EPA-670/2-73-099
February 1974
                       Environmental Protection Technology Series
 Investigation of Surface  Films -
Chesapeake Bay Entrance
                                 Office of Research and Development
                                 U.S. Environmental Protection Agency
                                 Washington, D.C. 20460

-------
            RESEARCH REPORTING SERIES
Research reports of the  Office  of  Research  and
Monitoring,  Environmental Protection Agency, have
been grouped into five series.  These  five  broad
categories  were established to facilitate further
development  and  application   of   environmental
technology.   Elimination  of traditional grouping
was  consciously  planned  to  foster   technology
transfer   and  a  maximum  interface  in  related
fields.  The five series are:

   1.  Environmental Health Effects Research
   2.  Environmental Protection Technology
   3.  Ecological Research
   4.  Environmental Monitoring
   5.  Socioeconomic Environmental studies

This report has been assigned to the ENVIRONMENTAL
PROTECTION   TECHNOLOGY   series.    This   series
describes   research   performed  to  develop  and
demonstrate   instrumentation,    equipment    and
methodology  to  repair  or  prevent environmental
degradation from point and  non-point  sources  of
pollution.  This work provides the new or improved
technology  required for the control and treatment
of pollution sources to meet environmental quality
standards.

-------
                                                EPA 670/2-73-099
                                                February 1974
INVESTIGATION OF SURFACE FILMS—CHESAPEAKE BAY ENTRANCE
                            by

                Dr. William 6.  Maclntyre
                Dr. Craig L.  Smith
                Dr. John C. Munday
                Mrs. Victoria M.  Gibson
                Mr. James L.  Lake
                Mr. John G. Windsor
                Dr. John L. Dupuy
                Dr. Wynam Harrison

                           and

                Dr. John D. Oberholtzer
                    Project 15080  EJO
                  Program Element  1BB041
                     Project Officer

                  Mr. Robert D.  Kaiser
     Office of Research and Development,  Region III
                Sixth and Walnut Streets
             Philadelphia, Pennsylvania 19106
                      Prepared for

           OFFICE  OF RESEARCH AND DEVELOPMENT
          U.S.  ENVIRONMENTAL PROTECTION AGENCY
                 Washington, D.C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $2.06

-------
                       ABSTRACT

Experimental point source oil releases have been conducted
in the Chesapeake Bay mouth area.  Predictions of oil
slick motion were tested, and slicks were sampled and
analyzed to measure their aging rates over periods, up to
32 hours.  Remote sensing techniques were used to detect
and measure the spreading rate of oil.  Some laboratory
oil film aging experiments were done to further document
and elucidate aging processes.  Results indicate a
reasonable motion prediction, an explanation of the ,
non-biological initial aging of oil films, and a fair
corroboration of a theoretical oil spreading model.

Indigenous surface films in the study area were analyzed
for lipid and chlorinated hydrocarbon content.  Hydro-
carbons were 300-500 /ug/liter and fatty acids and esters
700-7800 jug/liter in surface film samples.  Chlorinated  .,
hydrocarbons were generally less than 100 parts per	
trillion in surface films, in contrast to some earlier high
concentrations found in Biscayne Bay.  Surface film
analysis limitations imposed by sampling methods are
discussed.  Plankton in slick, non-slick, and subsurface
water were counted.  Populations were higher in surface
than subsurface water, and higher in non-slick than in
slicked surface water.

     This report was submitted by the Virginia, Institute of
Marine Science in fulfillment of Project Number 15080 EJ0
under the (partial) sponsorship of the Office of Research and
Monitoring, Environmental Protection Agency.   Work was
completed as of March 17, 1972.
                          ii

-------
                       CONTENTS
ABSTRACT
LIST OF FIGURES
LIST OF TABLES
ACKNOWLEDGMENTS
Page
  ii
  iv
  vi
viii
Sections
     I    CONCLUSIONS
    II    RECOMMENDATIONS
   III    INTRODUCTION
    IV    OIL SLICK MOTION
     V    INITIAL AGING OF OIL FILMS
               ON SEA WATER
    VI    REMOTE SENSING OF OIL SLICKS
   VII    PHYTOPLANKTON IN SURFACE SLICKS
               AND IN ADJACENT SUBSURFACE
               AND NON-SLICK WATER
  VIII    CHLORINATED HYDROCARBONS IN INDIGENOUS
               SURFACE FILMS
    IX    LIPIDS OF SURFACE FILMS ON CHESAPEAKE
               BAY
     X    REFERENCES
    XI    LIST OF PUBLICATIONS
   XII    APPENDICES
   I
   3
   5
  10
  24
  69

  81
  99
 109
 130
 140
 141
                          iii

-------
                        FIGURES
No.                                         ,           Page
 1.  Oil release experiment locations.                    8
 2.  Relative wind factor versus wind speed from
     data in Table 2.                                     14
 3.  Oil slick trajectories at Chesapeake Bay
     entrance.  Solid lines are observed trajectories,
     dotted lines are predicted paths with squares,
     circle and triangle indicating zero, three and  ten
     percent overall wind factor respectively.            16
 4.  Predicted (•-••) and observed (	) paths for
     Aug. 12, 1970, release.                              18
 5.  Predicted (••••) and observed (	) paths for
     Dec. 3, 1970, release.                               19
 6.  Predicted (••••) and observed (	) paths for
     April 26, 1971, release.                             20
 7.  Predicted (••••) and observed (	) paths for
     Aug. 3, 1971, release.                               21
 8.  Predicted (••••) and observed (	) paths for
     Sept. 14, 1971, release.                             22
 9.  Boiling range composition of No. 2 fuel oil.         26
10.  Boiling range composition of No. 4 fuel oil.         28
11.  Boiling range composition of No. 6 fuel oil.         29
12.  Bubbler assembly for oil film aging.                 32
13.  n-paraffin standards for gas chromatography.         34
14.  Loss of n-paraffins from No. 4 fuel oil release,
     Aug. 3, T971.                                        52
15.  Comparison of No. 4 fuel oil aromatic  fraction
     b.p. 190°-250°C with No. 4 fuel oil water
     solubles.                                            54
                            iv

-------
                   FIGURES (Cont'd)
No.                                                    Page
16.  Scrubbing of dissolved aromatic species in
     bubbler apparatus.                                  56
17.  Change in boiling range of evaporate fraction
     in bubbler apparatus.                               58
18.  Typical gas chromatograms of No. 2, 4, and 6
     fuel oil evaporate fractions.                       59
19.  Loss of volatiles in bubbler apparatus.             60
20.  Volatiles remaining in bubbler apparatus.           61
21.  Loss of volatile fraction of No. 6 fuel oil
     in bubbler and field experiments.                   64
22.  Loss of volatile fraction of No. 4 fuel oil
     in bubbler and field experiments.                   65
23.  Loss of volatile fraction of No. 2 fuel oil
     in bubbler and field experiments.                   66
24.  Spreading rate of No. 2 and No. 4 fuel oils
     on seawater.                                        79
25.  Plankton sample areas.                              83
26.  Comparison of surface and one meter diversity
     in slick areas.                                     94
27.  Comparison of surface and one meter diversity
     in non-slick areas.                                 95
28.  Sampling locations and dates.                      101
29.  Amoco oil spill trajectory.                        157
30.  Guinea Marsh oil sampling locations.  Black
     areas are oil exposed beach.                       160
31.  Boiling range composition of Amoco spill oil.      162
32.  Gas chromatogram of Amoco Spill Oil.  Integers
     are carbon numbers of normal hydrocarbon peaks.    164

-------
                        TABLES
                                                       Page
 !•          Experimental Oil Releases                    6
 2.          Oil Slick Motion and Relative Wind Factor   11
 3 (a)-(l).  Fuel Oil Aging - Loss of Normal Paraffins   37
 4.          Dissolved Oil Components from Bubbler
             Experiment                                  "
 5.          Boiling Range Composition Coefficients      62
 6.          Fuel Oil Aging Comparison                   67
 7.          Oil Slick Remote Sensing Experiments        70
 8.          Results of Sampler Comparison Experiments   85
 9.          X2 Test for Randomness                      87
10.          Counts of Five Replicate Subsamples from
             Five Samples                                38
11.          Counts of Ten Replicate Subsamples from
             One Sample                                  88
12.          Results of Sample Counts                    90
13.          Confidence Intervals of 997o Level for
             Sample Counts                               92
14.          Composition of Surface Samples Slick
             and Non-Slick                               93
15.          Affinity Index Values for Pairs of
             Samples                                     93
16.          Chlorinated Hydrocarbon Concentration
             Results                                    105
17.          Fatty Acid Sample Concentration and
             Composition                                113
18.          Percentages of Predominant Fatty Acids     121
19.          Hydrocarbon Sample Concentration and
             Composition                                122

-------
                    TABLES (Cont'd)

No.                                                    Page

20.          Fatty Acid Concentration in Slicks and
             Non-Slicks                                 129

21.          Hydrocarbon Concentrations in Slicks
             and Non-Slicks                             129

22.          Population Statistics of Marsh
             Intertidal Fauna                           167
                          vii

-------
                    ACKNOWLEDGMENTS
The authors are indebted to Mr. Paul Alfonsi of NASA,
Wallops, for his assistance in design and construction of
a surface film skimmer, and to Mr. Robert Long of NASA,
Wallops, and Captain Alan Hancock, U.S.C.G. Group, Norfolk
Commander, for their ship scheduling efforts.

Assistance in the project by the Region III personnel,
Environmental Protection Agency, especially to Mr. Ralph L.
Rhodes, Mr. Russel H. Wyer, and Mr. Robert D. Kaiser, the
Grant Project Officer, is gratefully acknowledged.
                         viii

-------
                       SECTION I

                      CONCLUSIONS

A computer program has been developed  to enable prediction
of oil slick trajectories from available wind and  tidal
current information.  The accuracy  of  prediction depends
on tidal current data quality and knowledge of steady
currents.  Significant errors may occur in locations where
current systems are poorly described.

The most important process in initial  aging of fuel oils at
sea is the loss of volatile compounds  through evaporation.
Loss of material by dissolution into seawater is quite small
and little change is produced in the oils by this  aging
mechanism.

The components of a fuel oil slick  which dissolve  at
greatest concentration into seawater under aerated, agitated
conditions are naphthalene, and the several methyl-
substituted naphthalenes.

Quantitative determination of the degree of aging  of a fuel
oil can be made from the boiling range composition of the
original oil and gas chromatographic analysis of n-paraffins
in original and aged oil samples.

Photography is useful for oil slick detection and
discrimination.  The near ultraviolet  band is best for
imaging edges and thin slicks.  The green band best
delineates thick oil regions.  Color film is useful, as
it distinguishes thick from thin oil,  and fuel oil from
recent lipid slicks.

No. 2 fuel oil is distinguished from No. 4 and 6 fuel oils
by its lack of negative contrast in blue and green band
photographs.

Oil slick spreading was found to fit a theoretical model
for small volume spills of No. 2 and 4 fuel oils.  No. 6
fuel oil did not spread in our tests.

In an estuarine environment the surface microlayer contains
more phytoplankton than water at one meter depth,  and species
diversity is lower at the surface than at one meter.
Phytoplankton population in the surface microlayer are
lower in slick than non-slick areas.   This result  was
obtained with both light fuel oil and  natural lipid slicks.
Samples were taken seasonally and show the usual temperate
zone species pattern with peak populations in March and
September.

-------
Chlorinated hydrocarbons (CHP) and polychlorinated biphenyls
(PCB) were detected in indigenous estuarine slicks and
subsurface water.  Total chlorinated hydrocarbon and CHP
concentrations were generally higher in surface than sub-
surface water, while PCB concentration was similar in
surface and subsurface water.

Lipid analysis of indigenous surface slicks from
Chesapeake estuarine locations showed them to be predomi-
nantly fatty acids and esters of recent biological origin.
The minor hydrocarbon amounts were petroleum products with
the G.C. characteristics of a light fuel oil.

-------
                      SECTION  II

                    RECOMMENDATIONS

Detailed current studies should be conducted  in probable
spill areas to permit more accurate prediction of oil
transport.  The computer program given here should be used
for oil spills where tidal currents are available, and
slicks will not beach for at least 12 hours.

A quick-deployment air dropped current meter  system should
be developed for use at spill  locations where prior
information on steady and tidal currents is unavailable.
This would make a rough prediction of slick motion possible,

The validity of the assumption of linear vector addition of
steady current and wind velocity should be tested by
experiment and theoretical analysis, so that  conditions
under which the predictive model fails may be understood
with consequent improvement of the model.

Large scale turbulence produces unsteady non-tidal currents
which are not included in the  present prediction model.
This phenomenon should be investigated, perhaps by remote
sensing methods, for coastal regions.

Long term aging experiments on air-barrier confined slicks
of crude and residual fuel oils should be designed to
investigate tar ball formation.  The aged synthetic tar
balls should be compared with  those collected in the
Atlantic Ocean.  This work would permit greater under-
standing of semi-permanent oil pollution in the world ocean,

Further oil component dissolution in seawater experiments
should be conducted using GC-MS analysis.  This work will
recognize the non-equilibrium  nature of release of oil
aromatics and their partitioning between oil  and water.
The effect of oil dispersants  in keeping otherwise
volatile toxic oil aromatics in the water column should be
investigated.

Photography should be used for daylight remote detection
and discrimination of oil spills.  An infrared scanner
should be used for night oil spill detection.  Photography
of oils should be conducted with films and filters
suggested in this report.

Further analysis of the oil spreading model verified here
should be made and the viscosity of oil included in the
model.

-------
Toxicity of oil and oil components on phytoplankton should
 ?ia ^OT concern, because there is prior evidence that
oil adversely affects the primary aquatic producers.  This
work indicates some effects of organic films on surface
phytoplankton populations.

Attempts to establish the importance of organic surface
films in concentration and transport of chlorinated
hydrocarbons should be discontinued until adequate
separation methods for compound classes are developed, a
means of sampling solely the organic surface layer without
water is devised, and methods to determine the age and prior
history of a slick are found.  These requirements seem
unsatisfiable in the near future.

Information on indigenous estuarine slick composition
obtained in this study indicates the relatively unpolluted
situation in the sampled region.  This work should be
repeated in other areas to establish the petroleum
hydrocarbon content of our coastal surface waters.

-------
                      SECTION III

                      INTRODUCTION

A series of experimental oil releases have been conducted
in the Chesapeake Bay and Virginia coastal waters.
Observations and conclusions from these and ancillary
laboratory studies are presented in this report.

The several objectives of this program were:

1.)  Description and prediction of the motion of point
     source oil spills under varying wind and tidal
     current conditions.

2.)  Collection of oil from slicks at serial times after
     release, and analysis of these samples to determine
     the loss of water soluble and volatile constituents.

3.)  Photographic and passive radiation imaging at serial
     times after oil release to establish oil spreading.
     rates, document slick shape, and fix its orientation
     relative to measured winds and currents.

4.)  Determination of water surface photoplankton
     population and species composition in film covered
     and film free areas.

5.)  Chemical analysis of natural films collected in the
     Chesapeake Bay estuarine system to establish their
     content of petroleum compounds, chlorinated organics,
     and recently formed lipid substances.

Table 1 is a chronological listing of the experimental oil
releases which yielded information relating to the first
three objectives above.  This tabulation should be referred
to subsequently in this report, because oil releases are
identified here by release date.

A map of the Virginia Maritime Area (Fig. 1) shows the
approximate locations of the oil releases.  Detailed
navigational data are available on request from the
reporting agency.

Oil sources were: No. 6 and-No. 4 fuel oil from the Norfolk
terminal of ESSO and originating from mixed residues and
distillates of Nigerian and Venezuelan crude processed by
the Lago Refinery, Aruba, N.W.I., No. 2 fuel oil from
various local heating oil company distributors, and
menhaden oil from Haynie Products Co., Reedsville, Virginia.

-------
Table 1.  Experimental Oil Releases

                                               Kind of     Volume   Observation
    Date       Oil Type   Release Location   Observation  Released    Period
                                                           (gal.)     (hrs.)
Sept. 12, '69 6 Ches. Lt. Tower
Sept. 25, '69 6/menhaden Bridge Tunnel
Sept. 26, '69 6/menhaden York River
Oct. 27, '69 6 Bridge Tunnel
Nov. 24, '69 6 Bridge Tunnel
Dec. 6, '69 6 Bridge Tunnel
Jan. 28, '70 6 Bridge Tunnel
Feb. 13, '70 6 Bridge Tunnel
Mar. 9, '70 2/6 Bridge Tunnel
May 6, '70 6 Bridge Tunnel
Aug. 12, '70 2 Ches. Lt. Tower
Sept. 14, '70 4 Ches. Lt. Tower
Nov. 6, '70 4 York River
Dec. 3, '70 6 Ches. Lt. Tower
Jan. 21, '71 2 Ches. Lt. Tower
B
A
B
B
A
A
A
A
A
A
B
A
ABC
B C
ABC
100
5/30
5/10
25
25
25
6
5
50
30
200
200
150
200
200

5
-
6
4
7
7
5
7
3
8
24
9
30
7

-------
Table 1. (Cont'd).  Experimental Oil Releases


    Date
                                Kind of     Volume   Observation
Oil Type   Release Location   Observation  Released    Period
                                            (gal.)	(hrs.)
Jan.
Mar.
Apr.
Jun.
Aug.
Aug.
Sept
Oct.
22,
18,
26,
23,
3, '
30,
. 14,
13,
'71
'71
'71
'71
71
'71
'71
'71
4
2
4
6
4
2
6
2/6
Ches
; ;•
. Lt.
Tower
A C
York River
Ches
Mid
Ches
Ches
Ches
. Lt.
Ches.
. Lt.
. Lt.
. Lt.
Tower
Bay
Tower
Tower
Tower
Ocean Station
A

A
A
A
A
C
B
C
B
B
B
B

C

C
C
C
C
200
200
250
65
200
200
200
200/200
9
6
16
31
32
10
25
7/24
         track    B = Remote sensing
                               Sampling

-------
                                         .-••   ;  &/

                                       fa  J /XVS
                                       i/!  ; ,'f /,::i
                                         (  t ./ \ / \
                      •BRIDGE TUNNEL^
Fig.  1.   Oil release experiment locations.
                       8

-------
Oil was obtained in separate batches for each release, and
each batch had a potentially different history of well
source, refining treatment, and aging in storage.  Thus,
chemical comparisons are made only within a set of oil
release samples, but it was assumed that bulk physical
properties of all batches of a particular type of fuel
oil were similar.

Releases were made from storage tanks or barrels carried
on board the vessel used to maintain station during
observation of the oil slick.  Vessels were provided by
N.A.S.A., the U. S. Coast Guard, and the U. S. Navy.  The
N.A.S.A. ship "Range Recoverer" was most often used.

The release vessel served as a navigation and weather
observation platform, a communication center for directing
activities of remote sensing aricraft, and a mother ship
for motorized inflatable boats used to sample the oil and
gather remote sensing ground truth information.

Direct observation and sampling of oil slicks was not
possible during hours of darkness, so flasher drogues
that floated in or near the slicks were deployed in the
evening and followed till morning.  The slick could then,
in some cases, be visually relocated and further studied.

A contingency plan, which stated action to be taken by
agencies involved in this study in the event an experi-
mental oil spill should endanger shorelines, was adopted.
Notification was given to E.P.A., U.S.C.G., the U. S.
Army Corps of Engineers, and the Virginia Water Control
Board prior to each release.  Weather-slick release point
combinations were selected to minimize the possibility of
beaching the oil.  Oil slicks were always followed until
no longer detectable or practical to sample, and, since
there were no unfortunate incidents, the contingency plan
was never applied.

Studies on the latter two of the stated program objectives,
plankton population and chemical composition of occurring
slicks of origin not related to the experimental oil
release program, were conducted entirely in the Chesapeake
Bay.  The work was concentrated in the York River and
Chesapeake Bay Bridge-Tunnel regions.

-------
                      SECTION IV

                   OIL SLICK MOTION

Effective control and cleanup of oil slicks often requires
predictions of oil slick motion.  In marine areas, slicK
trajectories are the net result of tidal currents, wind
generated currents, and motion due to local wind stress.
To establish the relative importance of these factors,
slug oil releases in this program were tracked throughout
each experiment by the navigational equipment available on
the oil release vessels.  These releases are listed in
Table 1.

A few field observations and theoretical treatments
indicate that "oil slicks move with the wind at 2.3 to 5
percent of the wind speed (Batelle, 1969, 1967; Kolpack,
1969; Smith, 1968).  This percentage value has been termed
a "wind factor."  Wind generated surface currents have a
similar wind factor (Laevastu, 1962; James, 1968; Doebler,
1966; Hela, 1952; Tolbert and Salsman, 1964), so a
distinction must be made between the overall wind factor,
and the relative wind factor.  Slick position data can be,
used to^calculate a total slicjc velocity S.  The wind
vector W can be divided into S to give the overall wind
factor.  If the local surface current vector C is measured,
the sj^ick velocity data can be reduced by subtracting C
from S to give R, the compcpienk slick motion due to local
wind stress.  The value  ]R|/|W| x 100 is referred to as
the relative wind factor.
                                    -'
In the experiments conducted in Chesapeake Bay on the dates
shown in Table 2, the relative wind factor was calculated.
Overall wind factors were calculatedifor those other
releases in which navigation data were available but
surface currents were not measured.

Local surface currents, C, were determined by tracking a
current drogue from the observation vessel, which provided
simultaneous position fixes.  The dorgue had four vertical
orthogonal 25 centimeter square vanes, and was weighted to
place the vanes 1 meter below the water surface.  A light,
bamboo pole carrying a diver's flag at 6 ft above the water
surface made the drogue easily visible.  The wind drag on
the drogue was 0.25 percent, measured against surface water
masses marked with Rhodamine WT with winds of 20 knots and
white capped waves 0.3 to 0.7 meter high.  Data were.not
corrected for this small wind factor.
                           10

-------
Table 2.  Oil Slick Motion and Relative Wind Factor.
Experiment
Date and
Vol. (gal.)
BUNKER C FUEL
24 Nov. 69
25
6 Dec.
25
28 Jan. 70
6
13 Feb.4
5
Water
Temp.
(°C.)
OIL
10.0

6.6

0.5

3.0

Time1
(hours MT)

1319-1440
1440-1534
1205-1320
1345-1545
1121-1325
1500-1545
1251-1604

Relative
Velocity
(knots)

0.163 NW
0.24 NNW
0.43 NW
0.0
0.02 ESE
0.02 ESE
0.0

Wind
Velod ty
(knots)

15 N
15 N
10 N
7 N
8 ESE
8 ESE
3.5 NW

Relative
Wind
Factor (%)

0.8 (45°Left)2
1.6 (10°Left)2
0.88 (50°Left)2
0.0
0.25
0.25
0.0

 Portions of experiments containing most reliable quantitative data.
^Orientation of slick vector relative to wind vector.
^Velocity magnitude value exaggerated by extensive slick spreading and disper-
 sion.  Simultaneous measurements on two drogues formed the basis for reducing
 the value in calculating a wind factor
    gallons of SAE 30 oil added to increase slick visibility.

-------
       table 2. (Cont'd.).  Oil Slick Motion and Relative Wind Factor,
ro
Experiment
Date and
Vol. (eal.)
BUNKER C FUEL
9 Mar.
505
6 May
30
Water
Temp.
(°C.)
OIL
2.2

15.0

Time1
(hours MT)

111-1152
1321-1404
1404-1417
1339-1408
1408-1448
1448-1510
1510-1533
Relative
Velocity
(knots)

0.3 NE
0.25 SE
0.0
0.184 NNW
0.185 NNW
0.135 NNW
0.257 NNW
Wind
Velocity
(knots)

17 NE
12 NNE
12 N
19.5 NNW
18 NNW
16 NNW
14.5 N
Relative
Wind
Factor (%)

1.7
0.0 (112°Right)2
0.0
0.95
1.03
0.85
1.75 (10°Right)2
       NO.  2 FUEL OIL
        9 Mar.          2.2
           505
       MENHADEN FISH OIL
1115-1517
0.0
14 NNE
        Reseeded with 50 gallons after three hours.
0.0
25 Sept. 69 21.0
25
1006-1200
1200-1235
0.0
0.4 NE
4 W
8 NW
0.0
0.0 (90°Right)2

-------
Wind velocities, W, were obtained from anemometers on the
vessel or, in a few cases, on  the northern end of Trestle
A of the Chesapeake Bay Bridge and Tunnel.  The anemometers
were not intercalibrated, but  were regularly serviced and
similarly mounted from 15 to 20 meters above the water
surface.

Slicks enlarged by spreading for a few hours after release
and took various shapes, sometimes linear or pseudopodal,
and often separated into sections.  In white cap sea
conditions, oil dissipation at slick edges was rapid, and_»
slicks remained small.  To track the slick and establish S,
the vessel was positioned at the downwind edge of the
slick which was taken as the slick position.  This
procedure yielded a maximum slick motion.

The Coriolis deflection of C from W is approximately 20°
in offshore areas at medium latitude, and decreases nearer
shore.  (Doebler, 1966; Hela,  1952; Mandlebaum, 1955).
Since all water depths occurring^ in tbJLs stuc^y are less
than 30 meters, deflections of C and R from W are assumed
to be zero.

Wind generated surface currents require a few hours to
r^each maximum speed (James, 1968), but a direct coupling of
C with W is assumed here.  The study area is not suited to
a check of the degree of coupling because there are
frequent wind vector changes and rather large tidal current
oscillations.  Calculated wind factors are slightly over-
valued, due to the wind-produced positive gradient of
horizontal velocities near the water surface.  The drogue
did not measure current at the air-water boundary where the
velocity is a maximum.

Figure 2 shows the relative wind factor plotted versus
speed.  There is apparently a  functional dependence, with a
wind speed intercept of approximately 7 knots.  This wind
speed must be exceeded to produce a significant difference
between the surface current motion and the slick motion.
For winds between 15 and 20 knots, the relative wind factor
is about 1.3 percent, and may  approach 2.0 percent at
higher wind speeds.

Additional oil release studies would permit more precise
establishment of the relative  wind factor, but this
precision will not be needed for oil cleanup purposes.
Given that overall slick speed should be known to 0.25
knot, sufficient accuracy is realizable from a knowledge
of surface currents, plus an approximate 1.3 percent of the
wind speed additional for winds greater than 15 knots.
                          13

-------
             2.0-
             1.5-
          o:
          o

          S
          2 i.OH

          o
          _J
          UJ
          o:
                         5        10       15

                        WIND   SPEED   (knots)
20
Fig. 2.   Relative wind factor versus wind speed  from data
         in Table 2.
                         14

-------
Accurate predictions of oil slick motion require knowledge
of local surface currents, tidal currents, oceanic
circulatory currents and wind conditions.  The error in
prediction of oil slick motion is directly dependent on the
quality of the available water current information and
weather predictions.  In areas of high potential oil spill
threat, detailed long term surface current studies should
be made.  The current data would permit better protection
of threatened coastlines.

Sufficiency of existing surface current information for
prediction of oil slick motion near  the Chesapeake Bay
Bridge-Tunnel has been tested using  the experimental
releases in this region.  Tidal currents were obtained
from predictions of the U. S. Coast  and Geodetic Survey
(1969, 1970), which include the effect of average seasonal
winds.  Wind velocities were taken as measured, and several
overall wind factor values were selected from the
literature and experience obtained in this program.

These data were supplied to a computer program (Appendix 1)
which predicted slick motion after a spill at a stated
location and time.  The current path is predicted under the
assumption of uniaxial sine wave tidal oscillations and
using inverse r2 weight-averaging of tidal currents at
U.S.C.&G.S. stations to estimate tidal currents at other
positions in the region.  Possible effects due to winds
preceding the spill (James, 1968; Harrison and Pore, 1967;
Johannessen, 1968; Jones and Bellaire, 1962; Bellaire,
1963) were not included in these predictions due to
unavailability of information for lower Chesapeake Bay.

Figure 3 is a chart of the Chesapeake Bay entrance, showing
observed slick motions and predicted current paths calcu-
lated with 0, 3, and 10 percent overall wind factors.  In
comparing predicted paths and actual slick motions, it
should be realized that slicks follow the current path and
are only slightly modified by the relative wind factor.
The predicted paths do not match or  bracket the observed
motions.  Discrepancies are not regular and amount to
several miles in a single flood or ebb tide.

For Chesapeake Bay entrance, tidal currents permit
predicting the general directions of slick travel but are
inadequate for accurate prediction of oil slick trajec-
tories.  Better prediction will require analysis based on
long time series observations of local currents and winds.
It is doubtful that acquisition of these data could be
financially justified for prediction of oil slick motion
alone.  If there are other environmental problems that also
                           15

-------
Fig. 3.  Oil slick trajectories at Chesapeake Bay
         entrance.  Solid lines are observed trajectories,
         dotted lines are predicted paths with square,
         circle and triangle indicating zero, three and
         ten percent overall wind factor respectively.
                        16

-------
require this series data, consideration  should be given  to
installation of long  term monitoring  stations.

The oil slick motion  prediction program  was used to predict
the path followed by  oil on releases  near Chesapeake Light
Tower on August 12, 1970, December 3,  1970, April 26, 1971,
August 3, 1971, and September  14, 1971.

Tidal current information for  this area  was obtained from
Haight (1942) and U.S.C.G. tidal height  tables for 1970
and 1971 by referring tidal currents  at  the light tower  in
hours after Greenwich transit  of the  moon to tidal heights
and times at Hampton  Roads on  the spill  dates.  It is
fortunate that tidal  currents  here are linear reversing  and
can be used directly  in the program.   The program is now
being modified to accommodate  radial  tides which occur in
other continental shelf areas.  The tidal currents used here
are based on more than a year  of observations, so transient
phenomena are quite well filtered.

The overall wind factor used in the program was 3.7
percent, a value suggested by  our observations and the
experiments of Schwartzberg (1971).   The predicted and
observed slick motion paths are given' for each slick in
Figures 4 through 8.

