DOC
EPA
United States
Department of
Commerce
National Oceanic and Atin-
Administration
Seattle WA 98115
United States
Environmental Protection
Agency
Uftio: of Environmental
Engineering and Technology
Washington DC 20460
EPA 600/7 80 140
July 1980
Research and Development
Recovery of
Strait of Juan de Fuca
Intertidal Habitat
Following Experimental
Contamination with Oil
Interagency
Energy/Environment
R&D Program
Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
RECOVERY OF STRAIT OF JUAN DE FUCA INTERTIDAL HABITAT
FOLLOWING EXPERIMENTAL CONTAMINATION WITH OIL
Second Annual Report
Fall 1979 - Winter 1980
J.R. Vanderhorst, J.W. Blaylock, P. Wilkinson,
M. Wilkinson, and G. Fellingham
Battelle, Pacific Northwest Laboratories
Marine Research Laboratory
Sequim, Washington 98382
Prepared for the MESA (Marine Ecosystems Analysis) Puget
Sound Project, Seattle, Washington, in partial fulfillment of
EPA Interagency Agreement No. D6-E693-EN
Program Element No. EHE625-A
This study was conducted as part of the
Federal Interagency Energy/Environment
Research and Development Program
Prepared for
OFFICE OF ENVIRONMENTAL ENGINEERING AND TECHNOLOGY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
March 1980
-------
Completion Report Submitted to
PUGET SOUND ENERGY-RELATED RESEARCH PROJECT
MARINE ECOSYSTEMS ANALYSIS PROGRAM
ENVIRONMENTAL RESEARCH LABORATORIES
by
BatteHe, Pacific Northwest Laboratories
Marine Research Laboratory
Sequim, Washington 98382
This work is the result of research sponsored by the Environmental
Protection Agency and administered by the Environmental Research Laboratories
of the National Oceanic and Atmospheric Administration.
The Environmental Research Laboratories do not approve, recommend,
or endorse any proprietary product or proprietary material mentioned in
this publication. No reference shall be made to the Environmental Research
Laboratories or to this publication furnished by the Environmental Research
Laboratories in any advertising or sales promotion which would indicate
or imply that the Environmental Research Laboratories approve, recommend,
or endorse any proprietary product or proprietary material mentioned
herein, or which has as its purpose an intent to cause directly or indirectly
the advertised product to be used or purchased because of this Environmental
Research Laboratories publication.
-------
FOREWORD
An anticipated increase in oil tanker traffic and proposals for
construction of subsea pipelines in the Strait of Juan de Fuca and northern
Puget Sound regions of Washington State are foreseen as part of the
national energy development plans. These activities increase the opportunity
for spillage of crude oil into the marine ecosystems of the region. The
U.S. Environmental Protection Agency has supported studies dealing with
biological characterizations, physical oceanography, trajectory modeling,
pollutant monitoring, and fate and effects of oil in the region. These
studies are being administered by NOAA's Marine Ecosystem Analysis
(MESA) Puget Sound Project Office. The research reported here deals
with recovery of intertidal and shallow subtidal communities in experi-
mental habitats contaminated with Prudhoe Bay crude oil. The studies
make comparisons in rate of recovery by communities in experimental
coarse and fine sand habitat, and hard substrate habitat. They examine
the role of vertical distribution of habitat in the tidal zone, site,
type of substrate, season, and duration for recovery in field experiments.
m
-------
ABSTRACT
This is a second year interim report on the effects of experimental
oiling with Prudhoe Bay crude oil on recovery of intertidal infauna and epi-
fauna of the Strait of Juan de Fuca, Washington State. It describes completed
studies of the recovery of infauna as recovery rate relates to the experimental
oiling, the site of study, tidal height, season of study, and duration of
recovery. The report also describes the methods and initial results of studies
of the effects from experimental oiling on epifauna colonization of hard
substrates.
Full recovery is defined within the experimental framework for infauna as
that composition and density of species which had colonized trays of untreated
coarse substrate within the 15-month study period. The relevance of this
definition is supported by presentation of data on composition and density of
infauna at adjacent baseline stations as measured by other investigators. In
terms of species composition, nearly full recovery of oiled substrates occurred
in 15 months. For individual species densities, as well as overall abundance,
however, oiled substrates had recovered only about one-half in 15 months.
Total hydrocarbons in treated substrates were reduced from initial concentration
by 85 and 97% for fine and coarse sediments, respectively, in 15 months.
Based on rate of loss between 3 and 15 months, it is speculated that total
hydrocarbons would have reached background levels in 18.5 months. Analyzed
saturate compounds appeared to be lost from treated sediments at a rate similar
to total oil. Analyzed aromatic compounds exhibited a much more rapid reduction
in concentration than did saturate compounds or total oil.
As analyzed experimental variables, the site of study, tidal height, and
sediment type, produced significant effects on the density if primary biological
species. Overall, there were much higher densities at two feet below Mean
Lower Low Water (MLLW) than at MLLW. Overall abundance appeared about equal
between sediment types. Although not analyzed statistically, there appeared
to be an order of magnitude higher density in the summer-fall experimental
period than in the spring-summer experimental period.
The most severe effects from oiling on infauna density, as an expression
of recovery, were seen for detritivorous and herbivorous species. The species
for which significant effects on recovery were demonstrated were among those
identified as having major trophic importance for a variety of bottom feeding
fishes by other Strait of Juan de Fuca investigators.
The experimental oil treatment, while perhaps a "worst" case in the sense
that the oil was mixed in sediment, was well within the concentration measured
in sediments following some actual oil spills.
iv
-------
TABLE OF CONTENTS
Foreword •••
Abstract '•'.'•'•'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. }l
1. Introduction . . 1
2. Conclusions 3
3. Recommendations . . . 6
4. Materials and Methods .......... 8
Infaunal Studies . . 9
Sediment Extraction and Chemistry '. '. ' 16
Epifaunal Recovery Studies ...... 16
Chemical Characteristics of Bricks '.'.'.'.'.'. 17
5. Results and Discussion '.'.'.'.'.' 20
Infauna Recovery '.'.'.'.'. 20
A Perspective '.'.'.'.'' 20
Recovery of Primary Species '.'.'.'.'.'.'.''' 28
Experiment I. Site, Sediment, Oil Effects, 3-month
Recovery-- '. 31
Experiment II. Tide Level, Oil Effects', 3-month
Recovery— 34
Experiment III. Sediment, Oil Effects, 15-month
Recovery— 37
Substrate Recovery Indicators for the Strait of Juan
de Fuca 37
Recovery of Other Community Members '.'.'. 40
Trophic Mode of Primary and Selected Other Species '.'.'.'. '. 51
Relevance of Findings to Upper Trophic Levels 54
Severity of Treatment in Infauna Experiment '. 56
Recovery on Hard Substrates 63
Species Composition and Density in Monthly Experiment .... 65
Severity of Treatment on Hard Substrates 69
References 72
-------
FIGURES
Number Page
1 Study sites , . 10
2 Composite food web characteristic of protected sand/
eel grass, shallow sublittoral habitats in northern
Puget Sound and the Strait of Juan de Fuca 55
3 Concentrations of total oil IR (CCH extractable
organics in oil treated substrates minus CCU
extractable organics in control substrates) 59
4 Concentrations of total oil IR (CCH extractable
organics in oil treated substrates minus CCU
extractable organics in control substrates),
Experiment II ,..,.....,...., 61
vi
-------
TABLES
Number page
1 Schedule of sampling for oil recovery experiments 11
2 Preparation of units and sampling schedule for "on-going"
experiments on effects of oil on recovery by commercial
clams and epifauna of rocky intertidal ........... 18
3 Summary of number of species in principal groups of
animals from baseline studies and 15-month controls
in recovery experiment 21
4 Summary of number of individuals in principal groups of
animals from baseline studies and 15-month controls in
recovery experiment 23
5 Group density/m2 in oil recovery experiments ,,....,.. 24
6 Summary of numbers of species in oil recovery experiments . . 25
7 Numbers of individuals for "primary" species as a
percentage of all individuals for Beckett Point and
Jamestown and recovery study situations 29
8 Density (No./m2), recovery (% final control), and
representation (% respective total), for primary
species in recovery experiments at Sequim Bay/MLLW/
coarse sediment 30
9 Mean density of primary species (No./tray) in
Experiment I 32
10 Hypothesis tests for density of primary species in
oil recovery Experiment I 33
11 The mean density of primary species in Experiment II
(tide level, oil, 3-month recovery, summary) 35
12 Hypothesis tests for density of primary species in
Experiment II (spring-summer, tide, oil) 36
13 The mean density of primary species in Experiment III
(Sequim Bay, 15-month recovery, sediment type, oil) .... 38
vn
-------
Number Page
14 Hypothesis tests of density of primary species in
Experiment III (15-month recovery, Sequim Bay,
sediment, oil) 39
15 Mean density for recovery indicators in Experiment I ...... 41
16 Estimated Density for recovery indicator species
at adjacent baseline sites ,.,,,,,.,... 42
17 Mean density of nonprimary species CNo./m2) of
polychaetes in Experiment I Sequim Bay example 44
18 Mean density of nonprimary species (No./m2) of
crustaceans in Experiment I Sequim Bay example 45
19 Mean density of nonprimary species (No./m2) of
"others" and mollusks in Experiment I Sequim Bay
example 46
20 Mean density of nonprimary species (No./m2) of
polychaetes in Experiment III 47
21 Mean density of nonprimary species (No./m2) of
crustaceans in Experiment III 49
22 Mean density of nonprimary species (No./m2) of
"others" and moll us ks in Experiment III 50
23 Trophic levels of primary species in oil recovery
experiments 52
24 Comparisons of total oil (IR) in Experiments I
and III, at Sequim Bay 57
25 Comparisons of total oil (IR) in Experiment II
at Sequim Bay 60
26 Means of summed analyzed saturate and aromatic
compounds (capillary GC) in Experiments I and III
at Sequim Bay 62
27 Means of summed analyzed saturate and aromatic
compounds (capillary GC) in Experiment II
at Sequim Bay 64
28 Mean numbers of individual polychaetes in hard
substrate oil recovery experiments 66
viii
-------
Number Page
29 Mean numbers of individual crustaceans in hard
substrate oil recovery experiments 67
30 Mean numbers of individual molluscs in hard substrate
oil recovery experiments , 68
31 Mean total CCU-extractable organics for whole brick
extractions in recovery , 70
32 Total hydrocarbons extracted from top surface of
hard substrates 71
-------
SECTION 1
INTRODUCTION
This is a second-year interim report on experimental studies of the
effects of crude oil on recovery of intertidal communities of the Strait
of Juan de Fuca. High interest in effects from oil and recovery of
communities following oiling, stems from anticipated increases in tanker
traffic, and superport and pipeline construction in the marine systems of
the region. The interest is national in scope because the region is
targeted as a through-point for crude oil shipment to midwestern markets.
