the beach, but the depth of this penetration
is uncertain.
Based on the limited success of the preliminary calculations first
approximation of the freshwater, mixing, and saltwater regimes in the beach
may be similar to the conceptual drawing in Figure 4.
CONCLUSIONS
1. Water movement, fresh and salt, in PWS beaches is very complex,
. beyond current model's ability to describe. Modifications of existing
models or development of improved models is necessary.
2. Significant nutrient concentrations were observed at 1.5 and
2.3 meters depth in beach well samples.
3. Salinities from well samples varied between those for open water
and fresh water. Model simulates which mechanisms affects or
controlled salinities but these interpretations are pending.
4. There is a definite relationship between nutrients and salinity
that defines the degree of mixing of nutrients in the beach. The
transport and mixing have not been analyzed, but clearly the spray
irrigation system 1s effective in delivering nutrients over the
complete oiled profile on at least two or three occasions each
day. This provides an important mechanise to put nutrients deep
into the beach where other application awthods may be inef-
fective.
REFERENCES
1. R.M. Atlas and R. Bartha, "Degradation and alneralization of
petroleum in seawater: limitation by nitrogen and phosphorus",
Biotechnol. Bioenq. 14:297 (1972).
10 x . - .
-------
2. R.M. Atlas and R. Bartha, "Stimulated biodegradation of oil slicks
using oleophilic fertilizers", Environ. Sci. Techno!. 7:538 (1973).
3. C.F. Gibbs, "Quantitative studies in marine biodegradation of oil.
I. Nutrient limitations at 14*C", Proc. R. Soc. Lond. Ser. B. 188-61
(1975).
4. A. Jobson, M. Mclaughlin, F.D. Cook and D.W. Westlake, "Effect of
amendments on the microbial utilization of oil applied to soil", ADD!
Microbiol. 27:166 (1974). ~^
5. R. Olivieri, P. Bacchin, A. Robertiello, N. Oddio, L. Deggen and A.
Tondo, "Microbial degradation of oil spills enhanced by a slow-release
fertilizer", ADD!. Environ. Microbiol. 31:629 (1976).
6. B. Tramier and A. Sirvins, "Enhanced oil biodegradation: a new
operational tool to control oil spills", in "Proceedings of the 1985 Oil
Spill Conference". U.S. Coast Guard, Amer. Petrol. Inst., Environ. Prot.
Agency, Los Angeles, CA, 1985, pp. 115-119.
7. R.M. Atlas, P.O. Boehm and J.A. Calder, "Chemical and biological.
weathering of oil from the Amoco Cadiz oil spillage within the littoral
zone", Estuarine Coastal Mar. Sci. 12:589 (1981).
8. M. Blumer, M. Ehrhardt and J.H. Jones, "The environmental fate of
stranded crude oil", Deep Sea Res. 20:239 (1972).
9. A. Sirvins and M. Angeles, Strategies and Advanced Techniques for
Marine Pollution Studies: Mediterranean Sea; C.S. Gram and H.O. Dou,
eds. Springer-Verlag, Berlin, 1986, pp. 357-404.
10. P. Sveum, Accidentally spilled gas-oil in a shoreline sediment on
Spitzbergen; natural fate and enhancement of biodearadation. N-7034,
Sintef, Applied Chemistry Div., Trondheim, Norway, pp. 1-16.
11. S.C. McCutcheon;"Proceedings of Beach and Nearshore Workshop for
Prince William Sound Bioremediation Project". U.S. Environmental
Protection Agency 1989.
12. T. R. Parsons, Y. Malta, and C.M. Lalli, A Manual of Chemical and
Biological Methods for Seawater Analysis. Pergamon Press, Inc., Maxwell
House, Elmsford, New York, 1984.
13. G.T. Yeh, FEMWATER: A Finite Element Model of Water Flow Through
Saturated-Unsaturated Porous Media-First Revision. Oak Ridge National
Laboratory, Oak Ridge, TN, pp. 258, 1987.
14. C. Voss, A Finite-Element Simulation Model for Saturated-
Unsaturated, Fluid-Density-Dependent Ground-Water Flow with Energy
Transport or Chemically Reactive Single-Species Solute Transport. U.S.
Geological Survey, Reston, VA, p. 409, 1984.
11
-------
90-22.5
Raven
Guillemot
Figure 1. Schematic diagram of Passage Cove and experimental
beaches. Raven,control; Tern, oleophilic and granular;
Kittiwake, water soluble; and Guillemot, oleophilic
and granular.
-12
-------
90-22.5
depth
gauge
7
3
10
mean high tide
mean low tide
Figure 2. Location of Wells for Beach Hydraulics Experiment at Passage Cove
•13
-------
90-22.5
Strap for removal
Aanaura sensor lor
conauctivity ana temperature
3 PVC tubes attached to inside of
casing extending to precise locations
Sensor package rests firmly
on the bottom screen
Aandura sensor lor pressure
conductivity, and temperature
6" diameter
Casing capped with
a screw-on cap
Beach Surface
Band of screen
excluding solids
Tygon luting lor
sample collection
Holes or slots in
band encircling pipe
Screen
Figure 3. Cross section, of well casing installed in Kittiwake
beach.
14
-------
90-22.5
Cd
§
pa
— o
— o
1
to -a
01 C
M O
ta cu
1
CO
1
1
^t
* '
1
o
t^
1
^
0)
CJ
S-i
3
CO
XI
CO
o
4J
tO
------- |