The predictions for the August 12, 1970, and December 3,
1970, releases are acceptably  close to the observed motion.
The predicted motion  for April 26, 1971, is totally wrong,
and on August 3, 1971, the direction  but not the total
displacement was correctly predicted.  The observed path
for September 14, 1971, made a sudden 90 degree course
change that did not appear in  the prediction.

Large deviations of predicted  from observed slick motion
were anticipated in the Chesapeake Light Tower vicinity,
because tidal information was  obtained 5 miles north of
the present light tower, large scale  current systems in
the region are poorly known, spring and  neap tide
situations were not considered, the wind history prior to
release was not available, and local  surface currents were
not obtained as current meters were not  deployed.  The
predictions were surprisingly  good in view of the above
limitations.  The simple program used here predicts motion
fairly accurately in  the absence of large scale current
fields, excepting tides, generated outside the area of
observation.

Research on oil slick motion is continuing, and releases in
proximity to fixed current meters are being planned.
                            17

-------
36°
                          FINISH 1530.
56'
      AUGUST  12,  1970
                                                FINISH
                                                 1530
36*
                                              LIGHT
                                              TOWER
54'
    START
    0742
75e
 45'
75'
Fig. 4.  Predicted  (••••)  and observed (.
        Aug. 12, 1970, release.
43'
                                        -) paths for
                          18

-------
       75*
45'
                   NOTE:
75*
           14.9 MILES  AT 14° 15' TRUE
           TO FINAL SLICK POSITION AT
                IMS  DEC. 4T 1970
                                  FINISH  1700
                          DEC.   3,  1970
                                      START  0903
Fig.  5.  Predicted (••••) and observed  (	) paths for
        Dec. 3, 1970,  release.
43'
                                           36*
                                                     55'
                                                    LIGHT
                                                    TOWER
                                                    36*
                                                    54'
                                                    36'
                                                    53'
                         19

-------
10
O

      s,
     »-»!*
      O
     N>rt
     VO
(D a.
&>
co o
(6 O*
•  CO
  (D
      13
      ft)
      CD

      Hi
      O
            36°

            56'
        FINISH
          1950
        36*
            54'
            36°

            52'
                75° 44
                                START 0801
                                       LIGHT
                                     • TOWER
                         APRIL  26,  1971
                    FINISH
                      1950
75*
42'
75"
40'

-------
FINISH
  1935
                                     FINISH 1935
  '54
                                                  LIGHT TOWER
       AUGUST  3f  1971
                                  75* 45'
        START
         0905
     Fig. 7.  Predicted  (••••) and observed (-
             Aug. 3, 1971, release.
-) paths for
                            21

-------
                                              FINISH
                                               1800
          LIGHT
          TOWER
                                                                            FINISH
                                                                             1800
       SEPTEMBER  14, 1971
10
NJ
         START  1000
    75° 43'
75° 41'
52.5'
                  Fig. 8.  Predicted (••••) and observed (-
                          Sept.  14, 1971, release.
                       -) paths for

-------
Parachute drogues may be used to measure,surface displace-
ment of oil from the underlying water mass.  The objective
of this work is to develop a more complete system of
equations for oil motion prediction.  These equations will
only be useful if concurrent wind and current data is
available.

To implement a more accurate prediction  than demonstrated
above, an instant delivery current meter-anemometer system
must be deployed at the spill site and data from the system
telemetered to a computer which will use the data to make
motion predictions.  The spill must be at least 6 hours
away from any beaching area in order for this type of
prediction to be achieved.

If such a prediction system proves too costly, current
surveys should be made to establish average large scale
currents.  Predictions based on these averages will be
imperfect, but certainly better than what can now be
achieved.  Prediction of slick motion on the continental
shelf is only possible where tidal currents are available,
which limits any present predicting ability to a few
selected regions.
                            23

-------
                      SECTION V

        INITIAL AGING OF FUEL OILS ON SEAWATER

                     INTRODUCTION

Fuel oils can be classified into two main types.  The
distillate fuels are the distilled fraction of _a selected
boiling range of crude petroleum.  Representative
distillate fuel oils are kerosene and diesel fuels, and
No. 2 fuel oil.  Residual fuel oils, on the other hand,
are generally mixtures of the asphaltic residue from the
distillation of crude petroleum with selected distillate
fractions of the crude petroleum.  The exact formula varies
from one producer to another and often varies between
batches from the same producer.  ASTM specifications for
these residual fuels, such as No. 4, 5, and 6 (Bunker C)
fuel oils are concerned primarily with the viscosity
range, and permit large variations in composition.
      e
Oil spilled on the water undergoes a process called "aging"
or "weathering," which has the effects of reducing the
total volume of oil, and of changing its physical
characteristics.  This aging is the result of at least four
distinct processes:

     1.)  Evaporation of volatile constituents.

     2.)  Dissolution of oil constituents into the
          water column.

     3.)  Microbial degradation or modification.

     4.)  Photochemical oxidation.

Little is known about the effect of light on oils, but
photochemical oxidation of individual hydrocarbons usually
requires considerable amounts of ultraviolet irradiation in
the vapor phase.  The effect of photochemical oxidation is
expected to be minimal for oils in relatively thick films
or clumps.  Microbial degradation (Kator, 1971), though an
important process in the ultimate fate of oil spilled on
the sea, requires seeding and addition of nutrients for
rapid effect.  This mechanism should not be an important
factor in the initial aging of oil spilled on the  sea for
at least the first 48 hours.  Thus, evaporation and
dissolution will be major mechanism of initial aging of
fuel oil on seawater.
                          24

-------
Information on the initial aging of fuel oils can be
valuable for the estimation of the remaining volume of an
oil slick by the time men and equipment can be assembled to
deal with it.  Additionally, one can predict the change of
character of the spilled oil from its residence time at sea,
and estimate the amount of oil components which have been
released to the atmosphere by evaporation, and to the water
column by dissolution.

Two different approaches have been utilized in this study
of initial oil aging.  The primary line of study was the
observation and analytical sampling of small volume spills
of No. 2, 4, and 6 fuel oils at sea.  Twelve releases,
ranging from 60-200 gal. of oil, four for each oil type,
were conducted under a variety of climatic conditions.
Nine releases were made in the Atlantic Ocean in the
vicinity of the Chesapeake Light Tower, and three releases,
one of each oil type, were made in the more sheltered
waters of the Chesapeake Bay, near the York River entrance.
Samples of oil were taken from the slicks' at regular
intervals and preserved for later analysis.

A laboratory aging experiment for each oil type was carried
out as an adjunct to the field studies.  The experiment
consisted of periodic sampling of the effluent air stream
and aqueous phase of a carboy bubbler apparatus in which
air was bubbled through an oil film floating on artificial
seawater.
                 METHODS AND MATERIALS

Fuel oils used in this study were procured locally.  No. 2
fuel oil was a straight run distillate derived, as far as
can be determined, from various Texas crudes.  It had a
boiling range of 170-370°C/760 torr, determined by actual
distillation under reduced pressure in a distillation unit
with a 24" Widmer column constructed according to
specifications from the ESSO Baytown Refinery.  Head
temperatures at reduced pressure ,were converted to
atmospheric, using tables from the ASTM Method for
Distillation of Crude Petroleum (D2892-70T).  Boiling
range was confirmed by the ASTM Test for Boiling Range
by Gas Chromatography (D-2887-70T).  A boiling range
composition is included in Figure 9.  The aromatic
character of this fuel was determined by the ASTM
Fluorescent Indicator Adsorption Test (D-1319-70).
                          25

-------
   O
   s
   O
   o
   CL

   CD


   O
   UJ
   h-
   o
   UJ
   or
   tr
   o
   o
      300-
      200-
             10  20 30  40  50  60  70 80  90  100

                     PERCENT  DISTILLED
Fig.  9.  Boiling range composition of No. 2 fuel  oil,
                     26

-------
No. 4 and 6 fuel oils, purchased  from the Humble Norfolk
Terminal, were  formulated by the  Lago Oil and Trmsport
Co., Aruba, Netherlands Antilles, as  C401 and C552,
respectively.   Both were derived  from tar produ/ed from
mixtures of Venezuelan crudes by  thermal crack^ig and
blended with diluents composed of a number of refinery
streams to achieve an appropriate viscosity.  /Soth fuel
oils had initial boiling points near  170°C/76f torr, and
actual reduced  pressure distillation  showed n6 major
discontinuities in the boiling range  composi/ion for
either fuel.  Boiling range  compositions are/shown in
Figures 10 and  11.  -The aromatic  character of each was
determined by the Fluorescent Indicator Ads/rption Test
(ASTM D-1319-70) on the distillate fractioi/, b.p.
170-270°C/760 torr.  This test is not possible for the
whole fuel oils, which contain considerable proportions
of high molecular weight components.   The/aliphatic and
aromatic portions of some fractions of the  oil distillate
were separated  on a silica gel column,  eZuted with pentane
to remove the aliphatics, then with benzene to remove the
aromatics from  the column.   Progress  of/the separation was
monitored with  gas chromatography.    /
                                       /
                                       /
In the field studies, oil samples were  collected in a
separator scoop which allowed the partial  separation of
water from the  oil.  Sample  collection was  biased toward
the thicker oil layers in the slick.  Oil  samples,
together with small volumes  of unseparated  water were
stored in quart Mason jars with aluminum foil lid liners.
Samples in the  jars were then frozen  and temporarily stored
in portable ice chests packed with dry  ice.  The samples
were later stored in a commercial freezer  chest until
analyses could  be made.            /

Analyses were carried out by gas  phromatography.  Oil
samples were treated in the  following manner.  No. 2 fuel
oil samples were chromatographed .directly.  No. 4 and 6
fuel oil samples, which were too/viscous to permit direct
syringe injection, were injected  in pentane solution.
To aliquots of  frozen samples weighing  approximately 5 g
each, 30 ml of  Baker "Baker  Grade  pentane  was added, and
the mixture was stirred and  triturated  until  a homogeneous
solution or suspension was achieved.  The solutions or
suspensions were each gravity filtered  through Whatman
#42 filter paper'to remove precipitated asphaltenes and
other pentane insoluble matter.   The  filtrates were then
stored in glass-stoppered flasks  and  portions  of these
filtrates were  then chromatographed.
                            27

-------
         40C-i
       O
       (0
       O
       e
       0.

       CD


       O
       UJ

       O
       UJ
       or
       or
       O
       u
         300-
         200-
                 10  20  30  40  50 60  70  80  90  100

                        PERCENT  DISTILLED
Fig. 10.   Boiling range composition of No. 4 fuel oil.
                      28

-------
       460-n
       400-
    o
    CO
    O
    o
    QD


    Q  300-
    UJ

    6
    kJ
    o:
    a:
    o
    o
       200
           0   10 20  30 40  50  60  70 60  90 JOO


                     PERCENT DISTILLED
Fig. 11.   Boiling range composition of No. 6 fuel oil.
                      29

-------
Gas chromatography was done using a Perkin-Elmer Model 900
Gas Chromatograph fitted with dual 1/8" x 6' copper columns
packed with 5% SE 30 or 60/80 mesh Chromosorb P-NAW, and
with flame ionization detectors.  The carrier gas was
helium, at a nominal flow rate of 20 ml/min, and the
temperature was programmed from 100-280°C at a rate of
4°C/min, and permitted to hold at 280°C until all volatile
components had been eluted.

The peaks in the chromatograms of the oil samples which
corresponded to the normal paraffins were determined by
comparison of retention times with the authentic samples of
normal paraffins.  Interpolation of retention times between
the even-carbon paraffins which were available was used to
estimate the retention times for the odd carbon paraffins.
Aging of the fuel oils was followed by calculating this
percentage loss of the individual normal paraffins, which
constitute a series of compounds in rather high initial
concentrations with boiling points evenly distributed
throughout the boiling range of the fuel oils.

Quantitative analysis of normal paraffin concentration
was accomplished by determination of the peak height for
each individual compound relative to that of n-eicosane
(n-C20H42)> which was chosen as an internal standard
because: 1) it has a sufficiently low vapor pressure at
ambient temperature that minimal loss by evaporation will
occur; 2) it is quite insoluble in water; 3) of the
possible compounds in the oil which could meet the first
requirement, it is present in all three oils in sufficient
concentration for the peak height comparison.

Peak heights were considered to be the vertical distance
from the top of the peak to a baseline constructed by
drawing a straight line connecting the shoulders of the
peak in question.  This baseline was generally, but not
in every case, tangent to the background of the
chromatograms.

Peak height for each n-paraffin in each chromatogram were
normalized by dividing the n-paraffin peak height by that
of n-eicosane.  The normalized peak heights of each
n-paraffin were then divided by the normalized peak
Height of the corresponding n-paraffin in the chromatogram
of the unaged oil to give (%Ci)t» the percentage of
n-paraffin remaining in the oil sample taken at time t.
The following equation expresses this procedure:
                           30

-------
               t - 100 x       x
where [Ci]t represents the peak height of the normal
paraffin CiH2i+2 at time t.

The reproducibility of this procedure was tested by
calculating (%Cj_)t using chromatograms from repetitive
injections of the same sample of No. 2 fuel oil.
Successive determinations showed that (%C^) varied ±3% of
the value determined from the initial sample.

The bubbler experiments were carried out using a specially
constructed apparatus (Figure 12).  Air from an aquarium
pump at the rate of 2 1/min was passed through a tubular
filter containing equal volumes of  16 mesh indicating
silica gel and activated carbon granules to remove
contaminating substances from the sweep gas.  The filtered
air was then bubbled up through a glass tube at the bottom
of a carboy containing 30 ml of oil on 10 1 of artificial
seawater (33%0 NaCl) .  An opening at the bottom of the carboy
permitted removal of the seawater for analysis without
danger of contamination from the surface film.  The effluent
air stream from the bubbler was then passed through a plug
of extra fine glass wool to remove  small droplets of water
and oil, and  then into a concentric-type vapor trap, which
was chilled in a dry ice/acetone cooling bath.

In operation, the bubbler was permitted to run until the
ice crystals  in the vapor trap began to reduce the sweep
gas flow rate.  The trap was then removed from the cooling
bath and warmed to room temperature, causing the
obstruction to melt and collect at  the bottom of the trap.
This procedure of trapping and warming was repeated as
necessary.  At selected intervals the trap contents were
removed, and  extracted 5 times with 1 ml of pentane.  The
combined extracts were allowed to evaporate to a volume of
4 ml and transferred to a 5 ml and  the sample used for
 fas chromatographic analysis.  Samples of the seawater
 ayer were removed at the termination of the bubbler aging,
filtered through a plug of extra fine glass wool to remove
suspended oil droplets, and extracts were allowed to
evaporate at  room temperature to a  volume of less than
1 ml, then transferred to a 3 ml volumetric flask and
adjusted to exactly 3 ml.  This sample was used for gas
chromatographic analysis.
                            31

-------
to
           FILTER^

      AIR
    2 liters/min.
    30ml.
    01 L
    10 liters
   33%o NaCI
   SOLUTION
SPRAY TRAP
                                VAPOR TRAP-^
                                                          EXHAUST
                                                 •t
                             DRY ICE, ACETONE
                                     ffi. '
                  Fig. 12.  Bubbler assembly for oil film aging.

-------
Samples from both the vapor  trap  and  the  seawater  layer,
prepared as above, were chromatographed on a Perkin-Elmer
Model 900 Gas Chromatograph  fitted with dual 1/4"  x 3'
copper column packed with 5% SE 30 on Chromosorb W-HMDS
treated support.  The temperature was programmed from
50-300°C at 8°/min, which produced the required resolution
for the ASTM Boiling Range Distribution of Petroleum
Fraction by Gas Chromatography  (D-2887-70T) .  Samples of
the vapor trap were analyzed in the following manner.
A plot of boiling point (atmospheric) versus retention
time was constructed from the chromatogram of a mixture of
authentic n-paraffins (Figure 13).  Then, initial  and final
boiling points were calculated according  to the ASTM
D-2887-70T test, by planimetry of chromatograms of vapor
trap samples.  By comparison of the total area under the
sample chromatograms with the area under  the peak  of the
n-octadecane concentration standard chromatogram,  the
total weight of the oil -derived hydrocarbons in the pentane
solution was calculated.  The following formula illustrates
the calculation:

oil weight = Peak area oil   _  x Volume Pentane extract
             3.69 Peak area//Ltg n-Cig   Volume injected
The factor 3.69 in the equation derives from a calibration
run, in which a standard solution of 9.704 g/1 of
n-octadecane in pentane produced a ratio of 3.69 peak area
units per yg n-octadecane.

Samples produced by extraction of the seawater layer were
chromatographed under similar conditions to those of the
vapor trap, and the area under the 'chromatogram was
integrated by planimetry.  The weight of oil hydrocarbons
in the pentane extract was determined by a formula analogous
to that above:
oil weieht = Pea^ area oil x Volume Pentane extract
       B     2.75 peak area/pg naphthalene x Volume injected

The factor 2.75 is derived from the calibration chromatogram,
in which a standard solution of 1.354g/l of naphthalene in
pentane produced an integrated peak area of 2.75 area units
per j/g of naphthalene.  Naphthalene was chosen for this
standard because the oil components extracted from the sea-
water appeared to consist of naphthalene and its derivatives.

In certain instances, large volumes of 10 pi or more of the
pentane extracts of the seawater layer were injected into
the gas chromatograph fitted with a fraction splitting
accessory.  Samples corresponding to major component peaks
                           33

-------
   350-1
   300-
o
o
O
QL


O

- 250
*j

o
CD
O
UJ
   200-
    150
                                     13
                                 '12
                             'II
                          '10
8
                          10  12  14   16  18  20  22  24  26
     0246


              RETENTION TIME (minutes)

Fig.  13.  n-paraffin  standards for gas chromatography


                         34

-------
were condensed in 2-3 mm diameter glass capillary tubing.
These sample tubes were sealed off, and used for mass
spectrometric analysis.  The tubes were broken open, and
fitted into the solid sample introduction device of a
CEC 21-104 Mass Spectrometer, and the  70eV mass spectrum
recorded.  Although the mass spectroscopic study of the
chromatographically separated components was incomplete,
due to poor separations, basic structural units were
identified.  Further investigation of  these dissolved
components should be carried out when  suitable instru-
mentation becomes available.

An additional experiment was designed  to determine the
rates of which certain representative  aromatic compounds
are scrubbed out of aqueous solution in the bubbler
apparatus.  Ten liters of 33i& aqueous  NaCl solution was
stirred for 24 hrs with a mixture of 5 g each of cumene
(i-propylbenzene), naphthalene, and 1-methyl naphthalene.
The excess organics were then removed, and the aqueous
solution filtered through a plug of extra-fine glass wool
to remove suspended droplets of organic compounds.  The
seawater was then returned to the bubbler apparatus.
After removing a time-zero sample, the air sweep, at a
rate of 2 1/min was initiated.  Samples were removed at
periodic intervals.  The samples, 1 1  each, were extracted
four times with 10 ml of pentane, and  the combined extract
reduced in volume by evaporation at room temperature to
less than 3 1.  The extracts were then transferred to 3 ml
volumetric flasks and adjusted to exactly 3 ml.  The
concentrations of the three organic components in the
extracts were determined by gas chromatography, comparing
the peak areas to those of known concentrations of
naphthalene.
                RESULTS AND DISCUSSION

Twelve experimental releases were conducted, employing the
three fuel oil types, No. 2, 4, and 6 (Bunker C) fiel oils.
Samples of oil from the slicks were collected at regular
intervals as long as possible.  Because of the low
solubilities of the oil components and the enormous
dilution factor involved, sampling of the water column
beneath the slick was not attempted.  The samples obtained
were analyzed by gas chromatography for the relative
amounts of the normal paraffin hydrocarbons, which were
selected to be model oil components.  The n-paraffins
consist of a homologous series of compounds in these fuel
oils, whose members produce distinctive, sharp peaks at
regular intervals throughout the boiling range in the gas
                            35

-------
chromatograms of oil samples.

Results of the n-paraffin analyses for the twelve field
releases are given in Tables 3(a) through 3(1).  The
entries in the tables are the percentage of the given
n-paraff in remaining in the oil at time t: (%Ci)t.  The
average local conditions during the sampling period are
also recorded.

The (%Ci)t for a given n-paraff in in an oil slick would be
expected to decrease in some sort of monotonic fashion with
time.  Inspection of the tables show, however, that fluctu-
ations occur, and in some instances, the (%Ci)t is greater
than 100%.  No reasonable mechanism can account for a net
increase of the n-paraffin content of an oil slick, so such
anomalously high values must not reflect the true concen-
trations.  Several factors inherent in the method of
analysis seem likely to contribute to such errors.  One
problem is the necessity to use four separate peak height
measurements in the calculation of (%C.£)t.  A test of
reproducibility using a single oil sample showed that
(%Ci)t could be determined to ±3%; however, since the
estimation of peak heights is not better than ±5% for most
peaks, the maximum uncertainty in the (%Cj[)^ would be
propagated to ±20% by the mathematical treatment.  Another
problem associated with the quantitative gas chromatography
of fuel oils is the fact that the n-paraffin peaks are not
completely separated from the background of aromatic,
isoprenoid, isoparaffin and naphthene hydrocarbons.  Peak
heights are measured from the peak maximum to an estimated
baseline, which depends upon the detailed nature of the
background.  Although a highly reproducible baseline
estimate can be made for imaged samples, problems arise
when the concentrations of the various petroleum compounds
change by aging.  Then, changes in overlapping and
underlying peaks greatly affect the estimation of the
n-paraffin baseline, and hence the determination of peak
Height.

A final possible source of error comes from the variation of
sampling location in the slick itself.  Since a thin oil
film will age more rapidly than a thicker one, the
thickness of the oil being sampled has considerable import.
In actual practice, the samples were collected from the
thickest layers of the slick that could be found.  It was
not possible to tell whether a particular layer had been
thick since the spill, or was the result of coalescence of
thinner films.
                            36

-------
      Table 3 (a)
Fuel Oil Aging - Loss of Normal Paraffin
Location: Chesapeake Light Tower
Oil: 200 gallons, No. 6 Fuel Oil
Release: 0900 hrs, EST, 12/3/70
Water Temp: 13°C.  Air Temp: 60°F.
u>
% Paraffin
C9
CIO
Cll
C12
C13
C14
C15
C16
C17
CIS
C19
C21
C22 .
C23-
C24
1:00
156.5
158.0
101.8
118.9
109.0
105.0
118.2
100.4
101.4
98.2
99.5
101.2
102.6
99.4
102.5
Time After Release (hrs)
2:00 3:00 4:00 5:45
70.7
65.3
69.9
97.8
106.0
102.4
109.8
101.2
105.5
102.2
102.9
102.4
103.8
129.0
n.a.
119.6
173.2
102.6
140.5
142.4
116.4
132.7
116.1
115.0
100.2
106.7
98.6
101.7
96.6
113.0
70.7
94.2
86.1
106.4
103.8
104.8
128.6
109.8
106.2
95.7
103.4
111.3
103.7
98.0
119.4
32.1
91.0
61.2
105.5
122.5
113.0
123.5
108.0
107.2
95.7
101.1
110.4
106.7
95.2
107.2
24:15
n.d.
n.d.
n.d.
n.d.
16.0
31.4
56.3
70.3
91.2
95.7
92.8
104.3
102.6
110.1
106.3
26:30
n.d.
n.d.
n.d.
n.d.
25.0
43.5
74.1
85.6
103.8
98.1
101.5
107.8
103.1
103.8
110.7
           n.a.  = not  available
           n.d.  * not  detected

-------
       Table  3(b).
Ui
00
Fuel Oil Aging - Loss of Normal Paraffin
Location: Mid-Chesapeake Bay/New Point Comfort
Oil: 66 gallons No. 6 Fuel Oil
Release: 0830 hrs, DST, 6/23/71
Water Temp: 76°F.  Air Temp:
°L Normal
Paraffin
C9
CIO
Cll
C12
C13
C14
C15
C16
C17
CIS
C19
C21
C22
C23
C24
C25
2:10
6 3". 6
71.4
88.1
102.4
107.5
61.6
104.2
105.9
103.5
100.8
103.2
96.7
99.5
113.2
102.3
104.2
4:10
4.2
10.1
19.2
41.1
51.3
145.9
94.1
96.8
101.7
93.1
97.0
95.5
100.4
111.0
98.7
94.1
Time From Release (hrs)
6:10 8:10 10:10 12:10
9.9
26.9
48.0
78.6
86.3
114.0
102.3
99.8
105.4
89.3
99.1
97.1
98.3
110.8
99.7
97.4
n.d.
n.d.
n.d.
11.1
30.7
118.3
87.6
97.0
106.6
94.4
100.6
101.7
105.3
121.9
105.1
105.1
n.d.
n.d.
5.4
32.7
52.1
167.2
99.4
98.3
100.9
79.6
88.7
90.5
95.2
105.4
90.9
86.6
n.d.
n.d.
n.d.
12.8
35.1
137.4
89.8
94.7
99.5
83.9
91.9
89.5
92.9
102.4
88.6
89.8
26:00
n.d.
n.d.
n.d.
n.d.
2.6
23.3
26.8
31.3
76.4
75.7
94.9
90.7
99.3
110.6
93.1
96.7
28:00
n.d.
n.d.
n.d.
n.d.
n.d.
7.7
26.1
39.5
71.5
72.9
87.1
92.3
99.4
112.4
96.2
95.1
30:00
n.d.
n.d.
n.d.
n.d.
n.d.
8.8
13.6
27.6
70.7
72.0
85.1
93.6
100.5
111.8
94.0
95.2

-------
       Table 3(c).   Fuel Oil Aging - Loss of Normal Paraffin
                    Location: Chesapeake Light Tower
                    Oil: 200 gallons No. 6 Fuel Oil
                    Release: 1000 hrs,  DST, 9/14/71
                    Water Temp:  23.8°C.  Air Temp:  74°F.

                                              Time  After  Release  (hrs)
       % Normal Paraffin       2:15        4:15      6:00      9:00      23:00     25:00
O)
vO
CIO
Gil
C12
C13
C14
CIS
C16
C17
CIS
C19
C21
C22
C23
C24
C25
24.5
14.1
37.6
60.0
85.2
88.0
98.0
94.9
95.6
100.6
101.5
99.1
101.4
n.a.
n.a.
n.d.
6.2
22.9
43.3
72.2
82.8
94.7
91.1
106.0
97.5
100.5
98.3
98.8
103.1
104.8
n.d.
n.d.
9.1
25.0
56.1
69.5
85.6
86.7
98.1
95.9
100.5
98.2
98.8
101.3
103.8
n.d.
n.d.
9.7
32.4
60.9
81.2
90.4
94.0
100.6
97.9
102.7
97.1
102.7
103.1
104.2
n.d.
n.d.
n.d.
n.d.
n.d.
11.3
18.2
36.8
70.7
85.9
104.4
101.7
107.4
110.6
112.5
n.d.
n.d.
n.d.
nid.
n.d.
5.2
18.1
27.8
61.4
81.7
108.3
105.7
108.3
110.3
109.8

-------
       Table 3(d).   Fuel Oil Aging -  Loss  of Normal  Paraffin
                    Location: Chesapeake Light Tower
                    Oil:  200 gallons  No. 6 Fuel Oil
                    Release:  1015 hrs,  DST, 10/13/71
                    Water Temp:  71°F. Air Temp:  75°F.