Intertidal and shallow subtidal benthic communities are especially
vulnerable to spilled oil because it is in their habitat that the surface
of the water, on which oil floats, comes into intimate contact with the
ground material, or substrate. Oil dissolved in water masses may have
short-term toxic effects on marine organisms, but the substrates have
been most clearly identified as sinks for spilled oil over longer periods
of time (e.g., Michael et al., 1975; Straughan, 1978; Clark et al.,
1973). There is variety in the substrate composition of the intertidal
zone of the Strait of Juan de Fuca, but rock and mixed substrate con-
sisting of rock and a matrix of mud, sand, or gravel predominate. Primarily
because of longevity of predominate organisms, Nyblade (1979) has indicated
that rock substrates will require the longest periods to recover (decades).
However, the retention time for the oil itself is much longer in fine-
grained substrate. In certain conditions of spillage, this retention of
oil may have an overwhelming influence on rate of recovery (Michael et
al., 1975).
In particular cases, recovery of communities depends not only on
organism longevity and substrate suitability but on a host of other
factors including specific site, position with respect to tidal influence,
season, and prevailing exposure. Additionally, recovery will depend on
amounts of oil spilled, the types of cleanup measures used, and the
degree of protection afforded critical populations, if these exist.
The overall objective in the present studies is to examine the
influence of Prudhoe Bay crude oil on rate of recovery by intertidal and
shallow subtidal communities in terms of substrate impairment by the oil
as it relates to specific controllable variables (substrate type, site,
tide level, season, and duration of recovery period). Information on the
relative influence of these factors is pertinent to choices which may
arise concerning physical protection of specific sites during a spill,
application of remedial measures (chemical dispersion and cleanup) or
siting of shore based facilities related to the through-point function.
To meet the overall objective, the studies have been divided into a
number of tasks. The tasks associated with infaunal recovery are now
complete and are as follows:
-------
A. Provide a time-series survey of species composition
and total hydrocarbon concentration in experimentally
prepared substrates.
B. Measure the effect from Prudhoe Bay crude oil on
reseeding by early colonizers as related to site
and substrate grain size during a late summer -
early fall recruitment period.
C. Measure the effect from Prudhoe Bay crude oil on
reseeding by later colonizers one year after initial
colonization during the late summer - early fall
recruitment period.
D. Measure at one experimental site the effect from
Prudhoe Bay crude oil on reseeding by early colo-
nizers during the late spring - early summer season.
E. Measure at one experimental site the effect from
Prudhoe Bay crude oil on reseeding by early colonizers
as related to tide height.
In addition, tasks have been initiated and some individual experiments
have been completed for attached epifaunal studies on solid substrates as
follows:
F. Investigate the suitability of some artificial hard
substrates for attachment in the exposed rocky
intertidal of the Strait of Juan de Fuca for future
experimental recovery studies.
G. Investigate recovery of a rocky intertidal community
in terms of larval reseeding rate for key species.
The following two tasks are scheduled for initiation during April 1980 and
completion during August 1980:
H. Investigate recovery of a commercial clam bed in
terms of larval reseeding rate.
I. Investigate recovery of a rocky intertidal community
in terms of mortality and removal of key species.
-------
SECTION 2
CONCLUSIONS
Experimental studies of the effects of Prudhoe Bay crude oil on the
recovery of infauna were conducted over a 15-month period. The studies
involved field placement of oil-treated and untreated sediment trays at
two sites, two tide levels, in two seasons and using two types of native
sediment. At the conclusion of 15 months of field colonization, the
numbers and kinds of animals in control trays closely resembled numbers
and kinds of animals reported in similar habitat at two adjacent baseline
stations. The similarity extended beyond overall numbers and kinds of
animals since the relative distribution of numbers and kinds within the
major taxa also closely paralleled that reported for baseline stations.
From these data it is concluded that the 15-month control sediments were
fully recovered and they reasonably represent what one would find by
sampling uncontaminated sites on the Strait of Juan de Fuca with similar
habitat.
Using the 15-month recovery of untreated sediments as a definition
of full recovery, 3-month recoveries in similar sediment type, site and
tide conditions were 69% for summer and 82% for fall in terms of numbers
of species, and only 11% for summer and 18% for fall in terms of numbers
of individuals. In 15 months, oil-treated sediments had recovered more
than 90% in terms of numbers of species but had recovered only 48% in
terms of numbers of individuals.
To protect the validity of statistical procedures, 13 species were
selected as "primary" to evaluate effects on recovery in terms of individual
species density. The numbers of individuals of the primary species
comprised a very substantial proportion of all individuals in this study
(78%) as well as at the Beckett Point baseline station (73%). They
represented 33% of all individuals at the Jamestown baseline station for
comparable conditions. Three-month control recovery for these primary
species in terms of numbers of individuals closely paralleled that seen
for all individuals (19% fall; 8% summer). The primary species represent
the three major taxonomic categories quite well (polychaetes, crustaceans,
mollusks).
Statistically significant effects on density (in this framework,
recovery) were seen for individual primary species in each of the seasonal
3-month experiments and in the 15-month recovery period. The species
affected were principally within the polychaete and crustacean groups.
The mean differences were greater than those indicated for all species
and indicate reductions in oil-treated sediments with one exception. The
exception was a high density mollusk after 15-month recovery. Because of
the differential response, here indicated by a statistically significant
finding and elsewhere in the study indicated by mean differences, we
-------
conclude that using total numbers of individuals to measure recovery from
oil contamination will lead to highly conservative estimates of actual
effects on recovery.
Effects from the oiling on recovery is strongly related to feeding
type. Detritivorous and herbivorous species were almost universally
influenced by the oiling. Carnivorous species were about evenly divided
in their response to the oiling and, with one exception, no significant
effect was seen on the recovery of a suspension feeder.
Although effects from oiling on recovery were found at each of the
tide levels (MLLW and -21- (.61m)), at each of the sites (Protection
Island and Sequim Bay), in each of the sediment types (Sequim Bay native
and Protection Island native), and in each of the seasons (spring-summer
and summer-fall), these physical factors, nevertheless, influenced gross
density in both treated and control sediments. Thus, much higher densities
were found in the summer-fall season than in the spring-summer season;
much higher densities were found at Sequim Bay than Protection Island;
and much higher densities were found at -2' below MLLW as compared to
MLLW tide level. Density related to sediment type tended to be species
specific and was about equal overall. Studies designed to elicit effects
on recovery in an actual oil spill event will, thus, need some form of
experimental control for these variables.
Two species, because of their nearly ubiquitous occurrence within
the recovery experiment and the north Puget Sound region, generally have
been identified as good recovery indicators when used in an appropriate
experimental framework. These species are the crustacean, Leptochelia
dubia, and the polychaete, Exogone lourei. A taxonomic problem with the
former and an important oil-resistant congener of the latter were identi-
fied.
Based on adjunct MESA studies of trophic relationships, it appears
that the severity of influence on recovery of species in this study could
be expected to have a deleterious effect on important fish populations,
and that this effect would extend somewhat beyond the 15-month period
studied here.