                                            Time  After Release  (hrs)
       % Normal Paraffin       2:15        4:30        6:15         22:15        24:15
•P*
o
CIO
Cll
C12
C13
C14
C15
C16
C17
CIS
C19
C21
C22
C23
C24
C25
n.d.
69.9
20.1
37.2
61.8
73.3
85.1
91.5
102.4
98.7
102.2
99.2
106.0
117.7
117.7
n.d.
n.d.
11.2
27.2
56.3
74.7
86.2
95.4
96.4
96.5
101.2
97.0
100.2
107.7
111.0
n.d.
n.d.
9.7
35.6
63.4
81.5
82.3
94.5
93.6
96.0
102.1
100.3
100.0
107.5
104.5
n.d.
n.d.
n.d.
9.2
36.8
61.9
76.7
91.9
92.8
97.2
101.6
99.8
110.6
118.8
110.9
n.d.
n.d.
n.d.
14.5
40.8
66.9
79.4
89.7
93.1
97.3
102.1
98.2
101.2
102.9
104.8

-------
Table 3(e).  Fuel Oil
 Aging
: York
- Loss of Normal  Paraffin
             Location:  York River Mouth
             Oil: 200 gallons No. 4 Fuel Oil
             Release: 0900 hrs, EST, 11/6/70
             Water Temp: 15.4°C.  Air Temp:  60°F.
% Normal
Paraffin
CIO
Cll
C12
C13
C14
C15
C16
C17
CIS
C19
C21
C22
C23
C24
C25
1:00
117.8
93.0
98.9
99.9
126.0
105.6
106.7
108.5
107.5
106.4
102.8
99.2
101.1
105.0
95.4
1:50
73.7
87.8
103.5
100.6
115.3
105.7
104.2
103.5
117.2
104.3
101.7
99.3
101.6
109.7
104.3
Time After Release (hrs)
3:00 4:00 5:00 6:00
71.8
93.9
106.4
100.9
118.2
103.8
107.3
95.9
105.6
105.3
106.0
102.3
101.8
112.2
96.9
55.9
84.7
91.4
100.8
116.9
103.0
104.6
104.6
105.8
106.9
105.8
101.8
102.6
109.0
99.1
59.5
100.9
147.7
107.3
124.1
114.9
109.6
100.2
99.4
97.7
98.7
93.2
92.9
98.9
92.3
9.8
45.1
94.3
95.7
103.8
106.4
103.6
97.5
99.9
97.8
101.5
98.3
97.5
116.5
98.2
7:00
n.d.
31.6
50.0
74.2
113.0
95.2
104.8
109.7
103.8
109.5
106.5
103.5
100.5
102.2
101.8
8:00
n.d.
26.3
34.5
63.3
125.7
78.9
117.3
125.9
107.1
111.4
109.8
104.8
91.5
99.5
112.2

-------
Table 3(f).
Fuel Oil Aging - Loss of Normal Paraffin
Location: Chesapeake Light Tower
Oil: 200 gallons No. 4 Fuel Oil
Release: 0730 hrs, EST, 1/22/71
Water Temp: 3.85°C.  Air Temp:  47°F.
7o Normal
Paraffin 1:00
Cll
C12
C13
C14
C15
C16
C17
CIS
C19
C21
C22
C23
C24
C25
20.0
54.3
49.4
94.9
96.7
101.7
88.6
86.0
89.8
99.7
104.8
128.0
n.a.
n.a.
2:00
155.8
102.0
97.8
96.3
88.7
94.4
84.3
88.6
89.7
98.5
102.8
156.1
100.0
100.0
3:00
20.0
50.0
76.8
91.2
95.1
101.3
90.8
90.8
92.9
97.1
104.3
124.2
95.2
94.3
Time After Release (hrs)
4:00 5:00 6:00 7:00
n.d.
2.4
45.5
75.8
88.9
100.6
85.3
90.8
84.3
93.7
99.1
120.7
94.8
84.4
75.1
80.5
87.9
93.9
89.1
95.6
88.5
85.3
90.7
99.7
100.7
124.6
92.6
87.6
144.8
130.5
84.4
97.2
96.3
97.9
90.1
91.4
95.2
99.4
107.8
128.3
96.4
102.3
79.3
91.4
93.5
83.3
88.3
91.4
89.5
102.2
99.0
99.4
112.7
129.8
96.8
92.3
8:00
5.0
74.3
91.1
87.3
89.3
97.0
93.7
114.3
101.3
106.4
109.6
124.6
102.6
89.9
9:00
5.2
81.0
85.3
89.0
84.0
80.6
91.9
118.8
105.4
104.5
113.4
124.4
93.8
97.2

-------
Table 3(g).   Fuel Oil Aging - Loss of Normal Paraffin
             Location: Chesapeake Light Tower
             Oil: 200 gallons No. 4 Fuel Oil
             Release: 0800 hrs, EST, 4/26/71
             Water Temp: 12°C.  Air Temp:  62°F.
70 Normal
Paraffin 1:00
Cll
C12
C13
C14
C15
C16
C17
C18
C19
C21
C22
C23
C24
C25
105.5
152.8
141.5
112.6
106.0
104.6
100.5
104.4
100.8
99.9
112.9
95.8
103.2
99.7
3:00
49.7
89.0
99.9
104.0
105.0
99.2
101.5
101.3
100.3
99.9
113.5
98.4
111.2
98.1
4:00
68.1
108.0
107.5
105.4
106.1
96.0
100.4
98.4
99.6
95.3
107.2
87.3
103.4
89.3
Time After Release (hrs)
5:00 6:00 7:00 8:00
55.2
93.4
103.9
97.8
104.8
96.8
103.8
98.2
101.2
95.7
109.0
89,9
90.5
84.5
74.7
105.0
105.9
104.6
101.8
95.1
98.0
95.6
95.3
97.1
111.4
96.7
99.1
91.5
18.0
46.8
84.2
87.5
99.3
89.0
97.0
91.0
95.3
84.7
112.5
97.2
102.3
90.1
22.7
64.8
93.2
87.5
102.1
86.6
98.4
92.5
96.8
96.6
115.0
102.4
99.5
94.5
9:00
n.d.
28.8
72.7
81.7
97.5
86.8
96.6
92.2
98.2
93.1
110.6
96.6
95.0
85.7
10:00
11.0
49.5
78.4
86.1
101.3
94.9
100.3
100.0
100.2
96.8
109.5
95.5
93.9
88.7
11:00
n.d.
42.9
70.7
76.0
92.2
80.2
95.0
89.2
95.1
98.3
111.5
96.0
106.0
94.2

-------
Table 3(h).  Fuel Oil Aging - Loss of Normal Paraffin
             Location: Chesapeake Light Tower*Avg.  of two  analyses
             Oil: 200 gallons No. 4 Fuel Oil
             Release: 0830 hrs, DST, 8/3/71
             Water Temp: 25.5°C.  Air Temp:
% Normal
Paraffin
CIO
Gil
C12
C13
C14
C15
C16
C17
C18
C19
C21
C22
C23
C24
C25
1:00
75.7
89.7
96.1
109.3
84.4
92.4
95.7
95.5
98.0
96.1
103.0
101.8
101.2
98.5
100.5
2:00
53.6
77.7
84.1
108.5
89.5
96.9
99.7
92.2
98.6
100.5
100.5
103.4
96.8
94.8
96.9
Time After
3:30
9.4
50.3
64.0
101.2
83.5
95.8
99.9
96.6
98.7
100.4
100.7
99.8
89.0
95.8
100.8
Release
4:45
13.1
46.1
65.3
90.2
89.5
95.8
99.6
96.9
96.1
99.6
98.7
98.4
97.6
96.5
93.9
(hrs)
6:00
3.8
34.0
50.4
83.6
79.5
92.4
96.3
95.0
94.7
99.7
101.8
101.4
99.7
98.9
102.5
6:45
n.d.
5.0
46.6
75.9
80.8
92.3
98.5
96.6
98.6
98.9
100.0
99.3
98.7
100.0
100.9
9:00
n.d.
n.d.
32.8
71.2
74.1
87.8
95.3
92.8
94.7
95.6
98.5
97.1
96.0
93.7
98.6

-------
Table 3(h). (Cont'd).
Fuel Oil Aging - Loss of Normal Paraffin
Location: Chesapeake Light Tower
Oil: 200 gallons No. 4 Fuel Oil      *Avg.  of two analyses
Release: 0830 hrs, DST, 8/3/71
Water Temp: 25.5°C.  Air Temp:
% Normal
Paraffin
CIO
Cll
C12
C13
C14
C15
C16
C17
C18
C19
C21
C22
C23
C24
C25
10:00
n.d.
n.d.
21.9
28.9
65.9
87.3
96.4
93.5
94.8
97.6
99.0
96.5
95.8
95.2
99.1
Time
11:00
n.d.
n.d.
13.7
24.8
68.4
86.8
92.6
91.7
92.8
96.9
103.6
104.5
102.8
104.7
103.5
After Release
25:00
n.d.
n.d.
n.d.
3.8
21.8
58.9
83.8
86.7
91.8
95.0
101.2
100.3
100.4
100.4
100.6
(hrs)
27:30
n.d.
n.d.
n.d.
2.4
15.8
52.1
81.2
85.9
92.6
96.7
97.8
96.9
95.6
97.4
96.1
28:45
n.d.
n.d.
n.d.
1.4
14.5
42.0
75.8
82.9
92.6
94.4
98.9
98.7
95.0
96.7
98.2

-------
       Table 3(i).   Fuel Oil Aging -  Loss  of Normal Paraffin
                    Location:  Chesapeake Light Tower
                    Oil: 200 gallons  No. 2 Fuel Oil
                    Release: 0730 hrs,  EST, 1/20/71
                    Water Temp:  5°C.  Air  Temp: 40°F.
ON
% Normal
Paraffin
CIO
Cll
C12
CIS
C14
C15
C16
C17
CIS
C19
C21
C22
1:00
28.5
76.5
92.8
93.4
103.1
85.7
99.9
99.2
111.9
99.0
100.6
79.9
Time After Release (hrs)
3:00
17.1
73.6
92.2
n.a.
n.a.
n.a.
n.a.
101.9
114.1
100.1
100.0
86.7
5:00
n.d.
13.4
44.7
63.1
107.8
95.9
112.9
114.2
126.9
111.2
109.4
103.1

-------
Table 3(j).  Fuel Oil Aging - Loss  of Normal Paraffin
             Location: Mobjack Bay
             Oil: 200 gallons No. 2 Fuel Oil
             Release: 0845 hrs, EST,  3/16/71
             Water Temp:  8.3°C. Air Temp: 40°F.
7o Normal
Paraffin
C9
CIO
Cll
C12
C13
C14
CIS
C16
C17
CIS
C19
C21
C22
0:20
91.1
96.7
112.0
107.6
101.3
119.8
116.6
113.0
115.3
111.8
107.7
98.2
101.2
0:50
19.9
74.7
118.5
126.2
127.3
140.5
139.6
139.0
134.6
129.7
115.8
90.4
92.5
Time After Release (hrs)
1:25 1:45 2:00 3:00
n.d.
23.6
56.4
50.0
104.2
130.8
124.0
118.5
139.2
118.8
120.4
103.1
85.4
n.d.
3.8
13.1
23.4
59.2
100.8
112.5
121.3
139.8
121.6
122.5
89.0
87.2
n.d.
8.3
7.2
17.0
50.1
114.6
120.8
125.0
152.8
131.8
132.5
82.7
77.8
n.d.
n.d.
n.d.
n.d.
4.3
37.5
75.2
99.3
131.9
121.0
121.5
85.7
79.7
3:25
n.d.
n.d.
n.d.
n.d.
2.9
11.3
44.9
98.7
137.3
127.7
145.5
84.7
78.6
3:40
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
18.1
85.8
154.0
151.3
141.2
84.7
n.a.
5:25
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
0.9
23.6
88.5
113.1
117.8
85.8
82.7

-------
       Table 3(k)
CO
Fuel Oil Aging - Loss of Normal Paraffin
Location: Chesapeake Light Tower
Oil: 200 gallons No. 2 Fuel Oil
Release: 0900 hrs, DST, 8/30/71
Water Temp: 23°C.  Air Temp: 76°F.
7o Normal
Paraffin
CIO
Cll
C12
C13
C14
C15
C16
C17
CIS
C19
C21
C22
1:30
42.3
68.6
74.9
84.9
88.2
87.6
91.5
91.9
95.9
95.3
93.1
103.6
3:20
36.4
69.8
73.5
86.5
91.5
92.6
97.3
96.7
98.3
96.6
95.7
90.6
Time
4:45
8.0
28.6
43.8
72.8
87.7
88.9
98.5
95.4
97.2
96.6
86.2
103.6
After Release
5:45
18.6
55.9
61.7
83.5
90.9
90.5
95.6
93.6
98.1
94.0
80.5
80.5
(hrs)
6:45
n.d.
7.0
12.6
41.1
72.4
80.1
91.6
83.7
92.3
90.6
90.6
97.2
7:40
4.3
27.9
34.7
64.4
81.4
85.2
93.3
86.7
91.7
90.1
96.5
96.8
10:30
n.d.
20.9
31.5
74.6
94.9
110.5
124.1
109.3
113.4
106.0
80.5
80.5

-------
Table 3(1).  Fuel Oil Aging - Loss of Normal  Paraffin
             Location: Chesapeake Light Tower
             Oil: 200 gallons No. 2 Fuel Oil
             Release: 1045 hrs,  DST, 10/13/71
             Water Temp:          Air Temp:
% Normal
Paraffin
CIO
Cll
C12
C13
C14
CIS
C16
C17
CIS
C19
C21
C22
1:00
83.0
101.4
105.9
109.5
106.4
107.4
106.5
107.4
104.0
106.6
93.3
98.3
2:09
28.6
70.3
97.7
109.0
106.8
108.6
110.1
111.0
108.8
110.1
89.7
67.5
4:30
8.3
54.9
103.1
121.8
123.0
127.0
128.7
128.2
123.7
119.7
0.765
0.572
5:30
n.d.
28.8
77.7
94.8
103.4
107.5
106.3
107.4
106.7
109.4
90.3
81.3
6:30
n.d.
30.2
69.6
92.1
99.9
104.9
107.5
106.9
105.5
105.5
93.6
93.5

-------
The loss of oil components from an oil film is a complex
problem to treat mathematically.  A simple model is
considered here, as a basis for comparison with the aging
study.  The model system chosen is an oil film of fixed
volume and surface area which is losing components by
evaporation to an air stream passing over it.  Preliminary
assumptions are made that the heat input is sufficient to
maintain the oil film at constant temperature, and that^the
oil components are in continuous equilibrium with a finite
air volume above the film.  A further assumption is that
the total number of moles of all compounds in the oil film
is not appreciably changed by the evaporative loss of
volatile compounds.

Define:

     Cfc  = number of moles of component C in the oil
           at time t.
     K   = Henry's Law constant.
     M   = Total number of moles of all components in
           the oil.
     F   = flow rate of the air stream.
     V   = volume of air in equilibrium with the oil.
     P   = partial pressure of C in V at time t.

The rate of loss of C from the oil will be given by the
amount of C evaporated into V  per unit time.  Therefore,
using the ideal gas law:
Henry's Law states that the partial pressure of a component
in the vapor phase is directly proportional to the mole
fraction of that component in the liquid phase with which
the vapor is in equilibrium.  Thus:


                    p t = CfcK
                    *c     M

and since  V = Ft, the equation becomes:


                      dC m RFC
                      Dt   MRT

Integrating from t = 0 to t = t ,


                          = In Cfc - In C°
                          50

-------
And the ln(70C)t, that  is,  the  log  percent  C  remaining  at
time t, will be a linear function  of  t:

              ln(%C)t  - -  H|  + ln(100)
For aging experiments  in  the  field,  then one might  expect,
to a first approximation,  that  linear relationships between
log (%C*)t vs time would  be obtained.   Data of several
n-paraffin analyses  from  one  of the  field experiments are
shown plotted in the above manner  (Figure 14) .  These plots
of log  (%C^)t vs t are not linear, but curve downward.
This non-linearity may have resulted from a number  of
causes:

     1.)  The temperatures of the  oil films and air
          are not constant.

     2.)  The wind speed  in the field is not constant.

     3.)  Neither the  surface area nor the thickness of
          the oil film is constant,  because of the
          spreading  tendency  of unconfined oil on the
          sea surface.

The temperature of the oil film is not explicit in  the
model equation of oil  aging,  but the Henry's Law
proportionality constant  is an  exponential function of
temperature.  The gigher  the  temperature of the oil film,
the larger the value of K becomes.   The effect of
increasing wind speed  will be to increase the  magnitude  of
the factor F, but the  relation  may not be a linear  one at
high wind speeds.  Finally, there  is the problem  of
changing surface area  and thickness  of the oil film.
Although the assumption was made that the oil  components
are in  continuous equilibrium with the passing air  volume,
this situation will  be approached  only at low  wind  speeds
and very thin film thickness.  The effect of oil  spreading,
then, will be to increase the magnitude of the factor KF.

The effect of increased temperature, wind speed,  and
surface area, then,  is to increase the magnitude  of the
slope coefficient -  J™ ,  and to increase the  rate  at which

C  is lost from the  oil film.
                            51

-------
    0.4
                                       15
                       10      15      20
                       TIME (hrs.)
Fig. 14.  Loss of n-paraffins from No. 4 fuel oil release,
          Aug. 3, T971.
                          52

-------
In the field releases, which were generally initiated in
the cool, still, early morning, the usual conditions
experienced were increasing temperature, wind speed, and
surface area.  Thus, plots of log (%Ci)t show increasing
curvature as a function of time.

Laboratory aging of the three fuel oil types in the bubbler
aging apparatus was conducted to get information on the
relative importance of loss of oil components by
evaporation and by dissolution in seawater.  The amount of
material from the oil film which was dissolved in the
seawater after a period of aging was determined by
quantitative gas chromatography of pentane extracts of the
seawater.  This information is listed in Table 4.  The
extracts of the seawater which was in contact with the
three oil types produced qualitatively similar gas
chromatograms of the aromatic portions of the oils.
Figure 15 compares the chromatogram of the seawater extract
from the No. 4 fuel oil aging experiment with the aromatic
portion of the No. 4 fuel oil distillate fraction, b.p.
190-250°C/760 torr.


Table 4.  Dissolved oil components from bubbler experiment.

                  Concentration of           Percentage of
Fuel Oil           oil components              total oil

 No. 2            1.425 x 1CT3 g/1              0.048%

 No. 4            7.47 x 10'4 g/1               0.025%

 No. 6            1.26 x 10'4 g/1               0.004%
The major water-soluble components from all three oil types
were naphthalene and the isomeric monomethyl- and dimethyl-
naphthalenes.  The naphthalene and monomethylnaphthalene
peaks were identified from their mass spectra, and by
comparison of their gas chromatographic retention times
with those of authentic samples.  The identity of the
isomeric dimethylnaphthalenes was inferred from their
retention times relative to that of naphthalene and the
monomethylnaphthalene, with reference to reported gas
chromatograms (Boylan & Tripp, 1971).

The differences noted in the weights of dissolved oil
compounds among the three oil types is probably related to
concentrations of water-soluble aromatic compounds in the
original oil.  The No. 2 fuel oil contains the highest
                           53

-------
         #4  fuel oil
       aromatic  fraction
        #4  fuel oil
       aqueous extract
                 I
                200°C
240°C
 I
280° C
   SIMULATED  BOILING  POINT  (ASTM  D2887-70T)


Fig.  15.  Comparison of No. 4 fuel oil aromatic fraction
         b.p.  190°-250°C with No. 4 fuel oil water
         solubles.
                         54

-------
naphthalenes found  (218-262°C), whereas No. 4  and No. 6
fuel oils contain respectively less.

Several interesting features were noted of the character of
the seawater extracts from  the bubbler aging experiments.
One is the relatively low concentration of naphthalene in
the seawater.  In all three experiments,  the concentration
of naphthalene was  one  to two orders  of magnitude lower
than the maximum calculated solubility of naphthalene in a
33&o aqueous sodium  chloride solution, 1.9 x 10-2g/l  (Gordon
and Thome, 1967).  The other is the  noticeable absence of
the lower boiling,  rather water-soluble alkyl-benzenes.
Boylan & Tripp (1970) found considerable  quantities  of
these species in seawater which had been  equilibrated with
various types of fuel oils  and crude  oils in closed  systems.
It is postulated that in the bubbler  apparatus, and  also in
open water oil spills,  the  volatile water soluble aromatic
compounds are lost  preferentially by  evaporation.  The
evaporative loss might  be occurring directly from the oil
film, or could come from the scrubbing of the  seawater by
the bubbler, or wave action.

The possibility that dissolved aromatic compounds could be
scrubbed out of seawater solution by  interaction with the
air was demonstrated in a quantitative manner.  A solution
of cumene (isopropyl benzene, b.p. 152°C), naphthalene
(b.p. 218°C), and 1-methylnaphthalene (b.p. 241°C) was
allowed to age in the bubbler apparatus.  The  content of
these species in the seawater was determined by gas
chromatographic analysis of the pentane extracts of  the
seawater as a function  of time.  Figure 16 shows the plot
of log (% remaining) vs time for the  three species.  The
scrubbing action of the bubbler followed  an exponential
decay, and produced straight line plot.   The lowest  boiling
component, cumene,  was  lost beyond detectability by  the
process in less than four hours.  The naphthalene and
1-methylnaphthalene were lost at slower,  though appreciable
rates.

Therefore, in the bubbler experiments, and presumably in
the oil releases at sea as  well, the  lower boiling aromatic
compounds, though rather soluble in seawater,  are lost
through an evaporative  process, either directly, or  by
scrubbing of the water  column, leaving the less volatile,
higher molecular weight aromatics as  the  major dissolved
species.  Since solubility  in water of aromatic compounds
generally decreases as  the  molecular  weight and size of the
compound increases, it  seems that the naphthalene,
monomethylnaphthalenes, and dimethylnaphthalenes occupy a
prominent position  of favorable compromise between the
                           55

-------
                          I - METHYLN APHTH ALENE
                         NAPHTHALENE
                   4567
                   TIME  (hrs.)
Fig.  16.  Scrubbing of dissolved aromatic species in
         bubbler apparatus.
                    56

-------
opposing characteristics of volatility and  solubility  in
seawater.

The boiling ranges of the trapped evaporate fractions  from
the bubbler apparatus for the  three  oil  types were
determined by ASTM D2887-70T Boiling Range  Test by Gas
Chromatography.  Figure 17 shows the changes in the initial
and final boiling points with  time.  The results from  all
three oil types were similar,  so that single lines were
drawn to fit points from the three experiments.  The final
boiling point of the fraction  levels off in the vicinity of
270°C, while the initial boiling point continues to
increase.  Presumably, the lines will eventually converge.
Because of this behavior, the  term volatile fraction is
introduced to describe that fraction of  the fuel oil which
has a boiling point of 271°C or less.  The  boiling range
of the volatile fraction was defined to  include
n-pentadecane, b.p. 271°C as its upper limit, as this
compound was detected in the evaporate fraction.

Although the boiling ranges of the evaporate fraction  is
similar for the three oil types, the detailed nature of
their components is not.  Figure 18  compares the gas
chromatogram of the evaporate  fractions  from the three oil
types at comparable aging times.  It should be noted that
the three were recorded at different relative amplifications,
however.  The horizontal axis  is in  units of simulated
boiling point, derived from ASTM D2887-70T  instead of  the
more usual retention time.

The loss of the components from the  oil  is  shown in
Figure 19.  The loss of weight is greatest  for No. 2 fuel
oil, and respectively less for No. 4 and No. 6 fuel oils.
This result is no doubt related to the relative sizes  of
the volatile fractions, i.e.,  b.p. £ 271°C,  in the three
oils.

Although the derivation for model aging  which was treated
above considered the loss of a single compound, its main
point, that the aging process  should be  an  exponential
function of time, may be applied here.   If  the volatile
fraction is treated as a single entity rather than a
variety of compounds, one can  plot log (% volatile fraction
remaining) vs time instead of  (*/£$)t-  Such a plot is  shown
in Figure 20.  Here, the loss  of the volatile fraction is
not quite linear with respect  to time, and  shows curvatures,
particularly in the early hours of aging.   This presumably
is due to the loss of the extremely  volatile compounds.
The slopes of No. 4 and No. 6  fuel oils  become quite similar,
though that of No. 2 fuel oil  is larger, reflecting its
                            57

-------
       3001
     o
     o
     O
     O.
     d 200
     O
     ffi
     o
     UJ
       100
                                      INITIAL
                  10      20     30     40

                         TIME  (hours)
Fig. 17.   Change in boiling range of evaporate fraction
          in bubbler apparatus.
                       58

-------
                  n-Cn
     I50°C         200°C          250°C

SIMULATED  BOILING  POINT  (ASTM  D2887-70T)

Fig. 18.  Typical gas chromatograras of No. 2, 4,  and 6
        fuel oil evaporate fractions.
                    59

-------
   100
    86
                     20     30      40
                     TIME  (hrs.)
50
Fig.  19.  Loss of volatiles in bubbler apparatus.
                   60

-------
     2.00
      1.86
                                 NO. 4
                                     FUEL OIL
                       NO. 2
                          FUEL OIL
                 10      20     30     40
                       TIME (hrs.)
Fig.  20.  Volatiles remaining in bubbler apparatus.
                     61

-------
greater content of more volatile compounds.

A scheme has been devised to allow comparison of the  aging
of fuel oils in the bubbler apparatus with the open ocean
aging experiments.  Assume that, to a first approximation,
the evaporative aging is restricted to the loss of the
volatile fraction, i.e., b.p. ^ 271°C.  The percentage of
the fuel oil volatile fraction remaining in the oil film
at any given time t can be calculated from the values of
(%Ci)t listed in Tables 3(a) to 3(1) in the following
manner.  Coefficients a through e are calculated from the
boiling range compositlon diagram, Figures 9 through  11:

     a = % volatile fraction, b.p. * 196°C
     b = % volatile fraction, 196°C <* b.p. ^ 216°C
     c = % volatile fraction, 216°C <  b.p. ^ 235°C
     d = % volatile fraction, 235°C.<  b.p. * 253°C
     e = % volatile fraction, 253°C <  b.p. ^ 271°C

The actual values of the coefficients are listed in Table 5
for each type of oil.


Table 5.  Boiling Range Composition Coefficient
Fuel
No.
No.
No.
Oil
2
4
6

0.
0.
0.
a
080
000
000

0.
0.
0.
b
140
067
300

0.
0.
0.
c
230
233
250

0
0
0
d
.310
.367
.200

0.
0.
0.
e
240
333
250
Now, the percentage in each of the above portions of the
volatile fraction at time t is approximated by the gas
chromatographically determined (/>Ci)t, where C^ is the
n-paraffin which has its boiling point in that portion:
   volatile fraction remaining) t = a^C-ii)*. + b(%Ci9)r
  c(%C13)t + d(7oC14)t + e(7oC15)t.
The tabulated values of (%C4)t are used except
7oC^ > 100%, in which case the value is assumed
                                               where
                                               to be 100%,
The resulting data are plotted in the usual manner as log
(% volatile fraction) vs time.  The plots from each oil
release, along with the corresponding bubbler aging plot
are shown in Figures 21 through 23.  The slope of the
visually fitted straight line through the points for each
experiment was determined and recorded in Table 6.  The
values of the slopes gives a rough estimate of relative
                           62

-------
aging.  When correlated in this manner, the open water oil
films aged consistently faster than the bubbler films,
ranging from 2-180 times as fast.
                            63

-------
  0       5      10      15      20     25     30

                  TIME   (hrs.)
Fig. 21.   Loss of volatile fraction  of No. 6 fuel oil
          in bubbler and field experiments.

                          64
35

-------
o. 2.0
                                             BUBBLER
                                          4/71 j  11/70}  1/71 (A)
                     10       15      20
                         TIME  (hrs.)
  Fig. 22.  Loss of volatile fraction of No. 4 fuel oil in
            bubbler and field experiments.
                             65

-------
                                       BUBBLER (•)
0.0
                  10     15      20
                    TIME  (hrs.)
 Fig.  23.  Loss of volatile fraction of No.  2  fuel oil in
          bubbler and field experiments.
                          66

-------
Table 6.  Fuel Oil Aging Comparison.

                                  Slope                        Aging Rate
  Experiment            log (% remaining)  Units/hr          Relative to Bubbler
No. 2 Fuel Oil
Bubbler
1/71 Release
3/71 Release
8/71 Release
10/71 Release
No. 4 Fuel Oil
Bubbler
11/70 Release
1/71 Release
4/71 Release
8/71

2.94 x 10"3
2.45 x 10'2
5.28 x 10"1
2.09 x 10"2
6.67 x 10"3

1.96 x 10"3
7.02 x 10"3
7.02 x 10'3
7.02 x 10~3
2.33 x 10"2

1.00
8.33
179.59
7.11
2.27

1.00
3.58
3.58
3.58
11.89

-------
       Table 6. (Cont'd).  Fuel Oil Aging Comparison.
                                         Slope
Aging Rate
Experiment
No. 6 Fuel Oil
Bubbler
12/70 Release
6/71 Release
9/71 Release
10/71 Release
log (% remaining) Units /hr

2.27 x 10"3
2.13 x 10'2
3.66 x 10"2
7.66 x 10"2
2.50 x 10"2
Relative to Bubbler

1.00
9.38
16.12
33.74
11.01
00

-------
                      SECTION VI

             REMOTE SENSING OF OIL SLICKS

Oil remote sensing techniques have been applied to the
deliberate oil releases, natural films, and accidental
oil spills observed in this project.  Oil types included
menhaden fish oil and numbers 2, 4, and 6 fuel oils.

Overflights of oil slicks have been made by a C-54 and
a helicopter provided by NASA Wallops Station.  These
aircraft were equipped with multiple nadir view 9-in.
format Fairchild T-ll cameras.  In one experiment,
coverage was also provided by the University of Michigan
C-47 aircraft, which carried two 70 mm format cameras and
two multispectral scanners to collect data in 17 spectral
channels.

Remotely sensed and surface collected data have been used
to analyze oil slick temperature, thickness, and spreading
rate, and to evaluate techniques for remote detection and
discrimination of various oils.  Accurate surface measure-
ments have made the evaluation of remote sensing data
especially reliable.  The oil slick remote sensing program
is continuing and other observation frequencies and modes
are being evaluated.  Passive microwave radiometric oil
thickness measurements are now being processed by the Naval
Research Laboratory, which operates the instrument.


                REMOTE SENSING COVERAGE

There were twelve experiments with remote sensing coverage.
Film-filter combinations and non-photographic detectors are
listed in Table 7, which includes photography of No. 6 oil
from an accidental 10,000 gallon spill on May 6, 1971, from
the Amoco Refinery at Yorktown, Virginia.  Remote sensing
records are available at NASA Wallops Station.  Surface
data are located at the reporting agency.


                SURFACE DATA COLLECTION

Surface data was collected from small inflatable boats
stationed in the slicks, so observations were generally
conducted in fair weather with winds less than IS^knots
and wave heights less than one meter.  Meteorological data
were collected from the oil release vessel.  Extent of haze,
water color and turbidity, apparent current boundaries, and
biological features were noted.
                          69

-------
Table 7.  Oil Slick Remote Sensing Experiments

   Date	Oil Type	Coverage	

12 Sept. 69            6                  Tri-X/47B
                                          Kodachrome II
                                          8443/15

26 Sept. 69      6 and menhaden           8403/47B
                                          8443/1A

27 Oct. 69             6                  8403/47B
                                          Kodachrome II

12 Aug. 70             2                  2405/15
                                          8442/1A
                                          2403/47
                                          8443/15

6 Nov. 70              4                  2492/18A
                                          2403/57
                                          2403/47
                                          8443/12
                                          2424/89B
                                          SO-397/1A
                                          2443/12
                                          multispectral
                                            scanner

3 Dec. 70              6                  2403/47
                                          2448/pol.
                                          2448/1A

21 Jan. 71             2                  2403/47+pol.
                                          2445
                                          2445/pol.
                                          2443/12
                                          2403/18A

26 Apr. 71             4                  2403/47
                                          2403/57
                                          SO-397
                                          2443/12

6 May 71               6                  2443/1A
                                          2443/15,20B
                                          SO-397/1A

3 Aug. 71              4                  SO-397/1A
                                          2443/12
                                          2403/47
                                          2403/57
                                     microwave  radiometer
                                     infrared scanner


                          70

-------
Table 7. (Cont'd.).  Oil Slick Remote  Sensing Experiments

   Date	Oil Type	Coverage	

30 Aug. 71              2              microwave radiometer
                                           2443/1A
                                           SO-397/1A
                                           2403/57
                                           2403/47

4 Sept. 71              6                  2443/1A
     V                                      2403/47
                                           2403 /

 13  Oct. 71            2 and 6          microwave radiometer
                             71

-------
A floating thermistor device was used to measure water
surface and oil slick temperatures.  The thermistor
(L type, 5000 Q at 25 C, Applied Research Austin, Inc.,
Austin, Texas) was attached to the end of a spinal tap
needle mounted with vertical adjustment in a triangle
frame of stainless steel rod.  Flotation of the frame was
provided by a ping-pong ball glued to each apex.  The
design was similar to MarlattYs (1967), but did not damp
capillary waves, and should indicate more closely the
undisturbed surface temperature.  A fishing pole was used
to place the device at least one meter away from the boat
hull, and carried the wires leading from the thermistor
to the temperature readout unit.  The thermistor was
calibrated against a mercury stem thermometer, using the
freezing point of distilled water as a reference.  This
technique measured temperature in the upper 2 mm of the
surface layer.  Subsurface temperatures with the unit were
compared occasionally with those taken by stem thermometer.

Oil film thickness was calculated from the volume collected
in a 10 ml graduated pipette blown to the stem of a glass
funnel.  The funnel, when pushed through the slick, acted
as a thickness amplifier, and was calibrated over the
thickness range 0.5-2.5 mm with No. 2 oil.  Other funnel
pipettes were calibrated down to 30 /LOU thickness.  The
funnels were limited to use in calm water, produced erratic
field data despite good prevision in the laboratory, and
worked poorly with No. 6 oil which adhered to glass surface
,even when they were precoated with photo-flow solution.
More accurate thickness data for heavy oils can be obtained
by weighing retrieved film samples, or by solvent
extracting them for colorimetric analysis.  For thickness
less than 50 jum, appearance was converted to thickness
using the graph of Allen (1969), or Allen and Schleuter
(1969).

Each oil type used had its own visual characteristics,
which are described in the following paragraphs.  In
general, these characteristics were present in aerial color
photographs, but much of the fine detail was lost, for
example, the distinction between No. 4 from No. 6 fuel oil
was lost.

Number 2 fuel oil spread rapidly.  The slicks were thin
and symmetrical with interference rings near the edge, and
a large lens area in the center with an obvious yellow
color.  This lens was broken only by vigorous wave action.
Aggregation into lumps or fibrils was never observed.

Number 4 fuel oil spread more slowly, and the resulting
slicks were much less regular in shape.  Interference
colors were present around the slick edge and between areas
                           72

-------
where there were lenses of thicker  oil.  These  thin regions
were compressed by wave action and  formed  thin  brown
fibrils of oil that mixed below  the water  surface and
spread again when the forces  leading to  their compression
were removed.  The lenses were visually  thicker than with
No. 2 oil and were very dark  brown.  These slicks did not
fragment and clump.

Number 6 fuel oil was not observed  to  spread after initial
hydrodynamic energies of introducing the oil were
dissipated.  The interference portions of  the slick were
much smaller in area than for the other  two fuel oils.
The lens portions of slick were  difficult  to disturb and
appeared like a crinkled warm asphalt  when waves passed
under them.  Lens were of irregular shape  and their number
and extent depended on the manner of introduction of the
oil.  Number 6 oil had a high coherence  with itself, but,
when the lens were broken by  breaking  waves, small frag-
ments of oil did not reform the  lens structure.  The oil
mixed down into the water column as irregular fragments,
with the tendency to fragment increasing in experiments
conducted at lower water temperatures.

The appearance of these oils  should vary with the source,
but the physical characteristics, such as  viscosity, pour
point and other properties of the fuel oil types are
sufficiently different to allow  visual distinction on water.
This situation might be changed  as  the oil ages, but the
oils generally dissipated before obvious changes occurred
in these small releases.  The only  aging change was that
No. 6 fuel oil aged over 1 day at sea  was  present in
fragments with much less thin oil and  interference rings
than was present immediately  following release.


                REMOTE SENSING ANALYSIS

Photographic film was developed  by  the laboratory at NASA
Wallops Station and their contractors.   Images  were not
rectified for aircraft attitude.  Multispectral scanner
data were reduced at the Willow  Run Laboratories of the
University of Michigan.  Data from  all channels were
processed to a continuous-tone photographic format.
Ultraviolet and thermal infrared signals were processed
by voltage-slicing to give a  radiance-contoured
photographic format.  The contour intervals were assigned
ranges of equivalent black body  temperature based on
scanner readings of two thermostated black body reference
plates (Hasell and Larsen, 1968).
                            73

-------
             INFRARED SCANNER CALIBRATION

Infrared scanner temperatures from the oil release of
November 6, 1970, have been compared with temperatures
simultaneously measured on the surface.  The scanner was
operated at 2000 ft altitude at 0900 and 1145, after oil
release at 0840.  Corresponding surface temperatures were
measured at 0900 and 1200.  The tabulation below shows the
scanner temperature (Teq), and the temperature of the top
2 mm layer of water (Tsw) .

Aircraft Pass      Time         Teq        Tsw         AT

    1              0858      14.2-14.5    15.4       0.9-1.2

    2              0902      13.2-13.6    15.4       1.8-2.2

    3              1144      13.5-14.3    15.5       1.2-2.0

Between passes 1 and 2, Tsw was constant, Teq varied by 1 C,
the instrument reference temperature varied by 2.5 C and
there were apparent power supply problems.  Passes 1 and 2
can be used only as indicators of relative temperatures.
There was no difficulty in calibration of pass 3, so its
AT value should be attributable to water emissivity value,
reflection of sky radiance, absorption and emission in the
atmospheric path to the passive infrared sensor, and
vertical temperature gradients near the water surface.

The surface air temperature was generally lower than Tsw
and the water temperature increased to 16.7 at 20 mm depth.
This temperature gradient is caused by evaporation, and
water-to-air heat conduction (McAlister, 1969).  Assuming
the gradient continues linear from 2 mm depth to the 20 mm
infrared measurement depth, 0.15 C of the observed AT is
explained.

The Teq value is uncertain because of poor accuracy of
water emissivity and sky radiance information.
Emissivities reported range from 0.970 to 0.993 (Buettner
and Kern, 1965; Buettner, Kern, and Cronin, 1965; Saunders
and Wilklns, 1966; Saunders, 1967; Griggs, 1968; Lee, 1969;
Anding and Kauth, 1970) and caused an uncertainty in Teq
greater than the observed AT.  From the field data of Weiss
(1963), Oshiver, et al. (1965), and Saunders (1967), 0.5 C
of AT  is explained" By the above uncertainties.

Atmospheric radiation absorption and emission a 1.0 C
error in Teq at 2000 ft altitude, as indicated by Oshiver,
et al. (1965), Gamier (1971), and Marlatt and Harlan (1971)
                           74

-------
This explains 1.0 C of AT.  The total explained  AT is
1.65 C, which is close to the observed value  and  the  AT of
2.2 C given by the equation of Pickett  (1966).   It is
apparent that Run 3 contains a consistent  temperature data
set.
                OIL SLICK TEMPERATURES

In slicks of No. 4 and 6 oils, oil  surface  temperatues
(Ts°) were equal to Tsw for thin  films.  Tso was up to
7.5 C higher than Tsw for millimeter  thick  portions of the
slicks.  The greatest temperature differences were at
mid-day for slicks several hours  old.  Subsurface open
water temperatures were lower than  the corresponding values
under warm oil.

During the 0848 pass on November  6, 1970, Ts° equaled Tsw.
Ts° for thick regions was measured  80 minutes later, but
still was only 0.3-0.6 C above Tsw.   Comparison of infrared
and ultraviolet scanner outputs showed that the infrared
did not record slick edges.  The  scanner temperatures for
oil and water (Teq° and Teqw) differed by up to about 3.0 C,
but differences were not consistent due to  scanner
calibration problems.

For the 1145 pass Ts° ranged from 15.5 C in thin films to
23.0 C in thick oil lenses.  In moderate oil thicknesses,
TeqO was less than Teqw by 0.9-2.5  C.  This difference was
reversed for very thick oil.

The Teq° - Teqw differences can be  explained in terms of
the temperatures and emissivities (eo and  ew) of oil and
water.  Buettner, et al. (1965) reported eo for a oil film
of unknown type ancTthTckness as  0.972, compared to
ew = 0.993.  In view of the uncertainty in  ew,  eo = 0.972
must be regarded as a rough estimate, and using this value
with Ts° - Tsw gives calculated differences considerably
smaller than those observed.  The proposed  eo value is
evidently too large, and a dependence of eo on oil
thickness is indicated, as suggested  by Chandler (1970).

The rise in Tsw with time due to  solar heating was less
than for Ts° of thick oil, because  water has a lower
visible light absorbance than oil.  It has  been suggested
(Chandler, 1969; Estes and Golomb,  1970; Stewart, et al.,
1970) that the thermal infrared information might correlate
with slick thickness, but such correlations must be
ambiguous because of surface temperature variations and
uncertainty in e values and in the  unknown  eo - film
thickness relation.
                           75

-------
                  OIL FILM THICKNESS

Thickness of films was measured on November 6, 1970,
December 3, 1970, and January 21, 1971, and in later
experiments which involved the NRL microwave radiometer.
Stewart, et al (1970) have shown that oil spectral radiance
changes with thickness are small for films greater than
0.5 mm thick and that dark oil should be at least this
thickness.  Thickness of dark regions were 0.1-0.8,
0.1-2.4, and 0.3-0.6 mm for numbers 2, 4, and 6 oil,
respectively.  These values permit a photographic estimate
of the minimum volume of oil in dark areas of heavy oil
spills.


       DETECTION AND DISCRIMINATION OF OIL TYPES

Photographic and scanner imagery have been examined
according to spectral region and observed oil type, but
number- 4 oil was the only type studied in all bands.
Photographic sensing bands are identified here by the
Kodak film - Wratten filter combination used.  The term
positive contrast means that oil displayed more radiance
than water.  Some photographs from this program are given
by Munday, et aL (1971).

In the near ultraviolet (0.32-0.40 jum, 2403, 2492/ISA),
number 2 oil showed positive contrast.  Interference rings
were expected, but not detected.  Number 4 oil showed
interference rings, sharp slick edges, large positive
contrast for films thinner than 50 jum, and small negative
contrast for thicker regions.

In the blue band (0.40-0.50 jjn; 2403/47), No. 2 oil showed
strong interference rings, and moderate positive contrast
in the main body of the slick.  There was little radiance
variation across the slick, so thickness regions were not
distinguishable.  Numbers 4 and 6 oil also showed strong
interference colors.  Thin regions showed moderate positive
contrast and thick regions moderate negative contrast.

In the green-yellow band (0.50-0.62 iim, 2403/57), numbers
4 and 6 oil showed moderately.  Thick regions showed large
negative contrast.

In the red-near infrared band (0.62-0.80 jum, 2424/89B),
number 4 oil was barely visible and thick regions showed
small negative contrast, so oil detection was not
practicable.
                          76

-------
Ordinary color (SO-397, 8442, 2448/1A), infrared color
(8443, 2443/12i 15) and infrared color with haze filter
(8443/1A) photography of oils was tested.  Infrared color
was difficult to expose properly and gave poor contrast.
Ordinary color for No. 2 oil showed interference rings but
did not distinguish thick and thin regions.  For No. 4 and
6 oil the interference rings were seen, edges appeared
light blue, thin regions appeared blue, yellow or brown, and
thick regions appeared gray or black.  The infrared color
with haze filter gave the best image and color contrast.
For number 6 oil, interference rings showed well, thin
regions were pink with positive contrast, thick oil dark
red.  Menhaden fish oil was white on a magenta water
background, thin oils were not imaged, and, as noted by
Barringer (1968), intermediate thicknesses were not
present.

Photography is useful for oil slick detection and
discrimination, a conclusion at variance with that of Estes
and Golomb (1970), who state that oil  slicks do not
photograph well on color film.  Photographic results here
agree with Lowe and Hasell  (1969) and  with the radiance
model of Stewart, et aL (1970) which predicts a change from
positive to negative oil/water contrast at 0.4 jum wave
length, a value close to the value 0.46 /urn derived from
observations in the present study.  The results also agree
with Catoe's (1970) statement that thin slicks are detected
efficiently by UV-blue band photography.

The near ultraviolet band is best for  imaging edges and
thin slicks, while the green band is best for delineating
thick oil regions.  Number  2 oil can be distinguished from
No. 4 and 6 because it does not show negative contrast in
the blue and green bands.  Menhaden oil can be distinguished
from the fuel oils by its lack of interference colors.

Ordinary color film is a good accessory record for all oil
spills.  It distinguishes thick from thin oil, and helps
differentiate between oil and natural  slicks.  Color
infrared film may sometimes be needed  to distinguish
between oil and floating vegetation.


                  OIL SLICK SPREADING

Slick areas were measured with a polar planimeter on Tri-X
film with a Wratten 47 filter or on color film.  The slick
edge was taken as the edge  of the outer dark interference
ring.  True areas were calculated from vessel lengths in
photographs.  Hourly area values were  used to calculate oil
spreading rates, which were compared with rates predicted
                          77

-------
by the model of Fay (1969).  This model predicts three
phases of spreading of point oil releases on calm water.
In phase three, spreading is caused by surface tension,
and retarded by viscous drag of the boundary water layer.

A small spill should quickly enter, phase three, whose
spreading law is  r = k (a2t^/p2v)^ , where r is the slick
radius, k is a constant, a is the surface tension spreading
coefficient, t is time since release, p is water density
and v is water kinematic viscosity.  Assuming slick area is
7rr2 and using the above equation,

                      A =  at 3/2

where a represents a group of constants including k.  Thus,
log A versus lot t should have an intercept of log a and a
slope of 1.5.

A vs t has been plotted on a logarithmic scale for releases
on August 12, 1970, November 6, 1970, December 3, 1970,
January 21, 1971, April 26, 1971, August 3, 1971, August 30,
1971, September 14, 1971.  This plot is shown in Figure 24.
A line of slope 1.5 has been drawn through the points.  The
intercept of this line, -1.072, was calculated from the
average of log a. values for all data points and differs by
a factor of 6 from that used by Fay.  However, his value
involved an estimation of k without experimental verification,

The fit of the line to the data suggests that the phase
three spreading law is applicable.  Guinard (1971) found
that a linear spreading law fit his observations, but no
linear relation has appeared in experiments here.  At short
times after release, there is scatter in the data points
because the releases actually were not point sources of
oil, but were released over a 5 to 10 minute interval, and
because near the start of spreading the phase two conditions
stated by Fay may be more realistic than the phase three
assumption.

The number 2 oil spill on January 21, 1971, was made in high
wind conditions.  Only one data point was obtained before
the slick was broken up.  Number 6 oil did not spread in a
regular and monotonic manner.  Much of the oil remained in
fixed patches, perhaps due to the high oil viscosity.
Fay's model does not treat the effect of oil viscosity, and
he acknowledges (personal communication) that there is some
oil viscosity limit above which his model fails.  The lack
of spreading of number 6 oil is a useful criterion for its
remote sensing detection, particularly in the case of
recent spills.  Number 4 oil, spilled on November 6, 1970,
                           78

-------
N
E
(T
           ©
           o
           D
#2 OIL (Aug '71)
    OIL (Aug '70)
    OIL (Jan '71)
-#4 OIL (Apr '71)
•#4 OIL (Nov '70)
    OIL (Aug '71)
                      a
      10s
                                             10=
                          TIME   (seconds)
    Fig. 24.  Spreading rate of  No.  2 and No. 4 fuel oils
              on  sea water.
                            79

-------
failed to spread regularly after the first hour.  This lack
of spreading may be caused by chemical confinement by
natural surface-active organic material or by physical
confinement between current convergence zones.  These
factors are likely to be operative in the estuarine water
of the York River, where the release was made.  All other
number 4 and 2 oil spills spread regularly.

Other spreading models (Murray, et aL, 1970; Blokker, 1964;
Abbot and Hayashi, 1967; Berridge, et aL, 1968)  should be
applied to the spreading data, but lEEe nearly calm
conditions during experiments indicated that Fay's model
was most applicable to the present work.
                          80

-------
                      SECTION VII

    PHYTOPLANKTON IN SURFACE SLICKS AND IN ADJACENT

            SUBSURFACE AND NON-SLICK WATER

The surface microlayer of the ocean is being recognized as
an environment which possesses certain unique chemical,
physical, and biological properties.  Of these, the first
two have, to date, been given the most attention.

Both naturally occurring and man-made surface films have
been investigated.  Blanchard (1964) proposed that the
surface active organic material has its source in the
dissolved organic matter of the sea, and that it is
transformed into particulate form and transported to the
surface by bubbles.  Sturdy and Fischer (1966) studied the
surface tension of slick patches produced by large beds of
kelp.  Jarvis, Garrett, Schieman and Timmons (1967)
collected samples of surface active material from different
locations and found that all the material collected was
similar in surface properties, indicating chemical
similarities.  Further study showed slick material to
contain fatty esters, acids, alcohols, and hydrocarbons
(Garrett, 1967a), as well as high concentrations of organic
carbon, nitrogen, and phosphorous.

The need for better understanding of the biology of the
surface microlayer has only recently been stated.  Parker
and Barsom (1970) stressed the probability that the micro-
layer was of considerable ecological importance because of
the interaction between this layer and the air-sea
interface.  Ahlstrom (1969) stated that although surface
films and slicks must contain communities different from
those in the rest of the water column, there is little
known about these organisms.  David (1965) pointed out
that the surface layer is of particular significance
because the local environmental conditions are liable to
such great and rapid changes.

Biological investigations of surface layers have been
limited by problems associated with obtaining an adequate
surface sample.  Recently, some sampling devices have been
developed which helped overcome this difficulty.  Harvey
(1965) originated a rotating drum sampler capable of
collecting a layer of water and surface film from the upper
60 microns of sea surface.  He compared samples taken by a
bucket dipped to a ten centimeter depth with samples taken
by his drum sampler, and found a higher concentration of
organic material, including live phytoplankton, in the
                          81

-------
surface microlayer.  A second type of sampling device, a
screen apparatus, has been successfully used by Garrett
(1965).

The present study investigates the phytoplankton found in
the surface microlayer of the water column of an estuary.
A Garrett-type screen collection is used to make a
quantitative comparison between the phytoplankton present
in this thin layer, and that found in the water one meter
below.  It also attempts to discover phytoplankton changes
brought about by the presence of a film of slick material
over the water surface.
                  SAMPLING PROCEDURE

Collections were made in two areas (Figure 25):  the York
River (I), near the Virginia Institute of Marine Science
(VIMS), and, in one case, close to the Chesapeake Bay
Bridge-Tunnel (II).  Collection intervals were irregular
because sampling was possible only in calm weather.  A
total of 10 sets of samples were collected on the dates
shown in Table 5.

Each set of collections consisted of four samples: a pair
of surface and subsurface samples in an area of water
covered by a monolayer (designated a "slick" area); and a
similar pair of samples in an adjacent "normal" or
"non-slick" area.

One set of samples departed from this procedure.  In
Collection 3 a non-slick environment was sampled, then a
quantity of No. 2 fuel oil was released in the area.
Another set of samples (corresponding to the usual slick
samples) was taken after 30 minutes.


                   SAMPLING DEVICES

Surface samples were taken with a screen device constructed
and used according to Garrett (1965).  It consisted of a
24 in. square of 16 mesh monel screen in a brass frame with
upright handles and cross-bar.  It is claimed that the
upper 0.15 mm of surface water is sampled by this method,
that sea slick material can be collected in sufficient
quantity for study, and that the method is about 70 percent
efficient after the first dip of the screen.

The screen was dipped into the water, withdrawn, and
drained through a funnel into a wide-mouthed polyethylene
jar.  Ten dips were required for a sample volume of
                          82

-------
10    20    30
   I   I  I
 NAUTICAL MILES
Fig.  25.   Plankton sample areas
               83

-------
approximately one liter.

Subsurface samples were taken at a depth of one meter with
a one liter Frautschy bottle.

The sampling problems presented by the irregular horizontal
distribution and abundance of phytoplankton are well known
(Ahlstrom, 1969; McAlice, 1970).  Patchiness makes it
nearly impossible to obtain an accurate representation of
the phytoplankton in a given area from a single sample.
Holmes and Widrig (1956) stated that in order to reduce the
time required for the analysis of additional samples, the
samples may be pooled and treated as one, without damaging
the precision of the estimate.  Sample pooling has been
used in this study.  In each surface collection the ten
dips of the screen were combined into a single one liter
sample, and for every subsurface sample three Frautschy
bottle casts were made and pooled into one sample.


                COMPARISON OF SAMPLERS

In order to make quantitative comparisons of surface and
one meter collections, it was necessary to compare
Frautschy bottle and screen sampler cell counts.  A 420
liter tank was filled with filtered seawater, and a
unialgal culture of the diatom Phaeodactylum tricornutum,
of a known concentration, was added.  The water was
agitated to insure a homogeneous distribution of cells
throughout the tank, and samples were taken at the
surface with the screen, and at one meter with the
Frautschy bottle.  A second set of samples was taken after
remixing.  The experiment was later repeated with a
different volume of algae.  All samples were counted on the
inverted microscope.  There were no significant differences
between the samples obtained with the two devices when the
standard F, or Variance Ratio, Test was applied.  Both
samplers gave good estimates of the population in the tank
(Table 8).
             QUANTITATIVE SAMPLE TREATMENT

All field samples were stored in wide-mouthed polyethylene
jars, preserved with five percent neutralized formalin,
and returned to the laboratory for concentration.  After
three days of settling the supernatant was siphoned off,
leaving a volume of approximately 500 ml.  Further
concentration reduced the volume to 150 ml.  An additional
procedure was carried out for Collection 3A (the surface
slick sample taken 30 minutes after an oil spill) .  After
                          84

-------
        Table 8.  Results of Sampler  Comparison Experiments-?-
oo
Exper .
I
II
III
Screen F. Btl. Actual #
Cell Cell Cells Screen F. Btl. F
Counts Counts in Tank Est. Est. Value
Cells /ml Cells /ml Cells /ml Cells /ml Cells /ml
2,014
1,862
1,475
2,038
1,593
1,421
1,681
10,010
12,118
11,760
1,912
1,970
1,446
2,102
2,372
2,102
2,087
12,871
13,661
10,932
2,000 1,847 1,857 81,579 ^ i o06 N s
67,659 ' '
2,000 1,565 2,187 25,725 _ -, 471 „'
17,48> i'4/i N-0>
10,000 11,296 12,488 197,187,700 _ -, ,, 7 „ q
136,274,535 L'^! n§s'

-------
settling, the No. 2 fuel oil formed a layer at the top of
the sample jar.  A portion of the oil was pipetted off and
examined separately from the rest of the sample.

Phytoplankton cells were enumerated quantitatively by the
Utermohl method, which the National Academy of Sciences
(Ahlstrom, 1969) called "probably the best method known
for this purpose.

A detailed treatment of the statistical procedures involved
in the Utermohl method is found in Lund, Kipling and
Le Cren (1958).  Assuming a random distribution of cells,
the counting and sampling errors can be estimated.  The
Chi Square (X^) test for randomness was applied to ten sets
of five replicate counts each.  Since none of these had a
significant X^ value (Table 9) the hypothesis of random
distribution was supported.  An estimate of the variation
due to subsampling was made by counting five subsamples
from each of five different samples.  Confidence limits at
the 95 percent level were ±3.7, ±3.3, ±2.5, ±4.8, ±3.6
percent (Table 10) .  The personal counting error was
determined by ten replicate counts of a single subsample
(Table 11) , and the confidence interval at the 95 percent
level was ±2.9 percent.  Since this variation was within
the range expected from a series of random samples, this
source of error was insignificant.

For every concentrated field sample three one ml subsamples
were counted.  Each subsample was pipetted into a
cylindrical chamber with an inside diameter of 25 mm, and
allowed to settle overnight.  Counts were made of a 100 mnr
section of each chamber.  The median number of phytoplankton
cells per ml of water sample was calculated by the formula:
where N = number of organisms per ml of original water
          samples
      n = median number of organisms from the three
          counts
      c - volume of the concentrate in ml
      s = original volume of sample before concentration
      x = area of bottom of counting chamber
               area of counted portion

Ricker (1937) gave a formula for calculating the confidence
intervals of a count, and Lund, et al (1958) stated that if
the organisms have been shown to be randomly distributed,
the confidence limits can be used to compare counts.  There
is a significant difference between two counts if their
                          86

-------
00
•-J
        Table 9.  X2 Test for Random Distribution.


               I       II     III      IV       V      VI     VII      VIII     IX      X
1.
2.
3.
4.
5.
X2
163
166
173
178
167
.86
N.S.
1130
1212
1197
1222
1190
4.3
N.S.
982
1002
1032
1028
1016
1.7
N.S.
345
369
378
375
357
1.3
N.S.
1328
1446
1334
1372
1361
6.5
N.S.
1189
1198
1200
1155
1221
1.9
N.S.
1772
1790
1718
1792
1834
5.2
N.S.
337
347
327
360
332
2.0
N.S.
2586
2398
2551
2546
2537
8.3
N.S.
1902
1912
1868
1931
1899
1.1
N.S.

-------
Table 10.  Counts of Five Replicate Subsamples from
           Samples.  Figures represent cells/ml.

                        B
Five
1.
2.
3.
4.
5.
Mean
Stand, dev.
Stand . error
Interval est.


1130
1212
1197
1222
1190
1190
35.89
16.05
±44.55
or
± 3.7%
1790
1718
1792
1834
1732
1773
37.63
21.30
±59.13
or
± 3.3%
982
1002
1032
1028
1016
1012
20.45
9.10
±25 . 26
or
± 2.02%
337
347
327
360
332
341
13.13
5.87
±16.29
or
± 4.8%
354
369
378
375
357
367
10.70
4.79
±13.29
or
± 3.6%
Table 11.  Counts of Ten Replicate Subsamples  from One
           Sample.  Figures represent cells/ml.










Mean
Stand, dev.
Stand, error
Interval est


1.
2.
3.
4.
5.
6.
7.
8.
9.
10.



•


770
763
812
798
733
801
774
807
746
724
773
31.
9.
±22.
or
± 2.











5
9
39

9%
                          88

-------
confidence intervals do not overlap.  This criterion was
used to compare samples.


                 POPULATION STATISTICS

The phytoplankton cells in each sample were enumerated and
identified, where possible, to species.  The percentage
composition of the samples was determined by a method
recommended by Morse (1947) .  The first 200 cells of each
sample were identified and the total number of each species
divided by two to find the percentage for each species.

Diversity was calculated according to Shannon (1948):

                 H1 = - E p± Iog2 p±

where H1 = a proportion between the number of individuals
           of each species and the total number of
           individuals in the sample.

      Pi * n£ / N.

      n^ = number of individuals in the ith species.

      N =  total number of individuals in the sample.

This index is cited by Pielou (1966) as appropriate for
situations, such as plankton samples, in which a collection
is too large for all of its members to be counted.  It is
improbable that all species present in the samples were
identified since the number of cells enumerated was limited
to 200, but, when H1 is used as a measure of diversity, the
rare species have little influence on the result.

Further comparison between samples was made using Sander's
(1960) dominance-affinity index.  Two samples are compared
by computing the percentage of the total sample represented
by each species present in both samples, and summing the
smaller percentage for each species.  The resultant value,
the index of affinity, is a measure of the percentage of
organisms common to the pair of samples.


                RESULTS AND DISCUSSION

Table 12 summarizes results of plankton counts.  With one
exception (Collection 7) all non-slick samples showed an
abundance of phytoplankton that was 1.5 to 5.4 times greater
than in samples from adjacent slicks.  Similarly, nine of
                          89

-------
Table 12.  Results of Sample Counts, in cells/ml.
Collection    Location       Date                Slick
Non-Slick
#
1 I
2 II
3 I
4 I
5 I
6 I
7 I
8 I
9 I
10 I

29 VIII 69
25 IX 69
12 XII 69
22 IV 70
1 VI 70
28 VII 70
28 VII 70
17 IX 70
17 IX 70
17 IX 70
Surface
1561
250
284
4700
421
665
870
1356
1686
62
1 Meter
674
245
351
1920
1268
658
891
523
460
55
Surface
2074
492
544
8281
2257
1000
732
2202
3026
131
1 Meter
1312
207
316
1555
781
483
619
355
721
64

-------
ten non-slick surface samples had a greater abundance of
phytoplankton than the-corresponding samples from one
meter--ranging from 1.2 to 6.2 times greater.  The
reliability of Collection 7 was doubtful because it was
collected in unfavorable weather.  Confidence intervals
(Table 13) show that in all cases these differences were
statistically significant, since their intervals never
overlapped.

The cell count in the fuel oil fraction of the slick sample
in Collection 3 was compared with that in the normal slick
sample.  It was found that the oil fraction contained a
higher number of cells.

The total number of cells in each sample is divided into
its component parts (Table 14) to show the percentage of
the sample that is composed of diatoms, dinoflagellates,
and other types of phytoplankton (including silicoflagel-
lates and Euglenophytes).  The diatoms displayed peaks of
population in April and August-September.  Dinoflagellates
reached a population maximum in August, with a lesser
increase in June.  The August dinoflagellate abundance is
not from the regular sampling program, but represents a
surface sample taken during a dinoflagellate bloom.