Because oil was mixed into sediment, the present case may be considered
a "worst" case situation in terms of treatment severity. However, the
sediment-borne concentrations of both total oil and analyzed aromatic and
saturate compounds were well within the range of concentration reported
for sediments exposed to actual spillages elsewhere. Initial target
concentrations of 2000 ppm for summer-fall, 3-month recovery and 15-month
recovery and 1000 ppm for spring-summer recovery were obtained. Reductions
in total hydrocarbons were about 35% in three months for summer-fall
regardless of sediment type. The comparable amount for MLLW in the
spring-summer period was 43%. There was a slightly more rapid loss at
the lower tide level (53%). At 15 months, total hydrocarbons were reduced
by 85 and 97% for coarse and fine sediments, respectively. Based on rate
of loss data between 3 months and 15 months, it is speculated that back-
ground levels would be reached in a total of 18.5 months. An important
contribution to sediment infrared spectra (most likely due to biogenic
materials and unrelated to oiling) was identified in analyses of control
sediments after 15 months.
-------
Capillary gas chromatography revealed a much more rapid loss of
analyzed aromatic compounds than saturate compounds. There were seasonal
differences in aromatic content of sediments. Aromatics were more than
80% reduced in the summer-fall, 3-month period and at background levels
in the spring-summer, 3-month period. As a percentage of preliminary
concentrations, analyzed saturate compounds closely paralleled concen-
trations of total oil as measured by infrared spectrophotometry.
-------
SECTION 3
RECOMMENDATIONS
The data in this study verify the utility of an experimental approach
to measuring recovery of infauna. There are some specific studies using
the approach which have high priority and are recommended based on the
present findings.
In the realm of oil pollution research, the following should be
investigated:
1. A less severe treatment with Prudhoe Bay crude oil should be
used to bracket potential effects on recovery. The reduction in severity
should relate to the method of applying oil and not necessarily the total
amount used which, in the present case, was slight. Thus, an approach
where oil is layered onto the surface of sediments, either from a seawater
surface slick or direct surface application would be appropriate.
2. Comparative studies of the effects on recovery from processed
petroleum products, i.e., light fuel oils and residual fuels should be
undertaken especially in areas of the north Puget Sound region especially
vulnerable to such spillage.
3. The relative severity on recovery effects following oil and oil
dispersant combinations should be investigated using the present approach
to assist decision making regarding the application of dispersants should
spillage occur in our region.
The methods used in this study appear particularly suitable for
investigation of other sediment-related pollutant.problems in the Puget
Sound region. Specifically, we recommend studies of the effects on
recovery from:
1. Heavy metal contamination.
2. Synthetic organic contamination.
3. Dredge spoil contamination.
4. Wood fiber and by-product contamination.
5. Combinations of the above.
Since the methodological sensitivities are a function of the prevailing
seed populations and the types of physical factors identified in this
study, there is good reason to believe that the approach used here would
be appropriate for these studies.
The present studies have clearly identified an urgent need for two
further types of investigation to assist in interpretation of the effects
demonstrated:
-------
1. Basic life history studies of especially oil-sensitive primary
species to include studies of behavior in response to pollutant contami-
nation. Two questions need to be addressed: (1) what is the zonal
distribution of the seed source? and, (2) where do organisms go that are
absent from oil treated sediments?
2. Experimental studies of feeding relationships, particularly of
bottom-feeding flatfishes. The experiments should have a field orientation.
-------
SECTION 4
MATERIALS AND METHODS
The approach in these studies used substrate units (trays of sediment
or concrete bricks) which were: (1) initially free of organisms; (2)
treated with oil (treated) or not (controls); and (3) allowed to colonize
in a selected array of intertidal habitat conditions. The power of this
approach lies in: (1) an equal starting point (organism free) for treated
and untreated substrates; (2) the use of strict random procedures (as
opposed to haphazard) for allocation of substrate units within and between
treatments; (3) the inclusion of important environmental variables (site,
tide level, substrate type, season) as treatment categories (receiving
both oil-treated and untreated units); (4) the use of a replication
scheme for which methodological sensitivity: (a) had been pre-evaluated
(Vanderhorst et a!., 1978); and (b) can be re-evaluated.
>
Balanced designs were used in the experiments reported here, and
independent controls were utilized in each phase of each experiment;
thus, a correct use of analysis of variance is indicated. This is in
marked contrast to use of analysis of variance in field surveys for which
time series data create dependencies between treatment categories and
alter error probabilities in an undefinable manner. In contrast to field
survey approaches, the present approach has a much smaller sampling
requirement for two important reasons.
The reasons for a smaller sampling requirement in the present approach
relate to: (1) habitat description; and (2) population characteristics.
In a survey context, it is necessary to relate the contribution of each
of these features to an index in which there is inherent interest (e.g.,
population density, community diversity, stock value). With respect to
habitat description, an adequate survey with the aim of defining effect
due to oiling must deal with the variance associated with mean index
magnitude contributed by all or at least "representative" habitat features.
In the case of population characteristics, the variance associated with
mean index magnitude also includes elements relating to seasonal spawning
cycles, competitive and predator-prey relationships in the water mass,
and both passive and active migrations of larval forms. Crude, but
conservative, estimates of contribution to variance of mean density by
the latter features range (expressed as coefficient of variation) from
150% to several hundred percent of the mean in rock habitat of the Strait
of Juan de Fuca, and 30% to more than 100% for infauna (based on data
from Nyblade, 1979). Contribution to the habitat mean density is even
greater for many species. Against this background, even rather substantial
(and perhaps detectable by visual inspection) effects from oiling cannot
be statistically validated with a practical number of samples. The
present approach avoids this problem by providing controls within each
treatment category of interest.
8
-------
INFAUNAL STUDIES
Details of the experimental methods including criteria for site
selection, preparation and placement of substrates, chemical and bio-
logical characterization, and sampling rationale and procedures have been
previously reported for the infaunal studies (Vanderhorst et al., 1979).
In summary, native substrate was collected from two sites (Figure 1),
brought to the laboratory, and given a repeated freezing-thawing treatment
to kill macrofauna. Both native sediments were principally sand. At the
time of collection, sediments dug from each of the experimental sites
were thrown through a 13 mm (1/2-in) mesh hardward cloth screen to remove
cobble and larger pebbles. Based on visual inspection, sediments from
Protection Island were classified as "fine" and those from Sequim Bay
were classified as "coarse," relative to each other. The evaluation of
effect of sediment type in this report is, in fact, an evaluation of
effect from native Protection Island sediment compared to native Sequim
Bay sediment as represented by the respective initial collections.
Samples for particle size analysis, which will more clearly place these
relative sediment textures in perspective to the region as a whole, were
taken and preserved at the time of initial sediment collection. Results
of these analyses will be presented in the final report on this project.
Half of the substrate from each site was treated with Prudhoe Bay
crude oil by mixing in a commercial cement mixing truck for Tasks A, B,
and C, and a motor-driven portable cement mixer for Tasks D and E.
Because of the difference in mixing method and availability of results
from Task B, a target concentration of 2,000 ppm total oil was sought in
the former case, and a target concentration of 1,000 ppm total oil was
sought in the second. Total amounts of oil and concentrations of selected
petroleum compounds were measured in treated and untreated sediments
prior to field installation, at intervals between installation and com-
pletion, and upon completion of a given experiment. The other half of
the substrate from each site served as control. It received the mixing
just as the oiled substrate but did not receive an application of oil.
The prepared substrates were placed in PVC trays (30 x 15 x 15 cm). The
bottom of the trays were provided with eight 2.5 cm diameter holes for
drainage. Experimental substrates were retained in trays by placing a
fiberglass screen over these holes. For Tasks A, B, C, and E, trays were
buried with top surface flush with the ground surface at Mean Lower Low
Water at each of the sites. For Task D, trays were buried in a similar
fashion at -2 feet below Mean Lower Low Water at the Sequim Bay site.
Field installations for Tasks A, B, and C were during August 1978. Field
installations for Tasks D and E were during April 1979. Task B terminated
in November 1978. Tasks D and E terminated in August 1979. Tasks A and
C terminated in November 1979.
A relatively high amount of replication was used in both the place-
ment and sampling of substrate units. This was based on a predesign
study using similar units (Vanderhorst et. al., 1978). See Table 1 for
list of sampling. This replication was done to permit evaluation of
methodological sensitivity and to attain a quantitative measure of the
density of individual species colonizing the trays. Because of interest
in a large number of species and a limited number of independent units in
even this rather large design, it was necessary to a priori select species
of special interest for hypothesis testing with valid probability statements
-------
STRAIT OF JUAN DE FUCA
SAMPLING
SITES Sequim
Bay
FlGURE 1. STUDY SITES
PROTECTION
•••; ISLAND""
ITE
STRAIT OF JUAN DE FUCA
PENINSULA''.
Ma
Lab
0 Nautical Miles 1
SITE
TT 1SEQUIM
BAY
0 Yards
2000
I
4000
-------
Table 1. Schedule of Sampling for Oil Recovery Experiments.