Ninety-eight species were identified.  Of these, 36
occurred with sufficient frequency to be considered major
components of the flora.  Skeletonema costatum was the
dominant organism in 92.5 percent of the samples, and the
second most abundant species in an additional 27.5 percent.
It was totally absent from only two samples, or five
percent of the total.
                                    T[
Comparison of diversity values for slick and non-slick
areas did not show any significant trend.  Comparison of
values from surface and one meter samples showed that the
diversity in the surface microlayer was generally lower
than at one meter, as can be seen in Figures 26 and 27.
Comparison of sample pairs (surface slick and non-slick;
non-slick surface and one meter) by the affinity-dominance
index are given in Table 15.  Values range from 53.5 to
100 percent.  Over half of the sample pairs have affinities
over 75 percent, indicating a very homogeneous flora.

Chemical analysis of two of the slicks showed that they
consisted of naturally occurring slick material.  It is
probable that the other samples (except Collection 3) were
of similar composition.
                          91

-------
         Table 13.  Confidence Intervals, at the 99% Level, for Sample Counts.
         Collection
Surface
Non-Slick
ro
# Slick
1
2
3
4
5
6
7
8
9
10
1172
212
243
4526
371
601
797
1264
1583
44
- 1357
- 295
- 331
- 4880
- 478
- 735
- 950
- 1455
- 1796
- 86
Non-Slick
1959
438
487
8049
2137
921
665
2085
2887
104
- 2195
- 553
- 608
- 8519
- 2383
- 1085
- 806
- 2328
- 3172
- 165
Surface
1959 -
438 -
487 -
8049 -
2137 -
921 -
665 -
2085 -
2887 -
104 -
2195
553
608
8519
2383
1085
806
2328
3172
165
1 Meter
1069 -
173 -
273 -
1456 -
712 -
429 -
557 -
309 -
665 -
46 -
1245
248
366
1661
857
543
686
408
794
89

-------
Table 14.  Composition of  Surface  Samples,  Slick and
           Wnrn-ol -I r»tr
Non-slick.
Sample #
1A
1C
2A
2C
3A
3C
4A
4C
5A
5C
6A
6C
7A
7C
8A
8C
9A
9C
10A
IOC
Diatoms
1210
1588
194
475
275
506
4606
8199
246
902
572
580
644
681
1288
2203
1686
3026
53
112
Dinoflagellates
226
238
56
17
9
30
71

139
1140
93
420
226
51
68
_
_
.
7
14
Other
125
•&» «^
248
*•» V \f


8
23
•M*^
82
\J AM
36
214

^
_
_
_
^
_
—
2
5
           Slick      C = Non-slick
Table 15.  Affinity  Index Values  for  Paris  of  Samples.

Collection #         Slick/Non-Slick        Surface/1 Meter
                        (Surface)    	(Non-Slick)
1
2
3
4
5
6
7
8,
9
10
75.0%
74.5%
64.5%
89.5%
60.5%
71.0%
78.5%
94.0%
94.0%
71.0%
53.5%
99.0%
71.5%
77.0%
80.5%
74.0%
82.5%
100.0%
100.0%
72.5%
                           93

-------
     3.51
     3.0-
     2.5-
  >


  5  2.0-
  o
  t  1.5-
  0)
  DC
  UJ
     1.0-
     0.5-
     0.0-
         SURFACE

         I  METER
I    I    |    I
i   n   ni  12
                               3Zi  3zn. 3znr TK
                     COLLECTION  NO.

Fig.  26.   Comparison of surface and one meter diversity
          in slick areas.
                       94

-------
      3.5-1
  _i
  <
      3.0-
      2.5-
      2.0-
  ^  1.5
  CO
  o:
  UJ
      1.0-
     0.5-
     0.0
n
     SURFACE
     I  METER
                        rz:  2  si mr snr ix  z
                     COLLECTION  NO.
Fig.  27.  Comparison of  surface and  one meter diversity
         in non-slick areas.
                          95

-------
This study shows that, in an estuary, the surface micro-
layer of the water column differs substantially from the
underlying subsurface water (one meter), both in the
phytoplankton concentration and species diversity.  The
data support Harvey (1965), who reported that microlayer
samples had at least four times the total number of algae
and protozoans as samples from 10 cm.  This information on
the vertical distribution of phytoplankton should be
considered when designing a phytoplankton sampling program,
and is significant in studies attempting to estimate the
standing crop of phytoplankton.  This estimate requires
accurate determination of the amount of phytoplankton in a
sample, and collection of samples that are representative
of the area being studied (Small, 1961).  A typical
"surface" sample consists predominantly of water lying at
some depth beneath the actual surface layer.  This extra
volume of water dilutes the number of cells present and
results in an underestimation of the abundance of
phytoplankton per unit area of surface.

It is possible that previous estimates of primary
production have erred in the same way.  Depth samples for
Carbon-14 measurements standardly include the surface
(Steeman Nielsen, 1952a), and in some cases are restricted
entirely to the surface (Doty, 1956).  It remains to be
shown if the cells in the surface microlayer are photo-
synthesizing at a maximal level.  David (1965) suggested
that there must be a permanent flora in the surface layer
despite high light intensity, and stated two hypotheses:
first, that the cells of this population would be highly
specialized and able to utilize intensities that are
normally thought to be inhibiting; and second, that
photosynthesis would take place only when light intensity
was low.  There is a third possibility—that the inhibitory
effect of high light intensity has been overemphasized,
since most data has been obtained through laboratory
studies in simulated conditions (Steeman Nielsen, 1952b,
1962; Sorokin and Krauss, 1958; Brown and Richardson,
1968) .

All observations here indicate that the cells of the micro-
layer are healthy and active.  When samples were examined
live the cells appeared in good phytiological condition.
Chloroplasts were abundant, and organisms capable of
locomotion (dinoflagellates and some diatom species) were
moving.  In examination of preserved material, cells were
often observed that had recently divided, or had been
preparing to do so.

The lower diversity in the surface microlayer was caused by
an increase in abundance of the one or two most dominant
                          96

-------
organisms (most often Skeletonema costatum  but
occasionally Cerataulina bergonTI or Rhizosolenia
faeroense) .  The rare species that occurred in the one
meter samples did not appear in the surface samples.  This
implies that certain species select the surface environ-
ment, possibly due to the light conditions.

A normal environment is assumed to support a diverse
assemblage of species, while one that has been subjected
to any type of pollution suffers a reduction in diversity,
with an increase in numbers of individuals of the remaining
species.  Patrick (1949) stated that this is the first
effect produced by a pollutant, and described a system
which employs the distribution of species and specimens of
the diatom population as an index of water pollution.  This
system has been successfully applied in several studies
(Patrick, et al, 1954; Hohn, 1959; Patrick and Strawbridge,
1963).  Wilhm and Dorris (1968) agreed that pollution
results in depressed diversity, and stated that the
properties of diversity indices (the fact that they are
dimensionless, independent of sample size, etc) make them
well suited for use as criteria for judging water quality.
A study of the effect of oil pollution in a river
(McCauley, 1966) found that oil produced the typical result
by eliminating the plankton organisms sensitive to the
toxicity, while permitting the more tolerant species to
thrive.

It seems significant that there is no pattern discernible
in diversity values of the four samples in each of the
collections.  This indicates that, although the presence of
a slick does have a measurable influence on the phytoplank-
ton in that area, it does not act in the manner usually
associated with pollutants.  Comparison of sample pairs by
means of the affinity-dominance index supports the data
obtained from diversity comparisons.  In all cases the
slick and non-slick pairs of samples had high affinity
values, showing that their populations were very similar
in composition.

Slick material might produce its effect through purely
physical factors.  The fact that more cells were discovered
in the oil fraction of Collection 3 than in the rest of the
slick sample suggests that some phytoplankton cells may
have been physically entrapped in the oily layer. ^This^
would explain the absence of a reduction in diversity since
the cells would, presumably, be caught at random, rather
than selectively by species.  Mironov and Lanskaja (1958)
stated that floating organisms, such as zooplankton and
phytoplankton, are particularly susceptible to injury from
oil because they cannot move actively to avoid a polluted
area.
                           97

-------
There is some disagreement as to the effect of oil on the
organisms of the phytoplankton, but few actual experiments
have been carried out.  Zobell (1964) stated that diatoms
seemed to be harmed only by "prolonged exposure to large
amounts of oil."  In one experiment (Galtsoff, Prytherich,
Smith, and Koehring, 1963) the diatom Nitzschia closterium
was found to grow almost as well in medium overlayed with
various kinds of oil as in the controls.  However, a
water-soluble extract from 25 percent crude oil was shown
to retard growth of diatom cultures, and a !50 percent
extract stopped growth entirely.  Mironov and Lanskaja
(1966) conducted experiments on 20 species of planktonic
algae and showed that various oils exhibited a definite
toxic effect, causing retardation in cell division and,
ultimately, death.

Further investigation is necessary to determine what is
causing the reduction in number of phytoplankton cells
in the presence of a slick, and if there actually is a
physical entrapment of the cells.  Another area that
requires examination is the probable correlation between
the percentage of the population found in the surface
microlayer on a given day, and the amount of incident
radiation on that day.  One would predict an inverse
relationship: on a cloudy day more cells would be at the
surface in order to obtain sufficient energy for
phytosynthesis, especially in murky inshore waters.
                          98

-------
                     SECTION VIII

 CHLORINATED HYDROCARBONS IN INDIGENOUS SURFACE FILMS

Ecological and human health problems resulting from the use
of chlorinated hydrocarbon pesticides and polychlorinated
biphenyls make it desirable to know the role of surface
films in concentrating and transporting these substances.

The distribution of chlorinated hydrocarbons between
surface slick and subsurface (1 meter) water samples from
the York River estuary has been evaluated and compared with
data obtained by Seba and Corcoran (1969) in an attempt to
establish the behavior of these compounds in a relatively
unpolluted environment.  Slicks were sampled by a drum
skimmer (Harvey, 1965), but, as in Seba and Corcoran's
work, the ratio of dissolved organics to slick organics in
the sample was not measurable.  Preliminary analyses of
surface and subsurface samples indicated the presence of
detectable concentrations of polychlorinated biphenyls
(PCB) and chlorinated hydrocarbon pesticides (CHP).  The
similarities of chemical structure of PCB's and CHP's
(especially the DDT family) caused extensive masking and
interference of peaks when the samples were analyzed on the
gas chromatograph.  Several methods for total separation of
PCB's and CHP's were tested in an attempt to obtain
quantitative data on both groups.  Although several
researchers (Holden and Marsden, 1969, and Reynolds, 1969)
have claimed success in this separation, difficulties are
well known and documented (Zitko, 1971a).  To date,
consistent and practical total separation of PCB's and
chlorinated hydrocarbon pesticides at low concentration
levels (effective at 10~10g//il) has not been achieved at
this laboratory.  Without total separation of chlorinated
hydrocarbon pesticides and PCB's, the concentration
estimates for PCB's are less accurate and the concentration
of some pesticides cannot be evaluated.  In this work, the
DDT concentration was determined by converting it to DDE by
dehydrochlorination in methanolic KOH, and the PCB
concentrations were estimated from uninterfered peaks in
the chromatograms.


                   SAMPLING METHODS

Surface and subsurface samples were collected from the York
River estuary at irregular intervals from November 1970 -
November 1971.  Slicks sampled were observed and classified
as: (1) Light slick - capillary wave dampening under 0-5 ^
mph winds, (2) medium slick - capillary wave dampening with
                           99

-------
winds estimated 5-10 mph (usually contained light debris),
(3) heavy slick - capillary wave dampening with winds
estimated at 10-15 mph (usually contained much debris).
Non-slick surface and subsurface samples were taken for
comparison of chlorinated hydrocarbon content with the
different slick types.  Sampling dates and locations are
shown on Figure 28.

Twelve liter surface samples were taken with a boat-mounted
ceramic coated drum skimmer built by NASA Wallops Island
Station.  This apparatus avoided contamination while
sampling, and samples were taken easily and quickly.  When
sampling, the boat was operated in the upcurrent direction
to avoid collection of engine oils.  The water surface
layer was picked up in a stainless steel trough and drawn
under vacuum into an 18 liter glass bottle.  The sampling
apparatus was washed with chloroform before each day's use
and flushed with new sample at each station.  Only
stainless steel, glass, and teflon contacted the sample;
this prevented contamination of the sample by grease, oils,
fats, etc, that might have high chlorinated hydrocarbon
concentrations.

The subsurface samples were taken by a diver who immersed
an 18 liter glass bottle to approximately 1 meter and
allowed it to fill with 12 liters of water.
                 ANALYTICAL PROCEDURE

 In  the  laboratory, the 12 liter sample was transferred from
 the sample bottle to a 20 liter glass carboy with a bottom
 drain stoppered with a clamped teflon tube.  The sample
 bottle  was washed three times with 250 ml of pesticide-pure
 petroleum ether and these washings were also added to the
 carboy.  The carboy was stoppered and the sample and
 petroleum ether were shaken vigorously for 5 minutes; at
 the end of shaking the water layer had become cloudy.  The
 sample  was allowed to stand overnight in the carboy.  The
 next day the clear water layer was discarded through the
 bottom  spigot and the ether portion was collected in a
 2000 ml round bottom flask.  Three grams of anhydrous
 Na2S04  was added to remove the remaining water, and the
 sample  was redissolved and transferred to a 50 ml beaker by
 three 15 ml washings of pesticide pure petroleum ether.
 The petroleum ether was allowed to evaporate to a volume
 of  about 0.5 ml.  The 0.5 ml sample and two 1 ml ether
 washings of the 50 ml beaker were then added to a 3.0 ml
 centrifuge tube, evaporated to 0.25 ml in a jet of
 prepurified nitrogen, spotted on TLC plates, and developed
                          100

-------
               •8.SEPT. 9,1971
37
25'
20
37
                                                             II,OCT.5,I97I

                                                                  14,NOV. 3,1971
12,OCT. 5,1971
I3.NOV.3, 1971

     6, FEB. 6,1971
   2, NOV. 27,1970
    I, NOV. 27, 1970
                            5, JAN. 30, 1971
                                         3, JAN. 30,1971
                                             4, JAN. 30,1971 	
                                                 7, SEPT. 6,1971

             FIGURE  28.   SAMPLE  LOCATIONS  AND  DATES
                                     -101-

-------
in CC1-4 to the 10 cm line in a TLC chamber according to the
procedure of Breidenbach, et al (1966).  Silica gel on the
plate from 2 cm to 10 cm was scraped off and collected by a
vacuum silica gel collection assembly.  Chlorinated hydro-
carbons absorbed on the silica gel were eluted into a
graduated centrifuge tube by 15 ml of 50% petroleum ether-
acetone solution.  The samples were again reduced to 1.0 ml
in a jet of prepurified nitrogen and analyzed in a Perkin-
Elmer 900 gas chromatograph fitted with an electron capture
detector.  Two columns were used for sample analysis, one
was 3% D.C. 200 on Varaport 30 80/100 mesh, the other was
5% QF-1 on Chromosorb G AW DMCS 80/100 mesh.  Identifi-
cations of PCB's and DDT family pesticides were made by
comparing retention times of sample peaks with the
retention times of known standards on both columns.

The DDT content of the samples was determined by comparing
the reduction in the area of the DDT peak, following
dehydrochlorination of DDT to DDE in methanolic KOH, with
the DDT peak area of a known concentration standard.

The concentrations of PCB's were determined by measuring
the heights of four peaks that were not interfered with by
members of the DDT family and comparing these heights with
the four corresponding peak heights of a known concentra-
tion standard of Aroclor 1254.  Although this analysis
involves error in the quantitation of PCB's because the
electron capture detector response can vary widely for each
polychlorinated biphenyl, Aroclor 1254 is composed mostly
of tetra-, penta-, hexa-, and heptachlorobiphenyls (Koeman,
et al, 1970) and our error would be relatively small
because the electron capture detector response to these
compounds does not vary by more than a factor of 2 (Zitko,
1971b).

Pesticide and PCB standards of known concentration were
mixed into 12 liters of distilled water, extracted with
petroleum ether, and run through the procedure with the
samples to provide information on extraction efficiencies.
Twelve liters of distilled water was extracted and analyzed
as a blank.  A 750 ml petroleum ether blank was evaporated
in the rotary evaporator and treated as a sample extract.
No interfering substances were found in the petroleum ether
blank, but peaks corresponding to Aroclor 1254 were found
in the distilled water at a concentration 6 ng/liter.
These peak heights were subtracted from the pesticide and
PCB standards to obtain procedure efficiencies.  The
efficiencies were: PCB's (Aroclor 1254) - 50% (all peaks in
same ratios as standard), DDE - 6%, ODD - 20%, and DDT -
18%.  The appropriate efficiencies were used for
                         102

-------
determining the concentrations  of PCB's  and DDT  in the
samples.

All glassware was washed, rinsed with  distilled  water,
dried with acetone, and rinsed  with  petroleum ether before
each use.  Samples 6 and  7 were contaminated  by  lubricant
mistakenly applied to the rotary evaporator.   Because a
procedure efficiency was  calculated  and  care  was taken that
each sample was treated identically, the DDT  and PCS
concentrations and the total  chlorinated hydrocarbon count
(on Table 1) represent the chlorinated hydrocarbon concen-
trations found in slick,  non-slick,  and  subsurface samples.

Separations similar to Reynolds'  (1969)  Florisil column
separation of Aroclor 1254 and  CHP were  tested in the
laboratory and found unsatisfactory  because of solvent
contamination and the variability of Florisil and pesticide
pure solvents.  Volumes of 200  ml of hexane or 200 ml of
20% diethyl ether in hexane  solution from the elution of
the Florisil column in the Reynolds' procedure must be
concentrated to approximately 0.5 ml in  order to analyze
the small amounts of chlorinated hydrocarbons obtained in
this study.  Gas chromatographic  peaks of contaminants
interferring with chlorinated hydrocarbon analysis were
observed from evaporation of 200 ml  pesticide standard
hexane  (Nanograde, Mallinckrodt)  or  2070  diethyl  ether
hexane  solution to 0.5 ml.   Unpredictable and incomplete
elutions resulting from the  variability  of Florisil and the
different elution properties  of pesticide grade  solvents
(Zitko, 1971a) made the Reynolds  separation  technique
further undesirable for this  study.

The Holden  (1969) silica  gel  procedure for separating PCB's
and CHP's uses smaller amounts  of solvents, avoiding
solvent purity problems,  but silica  gel  columns, like
Florisil columns, vary in their elution  properties.  Two
types of silica gel, different  column  loadings,  and
different silica gel deactivations  (by shaking with
distilled H20 for 30 minutes) ,  were  used in an attempt to
separate Aroclor 1254 from DDT  and ODD,  but no reliable
separation was found.

Thin layer chromatography procedures similar  to  Mulhern s
(1968)  for the separation and removal  of organochlorine
insecticides from thin layer plates  were tested  using
Aroclor 1254 and a DDE, ODD  & DDT standard, but  the
separations were ineffective.

Differentiation and identification  of  PCB's and  CHP's  in
the samples by gas chromatography of nitrated sample_
extracts was not attempted because  some  PCB s also nitrate
                          103

-------
and clear cut identification cannot be made (Reynolds,
1969).
                RESULTS AND DISCUSSION

The DDT concentrations shown in Table 16 are similar to
those found in western streams by other workers (Brown and
Nishioka, 1967).  Duke, et s& (1970) have shown that
Aroclor 1254 concentrations in water can at times be great
(275 ppb) due to accidental introduction by industry, but
they found values in the range of our data (100 ppt in
Escambia River to non-detection in Escambia Bay) following
correction of the 1254 leak.

There was an average of 16 ppt (range trace to 54 ppt) DDT
o,p, and p,p' in our transitory slicks; Seba and Corcoran
(1969) found 19 ppt DDT in a transitory slick in the main
axis of the Florida current.  The DDT content of the only
semi-permanent slick sampled by this study varied from
12 ppt - 41 ppt on the two sampling dates.  Seba and
Corcoran found 61 ppt to 3465 ppt DDT in semi-permanent
slicks associated with the mouths of drainage canals in the
Biscayne Bay.  They found 2-35 ppt Dieldrin and 5-34 ppt
Aldrin in slick samples, but neither of these pesticides
were detected here.

Corcoran (personal communication) states that PCB's
occurred in very low concentrations in their samples.
PCB's corresponding in retention times to Aroclor 1254
were present in our samples as indicated on Table 16.

Seba and Corcoran (1969) sampled with jars by immersing
them just below the surface.  Neither this method nor the
drum skimmer permitted determination of the relative
amounts of surface film material and water obtained in the
samples.  Depending on whether the slick is spread out
thinly over a large area or is relatively thick but
covering a small area, the chlorinated hydrocarbon
concentrations in the samples could vary according to the
slick area exposed to the environment.  The first
priorities for any future work on surface films should be
the development of a field surface film thickness measuring
device.

Although the data show that our classification of slick
type cannot be used to determine the amount of chlorinated
hydrocarbons present in the surface film, it does indicate
that the concentration of total chlorinated hydrocarbons in
surface films are generally higher than the subsurface
concentrations.  The DDT concentrations were generally
                         104

-------
o
Ul
        Table 16.  Chlorinated Hydrocarbon Concentration Results


                                               Total
                                            Chlorinated
Sample #
1 Surface
I Subsurface
2 Surface
2 Subsurface
3 Surface
3 Subsurface
4 Surface
4 Subsurface
5 Surface
5 Subsurface
Date
11/27/70
ii it ii
11/27/70
H it ii
1/30/71
ii ti H
1/30/71
ii ii ii
1/30/71
H H II
Slick Hydrocarbon PCB**
Type Count* (ppt)
1
-
3
No
Slick
3
No
Slick
530
156
640
202
28
34
256
264
338
302
51
16
37
17
2
3
21
26
33
29
DDT
(ppt)
17
1
54
3
3
6
26
5
3
2
Slick Origin
Unknown

Probably
West Point
Pulp Mill

American Oil
Pier
(recent origin)

         6  Contaminated
         7 Contaminated

-------
Table 16. (Cont'd).  Chlorinated Hydrocarbon Concentration Results
                                       Total
                                    Chlorinated
Sample #
8 Surface
8 Subsurface
9 Surface
9 Subsurface
10 Surface
10 Subsurface
11 Surface
11 Subsurface
12 Surface
12 Subsurface
13 Surface
13 Subsurface
Date
9/9/71
n n n
9/9/71
n n n
9/9/71
ii n n
10/5/71
n n n
10/5/71
n it n
11/3/71
n n n
Slick
Type
3
-
2

2
-
3
-
3
3
Hydrocarbon
Count*
178
108
Spilled
86
43
59
120
58
118
152
325
172
PCS**
(ppt)
12
8

7 Not
4
5
5
9
7
11
24
12
DDT
(ppt)
Trace
Trace

Measured
Trace
Trace
Trace
Trace
12 ,
4
41
27
Slick Origin
Pulp Mill

Pulp Mill
•
Unknown

Unknown

Mouth of
Queen ' s
Creek
Mouth of
Queen ' s
Creek

-------
Table 16 (Cont'd).  Chlorinated Hydrocarbon Concentration Results.
Sample #
14 Surface
14 Subsurface
Date
11/3/71
II It M
Slick
Type
2
Total
Chlorinated
Hydrocarbon
Count*
208
124
PCB**
(ppt)
13
6
DDT
(ppt)
45
44
Slick Origin
VIMS
Ferry
Pier
Distilled H20
  (Blank)
66
                                                           Trace
 * Peak height (in graph units) of all identifiable chlorinated hydrocarbons in
   Chromatograms

** Peaks corresponding in retention time to Aroclor 1254.

-------
higher in surface samples and lower in the subsurface
samples; the PCB's seem to be more uniformly distributed.
Possible explanations for this difference are: PCB's and
DDT are likely carried in different physical states in the
atmosphere, PCB's as a gas and DDT on dust particles
(Harvey, 1971); PCB's and DDT enter the environment by
different methods; PCB's and DDT have differing surface
chemical properties.  Concentration mechanism, for PCB's
and CHP's in surface films and subsurface water, and the
exact role the surface film has in concentrating and
transporting chlorinated hydrocarbons can only be determined
by extensive and expanded further studies.
                         108

-------
                      SECTION IX

       LIPIDS OF SURFACE FILMS ON CHESAPEAKE BAY

The objective of this work was to provide more complete
information regarding lipid material in surface water and
surface films.  The 100 micron surface layer was sampled
in the York River and Chesapeake Bay, and the samples were
analysed for hydrocarbons and fatty acids by thin layer
and gas-liquid chromatography.

In the past, sampling of surface water was carried out by
dipping a container under the surface of the water.  In
Garrett's work (1965), sampling techniques were greatly
refined through the use of stainless steel mesh screens
that sampled the top 0.15 mm  surface layer.  Again using
screens, Garrett (1967) took  samples for chemical analysis
and estimated the total lipid material to be 0.2-1.0
milligrams per liter.  He tentatively identified one
hydrocarbon (C29H52) and reported the existence of several
fatty acids at eight stations.  Garrett's work was
primarily involved with the total dissolved organic
material that could be extracted from seawater with
chloroform.  The analyses were not quantitative and
comparisons between fresh and salt water were not
considered.

Jarvis, et al (1967) postulated the existence of fatty
esters, free fatty acids, fatty alcohols and hydrocarbons
in samples taken with screens.  He made no chemical
analyses.  From Garrett's work  (1967) and from physical
parameters  (film pressures, surface potentials, surface
viscosities, and damping coefficients) Jarvis arrived at
his conclusions.
                 MATERIALS AND METHODS

 Samples were  taken along the York River and nearby areas
 of  the Chesapeake  Bay.   Dates, positions and descriptions
 are given in  Appendix 2.  Some preliminary samples were
 taken along the Chesapeake Bay Bridge-Tunnel, but the
 sampling  boat was  unable to sample regularly due to rough
 sea conditions.

 A rotating drum mechanism was used to take the samples.
 It  was designed by Harvey (1965), who reported it collected
 larger plankton populations in surface samples in slick
 areas than in subsurface water.  Monolayers can be taken
 from the  water's surface and transferred to a rotating
                          109

-------
cylinder with an appropriate hydrophilic surface (Ries and
Grutsch, 1968, and Ries, 1968).  The drum used here was
designed and built by the National Aeronautics and Space
Administration's Wallops Station.  It was necessary that
no organic contamination be present, so all surfaces in
contact with the sample had to be stainless steel and
coated with Solaramic S5210-ZC (a product of the Solar
Company, San Diego, California).

It is supported 3 feet off the bow of an 18 ft Thunderbird
Cheyenne outboard boat by a tubular aluminum structure and
a boom and winch.  A stainless steel trough with a teflon
blade is mounted in such a fashion as to scrape the drum
as it rotates through the water, in a counterclockwise
sense as viewed from the right side of the boat.  The drum
is normally immersed to a depth of about 4 inches in the
water.

The drum is turned by an electric motor at a slightly
faster rate than the boat speed, which usually requires
between 8 and 10 drum revolutions per minute.  Water
scraped from the drum is deposited in the stainless steel
trough and pumped to a sample bottle by a hand operated
vacuum pump.

Slightly more than 18 liters of seawater were taken at each
sampling site.  Immediately upon return to the laboratory,
the samples were gravity filtered through Gelman Type A
glass fiber filter pads.  Gravity filtration was chosen
over vacuum filtration to avoid rupture of cells and
release of cellular material to the filtrate.  The pH of
the filtrate was adjusted to 2.0-2.5 with 12N HC1 to
convert salts of fatty acids to the free fatty acids.  At
this point, 50 cc of chloroform was added to retard
bacterial degradation of the sample, since it is sometimes
several days before extraction can be carried out.

Next, the filtrate was placed in a continuous extraction
device and bubbled through 1500 milliliters of chloroform
for approximately three hours.  The chloroform was then
drained off and reduced in volume to about one milliliter
in a vacuum rotary evaporator.

The extract was then subjected to preparative thin-layer
chromatography according to Stahl (1969).  This preparative
thin-layer chromatography was used to "clean up" the sample
and separate the various classes of compounds.  Plates were
coated with Silica Gel G (Merck, Darmstadt, Germany).  A
rapidly prepared mixture of silica gel and water was placed
on 20 cm x 20 cm glass plates in a 0.25 millimeter layer by
means of a spreading device (Brinkmann, Westbury, New York) .
                         110

-------
The sample spotted plate was  then developed with a  90:10:1
hexane, diethy1 ether, acetic acid  solution.  The hydro-
carbon band migrated with  the solvent  front while the fatty
acids migrated slightly less  than halfway up  the plate.

The bands of organic material was made visible by a 0.354
nanometer light.  The hydrocarbon band was then vacuumed
off the plate with a thin-layer  plate  vacuuming device
(Brinkmann).  The vacuumed silica gel  was washed with
methanol to remove the hydrocarbons from the  relatively
polar base.  The methanol  was evaporated to a very  small
volume and then extracted  with hexane  to remove the
hydrocarbons from the methanol.  This  micro extraction was
necessary to remove the methanol soluble calcium sulfate
binder.  The hydrocarbon in hexane  solution was reduced to
dryness by air evaporation and then redissolved in  a known
volume of hexane for gas chromatographic analysis.

The gas chromatographic analyses were  carried out on a
Perkin-Elmer Model 900 Gas Chromatograph, equipped  with
dual columns and flame ionization detectors,  and operated
under dual column compensated conditions.

For work with hydrocarbons, a 6  ft  x 1/8 in.  O.D. copper
column was used.  It was packed  with a 10% SE-30 on a
non-acid washed Chromosorb P  (mesh  60/80) both from Applied
Science, State College, Pa.  A temperature program  of 4 C
per minute was used from 100  C to 280  C with  an initial
time of two minutes and final time  of  eight minutes.  The
temperature of the injectors  was set at 320 C and of the
manifold was set at 290 C. Helium  was used as carrier gas.

The fatty acid band was also  vacuumed  off the plate and the
silica gel washed with methanol  to  remove the free  fatty
acids which were converted to their methyl esters by the
method of Metcalf and Schmitz (1961).   The esters were then
redis solved in a known amount of hexane and were now ready
for injection in the gas chromatograph.