MONTH/YEAR
August/1978
September/1978
October/1978
November/1978
SITE/STATUS
TIDE LEVEL ( ) TASK
Preliminary ABC
Protection Is. A
(O1)
Sequim Bay A
(O1)
Protection Is. A
(O1)
Sequim Bay A
(O1)
Protection Is. B
(O1)
TREATMENT
STATUS
Oiled
Unoiled
Oiled
Unoiled
Oiled
Unoiled
Oiled
Unoiled
Oiled
Unoiled
Oiled
SUBSTRATE
TYPE
Coarse
Fine
Coarse
Fine
Fine
Fine
Coarse
Coarse
Fine
Fine
Coarse
Coarse
Coarse
Fine
SAMPLE
TYPE
Infrared
Gas chromat.
Infrared
Gas chromat.
Infrared
Gas chromat.
Infrared
Gas chromat.
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
NUMBER
TRAYS
3
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
3
3
5
3
3
5
NUMBER
CORES
9
3
9
3
9
3
9
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
9
3
35
9
3
35
-------
Table 1. (Continued)
MONTH/YEAR
November/1978
November/1978
December/1978
January/1979
SITE/STATUS
TIDE LEVEL ( )
Protection Is.
(O1)
Sequim Bay
(O1)
Protection Is.
(O1)
Sequim Bay
(0')
Protection Is.
(O1)
Sequim Bay
(O1)
TREATMENT
TASK STATUS
B Unoiled
B Oiled
Unoiled
A Oiled
Unoiled
A Oiled
Unoiled
A Oiled
Unoiled
A Oiled
Unoi led
SUBSTRATE
TYPE
Coarse
Fine
Coarse
Fine
Coarse .
Fine
Fine
Fine
Coarse
Coarse
Fine
Fine
Coarse
Coarse
SAMPLE
TYPE
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
NUMBER
TRAYS
3
3
5
3
3
5
3
3
5
3
3
5
3
3
5
3
3
5
1
1
I
I
1
1
1
1
I
1
1
1
1
1
1
1
NUMBER
CORES
9
3
35
9
3
35
9
3
35
9
3
35
9
3
35
9
3
35
I
I
I
1
1
1
1
1
1
1
1
1
1
1
1
1
-------
Table 1. (Continued)
MONTH/YEAR
April/1979
Apr: 1/1979
June/1979
July/1979
SITE/STATUS
TIDE LEVEL ( )
Sequim Bay
(O1)
Preliminary
Sequim Bay
(O1)
Sequim Bay
(O1)
Sequim Bay
(O1)
Sequim Bay
(O1)
Sequim Bay
(-21)
Sequim Bay
(-21)
Sequim Bay
(O1)
Sequim Bay
(O1)
Sequim Bay
(O1)
Sequim Bay
(O1)
Sequim Bay
(-21)
Sequim Bay
(-2)
TASK
A
D,E
A
D(A)
A
D(A)
E(A)
E(A)
A
D(A)
A
0(A)
OCA)
ECA)
TREATMENT
STATUS
Oiled
Oiled
Oiled
Oiled
Unoiled
Unoiled
Oiled
Unoiled
Oiled
Oiled
Unoiled
Unoiled
Oiled
Unoiled
SUBSTRATE
TYPE
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
SAMPLE
TYPE
Infrared
Biological
Infrared
Gas chromat.
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
NUMBER
TRAYS
1
1
3
3
I
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
NUMBER
CORES
1
1
9
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
-------
Table 1. (Continued)
MONTH/YEAR
August/1979
August/1979
August/1979
September/1979
October/1979 •
November/1979
SITE/STATUS
TIDE LEVEL ( )
Sequim Bay
(O1)
Sequim Bay
(0)
Sequim Bay
(-2')
Sequim Bay
(0)
Sequim Bay
(0)
Sequim Bay
(0)
TREATMENT
TASK STATUS
A Oiled
Unoiled
D Oiled
Unoiled
E Oiled
Unoiled
A Oiled
Unoiled
A Oiled
Unoiled
C Oiled
Unoiled
SUBSTRATE
TYPE
Coarse
Coarse
Coarse
Coarse
Coarse
V
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
Coarse
SAMPLE
TYPE
Infrared
Biological
Infrared
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Biological
Infrared
Gas chromat.
Biological
Infrared
Gas chromat.
Biological
NUMBER
TRAYS
1
1
1
1
3
3
5
3
3
5
3
3
5
3
3
5
1
1
1
1
1
1
1
1
3
3
5
3
3
5
254
NUMBER
CORES
1
1
1
1
9
3
35
9
3
35
9
3
35
9
3
35
1
1
I
I
1
1
I
1
9
3
35
9
3
35
788
-------
concerning statistical error. For these species we use a criterion of
a ~ 0.01 to deem "significant" effect on density. The maximum real
probability for Type I error for any one of the seven species (Task B)
was 7%, and for any one of the ten species (C, D, E) it was 10%. Task A
data were outside the experimental framework. To meet the objectives of
Tasks B, C, D, and E, three independent experiments were conducted and a
priori selection of target species was made. These species were designated
"Primary" species and consist of the following:
For Experiment I
(Task B): Mollusks
Mysella tumida
Transennella tantilla
Lacuna sp.
Polychaetes
Platynereis bicanaliculata
Armandia brevis
Ophiodromus pugettensis
Capitella capitata
For Experiment II
(Tasks D and E) and
For Experiment III
(Task C): Mollusks
Mysella tumida
Protothaca stami nea
Lacuna variegata
Polychaetes
Platynereis bicanaliculata
Armandia brevis
Polydora social is
Exogone lourei
Crustaceans
Leptochelia dubia
Corophium ascherusicum
Photi s brevipes
The basis for a priori selection for Task B rested on: (1) results
from the predesign study (Vanderhorst et al., 1978); and (2) for two
species, Lacuna sp., and Capitella capitata, reported perturbations
following oil spills elsewhere. The basis for a priori selection of
species in Tasks C, D, and E were: (1) results of Task B studies; and (2)
examination of baseline data for nearby stations (Nyblade, 1979). The
danger in a priori selection in any experiment involving field colonization
is that selected species may not occur in the future experiment, or may
be relatively unimportant constitutents. In general, this was not the
15
-------
case for the species selected here. However, for one species of consider-
able commercial importance, Protothaca staminea, a low frequency of
occurrence was a problem. That species is to be specifically addressed
in a yet to be completed task (H).
In addition to the 13 species which were a priori selected to protect
the validity of error probability statements, quantitative data were also
collected on the density of nearly 200 other species which colonized
Experiments I, II, and III. Analyses of variance were computed for these
data for descriptive purposes. Since the densities of these many species
cannot realistically be assumed to be independent from one another, we do
not attribute statistical significance to the results obtained from these
analyses. However, the mean densities computed for these species in the
several treatment categories in each experiment do serve to better describe
the communities involved and are included in the results on recovery of
other community members.
Sediment Extraction and Chemistry
Sediment cores collected for chemical analysis were frozen immediately
after collection. The frozen samples were thawed at room temperature and
thoroughly mixed for subsampling. Twenty grams (wet) of sediment were
placed in a 250 ml teflon-capped bottle with 20 g of anhydrous sodium
sulfate and thoroughly mixed to hydrate the water from the sediment. To
the samples for I.R. analysis, 50 ml carbon tetrachloride was added, and
for capillary G.C. analysis 25 ml hexane were added. The bottles were
shaken overnight on a reciprocal shaker. The solvents were decanted from
the sediment. Carbon tetrachloride was extracted into a 25 ml scintil-
lation vial, and hexane was extracted into a 50 ml graduated cylinder.
The CC14 subsample was analyzed by infrared spectroscopy (Simard et al.,
1952). The samples for G.C. analysis were extracted for an additional 2
hours with 25 ml hexane and decanted into the graduated cylinder with the
first extraction. The sediment sample was then extracted with 5 ml
volumes of hexane and decanted until the total extract volume was 50 ml.
Twenty-five ml of the hexane extract was concentrated under nitrogen to 3
ml, and separated into aliphatic and aromatic fractions by silica gel
chromatography (Warner, 1976). These fractions were concentrated to 1 ml
and analyzed for individual hydrocarbons by capillary gas chromatography.
Concentrations were calculated by using standard addition and internal
standard compounds with comparison to authentic standards. For purposes
of this report, concentrations of individual aliphatic and individual
aromatic compounds were summed to represent the respective groups.
EPIFAUNAL RECOVERY STUDIES
In summary, the method involves four steps: (1) preconditioning of
concrete construction bricks by placement in a flowing seawater system
for two weeks; (2) treatment of one-half of the pool of conditioned
bricks with Prudhoe Bay crude oil; (3) characterization of treatment
severity by extraction of oil from bricks and chemical analyses; and (4)
field testing in the intertidal zone at Site II (Figure 1). Independent
experiments were conducted monthly commencing with October 1979.
Preconditioning of bricks in flowing sea water was done to leach out
any foreign materials which might have been associated with the manufacture
of the bricks and to allow some chance for colonization with a microflora.