Fatty acid methyl esters were run with the same program as
the hydrocarbons (100 C  -  280 C  at  4 C per minute,  injector
at 320 C, manifold at 290  C,  initial time 2 minutes, final
time 8 minutes).  Glass columns  (1/4 in. O.D. x 6 ft) were
used to reduce any possible oxidation.  The columns were
packed with 5% SE-30 on Chromosorb  G (AW DMCS, 80/100 mesh)
prepared by Perkin-Elmer  (Norwalk,  Conn.).

Peaks were identified by retention  time comparison  with a
series of standards purchased from  Applied Science  of State
College, Pa.  Areas under  peaks  were determined on  both the
standards and unknowns with the  areas  of each being
                           111

-------
directly proportional to their concentrations.  Repeat
chromatograms were made frequently on standards and blanks
were also run.
                        RESULTS

Good separations were obtained for saturated and unsaturated
mixtures of the fatty acid standards.  On blank runs base-
line drift was held at a minimum.

The fatty acids from 10:0 to 23:0 are reported (see Table
17).  The 14:0, 16:0, 16:1, 18:0, 18:1, and 22:0 fatty
acids predominate.  The percentage contribution of these
acids to the total free acid concentration ranged from 55
to 98% with most falling in the 70% to 90% range, as can be
seen in Table 18.  The amounts of fatty acid material
ranged from 700 micrograms per liter to 7800 micrograms per
liter. These results are very dissimilar to the results of
Stauffer (1969), who investigated the fatty acids at 2
meter depths in the James River.  Differences may, in part,
be a consequence of high surface concentrations of lipid
material.  The unidentifiable material ranged from 2% to
187o with an average of less than 1070.

Normal paraffinic hydrocarbons from CIQ to C24 are listed
in Table 19.  No concentration data was available for the
first eleven samples due to difficulties with sample
handling.  The concentrations of the last eight samples
ranged from 300 to 500 micrograms per liter.  The unknown
materials ranged from 20 to 45 percent with most in the 25
to 40 percent category.  All the gas chromatograms of the
hydrocarbon samples were rather similar to gas chromatograms
of number 2 fuel oil.

The total amount of organic material collected varied from
sample to sample but the variations were not related to
cause in this work because fluctuations in biological
communities or pollutants were not measured.  The concen-
trations were independent of salinity.  Samples were taken
from the Chesapeake Bay to the Pamunkey River with
salinities varying more than 20 &,, but no geographical
pattern of hydrocarbon concentrations was apparent.
     o
A 50 A thick slick or surface film represents about 0.005
percent by volume of a 100 micron thick water sample.
Depending on the state of compression of the organic
compound in the surface film, this layer should contain
sufficient amounts of organics to considerably affect, by
its presence or absence, the concentrations measured in the
100 micron thick sample.
                         112

-------
Table 17.  Fatty Acid Sample Concentration and Composition.
Sample #1
IK/1 %
10:0
11:0
12:0
13:0
14:0
14:1
15:0
16:0
16:1
16:2
17:0
18:0
18:1
18:2
19:0
20:0
21:0
22:0
23:0
?
Total
-
Oil
104
147
567
112
207
2510
720
-
095
874
1142
-
020
043
046
363
106
742
7861
-
0.14
1.32
1.87
7.21
1.43
2.64
31.93
9.80
-
1.21
11.12
14.53
-
0.25
0.55
0.58
4.62
1.35
9.44

Sample #2
-
017
024
tr
248
-
094
1503
135
-
052
728
628
-
050
013
033
052
Oil
346
3935

0
0

6

2
38
3

1
18
15

1
0
0
1
0
8

-
.44
.61
tr
.31
-
.38
.21
.43
-
.33
.50
.95
-
.27
.33
.83
.33
.27
.80

Sample #3
UK/1 %
_
-
010
-
173
mm
091
1754
159
tr
062
782
627
-
Oil
Oil
024
024
015
314
4058


0

4

2
43
3

1
19
15

0
0
0
0
0
7

_
-
.24
-
.27
-
.25
.22
.92
tr
.53
.27
.46
-
.27
.27
.59
.59
.37
.73

Sample #4
ug/1 %

-
028
148
148
-
052
852
116
-
048
399
379
-
013
028
013
205
009
063
2501


1.
5.
5.

2.
34.
4.

1.
15.
15.

0.
1.
0.
8.
0.
2.

—
-
13
92
92
-
09
06
62
-
92
94
16
-
52
13
52
19
35
53

                         113

-------
Table 17. (Cont'd).   Fatty Acid Sample Concentration and
                     Composition.
Sample #5
LlR/1 7o
10:
11:
12:
13:
14:
14:
15:
16:
16:
16:
17:
18:
18:
18:
19:
20:
21:
22:
23:
0
0
0
0
0
1
0
0
1
2
0
0
1
2
0
0
0
0
0
Total
005
005
037
135
228
-
077
1316
274
-
061
625
603
-
017
Oil
013
227
015
113
3764
0
0
0
3
6

2
34
7

1
16
16

0
0
0
6
0
3

.12
.14
.98
.59
.05
mm
.05
.96
.29
-
.62
.61
.03
-
.46
.29
.35
.02
.41
.01

Sample #6 Sample #7 Sample #8
ttg/1 % us/1 % us/1 "1°
- No concen- - -
trations*
004
051
085
-
061
1142
222
-
035
407
333
-
008
030
010
101
023
151
2665
0.
1.
3.
-
2.
42.
8.
-
1.
15.
12.
-
0.
1.
0.
3.
0.
5.

16
92
19

29
85
34

31
29
51

29
14
37
80
86
60

030
tr 013
2.89 370
039
0.72 035
31.05 1052
1.44 314
-
1.08 018
22.74 166
31.05 383
-
-
0.36 tr
_
8.66 061
-
261
2743
1
0
13
1
1
38
11

0
6
13




2

9

.11
.48
.50
.43
.27
.36
.44
-
.64
.04
.98
-
-
tr
-
.22
-
.53

     *No concentration values available.
                         114

-------
Table 17. (Cont'd)
Fatty Acid Sample Concentration and
Composition.
Sample #9
UK/I 7o
10:0
11:0
12:0
13:0
14:0
14:1
15:0
16:0
16:1
16:2
17:0
18:0
18:1
18:2
19:0
20:0
21:0
22:0
23:0
?
Total
-
050
040
-
019
339
040
-
014
130
162
-
tr
016
tr
064
tr
023
897
i
i
i
5
4

2
37
4

1
14
18


1

7

2

•»
••
M
.58
.49
-
.06
.74
.49
-
.58
.44
.08
-
tr
.82
tr
.16
tr
.55

Sample #10
«/l %
tr
tr
071
-
029
615
tr
-
061
361
050
-
019
040
tr
117
tr
029
1392

tr
tr
5

2
44


4
25
3

1
2

8

2

.09
-
.11
.21
tr
-
.38
.90
.60
-
.33
.89
tr
.37
tr
.11

Sample #11
MB/1 %
013
tr
111
-
043
342
026
-
048
249
122
-
019
028
tr
117
tr
035
1151
4
I
1
•*
V
.13
tr
9

3
29
2

4
21
10

1
2

10
.65
-
.69
.71
.27
-
.16
.67
.60
MB
.70
.27
tr
.12
tr
3

.03

Sample #12
MR/1 %
*•
016
020
102
009
037
428
027
-
075
287
117
-
016
075
388
424
010
453
2485
0.63
0.79
4.10
0.35
1.49
17.23
1.10
-
3.02
11.57
4.73
-
0.66
3.02
15.62
17.05
0.39
18.24

                          115

-------
Table 17. (Cont'd)
              Fatty Acid Sample Concentration  and

              Composition.


Sample #13   Sample #14   Sample #15    Sample  #16
 _ l\    oi     _ /i    et  '   _ /I    07      _/1     "I
10:0
11:0
12:0
13:0
14:0
14:1
15:0
16:0
16:1
16:2
17:0
18:0
18:1
18:2
19:0
20:0
21:0
22:0
23:0
9
Total
tr
tr
031
017
220
tr
083
806
tr
-
157
577
146
-
tr
660
128
603
tr
636
4065
tr
tr
0.75
0.43
5.41
tr
2.04
19.83
tr
-
3.86
14.20
3.59
-
tr
16.24
3.16
14.84
tr
15.65

-
-
tr
-
059
tr
028
311
069
-
059
212
718
-
013
247
024
204
tr
114
2058




2

1
15
3

2
10
34

0
12
1
9

5

-
-
tr
-
.86
tr
.38
.09
.33
-
.86
.32
.89
-
.63
.02
.16
.90
tr
.56

-
-
010
026
063
015
017
183
028
-
078
094
138
-
008
023
025
357
010
146
1222


0
2
5
1
1
14
2

6
7
11

0
1
2
29
0
11

-
-
.80
.14
.17
.25
.43
.97
.32
-
.42
.66
.32
-
.62
.87
.05
.23
.80
.94

-
-
tr
tr
035
005
Oil
098
014
-
032
049
085
-
tr
Oil
008
205
008
099
659




5
0
1
14
2

4
7
12


1
1
31
1
15

-
-
tr
tr
.29
.74
.65
.89
.15
-
.80
.44
.90
-
tr
.65
.16
.10
.16
.05

                         116

-------
Table 17. (Cont'd).
       Sample #17
 Fatty Acid Sample Concentration  and
 Composition

Sample #18   Sample #19
ug/1
10:
11:
12:
13:
14:
14:
15:
16:
16:
16:
17:
18:
18:
18:
19:
20:
21:
22;
23;

0
0
0
0
0
1
0
0
1
2
0
0
1
2
:0
:0
;0
;0
:0
?
Total

-
009
019
052
-
018
179
019
-
042
097
129
-
007
028
010
386
004
199
1197
•
•
•
M
0.73
1
4
•
1
14
1

3
8
10

0
2
0
32
0
16

.55
.32
••
.50
.97
.59 i
-
.50
.15
.79
-
.55
.32
.86
.23
.32
.61


tr
Oil
026
063
014
033
113
342
-
091
209
551
-
013
098
044
536
Oil
309
2465
•
*
tr
0.44
1
.06
2.56
0
1
4
13

3
8
22

0
3
1
21
0
12

.57
.33
.59
.87
- •"''
.71
.48
.36
-
.53
.98
.77
.74
.44
.55


003
007
016
045
-
022
126
008
-
050
059
061
-
003
016
003
282-
010
117
827
•
•
0.39
0.86
1.91
5.47
•
2
•
.63
15.22
0
•
6
7
7

0
1
0
34
1
14

.92
•»
.06
.18
.38
-
.33
.91
.33
.12
.19
.10

                          117

-------
Table 17. (Cont'd).  Fatty Acid Sample Concentration and
                     Composition.

       Sample #20   Sample #21   Sample #22   Sample #23
       Ug/1    7o    Ug/1    %    UK/1
10:0
11:0
12:0
13:0
14:0
14:1
15:0
16:0
16:1
16:2
17:0
18:0
18:1
18:2
19:0
20:0
21:0
22:0
23:0
?
Total
1
1
3
12
33
-
7
67
21
16
234
14
242
-
1
13
8
-
-
41
716
0.18
0.19
0.42
1.66
4.58
-
0.97
9.43
2.95
2.29
32.71
1.99
33.76
-
0.18
1.76
1.15
-
-
5.77

9
10
16
148
59
-
14
270
26
11
2343
167
1697
-
40
39
-
-
-
116
4967
0.
0.
0.
2.
1.
-
0.
5.
0.
0.
47.
3.
34.
-
0.
0.
-
-
-
2.

18
20
32
98
19

28
40
52
22
17
36
16

81
79



33

3
-
7
-
42
-
16
83
37
32
57
52
135
-
21
114
64
-
-
68
732
0.
-
1.
-
5.
-
2.
11.
5.
4.
7.
7.
18.
-
2.
15.
8.
-
-
9.

40

02

77

18
38
00
39
76
06
44

92
52
78


28

16
11
8
29
30
-
17
36
45
23
27
25
-
-
5
15
16
-
-
90
395
3.
2.
2.
7.
7.
-
4.
9.
11.
5.
6.
6.
-
-
1.
3.
4.
-
-
22.

97
86
07
35
67

30
09
49
75
86
33


27
82
15


83

                         118

-------
Table 17. (Cont'd).
       Sample #24
 Fatty Acid Sample Concentration  and
 Composition.

Sample #25   Sample #26
                     to
10:0
11:0
12:0
13
14
14
15
16
16
16
17
18
18
18
19
20
21
22
23

:0
:0
:1
:0
:0
:1
:2
:0
:0
:1
:2
:0
:0
:0
:0
:0
9
Total
7
4
9
21
34
-
18
98
39
21
511
30
783
«••
16
57
tr
-
-
96
1745
0
0
0
1
1

1
5
2
1
29
1
44

0
3



5

.39
.25
.54
.19
.95
-
.01
.64
.24
.20
.27
.70
.86
-
.94
.25
tr
-
-
.49


10
4
6
45
-
6
69
16
18
4
1
45
-
5
19
19
-
-
37
305
(
3
1
2
14

2
22
5
5
1
0
14

1
6
6



-------
Table 17. (Cont'd).  Fatty Acid Sample Concentration  and
                     Composition.

       Sample #27   Sample #28   Sample #29
       ue/1    %    u.s./l
10:
11:
12:
13:
14:
14:
15:
16:
16:
16:
17:
18:
18:
18:
19:
20:
21:
22:
23:

0
0
0
0
0
1
0
0
1
2
0
0
1
2
0
0
0
0
0
?
Total
8
6
4
29
24
-
14
105
-
-
519
68
134
-
11
56
11
-
-
90
1080
0
0
0
2
2

1
9


48
6
12

0
5
1


8

.70
.58
.41
.69
.22
-
.29
.76
-
-
.05
.31
.42
-
.99
.20
.05
-
-
.36

7
15
11
23
53
-
8
109
24
11
-
119
145
-
6
49
8
-
-
78
665
1
2
1
3
7

1
16
3
1

17
21

0
7
1


11

.04
.18
.61
.42
.97
-
.23
.42
.61
.61
-
.85
.83
-
.85
.41
.23
-
-
.77

-
-
1
-
26
-
18
53
44
18
-
9
417
-
-
-
35
-
-
32
653
-
-
0.14
-
4.03
-
2.71
8.12
6.77
2.80
-
1.45
63.81
-
-
-
5.32
-
-
4.88

                         120

-------
Table 18.  Percentages of Predominant Fatty Acids.
  Sample   14:0   16:0   16:1    18;0    18:1   22:0   Total
    1      7.21  31.93   9.80   11.12  14.53   4.62   79.21
    2      6.31  38.21   3.43   18.50  15.95   1.33   83.73
    3      4.27  43.22   3.92   19.27  15.46   0.59   86.73
    4      5.92  34.06   4.62   15.94  15.16   8.19   83.89
    5      6.05  34.96   7.29   16.61  16.03   6.02   86.96
    6      3.19  42.85   8.34   15.29  12.51   3.80   85.98
    7      2.89  31.05   1.44   22.74  31.05   8.66   97.83
    8      13.50  38.36   11.44    6.04  13.98   2.22   85.58
    9      4.49  37.74   4.49   14.44  18.08   7.16   86.40
    10      5.09  44.21     tr   25.90    3.60   8.37   87.17
    11      9.65  29.71   2.27   21.67  10.60   10.12   84.02
    12      4.10  17.23   1.10   11.57    4.73   17.05   55.78
    13      5.41  19.83     tr   14.20    2.59   14.84   57.78
    14      2.86  15.09   3.33   10.32  34.89   9.90   76.39
    15      5.17  14.97    2.32    7.66   11.32   29.23   70.67
    16      5.29  14.89    2.15    7.44   12.90   31.10   73.77
    17      4.32  14.97    1.59    8.15   10.79   32.23   72.25
    18      2.56    4.59   13.87    8.48   22.36   21.74   73.60
    19      5.47    5.47   15.22    0.92    7.18    7.38   70.29
                          121

-------
Table 19.  Hydrocarbon Sample Concentration and Composition,

Carbon   Sample #1*   Sample #2*   Sample #4*   Sample #5*
No.              %            %           "%            %
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
?
Total
-
1.80
5.93
0.19
9.09
0.68
2.16
7.80
5.48
9.60

5.67
8.76
10.31
4.64
27.89
100.00
0.23
0.12
0.25
0.21
1.13
1.06
3.28
6.68
7.90
12.79
13.96
10.03
-
1.58
-
40.74
99.96
0.10
0.08
0.07
0.31
0.18
0.76
1.38
9.24
11.62
16.10
9.30
7.11
9.20
, 4.81
1.36
28.35
99.97
0.15
0.41
0.16
1.07
6.31
4.20
6.97
10.41
8.41
9.68
6.54
5.14
3.87
5.21
1.87
29.60
100.00
*Samples #1 thru #11 have no concentration values available
 for the hydrocarbons.


Sample #3 lost due to spillage.
                         122

-------
Table 19. (Cont'd).
Hydrocarbon Sample Concentration and
Composition.
Carbon
No.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Total
Sample #6*
0.26
0.11
0.33
0.61
2.58
4.58
7.40
9.00
8.37
14.29
7.19
5.37
7.50
4.49
3.39
24.60
100.00
Sample #7*
0.45
0.18
0.23
0.68
3.12
3.76
5.12
8.21
8.00
14.37
7.37
6.28
5.31
1 6.88
0.97
29.07
100.00
Sample #8*
0.32
-
0.10
0.47
2.50
3.61
4.66
8.21
9.70
10.74
5.95
4.32
4.93
3.72
0.88
39.88
99.99
Sample #9*
0.23
0.13
0.13
1.53
6.36
8.25
8.49
11.40
9.75
8.62
5.43
3.58
4.77
1.59
-
29.23
100.00
                          123

-------
Table 19. (Cont'd).  Hydrocarbon Sample Concentration and
                     Composition.

Carbon  Sample #10*  Sample #11*  Sample #12  Sample #13
No.                               tig/1    %   ug/1	%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
?
Total

0
0
0
-
.08
.06
.16
14.30
36
21
6
1






19
99
.47
.04
.75
.61
-
-
-
-
-
-
.52
.97


0
0
12
26
15
5
2
2
1
1
1
1

28

-
-
.20
.32
.40
.09
.54
.88
.58
.41
.13
.61
.93
.29
-
.62

tr
tr
-
-
-
001.
Oil.
028.
056.
017.
019.
Oil.
046.
025.
068.
166.
449.
tr
tr



0.
2.
6.
12.
3.
4.
2.
10.
5.
15.
36.
100
-
-
-
32
48
27
53
68
25
55
34
52
15
92
.00
004.
002.
-
-
-
002.
010.
017.
029.
019.
029.
Oil.
016.
008.
009.
133.
289.
1
0



0
3
5
10
6
10
3
5
2
3
46

.26
.27
-
-
-
.55
.41
.71
.16
.70
.22
.85
.49
.64
,08
.15
,
                         124

-------
Table 19. (Cont'd).  Hydrocarbon Sample Concentration and
                     Composition.
Carbon
No.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1
Total
Sample #14
UK/1 %
tr
001.
-
001.

007.
025.
027.
052.
016.
051.
014,
033.
012.
016.
166.
423.
0.

0.
0.
1.
5.
6.
17.
3.
17.
3.
7f
2.
3.
39.

tr
30
-
23
11
73
86
50
35
83
17
31
81
85
76
18

Sample #15
uR/1 %
tr
tr
-
002.

009.
018.
020.
044.
008.
031.
006.
034.
010.
027.
067.
276.
tr
tr
-
0.
0.
3.
6.
7.
15.
2.
11.
2.
12.
3.
9.
24.


78
14
22
67
13
92
76
26
18
41
68
77
07

Sample #16
U8/1 %
tr
-
001.
001.
007.
022.
019.
046.
023.
099.
017.
027.
013.
017.
239.
529.
tr
i
0
0
1
4
3
8
4
18
3
5
•
.09
.18
.32
.11
.63
.77
.26
.73
.12
.16
2.40
3
45

.12
.09

Sample #17
U8/1 %
tr
tr
001.
003.
001.
004.
019.
048.
063.
037.
065.
023.
041.
Oil.
020.
216.
552.
0.06
tr
0.09
0.55
0.13
0.66
3.53
8.75
11.40
6.68
11.75
4.15
7.37
2.07
3.69
39.14

                          125

-------
Table 19. (Cont'd).  Hydrocarbon Sample Concentration  and
                     Composition.
Carbon  Sample #18  Sample #19  Sample #20  Sample #21
No.     ug/1    %   ug/1    %     /l    %   itg/1    %
10                   tr    0.03   -
11                   tr     tr
12                   tr    0.04   -
13        tr    tr    -----     -
14        tr    tr    -     -     2    0.84  36   27.21
15      005.  1.19  004.   0.81   2    0.74  16   11.56
16      023.  5.14  036.   8.14  25   12.62  29   21.44
17      020.  4.54  025.   5.61  27   13.12
18      061. 13.69  052.  11.86  33   16.37
19      012.  2.70  020.   4.60  21   10.38  10    7.15
20      054. 12.06  029.   6.62   9    4.53   2    1.68
21      023.  5.11  Oil.   2.59   8    3.96   3    2.10
22      033.  7.38  046.  10.50  15    7.55   7    5.04
23      010.  2.27  020.   4.60   9    4.53
24      028.  6.24  059.  13.23  21   10.18
 ?      178. 39.69  139.  31.39  31   15.28  0.32 23.53
Total   448.        442.        202          1.36
                         126

-------
Table 19.  (Cont'd).  Hydrocarbon Sample Concentration and

                      Composition.



Carbon   Sample #22  Sample #23  Sample #24  Sample #25
vf—       ..«. /"i     oy    —./i    o/      /£     -*•      «—    —
                 /o   Ug/1    /o    (^ft/1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
7
Total

-
12
21
-
-
52
27
15
6
90
-
1.14
3.32
-
-
3.
6.
-
-
12.
8.
4.
1.
27.
-
34.



77
99


90
52
14
80
12

62

3
50
45
80
-
-
52
50
9
18
35
23
130
4.95
0.54
10.12
9.09
16.25
-
-
10.47
10.16
1.85
3.70
7.09
4.62
26.18


9
16
20
-
-
16
10
9
4
5
12
39
141

6.
11.
14.
-
-
11.
7.
6.
2.
3.
8.
27.


49
35
33


22
30
76
97
24
38
57

**"* -
3
65
71
77
-
-
51
11
11
17
25
44
153
528
0.
12.
13.
14.
-
-
9.
2.
2.
3.
4.
8.
28.

51
24
36
65


67
11
17
18
77
37
97

                           127

-------
Table 19. (Cont'd).  Hydrocarbon Sample Concentration and
                     Composition.

Carbon  Sample #26  Sample #27  Sample #28   Sample #29
No.     ug/1    7o   tig/1    7o   «g/l    7o    ug/1    7,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

7
Total
-
2
59
37
34
-
-
53
54
8
7
9
40

135
437

0
13
8
7


12
12
1
1
2
9

BO

—
.44
.52
.46
.79
-
-
.04
.39
.74
.59
.09
.07

.79

*
tr
23
13
47
-
-
19
18
9
14
3
11

65
222


10
5
21


8
8
4
6
1
4

29

*•
tr
.47
.84
.29
-
-
.41
.24
.12
.18
.37
.72

.14

-
9
75
31
86
-
-
81
75
27
197
93
31

296
1000

0
7
3
8


8
7
2
19
9
3
*•>
29

-
.86
.53
.13
.56
-
-
.08
.47
.74
.67
.30
.13

.64

•*
-
51
77
84
-
-
40
34
10
16
69
64

139
535
-
-
9.62
4.99
15.75
-
-
7.41
6.41
1.85
2.99
12.97
11.97

26.01

                         128

-------
Depending on the state of compression of the organic

                   f?Ce fUm'  this lflyer shouSFSEL
                   of organics to considerably affect, by
100                                                     the
In Tables 20 and 21 fatty acid and hydrocarbon concentra-
tions are reported for samples taken in a slick, out of the
slick and in the slick with a bottle dipped under the
surface.  The fatty acids are more concentrated in the slick
samples than in non-slick samples.  Just the opposite is
true for the hydrocarbon samples, indicating the slick may
force more hydrocarbons into solution outside and under the
slick.
The fatty acid concentrations for non-slick areas and
bottle samples are in  the  same range as reported by Slowey,
Jeffrey, and Hood (1962) for the Gulf of Mexico surface
waters.  At this point, it seems these results do not
support Garrett's claim (1967) that surface fatty acid
concentrations are independent of slicks in the area.
Table 20.  Fatty Acid Concentrations in Slicks and
           Non-Slicks.
Slick
(Skimmer)
mg/1
1
2
3
(Samples
(Samples
(Samples
21,
24,
27,
22,
25,
28,
23)
26)
29) '
4
1
1
.967
.745
.080
Non-Slick
(Skimmer)
mg/1
0
0
0
.732
.305
.665
Bottle
(SHck)
mg/1
3.
0.
0.
95
608
653
Table 21.  Hydrocarbon Concentrations in Slicks and
           Non-Slicks.
Slick
( Skimmer)
mg/1
1
2
3
(Samples
(Samples
(Samples
21
24
27
,22,
,25,
,28,
23)
26)
29)
0.
0.
0.
136
141
222
Non-Slick
(Skimmer)
mg/1
0
0
1
.332
.528
.000
Bottle
(Slick)
mg/1
0
0
0
.495
.437
.535
                         129

-------
                      SECTION X

                      REFERENCES
Abbott, M.B., and Hayashi, T. (1967).  jta Proc. 14th
     Coastal Engineering Conference in Japan, Japan Soc.
     Civil Eng., Tokyo, pp. 226-229.

Ahlstrom, E. (Chairman), (1969).  Recommended procedures
     for measuring the productivity of plankton standing
     stock and related oceanic properties.  Nat. Acad.
     Sci., 59 pp.

Allen, A.A. (1969).  Santa Barbara oil spill.  Statement
     presented to the U. S. Senate Interior Subcommittee
     on Minerals, Materials, and Fuels, May, 1969.
     General Research Corp., Santa Barbara, Calif.,
     11 pp.

Allen, A.A., and Schleuter, R.S. (1969).  Estimates of
     surface pollution resulting from submarine oil seeps
     at Platform A and Coal Oil Point.  General Research
     Corp., Santa Barbara, Calif., TM-1230, 43 pp.

Anding, D., and Kauth, R. (1970).  Estimation of sea
     surface temperature from space.  Remote Sensing of
     Environment 1(4), pp. 217-220.

Annual Book of ASTM Standards (1970).  Parts 17 and 18,
     American Society for Testing and Materials,
     Philadelphia, Pa.

Barringer, A.R. (1968).  The remote-sensing of spectral
     signatures applied to pollution measurements and
     fish detection.  In Technical Papers, 9th Meeting,
     Ad Hoc Spacecraft Oceanography Advisory Group,
     U. S. Naval Oceanographic Office, Washington, D.C.,
     13 pp.

Battelle Mem. Inst., Pac. Northw. Lab. (1967).  Oil
     spillage study; literature search and critical
     evaluation for selection of promising techniques to
     control and prevent damage.  Distr. by CFSTI, Rep.
     No. AD 666 289, Richland, Wash., pp. 4/6-4/7.

Battelle Mem. Inst., Pac. Northw. Lab. (1969).  Study of
     equipment and methods for removing oil from harbor
     waters.  Rept. No. CR.70.001, Richland, Wash.,
     pp. 4/6-4/8.
                          130

-------
Bellaire, F.R. (1963).  Relation of winds, water levels
     and water temperature to currents in lower Lake
     Michigan.  Univ. Mich., Great Lakes Res. Div.,
     Publ. No. 10, Ann Arbor, pp. 219-230.

Berridge, S.A.; Dean, R.A.; Fallows, R.G.; and Fish, A.
     (1968) .  The properties of persistent oils at sea.
     J. Inst. Petrol. 54(539), 300 pp.

Blanchard, D.C, (1964).   Sea-to-air transport of surface
     active material.  Science, pp. 396-397.

Blokker, P.C. (1964).  Spreading and evaporation of
     petroleum products on water.  In Verlagboek van het
     Vierde Internationaal Havenkongres (Proc. 4th Int.
     Harbor Conf.), Koninklijke Valaamse Ing., Antwerp,
     Belgium, pp. 911-919.

Boylan, D.B., and Tripp,  B.W. (1971).  Determination of
     hydrocarbons in  sea  water extracts of crude oil and
     crude  oil fractions.  Nature 230, pp. 44-47.

Breidenbach, A.W.; Lichtenberg, J.J.; Henke, C.F.; Smith,
     D.J.;  Eichelberger,  J.W., Jr.; Stierle, H. (1966).
     The identification and measurement of chlorinated
     hydrocarbon pesticides in surface waters.  U. S.
     Dept.  of Interior, F.W.P.C.A.

Brown, E.,  and Nishioka,  Y.A. (1967).  Pesticides in
     selected western streams - A contribution to the
     National Program.  Pesticide Monitoring Journal,
     Vol.  1, No. 2, Sept. 1967.

Brown, T.E., and Richardson, F.L. (1968).  The effect of
     growth environment on the physiology of algae:
     light intensity.  J. Phycol. 4, pp. 38-54.

Buettner,  K.J.K., and Kern, C.D. (1965).  The determination
     of  infrared emissivities of terrestrial surfaces.
     J.  Geophys. Res. 70(6), pp. 1329-1337.

Buettner,  K.J.K.; Kern, C.D.; and Cronin, J.F. (1965).
  , *  The consequences of  terrestrial surface infrared
  ""  "emissivity.  In  Proc. 3rd Symp. Remote Sensing of
     "Environment, IJniv. Michigan, Ann Arbor, pp. 549-561.

Catoe, C.E. (1970).   Results of overflights of Chevron oil
  ~"  spill  in Gulf of Mexico.  U. S. Coast Guard Head-
     quarters, Washington, D.C., Project 714104/A/004,
     unpag.
                           131

-------
Chandler, P.B. (1969).  Study of clean-up activity
     effectiveness of the Santa Barbara, California,
     oil spill.  North American Rockwell Corporation
     Space Division, Santa Barbara, SD 69-178.