16
-------
Prior to the treatment phase, all bricks, both control and treated,
received a thorough washing with a high-pressure hose both to remove
unseen but possible incidental occurrences of settled macroorganisms and
to aid in equalizing the effect of preconditioning on control versus
treated bricks. For treated bricks, treatment lasted five days. During
this time the control pool remained in the preconditioning tank. At the
end of the treatment phase, the control bricks again received a wash with
the high pressure hose.
The treatment of bricks with oil was designed to simulate repeated
exposure of intertidal rock during shifts of the tide. The procedure was
as follows. A pool of conditioned bricks was placed on the bottom of a
rectangular tank (0.54 x 4.88 m). The tank was then filled with sea
water (25 cm depth). Prudhoe Bay crude oil (20 2) was poured on the
surface of the sea water to provide a slick thickness of about 1 cm.
Continuous inflow of clean sea water was provided (2 A/min). Both the
inflow and outflow of the sea water were subsurface to prevent disturbing
the surface slick and to retain it in the tank. Twice each day, the
inflow of sea water was discontinued and the seawater level reduced to a
depth of 2 cm. The period of this reduced water level was for two hours
at each treatment. During the periods of low water, the slick was in
contact with the upper and side surfaces of the bricks. The twice daily
regime was repeated for five days. At the end of five days, a surface
outflow was provided, and the slick was washed into the laboratory oil-
treatment facility. Clean seawater flow (2 £/min) was provided for a
further 24 hours.
Chemical Characteristics of Bricks
Routine chemical characterization of the treatment severity has been
based on 5 brick subsamples of the pool (see Table 2 for schedule).
Analysis methods for both infrared spectrometry and capillary gas chroma-
tography follow those previously reported (Vanderhorst et al., 1979).
Three types of extraction procedures have been evaluated and two are
routinely used. The first procedure involved washing whole wet bricks
with 500 m2 CC14. The extraction efficiency was poor. The second pro-
cedure involves air drying bricks for a period of 48 hours before extraction
with 500 m£ CC14. This improved the extraction efficiency approximately
100%. To provide a better measure of the amount of oil actually "seen"
by colonizing organisms, a two-part extraction procedure has been adopted.
In this procedure the top surface of bricks are washed with 200 m£ CC14.
The amount of oil in this extract is measured. The bricks are then
air-dried for 48 hours and reextracted with 500 m£ CC14. The amount of
oil in this extract is measured. Diluted samples of the extracts measured
by infrared spectrophotometry are computed in terms of numbers of grams
of oil per brick. Samples analyzed by capillary gas chromatography are
reported in terms of number of milligrams per brick for individual compounds.
17
-------
Table 2. Preparation of Units and Sampling Schedule for "On-going" Experiments on Effects of Oil on Recovery
by Commercial Clams and Epifauna of Rocky Intertidal.v
00
TASK/SITE DATES UNIT TYPE
H/Discovery Bay 5/80 Infrared
Trays
Cores
Capillary GC
Trays
Cores
6/80 Infrared
Trays
Cores
Biological
Trays
Cores
7/80 Infrared
Trays
Cores
Biological
Trays
8/80 Infrared
Trays
Cores
Capillary GC
Trays
Cores
Biological
Trays
Cores
PRELIMINARY MLLW TIDE
6
18
6
6
2
2
2
2
2
2
2
6
18
6
6
10
70
+2 MLLW TIDE
2
2
2
2
2
2
2
6
18
6
6
10
70
TOTALS
6
18
6
6
4
4
4
4
4
4
4
12
36
12
12
20
140
Note: Core profile on capillary GC and Biological Cores means that discrete
samples will be 2X indicated number for 8/80 sampling.
-------
Table 2. (Continued)
TASK/SITE
DATES.
UNIT TYPE
PRELIMINARY MLLW TIDE +2 MLLW TIDE
TOTALS
G/Sequim Bay 9/79 Concrete
Infrared
Capillary GC
Biological2
10
10
10
10
30
10
10
30
30
30
60
G/Rocky Point
10/79 - 8/80
5/80
The pattern for Task B wi 11 be followed each month with exception of
Capillary GC, giving totals as follows:
Concrete
6/80
I/Sequim Bay 4/80
5/80
1 Experiment balanced, i.e.,
2 Farh nf fhp hinlnnir;il iinil
Infrared
Biological2
Concrete
Infrared
Biological2
Concrete
Infrared
Capillary GC
Biological2
Concrete
Infrared
Biological
for every treated
K<; wi t.h t.hi<; nnt.at.i
10
10
10
4
120
sample a control
on fwit.h t.hp pyri
10
30
10
30
10
4
60
10
30
sample
^nt.inn r
10
30
10
30
10
4
60
10
30
is also indicated.
>f nrpl inn narv^ shmilri
30
60
30
60
30
12
240
20
60
HP
multiplied by 5 for 5 daily observations within the month. Appropriate total unit samples are:
Capillary GC, 60; Infrared, 532; Biological, 3328.
-------
SECTION 5
RESULTS AND DISCUSSION
INFAUNA RECOVERY
A Perspective
Since marine systems are dynamic and differ greatly from place to
place, a definition of full recovery is nebulous. For purposes of these
studies, we defined full recovery of the infaunal communities as that
composition and density of organisms in native substrate control trays
after 15 months colonization. The experimental results presented in the
following subsections arrive at effects from oil on recovery by making
direct comparisons in density of individual species in trays with and
without oil treatment but identical in all other respects, i.e., prep-
aration method, site, sediment type, beach location, tidal height, season,
and duration of colonization. The purpose of this section is to demon-
strate the relevance of the above definition of recovery in terms of
baseline information available for nearby sites from related MESA studies
(Nyblade, 1979). Also, summary data will be presented to give perspective
to the relevance of using a defined full recovery in assessing impacts of
oil on recovery.
Table 3 presents data from two MESA Puget Sound baseline stations
geographically embracing the location of the present study and data from
15-month control coarse substrate at Sequim Bay. These data show the
numbers of species of animals within the three principal taxonomic groups
(polychaetes, crustaceans, mollusks), for all other taxonomic groups, and
the total number of species. The MESA baseline data for Beckett Point
(Discovery Bay) and Jamestown were derived from Appendix I (Nyblade,
1979) and apply to the Mean Lower Low Water (O1) tide level for fall
samples at those stations (collected October 1977). The data for the
recovery study are based on enumeration of infauna from coarse sediment
control trays at Mean Lower Low Water (0') collected November 1979, 15
months after placement at Sequim Bay. In a general sense the habitats
are similar.
The total number of species for the three situations is reasonably
similar. The indicated number for the recovery study is slightly lower,
but one must keep in mind the sensitivity of data collection methods and
annual influences. For example, if one consults the same category of
samples for Jamestown during 1976 (Nyblade, 1978), a total of 57 species
are indicated, or less than the recovery study. In each of the situations,
polychaetes have the highest representation with from 43 to 56% of the
total. There is quite a bit of difference in percent contribution by the
other groups. Beckett Point and the recovery study were similar in
having 21 to 25% contributed by mollusks, and relatively small contri-
butions by the "other species" group (7 and 8% for Beckett Point and the
recovery study, respectively). The recovery study and Jamestown had more
nearly equal representation by crustaceans with 28 and 17%, respectively.
20
-------
Table 3. Summary of Number of Species in Principal Groups of Animals
from Baseline Studies (Nyblade, 1979) and 15-Month Controls
in Recovery Experiment.*
GROUP
Polychaetes
Moll us ks
Crustaceans
All Others
Total s
NUMBER OF SPECIES AND
BECKETT POINT
52
23
11
7
93
(56)
(25)
(12)
(7)
(100)
PERCENT (
RECOVERY STUDY
25
13
17
5
61
(43)
(21)
(28)
(8)
(100)
) OF TOTAL
JAMESTOWN
41
6
13
16
76
(54)
(8)
(17)
(21)
(100)
*
Numbers of species, or species richness, is strongly related to area
sampled until enough area is covered to reach some "asymptote" mean
number. Data from the recovery studies are conservative with respect
to baseline. See text for details.
21
-------
We conclude from these data that in crude form both the total number of
species and the relative distribution of species in the principal taxo-
nomic groups in the 15-month recovery study coarse sediment controls are
similar to what might be expected in sampling a fully recovered nearby
beach with similar physical attributes.
There are several good reasons which preclude quantitative comparisons
to further refine the above judgment. First, year to year variation in
numbers of species at a given site may be great. The difference cited
above for the baseline studies at Jamestown between 1976 and 1977 (25%
fewer in 1976) is not unusual. Our study was conducted during the succeeding
year (1978). Jamestown and Beckett Point differ within the same year and
season (17% fewer at Jamestown). Perhaps most important to the present
discussion is the relationship of increases in numbers of species with
area sampled. In general, this has been shown to be a logarithmic function
for marine sediments. The baseline studies sampled about 2 times the
area sampled in the recovery study. For this reason we believe the
numbers of species in each of the categories of the recovery study are
conservative estimates of the state of recovery.
Data for the same categories on numbers of individuals contributed
by the principal groups are shown in Table 4. Both the baseline data and
recovery study data have been normalized to a per m2 basis. Distinctions
closer than an approximate order of magnitude should not be made. For
example, one polychaete species in the recovery study, Polydora social is,
contributed more than 49,000 individuals per m2, roughly 48% of polychaetes
or 40% of all individuals per m2. High density of this species was not
an anomaly of the experimental design or sediment trays. The species is
quite visible and literally covered the Sequim Bay beach during this
period. Similar disproportions in contribution by a single species to
total numbers can be found in the baseline information. For example,
Oligochaeta spp. comprised 39% of individuals at Jamestown during the
period of interest (Nyblade, 1979), and for Beckett Point, the tanaidacean,
Leptochelia dubia, contributed 46% of all individuals. We believe that
the data in Table 4 indicate high similarity between the numbers of
individuals in taxonomic groups from the recovery study coarse controls
and the baseline sites. They also make a case for study of recovery
under controlled conditions on a species by species basis.
In the recovery study, each of the factors (site, tide level, sediment
type, season, and duration of recovery) had independently allocated oiled
and unoiled units. For this reason, a comparison of summary statistics
on total numbers of individuals (Table 5) and total numbers of species
(Table 6) is appropriate and adds perspective to the use of a "full
recovery" concept in describing recovery.
Sequim Bay coarse sediment controls with 15 months intertidal coloni-
zation, defined above as fully recovered, showed a rank order of importance:
(1) polychaetes; (2) crustaceans; (3) mollusks; and (4) "other species,"
both in terms of individuals (Table 5) and species (Table 6). This rank
order of importance was frequently observed in the experiments although
it does not consistently relate to the primary treatment categories
(oiling status, sediment type, tide level, or site). Thus, in the three
experiments this control rank order applied to 7 of the 15 remaining
treatment categories in terms of individuals (Table 5) and 11 of the
remaining 15 categories in terms of species (Table 6).
22
-------
Table 4. Summary of Number of Individuals in Principal Groups of
Animals from Baseline Studies (Nyblade, 1979) and 15-Month
Controls in Recovery Experiment.
GROUP
INDIVIDUALS NORMALIZED TO M2
BECKETT POINT
RECOVERY STUDY
JAMESTOWN
Polychaetes
Mollusks
Crustaceans
All Others
Totals
18,640
21,644
46,136
10,692
97,112
102,590
1,042
15,780
472
119,884
41,956
2,636
4,452
1,416
50,460
23
-------
Table 5. Group Density/m2 in Oil Recovery Experiments.
INDIVIDUALS/M2
EXPERIMENT/CONDITION POLYCHAETES CRUSTACEANS MOLLUSKS OTHERS TOTALS
I. Fall 3-Month Recovery: Started 8/78; Completed and Sampled 11/78.
Two sites (Protection Island; Sequim Bay); Two sediment types
at each site (Coarse-Fine); One tide level (MLLW); Two oiling
categories (Oiled; Unoiled Control).
Protection Island
Coarse Sediment
Control 2,046 4,781 571 197 7,595
Oiled 807 866 256 0 1,928
Fine Sediment
Control 1,397 3,168 2,578 0 7,142
Oiled 669 1,181 2,597 0 4,447
Sequim Bay
Coarse Sediment
Control 18,259 2,400 590 236 21,486
Oiled 6,316 1,180 571 59 8,126
Fine Sediment
Control 9,976 964 2,282 79 13,301
Oiled 14,973 433 2,322 79 17,807
II. Summer 3-Month Recovery: Started 4/79; Completed and Sampled 8/79.
One site (Sequim Bay); One sediment type (Coarse); Two tide levels
(MLLW; -2' below MLLW); Two oiling categories (Oiled; Unoiled Control).
Sequim Bay - Coarse Sediment
Minus 2' below MLLW
Control 16,744 12,592 551 925 30,812
Oiled 12,711 5,549 236 59 18,544
Mean Lower Low Water
Control 11,825 1,200 197 315 13,537
Oiled 5,549 394 98 20 6,060
III. Fall 15-Month Recovery: Started 8/78; Completed and Sampled 11/79.
One site (Sequim Bay); Two sediment types (Coarse; Fine); One tide
level (MLLW); Two oiling categories (Oiled; Unoiled Control).
Sequim Bay, MLLW
Coarse Sediment
Control
Oiled
Fine Sediment
Control
Oiled
102,590
49,564
120,869
94,405
15,780
7,103
17,571
9,602
1,043
1,003
1,220
1,358
472
275
453
433
119,886
57,847
140,113
105,797
Each mean presented based on n = 35 cores (14.52 cm2) in groups of 7 with 5
replicate groups (trays).
24
-------
Table 6. Summary of Numbers of Species in Oil Recovery Experiments.
NUMBER OF SPECIES
EXPERIMENT/CONDITION POLYCHAETES CRUSTACEANS MOLLUSKS OTHERS TOTALS
I. Fall 3-Month Recovery: Started 8/78; Completed and Sampled 11/78.
Two sites (Protection Island; Sequim Bay); Two sediment types
at each site (Coarse-Fine); One tide level (MLLW); Two oiling
categories (Oiled; Unoiled Control).
Protection Island
Coarse Sediment
Control 12 12 61 31
Oiled 11 5 4 0 20
Fine Sediment
Control 16 13 50 34
Oiled 13 9 3 0 25
Sequim Bay
Coarse Sediment
Control 21 14 86 49
Oiled 14 16 51 36
Fine Sediment
Control 18 15 65 44
Oiled 16 6 4 1 27
II. Summer 3-Month Recovery: Started 4/79; Completed and Sampled 8/79.
One site (Sequim Bay); One sediment type (Coarse); Two tide levels
(MLLW; -2' below MLLW); Two oiling categories (Oiled; Unoiled Control).
Sequim Bay - Coarse Sediment
Minus 2' Below MLLW
Control 27 25 7 10 69
Oiled 19 21 30 43
Mean Lower Low Water
Control 21 14 36 44
Oiled 15 6 4 1 26
III. Fall 15-Month Recovery: Started 8/78; Completed and Sampled 11/79.
One site (Sequim Bay); Two sediment types (Coarse; Fine); One tide
level (MLLW); Two oiling categories (Oiled; Unoiled Control).
Sequim Bay, MLLW
Coarse Sediment
Control
Oiled
Fine Sediment
Control
Oiled
26
28
32
27
17
15
22
12
13
8
9
5
4
7
4
3
60
58
67
47
*
Each number is aggregate of species in 35 cores (14.52 cmVcore) distributed
7 cores per tray in 5 replicate trays.
25
-------
Some of the exceptions to the rank order of importance pattern are
of interest. In terms of numbers of individuals it never applied to
Protection Island due primarily to the fewer number of polychaete individ-
uals found there (Table 5). In sharp contrast, the numbers of species
always followed the above rank order of importance at Protection Island
(Table 6). Interestingly, in Experiment II, the summer tide level experi-
ment at Sequim Bay, the oiled sediments at each of the tide levels followed
the fully recovered pattern for individuals while the control sediments
did not.
In coarse, oiled sediments, there was a reduction of the total
number of individuals (Table 5) as compared to unoiled coarse sediments
in every instance. In only one case (Experiment I, Protection Island)
did this reduction result in a change in the "full recovery" ranking. A
disproportionate reduction in polychaete numbers placed that group second
in importance for this case. Total number of individuals in oiled fine
sediments was also reduced in comparison to unoiled controls in the
15-month data (Experiment III), and 3-month data (Experiment I) for
Protection Island. Three-month fall data at Sequim Bay (fine sediments)
did not follow this pattern.
For the 3-month recovery data (Experiment I), it is apparent that
the site of experimentation had quite an appreciable effect on both the
numbers of individuals per m2 (Table 5) and numbers of species (Table 6).
Overall, Protection Island had a much smaller number of species and
individuals. Protection Island had 31 versus 49 species for Sequim Bay
for coarse sediment controls, and about a third as many individuals. For
fine sediment, native to Protection Island, the differences are not quite
so great with 34 species versus 44 species, and about half as many indi-
viduals. The overall data trend for both species and individuals is set
by the polychaetes. Protection Island had a greater number of individual
crustaceans (slightly fewer species) and about the same number of indivi-
duals and species of mollusks. Protection Island was lower in the con-
tribution by "other species" both in terms of species and individual
numbers.
The type of sediment (coarse versus fine) also had an influence on
the total number of number of individuals and species. For the fall
3-month experiment (I) there were a greater number of individuals per m2
in coarse sediment controls as compared to fine at each of the sites.
For oiled sediments, the situation was exactly reversed. In the 15-month
recovery experiment (III) there "were a greater number of individuals per
m2 in fine sediments as compared to coarse. This was true in both the
oiled and unoiled condition.
The tide level experiment (II) in Sequim Bay-coarse sediment reveals
important differences in recovery reflected by numbers of individuals
(Table 5) and numbers of species. The minus two feet below MLLW level
had substantially more species and individuals. The trend in this case
is consistent among all control groups.