Chandler, P.B. (1970).  Remote sensing of oil polluted
     seawater.  In The Environmental Challenge of the
     70's, Proc.~T6th Ann. Techn. Mtg. Institute of
     Environmental Sciences, Boston, pp. 336-341.

Coast and Geodetic Survey (1969, 1970).   Tidal Current
     Tables, Atlantic Coast of North America.

David, P.M. (1965).  The surface fauna of the ocean.
     Endeavor 24(92), pp. 95-100.

Doebler, H.J. (1966).  A study of shallow water wind
     drift currents at two stations off the east coast
     of the United States.  U. S. Navy Underwater Sound
     Laboratory,  USL Rept. No. 755, New London, May.

Doty, M.S. (1956).  Current status of carbon-fourteen
     method of assaying productivity of the ocean.   Mimeo,
     Univ. Hawaii, Ann. Rep.

Duke, T.W.; Lowe, J.I.; and Wilson, A.J., Jr. (1970).
     A Polychlorinated Biphenyl (Aroclor 1254) in the
     water, sediment, and biota of Escambia Bay, Florida.
     Bulletin of Environmental Contamination and Toxi-
     cology, Vol. 5, No. 2.

Estes, J.E., and Golomb, B. (1970).  Oil spills: method
     for measuring their extent on the sea surface.
     Science 169, pp. 676-678.

Fay, J.A. (1969).  The spread of oil slicks on a calm sea.
     In Oil on the Sea, D. Hoult (Ed.),  Plenum Press, New
     York, pp. 53-64.
                                                         i*
Galtsoff, P.S.; Prytherich, H.F.; Smith, R.O.; and
     Koehring, V. (1936).  Effects of crude oil pollution
     on oysters in Louisiana waters.  Bull. U. S. Bur.
     Fish. 18, pp. 143-210.                            i-

Gamier, B.J. (1971).  The observation of topographic
     variations in surface radiative temperatures by
     remote sensing from a small aircraft.  In Proc. 7th
     Int. Symp. Remote Sensing of Environment, Univ.
     Michigan, Ann Arbor.
                          132

-------
Garrett, W.D. (1965).  Collection of slick-forming
     materials from the sea surface.  Limnol. Oceanogr
     10, pp. 602-605.

Garrett, W.D. (1967a).  The organic chemical composition
     of the ocean surface.  Deep-Sea Res. 14(2),
     pp. 221-227.

Garrett, W.D. (1967b).  Stabilization of air bubbles at
     the air-sea interface by surface active material.
     Deep-Sea Res. 14, pp. 661-672.

Garrett, W.D. (1968).  The influence of monomolecular
     surface films on the production of condensation
     nuclei from bubbled seawater.  J. Geophys. Res.
     73, pp. 5145-5150.

Gordon, J.E., and Thome, R.L.  (1967).  Salt effects on
     the activity coefficient of naphthalene in mixed
     aqueous solutions.  I. Mixtures of two salts.
     J. Phys. Chem.  71, pp. 4390-4399.

Griggs, M.  (1968).   Emissivities of natural surfaces in
     the 8- to 14-micron spectral region. J. Geophys.
     Res.,  73, pp. 7545-7551.

Guinard, N.W., and Purves, C.G. (1970).  The remote sensing
     of oil slicks by radar.  U. S. Naval Research
     Laboratory R07-02, Washington, D.C., 8 pp.

Guinard, N.W. (1971).  Radar monitoring of oil slicks.
     In Proc. 7th Int. Symp. Remote Sensing of Environ-
     ment, Univ. Michigan, Ann  Arbor.

Haight, F.J. (1942).  Coastal currents along the Atlantic
     Coast  of the United States.  Spec. Pub. 230, U. S.
     Coast  and Geodetic Survey, pp. 73.

Harrison, W., and Pore, N.A.  (1967).  Shelf waters off the
     Chesapeake Bight, Part III.  ESSA Professional Paper
     3, Washington,  pp. 43.

Hartung, R., and Klinger, G.W.  (1970).  Concentration of
     p,p* DDT by sedimented polluting oils.  Environ. Sci.
     Technol. 4, pp.  407-409.

Harvey, G.S. (1965).  Microlayer collection from  the sea
     surface: a new  method and  initial results.  Limnol.
     Oceanogr. 11, pp. 608-613.
                           133

-------
Harvey, G.R.; Bowen, V.T.; Backus, R.H.; and Grice, G.D.
     (1971).  Chlorinated hydrocarbons in open-ocean
     Atlantic organisms.  To be published in the Symposium
     Proceedings Nobel Symposium, "The Changing Chemistry
     of the Oceans," Sweden, August 16-20, 1971.

Hasell, P.G., Jr., and Larsen, L.M. (1968).  Calibration of
     an airborne multispectral optical sensor.  Univ.
     Michigan, Willow Run Lab., Ann Arbor, EGOM-00013-137,
     87 pp.

Hela, I. (1952).  Drift currents and permanent flow.
     Commentationes Physico-Mathematicae, Societas
     Scientiarum Fennica, Vol. XVI, No. 14, pp. 1-28.

Hohn, M.H. (1959).  The use of diatom populations as a
     measure of water quality in selected areas of
     Galveston and Chocolate Bay, Texas., Inst. Mar.
     Sci. 6, pp. 206-212.

Holden, A.V., and Marsden, K. (1969).   Single-stage
     clean-up of animal tissue extracts for organochlorine
     residue analysis.  J. Chromatog., 44, pp. 481-492.

Holmes, R.W., and Widrig, T.M. (1956).  The enumeration
     and collection of marine phytoplankton,   J. Cons.
     Int. Explor. Mer. 22, pp. 21-32.

James, R.W. (1968).  Wind drift currents.  Mariners
     Weather Log 12, pp. 8-10.                :

Jarvis, N.L.; Garrett, W.D.; Schieman, M.A. ;  and Timmons,
     C.O. (1967).  Surface chemical characterization of
     surface-active material in seawater.  Limnol.
     Oceanogr. 12, pp. 88-96.

Johannessen, O.M. (1968).  Some current measurements in
     the Drobak Sound, the narrow entrance to the       - ;
     Oslofjord.  Hvalradets Skrifter Norske Videnskaps-
    1 Akad. Oslo 50.

Jones, D.L., and Bellaire, F.R. (1962).  A numerical
     procedure for computing wind-driven currents on Tthe-,.;
     Great Lakes.  Univ. Mich., Great Lakes Res. Div.,
     Publ. No. 9, Ann Arbor, pp. 93-102.

Kator, H; Oppenheimer, C.H.; and Miget, R.J.  (1971).  ;
     Micirobial degradation of a Louisiana crude oil in
     closed flasks and under simulated field conditions.
     Proc. Joint Conf. on Prevention and'Control of Oil
     Spills, API-EPA-USCG, pp. 287-296. ,           ,;
                          134

-------
Koeman, J.H., and van Genderen, H.  (1970).  FAO Technical
     Conference on Marine Pollution and its Effects on
     Living Resources and Fishing, Rome, Italy, 9-18 Dec.,
     MP/70/E-21, 17 Sept.

Kolpack, R. (1969) .  Movement of oil and water in Santa
     Barbara Channel, California.  At the API-FWPCA Joint
     Conference on Prevention and Control of Oil Spills,
     New York, December 15-17.

Laevastu, T. (1962).  The causes and predictions of
     surface currents in sea and lake.  Hawaii Jnst.
     Geophys. Rept. No. 21, July.

Lee, K. (1969) .  Infrared exploration for shoreline springs
     at Mono Lake, California, test site.  In Proc. 6th
     Int. Symp. Remote Sensing on Environment, Univ.
     Michigan, Ann Arbor, pp. 1075-1100.

Lowe, D.S., and Hasell, P.G. (1969).  Multispectral
     sensing of oil pollution.  In Proc. 6th Int. Symp.
     Remote Sensing of Environment, Univ. Michigan, Ann
     Arbor, pp. 755-765.

Lund, J.W.; Kipling, C.; and Le Cren, E.D. (1958).  The
     inverted microscope method of estimating algal
     numbers and the statistical basis of estimates by
     counting.  Hydrobiologia 11, pp. 143-170.

Mandelbaum, H.  (1955).  Wind-generated ocean currents at
     Amrum Bank Lightship.  Trans. Amer. Geophys. Un.
     36, pp. 72.

Marlatt, W.E. (1967).  Remote and in situ temperature
     measurements  of land and water surfaces.  J. Applied
     Meteorology 6(2), pp.  272-279.

Marlatt, W.E., and Harlan,  J.C. (1971).  A study of the
     attenuation by atmospheric particulates of thermal
     infrared radiation.  In Proc.  7th Int. Symp. Remote
     Sensing  of Environment, Univ. Michigan, Ann Arbor.

McAlice, B.J. (1970).  Observations on the small-scale
     distribution of estuarine phytoplankton.  Mar. Biol.
     7, pp. 100-111.

McAlister, E.D. (1969).  Measurement of the total heat
     flow from the sea surface with an infrared two-
     wavelength radiometer: progress report.  In Oceans
     from Space, P;C. Badgley, L. Miloy, and L. Childs
     (Eds.), Gulf Publ. Co., Houston, pp. 188-201.
                           135

-------
McCauley, R.N. (1966).  The biological effects of oil
     pollution in a river.  Limnol. Oceanogr., 11,
     pp. 475-486.

Metcalfe, L.D., and Schmitz, A.A. (1961).  The rapid
     preparation of fatty acid esters for gas chromato-
     graphic analysis, Anal. Chem., 33, pp. 363.

Mironov, O.G., and Lanskaja, L.A. (1966).  The influence
     of oil on the development of marine phytoplankton.
     Int. Oceanogr. Congr. 2, Moscow.

Mironov, O.G., and Lanskaja, L.A. (1968).  The capacity
     of survival in the sea water, polluted with oil
     products, inherent in some marine planktonic and
     bentho-planktonic algae.  Botanicheskii Zhurnal 53,
     pp. 661-669.

Mulhern, B.M. (1968).  An improved method for the
     separation and removal of organochlorine insecti-
     cides from thin-layer plates.  J. Chromatog., 34,
     pp. 556-598.

Munday, J.C.; Maclntyre, W.G.; Penney, M.E.; and
     Oberholtzer, J.D. (1971).  Oil slick studies using
     photographic and multispectral scanner data.  Proc.
     7th Int. Symp. on Remote Sensing of the Environment,
     Willow Run Laboratories, Univ. of Michigan, pp. 1027-
     1043.

Murray, S.P.; Smith, W.G.; and Sonu, C.J. (1970).
     Oceanographic observations and theoretical analysis
     of oil slicks during the Chevron Spill, March, 1970,
     Louisiana State Univ., Coastal Studies Inst., Baton
     Rouge, Tech. Rep. No. 87, 106 pp.

Oshiver, A.H.; Stone, R.B.; Clark, J.R.; and Berberian,
     G.A. (1965).  Factors in measurement of absolute sea
     surface temperature by infrared radiometry.  In
     Proc. 3rd Symp. Remote Sensing of Environment, Univ.
     Michigan, Ann Arbor, pp. 737-762.

Parker, B., and Barsom, G. (1970).  Biological and chemical
     significance of surface microlayers in aquatic
     ecosystems, BioScience 20(2), pp. 87-93.

Patrick, R. (1949).  A proposed biological measure of  .-,'
     stream conditions based on a survey of the
     Conestoga Basin, Lancaster County, Pa.  Proc. Acad.
     Nat. Sci. Phila., 101, pp. 277-341.
                          136

-------
Patrick, R.; Hohn, M.H.; and Wallace, J.H. (1954).  Anew
     method for determining the pattern of the diatom
     flora.  Acad. Nat. Sci. Phila., Notulae Naturae,
     259, pp. 1-12.

Patrick, R., and Strawbridge, D. (1963).  Methods of
     studying diatom populations.  J. Water Poll. Contr.
     Fed., 35(2), pp. 151-157.

Pickett, R.L. (1966).  Environmental corrections for an
     airborne radiation thermometer.  In Proc. 4th Symp.
     Remote Sensing of Environment, Univ. Michigan, Ann
     Arbor, pp. 259-262.

Pielou, E.G. (1966).  Shannon's formula as a measurement
     of specific diversity and its use and misuse.  Amer.
     Natur., 100, pp. 463-465.

Reynolds, L.M.  (1969).  Polychlorobiphenyls (PCB's) and
     their  interference with pesticide residue analysis.
     Bulletin of Environmental Contamination & Toxicology,
     Vol. 4, No. 3.

Ricker, W.E. (1937).  Statistical treatment of sampling
     processes  in the enumeration of plankton organisms.
     Arch.  Hydrobiol. Plankt., 22, pp. 643-645.

Ries, H.E., and Grutsch, J.F.  (1968).  Removal of surface
     contaminants from refinery waste water, presented at
     155th  National Meeting of the American Chemical
     Society, San Francisco, California, April 5, 1968.

Ries, H.E., and Gabor, J.  (196 ).  Monolayer transfer to a
    , rotating cylinder: surface flow patterns, Nature 212,
     No.  5065,  pp. 917-918.

Sanders,  H.L. (1960).  Benthic studies in Buzzards Bay III.
     The  structure of the  soft bottom community.  Limnol.
     Oceanogr.  5, pp. 138-153.

Saunders, P.M., and Wilkins, C.H. (1966).  Precise ariborne
     radiation  thermometry.  In Proc. 4th Symp. Remote
     Sensing of Environment, Univ. Michigan, Ann Arbor,
     pp.  815-826.

Saunders, P.M.  (1967).  Aerial measurement of sea surface
     temperature in  the infrared.  J. Geophys. Res.
     72(16), pp. 4109-4117.
                           137

-------
Schwartzberg , H.G. (1971).  The movement of oil spills.
     Joint Conf. on Prevention and Control of Oil Spill
     Washington, D.C.,. Spon. API-EPA-USCG, pp. 489-494.
Seba, D.B., and Corcoran, E.F. (1969).  Surface slicks as
     concentrators of pesticides in the marine environment.
     Pesticide Monitoring Journal, Vol. 3, No. 3, December.

Shannon, C.E. (1948).  A mathematical theory of communi-
     cation.  Bell Syst. Tech. J. 27, pp. 379-423, 623-656.

Small, L.P. (1961).  An optical density method of measuring
     phytoplankton standing crop.  Iowa State J. Sci.
     35(3), pp. 343-354.

Smith, J.E., Ed. (1968).  Torrey Canyon Pollution and
     Marine Life.  Cambridge Univ. Press, pp. 150-161.

Sorokin, C., and Krauss, R.W. (1958).  The effects of
     light intensity on the growth rates of green algae.
     PI. Physiol. 33, pp. 109-113.

Stahl, E. (1969).  Thin-layer chromatography.  Academic
     Press, New York, N.Y., pp. 1-1041.

Stauffer, T.B. (1969).  The distribution of dissolved
     fatty acids in the James River estuary and adjacent
     ocean water, Ches. Sci. 11, pp. 216-220.

Steeman, Nielsen E.  (1952a) .  On detrimental effects of
     high light intensities on the photosynthetic
     mechanisms.  Physiol. Plant, 5, pp. 334-344.

Steeman, Nielsen E. (1952b) .  The use of radioactive
     carbon (C^) for measuring organic production in the
     sea.  J. du Cons. 18, pp. 117-140.

Steeman, Nielsen E. (1962).  Inactivation of the photo-
     chemical mechanism in photosynthesis as a means to
     protect the cells against too high light intensities.
     Physiol. Plant, 15, 505-517.

Stewart, S.; Spellicy, R. ; and Polcyn, F. (1970).  Analysis
     of multispectral data of the Santa Barbara oil slick.
     Univ. Michigan, Willow Run Lab., Ann Arbor, 3340-4-F,
     57 pp.

Sturdy, G., and Fischer, W.H. (1966).  Surface tension of
     slick patches near kelp beds.  Nature 211, pp. 951-952.
                          138

-------
Tolbert, W.H., and Salsraan, G.G. (1964).  Surface circu-
     lation of the eastern Gulf of Mexico as determined by
     drift-bottle studies.  J. Geophys. Res., 69, p. 223.

Weiss, M. (1963).  Sea and earth surface temperature
     measurement using infrared.  In Proc. 2nd Symp.
     Remote Sensing of Environment, Univ. Michigan, Ann
     Arbor, pp. 343-357.

Wilhm, J.L., and Dorris, T.C.  (1968).  Biochemical
     parameters for water quality criteria.  BioScience
     18, pp. 477-481.

Zitko, V. (1971a).  Effects of pesticide-grade hexanes on
     the silicic acid chromatography of polychlorinated
     biphenyls and organochlorine pesticides.  J. Chromatogr.
     59, pp. 444-445.

Zitko, V., and Choi, P.M.K. (1971b).  PCB and other
     industrial halogenated hydrocarbons in the environment.
     Fisheries Research Board  of Canada, Technical Report
     No. 272.

Zobell,  C.E.  (1964).  The occurrence, effects, and fate of
     oil polluting the sea.  Contr. Scripps Inst. Oceanogr.
     34(1), pp. 1257-1283.
                           139

-------
                      SECTION XI

                 LIST OF PUBLICATIONS
Munday, J.C.; Harrison, W.; and Maclntyre, W.G., 1970.
     Oil slick motion near Chesapeake Bay Entrance.
     Jour. Am. Water Res. Assn., j6, 879-884.

Munday, J.C.; Maclntyre, W.G.; Penney, M.E.; and
     Oberholtzer, J.D., 1971.  Oil slick studies using
     photographic and multispectral scanner data.
     Proc. 7th Int. Symp. on Remote Sensing of the
     Environment, Willow Runn Labs., Univ. of Michigan,
     p. 1027-1043.

Roy, V.M.; Dupuy, J.L.; Maclntyre, W.G.; and Harrison, W.,
     1970.  Abundance of marine phytoplankton in surface
     films: a method of sampling.  Proc. Nat. Symp. on
     Hydrobiology, A.W.R.A., Miami, Fla., p. 381-389.

Smith, C.L., and Maclntyre, W.G., 1971.  Initial aging of
     fuel oil films on sea water.  Joint Conf. on
     Prevention and Control of Oil Spills, Washington,
     D.C., API-EPA-USCG, p. 457-461.
                          140

-------
                      SECTION XII

                      APPENDICES

                                                       Page

A.   Program Description for Oil Slick Motion          142


B.   Lipid Analyxis                                    152


C.   Effects of an Accidental Spill of No. 6           156
     Fuel Oil on a Salt Marsh
                           141

-------
                      APPENDIX

                  PROGRAM DESCRIPTION
                          FOR
                    OIL SLICK MOTION

The program computes current paths from input wind and tidal
current data; and predicts the motion of surface water
masses or oil slicks on surface water masses, in the case
when the motion is due to wind and to surface tidal
currents.  Suitable tidal current data can be obtained from
tables of the U. S. Coast and Geodetic Survey.  In a single
run, current paths can be computed for any number of sets
of input data.  The program is processed by an IBM 1130
computer.

The program first reads housekeeping data and labels the
printer output page.  Then, tidal current data from up to
ten stations are read, consisting of station latitude and
longitude, times of slacks and maximum currents, and
current velocities.  Finally, the initial location of
interest, a start and stop time, wind factor, and wind
velocity versus time are read as input.

The bulk of computation takes place in a cycle within a
loop, in which each pass of the cycle accounts for a single
station.  For each station, the phase of the tidal cycle is
determined by comparing the time of interest with the times
of slack waters, maximum flood, and maximum ebb.  The
current magnitude between maxima is calculated by inserting
the phase in a sine wave and multiplying by the appropriate
current maximum.  The resulting magnitude is weighted by
the inverse square of the distance between the station and
the location of interest.  Orthogonal components (weighted)
of current velocity are formed from the product of this
magnitude and the sine or cosine of the appropriate input-
current direction.

Orthogonal (weighted) components are computed and summed
only for stations as close as 5 nautical miles to the
location of interest.  The sums are divided by the sum of
weights.  To these sums are added wind velocity components
multiplied by the wind factor.  The resulting sums are
multiplied by a time increment.  The distance components
are algebraically added to the coordinates of the initial
location to produce a new location, and the time increment
is added to the initial (start) time to produce a new time.
The new time and new coordinates are printed and control
                          142

-------
returns to the beginning of the Do-loop to allow repetition
of all computations using the new location and  time.
Repetition ceases when a stop time is reached.

If allowed by an input control number, control  returns to
nearly the beginning of the program in order that a new
current path can be computed based on a new set of input
data.
                      INPUT DATA

A.  Card, Column, Format, and Description

Card  Column  Format                Description

  1     1-10   2A5      Dl, D2, label or title to be speci-
                        fied by user.  A blank card is
                        permissible.

  2     1-4    14       K, number  of tidal current data
                        stations per data set.  The number
                        must be right-adjusted in the
                        field.  The number of stations
                        allowed is limited only by the
                        input format 14.

  3     1-4             KK, number of sets of data to be
                        processed, equal to the number of
                        current paths to be computed, where
                        each current path requires a data
                        set.  The  number of data sets
                        allowed is limited only by the
                        input format 14.  The number must
                        be right-adjusted.

  A           12F6.0    ((A  (I, J), I = 1, 12), J = 1, K),
                        tidal current data for each of the
                        K stations.  All data fields
                        requires a decimal point.

        1-6    F6.0     A(1,J), degrees latitude of the
                        station.

        7-12   F6.0     A(2, J), minutes (plus decimal
                        fraction)  latitude.

       13-18   F6.0     A(3, J), degrees longitude.
                           143

-------
Card  Column  Format                Description

       19-24   F6.0     A(4, J), minutes (plus decimal
                        fraction) longitude.

       25-30   F6.0     A(5, J), time of slack water in
                        hours.  All times must be in hours
                        of military time (with a decimal
                        point).  The slack water may be
                        that before either flood or ebb,
                        where the choice is made by the
                        user so that the input data times
                        bracket the desired start and stop
                        time.

       31-36   F6.0     A(6, J), time of the maximum
                        current succeeding the slack time
                        A(5, J), in hours military time.

       37-42   F6.0     A(7, J), magnitude of the maximum
                        current A(6, J) in knots.

       43-48   F6.0     A(8, J), direction of the maximum
                        current A(6, J) in degrees
                        (toward).

       49-54   F6.0     A(9, J), time of succeeding slack
                        water in hours military time.

       55-60   F6.0     A(10, J), time of succeeding
                        maximum current in hours military
                        time.

       61-66   F6.0     A(ll, J), magnitude of the maximum
                        current A(10, J) in knots.

       67-72   F6.0     A(12, J), direction of the maximum
                        current A(10, J) in degrees
                        (toward).

                        Additional cards A as needed,
                        controlled by K.

  B           12F6.0    (B(I), 1=1, 6), initial data, and
                        F, wind factor.

        1-6    F6.0     B(l), degrees latitude of initial
                        location.
                          144

-------
Card  Column  Format                Description

       7-12    F6.0     B(2), minutes  (plus decimal
                        fraction)  latitude of  initial
                        location.

      13-18    F6.0     B(3), degrees  longitude  of initial
                        location.

      19-24    F6.0     B(4), minutes  (plus decimal
                        fraction)  longitude of initial
                        location.

      25-30    F6.0     B(5), start time  in hours military
                        time.

      31-36    F6.0     B(6), stop time in hours military
                        time.

      37-42    F6.0     F,  wind factor, as a fraction
                        (F  £ 1.0)  of wind speed  assumed by
                        the moving object or water mass,
                        over and above the movement due to
                        tidal currents alone.

  W           12F6.0    «W(I,  J), I - 1, 2),  J  = 1, JJ),
                        the hourly wind velocity, where JJ
                        is  the  number  of  on-the-hour wind
                        velocities needed to bracket the
                        start and stop times.

       1-6     F6.0     W(l, J) , wind  magnitude  in knots
                        during  hour J-% to J + %.

       7-12    F6.0     W(2, J), wind  direction  in degrees
                        during  hour J-% to J + %.

                        Additional columns as  needed for
                        additional pairs  of wind magnitude
                        and direction, controlled by JJ.

                        Cards A, B, and W are  repeated  as
                        needed, controlled by  KK.


 B.   Preparation  of Input  Data

 Card 1:   The label is included  for convenience of  the user,
 It  can include the date or any  code number.
                           145

-------
 Card  2:   The number  of  tidal  current data  stations must be
 decided by  the  user.  The  decision  should  be based on the
 proximity of stations to the  initial location  and to
 estimated subsequent locations.  The density of stations in
 the area  of interest may need to be high,  if the area has a
 detailed  current  structure.   Stations  further  than 5
 nautical  miles  from  a location along the current path are
 neglected in the  computations.

 Card  3:   A  number of data  sets may  be  processed to yield a
 current path for  each set.  The number is  decided by  the
 user.

 Cards  4 to  3 +  K:  Tidal current data  can  be obtained from
 Tidal  Current Tables of the Coast and  Geodetic  Survey,
 published by the  U.  S.  Department of Commerce.   Data  from
 each  station are  input  on  a separate card.  The latitude
 and longitude are obtained from Table  2.   Slack water and
 maximum current times are  obtained  from Table  1 with  time
 difference  corrections  obtained from Table 2.   Magnitudes
 of maximum  currents  are obtained from  Table 1 with velocity
 ratio  corrections obtained from Table  2.   Directions  of
 currents  are obtained from Table 2.

 Card 4 +  K:  Initial location data, start  and  stop times,
 and wind  factor F are decided by the user.  The literature
 suggests  that F = 0.05  is  the maximum  wind factor
 encountered in  the field.  The initial location must  be
 within 5  nautical miles of (at least)  one  station.

 Card 5 +  K:  The  hourly wind  velocity  data must be supplied
 by the user.


                   OUTPUT DATA

A.  Line,  Column,  Format,  and Description

                                    Description

                        User label.

                        Column headings: TIME  (in) HOURS,
                        LATITUDE  (in)  DEGREES  (and) MINUTES,
                        LONGITUDE  (in)  DEGREES  (and)
                        MINUTES.

 4                      Blank

 D      1-6    F6.0     Time  in hours military  time.
Line  Column  Format
                          146

-------
Line  Column  Format                Description

       13-17   F5.0     Degrees  latitude.

       27-31   F6.2     Minutes  (plus decimal fraction)
                        latitude.

       40-44   F5.0     Degrees  longitude.

       54-58   F6.2     Minutes  (plus decimal fraction)
                        longitude.

                        The  first line D includes the
                        start  time plus the initial
                        location.  Subsequent lines D
                        include  time and current path
                        location at quarter-hour
                        intervals, up to the stop time.


           LIMITATIONS AND VALIDATION

The following details of  the program limit its applicability
to a general situation:

1) The program has not been  tested for extended precision.

2) Input tidal current data  and  the program method do not
account for rotary current fields.  Single directions are
assumed for flood and ebb flow.

3) The program converts (forward and backward) from degrees
latitude into 59.881 nautical  miles, and from degrees
longitude  into 48.031 nautical miles.  Hence, the program is
presently  suited  (in North America) for only the Chesapeake
Bay entrance.  For use at other  locations, the conversion
factors must be changed in statements 18+2, 70+6, and
70 + 7.

4) If a computed current  path  moves more than 5 nautical
miles away from all data  stations, points of the path are
determined from the last  available computations in which
the path was within 5 nautical miles of a data station.^
This peculiar feature is  a result of allowing only stations
within 5 nautical miles to influence the current path.

5) If the  stop time exceeds  the  latest maximum current  time
of any station, computation  of current components will  be
based on the latest maximum  current velocity.
                          147

-------
The operation was validated by 5 digit hand calculations on
simulated data.  The hand calculations were based on the
program method as described in OUTPUT DATA, but not on the
program statements themselves.


SYMBOLS DEFINITION, EXPLANATION OF CONSTANTS, AND LIBRARY
ROUTINES

A.  Definition of Symbols in Alphabetical Order:

    A(I, J)  Tidal current data for J stations.

    B(I)     Initial location data; start and stop times.

    Cl, C2   Longitude and latitude of a station in
             fractional degrees.

    Dl, D2   a) Label or title;
             b) Longitude and latitude of a point on a
                current path in fractional degrees.

    F        Wind factor.

    Gl, G2   Longitude degrees and fractional minutes of a
             current path point.

    G3, G4   Latitude degrees and fractional minutes of a
             current path point.

    H        Current direction in radians (from north).

    J        Number of time intervals, based on start and
             stop times, and intervals of 0.25 hours.

    JJ       a) Number of hourly wind velocities needed to
                bracket start and stop times.
             b) Numbered hour for selection of wind velocity.

    K        Number of tidal current data stations.

    KK       Number of data sets to be processed.

    KKK      Counter of data sets processed.

    N        Counter of stations processed.

    R        Distance between a current path point and a
             station.

    S        Current magnitude.
                          148

-------
   T         Time in fractional hours.
   TA    ,   Stop time in fractional hours.
   TB       The on-hour time preceding start time.
   TT       Time in military hours.
   Tl,  T3   First and second slack water times in
             fractional hours.
   T2,  T4   First and second maximum current times  in
             fractional hours.
   U         Sum of weighting factors, 1/R .
   W(I, J)  Wind velocities for J hours.
   WM       Wind magnitude in knots.
   WD       Wind direction in degrees.
   X         Sum of north-south components of weighted
             surface current speed.
   Y         Sum of east-west components of weighted surface
             current speed.