Data on Tables 5 and 6 also permit a comparison of 3-month recovery
periods in the spring-summer, and late summer-fall. The appropriate
26
-------
comparison is Sequim Bay-coarse sediment controls from Experiment I to
Mean Lower Low Water controls in Experiment II. The total number of
species and individuals was slightly greater in the fall recruitment
period. The greater number of species was contributed by the moll usks.
The greater number of individuals pervades
all groups with the exception of "other species." The data are consistent
with the baseline study findings (Nyblade, 1979) for Beckett Point and
Jamestown which also indicate fewer species and individuals in summer
collections versus the fall.
The data presented above indicate that attempts to evaluate the
effects of oil on recovery of infaunal communities in terms of a one-
condition "fully recovered" situation, or indeed to baseline data since
these too are available (Tables 3 and 4), will be strongly confounded by
tide level, site, sediment grain size, and possibly season. The data do,
however, illustrate that, where these factors can be fixed by experimental
design, a valuable perspective on recovery effects emerges. To clarify
this position, data are presented on Sequim Bay-coarse sediment from Mean
Lower Low Water, indexed as a percentage to the 15-month control substrates
of like kind.
From Table 6, recovery relative to 15-month unoiled controls in
terms of total number of species, was 70 to 80% in 3-month periods. For
oiled substrates the corresponding range is from 40 to 60%. Within the
taxonomic groupings, crustaceans and polychaetes were 80% recovered in
3-month controls for either season. MoHusks were relatively less re-
covered, although quite variable (20 to 60%). Overall, in 15 months,
oiled substrates were 92% recovered relative to unoiled controls in terms
of numbers of species.
Numbers of individuals (Table 5) were much less fully recovered for
all categories than were species, except for mollusks. Three-month
control recovery was from 11 to 18%. Fifteen-month oiled coarse substrate
recovery was slightly less than one-half of controls. The 3-month oiled
substrate recovery of polychaetes and crustaceans was only 2 to 7% and 21
to 45% for the 15-month recovery.
The data presented in Tables 5 and 6 indicate that there were clear
effects from the oiling on degree of recovery, in terms of numbers of
species, and that it had largely been compensated for by the end of 15
months. These data further indicate that there were marked effects from
oiling on the total number of individuals at the end of three months and
that these effects from the oiling still amounted to roughly half the
numbers of individuals by the end of 15 months. Interpretation of the
importance of effects on numbers of individuals can only come from an
analysis of individual species populations because of the highly dis-
proportionate contribution of individual species to total numbers
(previously discussed) and the fact that while the numbers of some species
are undoubtedly reduced by the oiling, others may be increased.
Recovery of Primary Species
From the preceding section, the recovery of total numbers of indi-
viduals required more time and was apparently more strongly influenced by
27
-------
the oiling compared to recovery of numbers of species. Because we wish to have
statistically valid probability statements in hypothesis tests concerning
the effects of oil, site, sediment type, and tide level, and want valid
comparisons of season and duration of recovery, subsets of the total
number of species were preselected and designated as primary. The first
data in this section relate to the question of how well the primary
species represent all species in terms of numbers of individuals. By way
of clarification, 13 species were designated as primary species. Seven
of these were designated as primary for Experiment I (site, sediment
type, oil variables); ten were designated as primary for Experiments II
(season, tide level, oil variables) and III (sediment type, oil variables,
15-month duration). In some cases, designating the same species for
the two groups as primary accounts for the total of 13 rather than 17
species. Of the 13 species, eight contributed to available data for
Beckett Point, and eight contributed to available data for Jamestown
(Nyblade, 1979). These were not the same eight species for the site
cases, so that the combined data for Beckett Point and Jamestown included
12 of the 13 primary species.
On Table 7, the numbers of individuals per m2 for all primary species
are shown as a percentage of total numbers of individuals per m2 for
baseline sites (Nyblade, 1979) and 15-month-coarse sediment control data
in the recovery study. Overall, the primary species encompassed 73% of
individuals per m2 reported for Beckett Point, 78% of all individuals per
m2 for the recovery study, and 33% of total individuals per m2 for Jamestown.
In other words, they comprised a very substantial contribution of all
individuals for Beckett Point and our "fully recovered" experimental
data. They comprised about a third of total individuals per m2 for
Jamestown. While this is not as great, it is still a large contribution.
Within the various taxonomic categories, primary species account for 97%
of Crustacea, 92% of mollusks and 36% polychaetes reported for Beckett
Point. For Jamestown, primary species contributed 29% for polychaetes,
40% for Crustacea, and 97% for moll usks. For the recovery study, they
represented 76% of polychaetes, 90% of Crustacea, and 51% of mollusks.
We conclude from these data that effects on the density of primary species
will indeed alter indicated recovery in terms of total numbers of indi-
viduals.
To illustrate the contribution of density for primary species within
a recovery framework, Table 8 lists the numbers of individuals per m2,
the percent of final control (recovery) for numbers of individuals per
m2, in coarse sediment, MLLW experimental situations, and the percent
that the primary species contributed to means for all species in a given
experimental situation. The data indicate a recovery pattern for primary
species similar to that presented for total species. Fifteen-month
recovery in coarse oiled substrates is 48 and 38% for polychaetes and
crustaceans, respectively, and 48% overall. Mollusks show an "over-recovery"
i.e., 174% of controls. Three-month recovery was 8 and 19% for controls;
and was 5 and 7% for oiled substrates.
28
-------
Table 7. Numbers of Individuals for "Primary" Species as a Percentage
of All Individuals for Beckett Point and Jamestown (from
Nyblade) and Recovery Study Situations.*
% OF ALL INDIVIDUALS
GROUP
BECKETT POINT RECOVERY STUDY JAMESTOWN
Polychaetes
Mollusks
Crustaceans
All Others
Total s
36
92
97
0
73
76
51
90
0
78
29
97
40
0
33
*
Proportions of individuals/m2 computed from Appendix I (Nyblade, 1979),
Fall 1977, +0' tide level, for Beckett Point and Jamestown. Data from
Recovery Study, November 1979, Sequim Bay, coarse sediment controls.
29
-------
Table 8. Density (No./m2), Recovery (% Final Control), and Representation (% Respective Total), for
Primary Species in Recovery Experiments at Sequim Bay/MLLW/Coarse Sediment.
CO
o
3-MONTH FALL
3-MONTH SUMMER
Oil
Control
Oil
Control
15-MONTH FALL
Oil
Control
POLYCHAETES
No./m2
(%) Final Control
% Respective Total
CRUSTACEANS
No./m2
(%) Final Control
% Respective Total
MOLLUSKS
No./m2
(%) Final Control
% Respective Total
TOTALS FOR ALL GROUPS
No./m2
(%) Final Control
% Respective Total
5706
(7)
90
511
(4)
43
472
(89)
83
6690
(7)
82
16055
(20)
88
1220
(9)
51
393
(74)
69
17688
(19)
82
4623
(6)
83
236
(2)
60
59
(11)
60
4919
(5)
81
7064
(9)
60
610
(4)
51
177
(33)
90
7850
(8)
58
37994
(48)
77
5351
(38)
75
924
(174)
97
44271
(48)
77
78378
(100)
76
14127
(100)
90
531
(100)
51
93036
(100)
78
-------
To this point we have presented data and discussed recovery of
individual numbers in terms of numbers per m2 to facilitate general
comparisons with the baseline studies and the recovery study. All data
which follow are in the original units (numbers of individuals of a
category per tray (experimental unit)). More explicitly, this is an
average of the total numbers of individuals of a given category removed
from 5 groups of 7 cores per tray. We do this to preclude inflating the
statistical error. Further, to better meet assumptions of normality and
discrete as opposed to continuous statistical distributions, for each
analysis of variance performed on the counts, supplemental analyses were
performed on the transformations of raw data, In (x + 1) and V * + 1 •
In general, the "significance" attributed to results did not differ when
these transformations were made. They will be mentioned again only in
cases where computed differences in "significance" did arise. Obviously,
intuitive interpretation of effects on the transforms would be much more
obscure.
Experiment I. Site, Sediment, Oil Effects, 3-Month Recovery--
Mean density for primary species in Experiment I is shown on Table 9.
A cursory examination of these data indicate that higher densities and
more decided differences between experimental conditions are seen for
polychaetes. In general, polychaetes were more abundant at the Sequim
Bay site as opposed to Protection Island. Ophiodromus pugettensis was
entirely absent at Protection Island. Densities of the remaining polychaete
species were considerably less there. The three moll usk species appeared
more equitably distributed at the two sites. For the most abundant
mollusks, the small bivalves, Mysella tumida and Transennella tantilla,
there appeared to be substantially greater numbers in the finer sediment,
independent of site and oil treatment.