B.  Explanation of Constants in Order of Appearance:
    48.031
    59.881
     1.5708
     0.0174533
     3.14159
   239.524
   192.124
Nautical miles per degree longitude.
Nautical miles per degree latitude.
7T/2, radians per 90 degrees.
tr/180, radians per degree.
TT , radians per 180 degrees.
59.881 times 4, where 4 results from a
numerator factor of 1/4 hour.
48.031 times 4, where 4 results from a
numerator factor of 1/4 hour.
                          149

-------
Program Listing

*IOCS(CARD,KEYBOARD,TYPEWRITER,113 2 PRINTER)
*LIST SOURCE PROGRAM
*ONE WORD INTEGERS
     DIMENSION A(12,10),B(6),W(2,10)
     KKK=0
     WRITE(3,4)
     READ(2,2)D1,D2
     WRITE(3,3)D1,D2
     READ(2,6)K
   7 READ(2,6)KK
   8 WRITE(3,5)
     READ(2,10) ((A(I,J),I=1,12),J=1,K)
     READ(2,10)B,F
     T=IFIX(B(5)/100.)
     T=T+(B(5)-T*100.)/60.
     TB=IFIX(T)
     TA=IFIX(B(6)/100.)
     TA=TA+(B(6)-TA*100.)/60.
     JJ=TA-TB+1.5
     READ(2,10)((W(I,J),I=1,2),J=1,JJ)
     WRITE(3,11) B(5),B(1),B(2),B(3),B(4)
     J=4.*(TA=T)+1
     DO 80 1=1,J
     U=0.
     X=0.
     Y=0.
     N=0.
  15 N=N+1
     IF(N-K)16,16,70
  16 IF (1-1)17,17,18
  17 D1=B(1)+B(2)/60.
     D2=B(3)+B(4)/60.
  18 C1=A(1,N)+A(2,N)/60.
     C2=A(3,N)+A(4,N)/60.
     R=((D2-C2)*48.031)**2+((D1-C1)*59.881)**2
     R=SQRT(R)
     IF (R) 19,19,21
  19 R=0.01
  21 IF(R-5.)20,15,15
  20 T1=IFIX(A(5,N)/100.)
     T1=T1+(A5,N)-Tl*100.)/60.
     T2=IFIX(A(6,N)/100.)
     T2=T2+(A(6,N)-T2*100.)/60.
     T3=IFIX(A(9,N)/100.)
     T3=T3+(A(9,N)-T3*100.)/60.
     IF(T-T2)25,29,30
  25 S=A(7,N)*SIN(1.5708*(T-T1)/(T2-T1))/R**2
     GO TO 50
                          150

-------
30 T4=IFIX(A(10,N)/100.)
   T4=T4+(A(10,N)-T4*100.)/60.
   IF (T-T3)35,39,40
35 S=A(7,N)*SIN(1.5708*(T-T3)/(T2-T3))/R**2
   GO TO 50
39 S=0
   GO TO 55
40 S=A(11,N)*SIN(1.5708*(T-T3)/(T4-T3))/R**2
55 H=A(12,N)*0.0174533
   GO TO 60
50 H=A(8,N)*0.0174533
60 X=X-S*SIN(H)
   Y=Y+S*COS(H)
   U=U+1/R**2
   GO TO 15
70 T=T+0.25
   TT=IFIX(T)
   TT=TT*100.+(T-TT)*60.
   JJ=*T-TB+1.5
   WM=W(1,JJ)
   WD=W(2,JJ)
   D1=D1+(Y/U+F*WM*COS (WD*0.0174533+3.14159) ) /239.524
   D2=D2+(X/U+F*WM+SIN (WD*0.0174533) ) /19 2.124
   G1=IFIX(D1)
   G2=(D1-G1)*60.
   G3=IFIX(D2)
   G4=(D2-G3)*60.
   WRITE(3,11)TT,G1,G2,G3,G4
80 CONTINUE
   KKK=KKK+1
   IF(KKK-KK) 82,83,83
82 WRITE (3,4)
   GO TO 8
83 CALL EXIT
 2 FORMAT(16A5)  -
 3 FORMAT(/16A5/)
 4 FORMAT(1H1)
 5 FORMAT(5H TIME,7X,8HLATITUDE,20X9HLONGITUDE/6H HOURS,
  6X,2(7HDEGREE IS7X7HMINUTES 7X)/)
 6 FORMAT(1014)
10 FORMAT(12F6.0)
11 FORMAT(F6.0,6X,2(F5.0,9X,F6.2,8X))
                        151

-------
                      APPENDIX B
                                          3
                    LIPID ANALYSIS

Sampling Site Locations, Dates, and Descriptions.

Sample #1:   August 18, 1970     0940 hrs
    Long. 76°32?05"              Lat. 37°15I33"
    Water temp. 28.0°C           Ambient temp. 27°C

The slick extended from the Yorktown Naval Weapons Station
(N.W.S.) to about 1/2 mile northeast of the N.W.S. dock.
The slick also extended to mid-channel.  Light slick
material, no particulate matter and no foam were noted.

Sample #2:   August 18, 1970     1105 hrs
    Long. 76°28?37"              Lat. 37°14f30"
    Water temp. 28°C             Ambient temp. 27°C

The slick was in the vicinity of the Coleman Bridge, and
was about 2 miles long and 40 feet wide.  Little
particulate material, no foam and heavy damping of
capillary waves was reported.  Increasing winds were
experienced.

Sample #3:   August 18, 1970     1215 hrs
    Long. 76°28?37"              Lat. 37015'29"
    Water temp. 29.5°C           Ambient temp. 30°C

Due to increasing winds, this sample was taken in the
Sarah's Creek area.  The slick was reported as very light
with some particulate material and very little foam.
Sampling was discontinued due to the wind conditions.

Sample #4:   August 27, 1970     1200 hrs
    Long. 76°34755"              Lat. 37°17'35"
    Water temp. 28°C             Ambient temp. 29°C

A slick was found off Cheatham Annex.  It was very large
but it was light, with little foam and particulate
material.  The slick extended in both directions from the
pier, up and down river and well out into the main channel.
The winds were light.

Sample #5:   August 27, 1970     1330 hrs
    Long. 76°3r48lf              Lat. 37°15f22fl

The slick was located off the Naval Weapons Station in
Yorktown in up and down river directions.  It was light and
had little foam and particulate matter.
                          152

-------
Sample #6:   August 28, 1970    1200 hrs
    Long. 76°48Y25"             Lat. 37°31'54"
    Water temp. 28.5°C          Air temp. 29°C

The slick was located just south of the Pamunkey River
Bridge at West Point near the Chesapeake Corporation Plant.
Particulate material and some foam was noted in a heavy
slick.

Sample #7:   August 28, 1970    1245 hrs
    Long. 76°45?10"             Lat. 37°29f8"
    Water temp.                 Ambient temp.

Down river from West Point a heavy slick with much
particulate material and foam was noted.

Sample #8:   August 28, 1970    1430 hrs
    Long. 76°32T57"             Lat. 37°16f37"
    Water temp. 29°C            Ambient temp. 29°C

A very large slick with much particulate material and foam.
It extended up and down river for several miles with a
width about 50 yds.

Sample #9:   September 24, 1970  1030 hrs
    Long. 76°26*39"              Lat. 37013'34"
    Water temp. 27°C             Ambient temp. 24°C

A slick  was sampled off the Amoco (Yorktown) pier.  It was
light with some foam and particulate material.

Sample #10:  September 24, 1970  1145 hrs
    Long. 76°30f18"              Lat. 37014'30"
    Water temp. 27°C             Ambient temp. 24°C

This sample was taken under the Coleman Bridge.  The slick
was light with no foam and little particulate material.

Sample #11:  September 24, 1970  1300 hrs
    Long. 37°15^34"              Lat. 76°32I07"
    Water temp. 27°C             Ambient temp. 25°C

This slick was light with some foam and little particulate
material.  The sample was taken in the Naval Weapons
Station  area.

Sample #12:  May 15, 1971        0915 hrs
    Long. 76°23V               Lat. 37°14'20M
    Water temp. 18°C             Ambient temp. 20°C
                           153

-------
The sample was taken in the Chesapeake Bay near the mouth
of the York River.  The slick was very light with no foam
or particulate material.

Sample #13:  May 15, 1971       0945 hrs
    Long. 76°251                Lat. 37°13f55"
    Water temp. 18°C            Ambient temp. 20°C

Down river from the Amoco (Yorktown) Refinery a slick was
noted.  There was some foam but the slick was light.

Sample #14:  May 15, 1971       1015 hrs
    Long. 76026133"             Lat. 37013'32"
    Water temp. 18°C            Ambient temp. 21°C

A very heavy foamy slick was sampled just west of the
Amoco Refinery pier.  Hundreds of dead jellyfish were noted
in the slick.

Sample #15:  May 15, 1971       1050 hrs
    Long. 76°29125"             Lat. 37014!17"
    Water temp. 18°C            Ambient temp. 21°C

A slick was noted close to the Virginia Institute of Marine
Science pier.  Winds were increasing.  There was some foam
and some particulate material.

Sample #16:  May 15, 1971       1125 hrs
    Long. 76°32'I2011             Lat. 37°17'
    Water temp. 19°C            Ambient temp. 21°C

This slick was light but had much organic material.  Winds
were increasing.

Sample #17:  May 15, 1971       1205 hrs
    Long. 76°35130"             Lat. 37°19'55n
    Water temp. 19°C            Ambient temp. 21°C

Conditions are very similar to a no-slick sample due to
increasing wind.  A very slight dampening effect was
reported.  There was much organic material and little foam.

Sample #18:  May 15, 1971       1245 hrs
    Long. 76°38fl                Lat. 32°21135"
    Water temp. 18°C            Ambient temp. 19°C

Conditions were the same as Sample #17.
                          154

-------
Sample #19:  May 15, 1971       1400 hrs
    Long. 76°44'50n             Lat. 37°29'
    Water tern p. 18°C           Ambient temp.  19°C

Conditions were the same as Samples #17 and  #18.
                           155

-------
                      APPENDIX C

           EFFECTS OF AN ACCIDENTAL SPILL OF
            NO. 6 FUEL OIL ON A SALT MARSH

An accidental release of about 800 barrels of a No. 6
fuel oil prepared by cutting a catalytic cracking residue
with a light distillate fraction occurred on or about 2000
hours on May 5, 1971, at the Yorktown, Virginia, Amoco
Refinery.  The oil issued from a ruptured pipeline directly
into the York River and was carried by wind driven currents
through the Guinea Marsh where considerable amounts of oil
beached.  A portion of the oil passed through channels in
the marsh and moved into Mob jack Bay where it was sampled
on May 6, 1971.  The fraction of the spill volume confined
by the marshes is unknown, but observations from vessels
and a U.S.C.G. C-130 on May 6 led to an estimate of  3/4 of
the total oil remaining in the marshes.  The spill trajec-
tory is shown in Figure 29.  Wind and tidal current
information were available near the spill point, but could
not be used to justify the slick trajectory because the
spill time was uncertain to ±4  hours.

From 1200 hours on May 5 to 1200 hours on May 6 the wind
blew from 230°, averaging 9 knots, using hourly data taken
at the Ft. Eustis, Virginia, weather station.  This wind
direction is in qualitative accord with the observed 50°
slick direction to beaching.  The tidal current influence
on the oil motion may not have been great because, if the
slick moved at 3.7% of the wind speed, almost exactly one
tidal cycle would have been required for it to travel the
4.5 nautical miles to the beaching site.

E.P.A. investigators arrived on site on May 6, and were
transported to Guinea Marsh and assisted in sampling by
project personnel.  An investigation to determine the
effects of a No. 6 fuel oil in the marsh was commissioned
by E.P.A. as a supplement and extension to planned project
activities.  Amoco used a dispersant, Jansolv-60, at the
release point.  The presence, potential toxicity, and
synergistic effects of the dispersant in the marsh were
studied, and a general analysis of dispersant composition
was conducted.

The composition of Jansolv-60 dispersant was not disclosed
upon request to the manufacturer, so analyses were
conducted at this laboratory.  Atmospheric pressure
distillation to 128°C showed these percentages by volume.
                           156

-------
                             ^GUINEA ,/
                               MARSHES
Fig. 29.  Amoco oil spill trajectory
                  157

-------
          Water                 51%
          Light hydrocarbons    33%
          Pine oil.,             trace
          Surfactant            16%

Silica gel and gas chromatography indicated the light
hydrocarbons were saturated aliphatics boiling between
125-235°C at 760 torr which approximates a deodorized
kerosene such as Jet B turbine fuel.

Infrared spectra of the surfactant indicated a predominantly
saturated hydrocarbon structure with carboxyl anion groups.
Aqueous solutions of the surfactant gave positive magnesium
urany! acetate tests for sodium.  The ignition residue of
the surfactant was strongly basic.  Tests for presence of
phosphate groups were negative.  A portion of the
surfactant residue from the distillation was subjected to
Kjeldahl digestion and found to contain at least 3% nitrogen
by weight.  Qualitative tests for the presence of tertiary
amines using the Hinsberg test, and N-bromosuccinimide were
positive.  A titration of 1 ml of surfactant in 20 ml of
water with 0.1 N HC1, using pH electrode H+ detection,
showed the neutralization of the amine, but did not clearly
indicate the proton addition to carboxyl groups at low pH.
Original pH of dispersant and surfactant solution was 8.9
±0.1  depending on dilution.

Tests for ion charge characteristics of the dispersant
(Greenberg, 1962) showed it was cationic or amphoteric.
The presence of amine and carboxyl groups implies that the
surfactant is a long chain amino acid.  A search of
commercial products uncovered Deriphat 160 (General Mills,
Inc.), an amphoteric compound (disodium N-lauryl beta
iminodipropionate) used as a down-hole petroleum surfactant.
It is suspected that this, or a compound of similar structure,
is the principal surfactant in Jansolv-60, but considerable
further purification and analysis would be required to prove
the exact surfactant structure.,
      (' •' • ;    '                              - — '.•","         __
Dispersant composition studies were discontinued, as,no   .
potentially toxic compounds or functional groups were
detected that would bring to question the manufacturer's
toxicity claims.

       '                                  '" '.'"• '\''-
                     OIL SAMPLING

Samples of oil, oiled grass, and oil exposed sediment were
taken for chemical analysis at the locations listed
below.  Location numbers and names refer to the detailed
                           158

-------
map of Guinea Marsh  (Figure  30).

Pate    Time       Location            Sample Description

May 6   1320  37°17l12ll-76°17l30lt           Slick
        1405  37°15I39"-76°20I06"           Slick

May 7   1230  37°15l36"-76°24f4811           Slick
        1240  37°15I36"-76°24t33lt           Slick
        1420            1                   Oil pool

May 9                   1                  Oiled grass
                        1              Intertidal sediment
                        1          Oil slick bordering marsh
                        1          Oil pool at high tide
                                    time

                        2                  Oiled grass
                        3          Snails on oiled grass
                        4          Intertidal  sediment core
                        5          Intertidal  sediment core
                        6          Intertidal  sediment core
                        7          Intertidal  sediment core
                        1          Oiled peat  core
                        1          Intertidal  sediment core

May 18                  1                  Oiled grass
                        1               Oil covered peat
              37°13I17II-76°25I1911   Goodwin  Is.  Oiled grass

June 1                  1             Grass and peat with
                                       entrained oil
                        8             Oiled grass and mud

Samples on June  1 were at the only places where oil was
visible,  so samples  for chemical  analysis were discontinued,
It is certain that oil remains absorbed in the peat and
buried in sediments,  but  problems relating to the sampling
and subsequent analysis of dispersed oil at  uncertain
locations made further study of a very slowly aging
residual  portion of  the oil  impractical.  Beaches were
surveyed  completely  for oil  exposure on May  11,  1971.
Areas with oil coverage are  blackened in Figure 30.  All
samples were frozen  in dry ice immediately upon collection
and returned to  the  laboratory for analysis.
                           159

-------
                                               MOBJACK  BAY
                               \....~, —  .....•"•*• "9 *!"• **\
    YORK  RIVER
                                                                 7—
Fig. 30.  Guinea Marsh oil sampling locations.  Black areas are oil exposed
          beach.

-------
                  BIOLOGICAL  SAMPLING

The sampling location 1 was the most heavily coated  above
the high tide line detected in the  survey.   Aerial
photography from a commercial helicopter  showed  the  most
heavily oiled beaches, but did not  show any  oil  not  seen in
the surface survey.  A program to sample  inter tidal
organisms near location 1 and at two areas of significantly
lower oil exposure was begun  on July 14-16,  1971.  The
biological collection stations are  shown  as  capital  letters
on Figure 30.  Station B is the most heavily exposed to oil,
while stations A and C were free of visible  oil.  At each
station a transect of five equally  spaced substations was
established in the intertidal zone.  Ten  cores were  taken
at each substation with a plexiglas corer (cores 15  cm long
x 8 cm diameter).

The cores were sieved through a 1.0 mm mesh  screen,  and the
material remaining on the screen was preserved in formalin
diluted with seawater for later species identification and
enumeration.  Counts from individual cores were  pooled and
averaged for each substation  for population  statistics.

In December 1971, the same transects were repeated to
determine the possible recovery of  regions apparently
adversely affected by oil.


           COMPOSITION AND AGING OF THE OIL

Distillation and gas chromatographic analysis showed the
oil to be a mixture of catalytic cracked  crude residue and
a light recycled oil.  The residue  consisted primarily of
aromatic compounds boiling above 300°C at 1  atm, and was
quite unlike residual oils from crude  distillation which
show obvious normal hydrocarbon peaks  on  gas chromatograms.
The May 18 sample taken near  Goodwin Island  was  found by
gas chromatography to be a distillation residual of  a
paraffinic crude and not related to the Amoco spill.  No
Amoco oil was detected in marshes on the  south bank  of the
York River.

A fresh sample of oil from the pipeline was  provided
by the Amoco Refinery.  This  oil was vacuum  distilled and
gave the boiling range composition  shown  in  Figure 31.  The
lighter cuts of  this oil were subjected to the F.I.A.
silica gel chromatography to  establish relative  aromatics
and saturates and portions of the aliphatic  and  of the
aromatic efflux were collected for  G.D. and  G.C.-M.S.
analysis.
                           161

-------
Q.
m
   450n
   400-
   350-
   300-
   250-
   200-
   150-
   100-
             10
20
30
40
50
60
70
            PER CENT OF TOTAL  OIL  DISTILLED

     Fig-.  31. Boiling range composition of Amoco spill oil.

                        162

-------
Fraction    b.p.  (°C.  1 atm)   . % Aliphatic    %  Aromatic

Gas Trap                              83              17
   1               209                  8              92
   2            209-221              < 1            > 99
   3            221-238                0             100

A gas chromatogram of  the unaged oil is  shown  in  Figure 32.
It was run on a 1/8" x 6f copper column  packed with 5%
SE-30 on Chromosorb G  AW DMCS  in a Perkin-Elmer 900 Model
Gas Chromatograph programmed from 100 C  to 250 C  at 4 C per
minute.  Operation was dual column compensated with F.I.D.
detection.  The unusual nature of this oil is  indicated by
the lack of easily distinguished normal  hydrocarbon peaks
and the bimodal distribution of peaks.  Silica gel column
chromatography was used to separate  aliphatics from
aromatics  in  the  pentane soluble fraction of the  oil.
Aliphatics and aromatics were  gas chromatographed separately.
It was  found  that components boiling between 216°  C and
317°C were approximately 70% aromatic with very high
aromatic contents (> 90%) up to 280  C.  Aromatics  in this
boiling range correspond to the naphthalene-anthracene
range including long  chain alkyl substituted benzenes and
substituted naphthalenes that  are not easily lost  from the
oil by volatilization  and are  low enough in molecular
weight  to be  reasonably water  soluble.

The lower molecular weight species in the first distillation
cut were identified.   The major components, in descending
order of concentration were meta-xylene, ortho-xylene,
para-xylene,  ethylbenzene, and toluene.  Lesser amounts of
the ethyltoluene  isomers, n-propylbenzene, and cumene
(jL-propylbenzene) were also present.  These analyses were
performed at  NASA Langley Research Center.  Samples were
chromatographed through a 100  ft DPP open tubular capillary
column, then  introduced into a Finnegan  quadrupole mass
spectrometer.

The nature of the Amoco spill  oil was such that little change
of character  could be  achieved, even by  lengthy aging.
Pentane extracts  of each of the samples  listed above were
analyzed by gas chromatography.  By  May  6, the day following
the spill, the sampled oil had lost  nearly all the light
recycle oil,  but  the  catalytic-cracked residue retained its
original composition.  Little  or no  further change could be
noted in samples  taken as late as June 1.  Hence,  the only
observable aging  of this oil spill was the rapid,  immediate
loss of the light recycle oil  by evaporation.
                           163

-------
                                                                             24
Fig. 32.  Gas chromatogram of Amoco Spill Oil.   Integers  are  carbon numbers of
          normal hydrocarbon peaks.

-------
The ^oil had disappeared  from sight by June  1,  1971, by
various processes,  including entrainment  in the  sediments,
absorption into the peaty  root network  of the  Spartina
grass, and attachment  to debris which was subsequently
washed out of the area.  Much of  the residue remained in
the area for the slow  process of  dissolution into the
surrounding water.

The catalytic cracked  residue appears to  consist of various
substituted benzenes and polynuclear aromatic  species such
as naphthalenes, anthracenes, etc.  A program  of research
into  the solution behavior of these aromatic species is
being designed at this laboratory.  However, it  is probable
that  the alkylbenzenes present in this  residue are less
soluble in water, and  less toxic  than polynuclear aromatics
of  similar molecular weight. Boylan and  Tripp (1971) have,
for example, compiled  data indicating that  the naphthalenes
are more toxic than the  alkylbenzenes.

Gas chromatograms of the Goodwin  Island oil sample showed
it was a normal No. 6  fuel oil unrelated  to the  Amoco spill
and of unknown source.  Only a small area (200 ft of beach)
was affected.  This oil  showed a  bluer  and  less  intense
fluorescence of the pentane extract with  350 mm  exitation
than  did the Amoco  oil.  The Amoco oil  has  never, to this
date  of reporting,  lost  its strong yellow-green  fluorescence
due to weathering.  This strong fluorescence is  in keeping
with  the highly aromatic residual character of the oil.

It  should be noted  that  because of the  highly  unusual
nature of the Amoco oil  spill, the results  of  its
contamination of marsh areas will not approximate that of
a "normal" No. 6 fuel  oil.  The Amoco oil,  because of its
catalytic cracking  treatment, does not  contain a large
fraction of normal  paraffins, and is almost entirely
aromatic.
           BIOLOGICAL EFFECTS  ON  GUINEA MARSH

Several workers  (reviewed  by  Cowell,  1971) have  observed
that heavy oil has  little  effect on marine grasses with
rhizome systems.  The predominant Guinea Marsh grass
exposed to oil was  Spartina alterniflora.  In areas with
heavy  oil,  leaf blades were coated and died apparently due
to respiratory failure.  The  rhizomes survived and produced
new young  shoots within  one month.  There were no permanent
kill areas  and the  rhizomes prevented erosion of oil
exposed marsh.  It  is doubtful that marsh grasses would
survive repeated dosing  (Cowell,  1971),  but recovery was
                           165

-------
prompt during the growing season.

Intertidal algae were not investigated, as there were few
attached forms present above the low tide line, probably
due to wave action on the sandy bottom.  There were
extensive Zpstera beds below the low tide line, but these
were not oil coated and always were at least 50 ft from
the oiled shoreline.

Survival of animals exposed to the oil appeared to be the
most sensitive indicators of effects, but highly motile
animals which avoided the oil or fled after being affected
could not be used as indicators.  The intertidal macrofauna
in near surface sediments were exposed to oil and unable to
escape, so were sampled at the lettered locations shown on
Figure 2.  Intertidal sediment cores taken at stations A
and B were extracted with pentane and the extracts gas
chromatographed to prove the presence of oil at station B
and its absence at station A.  This was not done for station
C because no oil beached on that island.

Organism counts at each substation were pooled for each
station and the number of species (S), the number of
individuals (N), and the number of individuals of the ith
species (nj_) were recorded for each station for the July
and December samplings.

The above count information was used to calculate:


Species Richness ~


                         O 001 Q            S
Species Diversity (H1) = *'2    (N log N - E  n* log n...
                           w              i-1         *

                            TT !
Evenness Component (J1) =
                          log/

The Affinity Index was calculated from data at pairs of
stations according to Sanders (1960).   Statistical parameters
for July and December are listed in Table 22.

Population indices for stations A and C are quite similar,
while those for station B show great reductions in number
of organisms, species, and species richness.  The affinity
index for stations A and C is high.  The situation was
nearly the same for July and December collections.  It is
assumed that the Amoco oil caused population reductions and
                          166

-------
Table 22.
Population Statistics of Marsh Intertidal  Infauna
for July (December).
                        Station A    Station B
                                      Station C
N

S

Species
Richness

H'

J'

Affinity
  Index
            482  (536)    88  (107)    611  (2481)

             32  (34)     14  (19)      34  (60)

            5.02  (5.25) 2.90  (3.85)  5.14  (7.55)


            3.02  (3.21) 3.28  (3.21)  3.02  (4.03)

            0.605 0.631 0.861 0.756  0.594 0.684

  to  Sta. A        -      38.7  (27.4)  62.0  (46.8)

  to  Sta. B  38.7  (27.4)      -        48.2  (46.8)

  to  Sta. C  62.0  (46.8) 48.2  (42.1)
 structure alteration of the  intertidal  community at station
 B, but  this  is  difficult to  prove  in  the  absence of a
 detailed  survey at the stations  immediately prior to the
 oil  spill (Foster,  et al., 1971).  Biological collections
 will be repeated in May 1972 to  demonstrate the recovery of
 station B intertidal fauna,  but  recovery  may require
 several years because the oil is still  present in the
 sediments and may exhibit the long term toxic effects of
 soluble aromatic compounds observed by  Blumer, et al. (1970)

 In conclusion,  immediate effects of the oil spill on Guinea
 Marsh were not  great.   Long  term effects  cannot be
 documented as there was no baseline information to describe
 the  previous condition of the marsh.  A before and after
 study of  a previously well sampled salt marsh which is
 exposed to controlled amounts of number 6 oil is being
 designed  at  this laboratory  to prove  long term effects.
 Results here, taken from similar sedimentary environments,
 strongly  indicate a toxic effect of the Amoco oil spill.
 There is  no present means of assessing  the ultimate marsh
 ecosystem damage,  and,  were  this possible, such damage
 could not be estimated in dollars.
                           167

-------
                      REFERENCES

Blumer, M., Souza, G., and Sass, J. (1970).  Hydrocarbon
     pollution of edible shellfish by an oil spill.  Mar.
     Biol., 5.» PP- 195-202.

Cowell, E.B. (1971).  Some effects of oil pollution in
     Milford Haven, U.K.  Joint Conf. on Prevention &
     Control of Oil Spills, Washington, D.C., pp. 429-435,

Foster, M., Neushul, M., andZingmark, R. (1971).  The
     Santa Barbara Oil Spill, Part 2: Initial effects on
     intertidal and kelp bed organisms.  Environmental
     Pollution, 2, pp. 115-134.
                          168

-------
  SELECTED WATER
  RESOURCES ABSTRACTS

  INPUT TRANSACTION FORM
                                   ,1. Report No.
                                                       w
           INVESTIGATION OF SURFACE FILMS - CHESAPEAKE  BAY
             ENTRANCE,
            Maclntyre,  W.  G.; Smith, C. L.; Munday, J.  C.;
Gibson, V. M.;  Lake,  J.  L.; Windsor, J. G.; Dupuy, J.  L;  et.  al.
                                                        5. Report Date

                                                       \6. ' .

                                                       [ S. Performing Organization
                                                          Report No,     • '~
  9.  Organization
"*"         Virginia  Institute of Marine Science
          Gloucester Point, Virginia 23062
                                                       !'"':. PtOJf'il NO

                                                          15080 EJO
                                                                  '13^  Type vf Repot ,. and
        ..JTTrT...r-, „  .—PH ««—»>, «.^«-««». *-...» »~M«—«JL»»C » -~  —  •   *••„..       *    Period Covered
 •^n^"80™^^                          Protectio'n}Agency,  WQO
  15. Supplementary Notes
            U.S.  Environmental  Protection Agency report
             number EPA 670/2-73-099

  16. Abstract

         Experimental  point source oil  releases have been conducted  in the  Chesapeake
    Bay mouth  area.  Predictions of  oil  slick  motion were tested, and slicks were sampled
    and analyzed to measure their aging rates  over periods up to 32  hours.  Remote sen-
    sing techniques were used to detect and  measure the spreading rate of oil.   Some
    laboratory oil film aging experiments  were done to further document and elucidate
    aging processes.  Results indicate  a reasonable motion prediction, an explanation
    of the non-biological initial aging of oil  films, and a fair corroboratiori  of a
    theoretical  oil spreading model.

         Indigenous surface films in  the study area were analyzed for lipid and chlori-
    nated hydrocarbon content.  Hydrocarbons were 300-500 microgram  per liter and fatty
    acids and  esters 700-7800 microgram per  liter in surface film samples.  Chlorinated
    hydrocarbons were generally less  than  100  parts per trillion in  surface films, in
    contrast to some earlier high concentrations found in Biscayne Bay.  Surface film
    analysis limitations imposed by  sampling methods are discussed.  Plankton in slick,
    non-slick, and subsurface water were counted.   Populations were  higher  in surface
    than subsurface water, and higher in non-slick than in slicked surface  water.
  i?a. Descriptors  *oil  Spills, *Estuarine Environment,  *Chesapeake Bay, Oil Pollution,
   Estuaries,  Currents, Sampling, Chemical Analysis,  Chromatography, Chlorinated
   Hydrocarbons,  Pesticides, Liquids
  i7b. identifiers   *Surface films, *0i 1 slicks, Remote  sensing,  Hydrocarbon analysis,
     oil aging, fatty acids
  17c: COWRR Field & Group   05A
  18.  A vailability
                  -Security Ctes$; s
                  fRsptnti) ••'- '

                  Se&triiy Ot,$$. ' «
                                                       Send To:
                                                       WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                                                       U.S. DEPARTMENT OF THE INTERIOR
                                                       WASHINGTON. D. C. 2O24O
Dr. Win. G. Maclntvre
                                      I  institution Virginia  Institute og Marine Science
 U.S. GOVERNMENT PRINTING OFFICE: 1974-544-317:311

-------