We made null hypotheses that individual densities for the primary
species would be equal at the two sites, in the two sediment types, and
in oiled and unoiled trays. The computed probability for Type I statistical
error (significance level from Analysis of Variance) in rejecting these
hypotheses are tabulated on Table 10. We preselected a rejection criterion
of 0.010 to deem "significant" effects. Since we are testing hypotheses
on seven species, the maximum real probability for error is 7% in this
experiment. Based on this criterion, we reject the hypotheses that
densities at Sequim Bay and Protection Island are equal for Ophiodromus
pugettensis, Platynereis bicanaliculata, and Armandia brevis. Furthermore,
we reject the hypotheses that Mysella tumida and Transennella tanti11 a have equal
densities in the two sediment types. With respect to effects from oil
treatment on density, we reject the hypotheses that densities for Platynereis
bicanaliculata and Armandia brevis are equal in oiled and unoiled trays.
In light of the results of these hypotheses tests, a reexamination
of the mean densities on Table 9 can provide insight into the general
recovery picture in terms of individual numbers previously discussed.
Recall from Tables 5 and 8, that 3-month fall recovery in terms of
numbers of individuals was quite low. For primary species (Table 8) it
amounted overall to 7 and 19% for oil-treated and untreated trays,
respectively.
31
-------
Table 9. Mean Density of Primary Species (No./tray) in Experiment I.*
CO
ro
o.
to
I
EXPERIMENTAL CONDITIONS S
Protection Island
Coarse
Control
Oiled
Fine
Control
Oiled
Sequim Bay
Coarse
Control
Oiled
Fine
Control
Oiled
1.2
0
1.6
1.4
2.4
0.6
1.6
1.2
MOLLUSKS
a -ti
ID -JJ
to
2.4
0.4
7.4
10.2
1.4
2.6
7.2
8.8
POLYCHAETES
M N
r^-S t-o
1.4
1.8
16.4
14.8
0.2
1.6
13.4
13.4
.CO
0
0
0
0
0.4
1.2
0.2
0.6
«
_a -^
7.4
1.0
2.6
1.4
78.6
11.2
57.2
2.2
Platynereis
biaanaliculata
2.4
0.2
0.8
0.8
16.0
6.8
9.6
4.6
Capi tel la
capitata
0.8
0.6
1.6
0.6
17.0
17.0
11.2
134.2
* This experiment included: Two sites (Protection Island; Sequim Bay); Two sediment types (Coarse; Fine);
One tide level (MLLW); Two oiling categories (Oiled and Unoiled Control). 3-Month Recovery (Started 8/78;
Completed and Sampled 11/78). Means are based on 5 replicates per condition (7 cores per replicate).
Hypothesis test for main effects on Table 10.
-------
Table 10. Hypothesis tests for Density of Primary Species in Oil
Recovery Experiment I.
PROBABILITY FOR ERROR IN REJECTING THE HYPOTHESES
PRIMARY SPECIES
Lacuna sp.
Mysella tumida
Transennella tantilla
Ophiodromus pugettensis
Platynereis bi canal iculata
Capitella capitata
Armandia brevis
SITE1
0.353
0.879
0.334
0.000*
0.000*
0.142
0.000*
SEDIMENT2
0.353
0.000*
0.000*
0.459
0.079
0.334
0.014
OIL3
0.042
0.178
0.973
0.220
0. 004*
0.305
0.000*
1 The site hypothesis is: Density in trays at Protection Island equals
density in trays at Sequim Bay.
2 The sediment hypothesis is: Density in trays containing coarse sediment
is equal to density in trays containing fine sediment.
3 The oil hypothesis is: Density in trays receiving oil treatment is
equal to density in trays not receiving oil treatment.
* We reject the indicated hypothesis with a maximum real probability
for error of 7%.
33
-------
The primary species "significantly" affected by oil are among those
of highest density within the group. Armandia brevis had the highest
density in coarse sediments from which we drew the general recovery
picture (previous section). In fine, oil-treated sediments at Sequim
Bay, density of A. brevis was far exceeded by Capitella capitata. The
latter species, an opportunistic polychaete for which density has been
suggested as an indicator of pollution generally, and oil pollution
specifically, demonstrated a high lack of consistency in its experimental
pattern of density with the one exception. That is, that it, like the
other polychaetes, was apparently more prevalent at Sequim Bay. Because
of relatively high contributions to total numbers of ".idividuals by
species like Capitella capitata, and the high density mollusks, Mysella
turnida, and Transennella tantilla, whose density was unrelated to oil
treatment in this experiment, total number of individuals even among
primary species, will tend toward a conservative estimate of the effects
of oil on recovery. Because of "significant" site effects on density,
neither Ophiodromus pugettensis nor Platynereis bicanaliculata, should be
deemed as general indicators of recovery following a spillage of oil.
Experiment II. Tide Level, Oil Effects, 3-Month Recovery--
Density of primary species in the setting of Experiment II is shown
in Table 11. Generally, densities for the primary species were greater
at the lower tide level, with the outstanding exception of the high
density polychaete, Exogone lourei. Consistently lower densities in
oiled versus unoiled sediments are observable for half of the primary
species including, Mysella tumida, Corophium ascherusicum, Leptochelia
dubia. Exogone lourei, and Polydora social is.
From the hypotheses tests, Table 12, tide level differences are
deemed significant for Leptochelia dubia, Armandia brevis, Platynereis
bicanaliculata, and Polydora social is. Significant density differences
due to oil treatment were found for Corophium ascherusicum, Leptochelia
dubia, Exogone lourei, and Polydora social is. For each of these species,
mean density was greater in control than in oiled trays (Table 11).
Seasonal aspects are of interest, but are in most cases, confounded.
Lacuna sp., Mysella tumida. Platynereis bicanaliculata, and Armandia
brevis were primary species common to Experiments I and II. Among those
species, Platynereis bicanaliculata, and Armandia brevis were deemed to
have significant effects on density due to site and oiling in Experiment
I, and due to tide level in Experiment II. Nonalignment of these factors
should therefore be excluded in seasonal comparison. Because of depen-
dencies in treatment description data and possible differences in the oil
treatment itself (p. 57), Analysis of Variance was not used to describe
seasonal differences. For the appropriate comparison (coarse sediment,
control, mean lower low water), the mean density (Tables 9 arid 11)
appears to be an order of magnitude lower for these two species during
the summer season as compared to fall. The mollusk density (Mysella
tumida, Lacuna sp.) appears inconsistent with respect to season.
34
-------
Table 11. The Mean Density of Primary Species in Experiment II
(Tide Level, Oil, 3-Month Recovery, Summary).
MEAN NUMBERS/TRAY
PRIMARY SPECIES
MOLLUSKS
Mysella tumida
Protothaca staminea
Lacuna sp.
CRUSTACEANS
Corophium ascherusicum
Photis brevipes
Leptochelia dubia
POLYCHAETES
Armandia brevis
Exogone lourei
PI atynerei s bicanal i cul ata
Polydora social is
MLLW
CONTROL
1.6
0
0
1.0
0.6
4.6
6.0
38.0
0
2.6
OIL
0.2
0
0.4
0.6
0
1.8
0.2
9.0
0
0.4
MINUS
CONTROL
2.8
0
0.2
3.0
0.8
106.6
43.6
16.2
0.8
32.0
2'
OIL
1.6
0
0
0.6
5.0
40.2
56.6
10.2
1.4
11.4
Means based on 5 replicates per condition (7 cores per replicate).
Analyses of Variance for main effects on Table 12.
35
-------
Table 12. Hypothesis Tests for Density of Primary Species in
Experiment II (Spring-Summer, Tide, Oil).
PROBABILITY
PRIMARY SPECIES
MOLLUSKS
My sell a tumida
Protothaca staminea
Lacuna sp.
CRUSTACEANS
Corophium ascherusicum
Photis brevipes
Leptochelia dubia
POLYCHAETES
Armandia b rev is
Exogone lourei
Platynereis bi canaliculate
Polydora social is
FOR ERROR IN REJECTING
TIDE LEVEL1
0.092
1.000
0.536
0.031
0.178
0.000*
0.000*
0.049
0.003*
0.000*
THE HYPOTHESIS
OIL*
0.092
1.000
0.536
0.005*
0.344
0.000*
0.640
0.002*
0.357
0.004*
1 The tide level hypothesis is: Density in trays at MLLW is equal
to density in trays at -2' below MLLW.
2 The oil hypothesis is: Density in trays receiving oil treatment is
equal to density in trays not receiving oil treatment.
* We reject the indicated hypothesis with a maximum real probability
for error of 10%.
36
-------
Experiment III. Sediment, Oil Effects, 15-Month Recovery--
Mean primary species densities in Experiment III are shown on Table
13. In this 15-month framework, the most easily observable phenomenon is
the marked increase in density for most species, especially crustaceans
and polychaetes, in contrast to the 3-month recovery period (Tables 9 and
11). There are greater numbers of individuals in unoiled sediments as
compared to oiled in both coarse and fine sediment for all of the crustacean
species, and all of the polychaete species with one exception (Polydora
social is in fine sediments). One mollusk, Lacuna sp., also had lesser
numbers in oiled sediments. Another mollusk, Mysella tumida, consistently
had a greater number of individuals in oiled versus control sediment.
Hypothesis tests in this experiment revealed no significant differences
due to sediment type (Table 14). Differences in density due to oiling
were significant for Mysel1 a tumida, Armandia brevis, and Platynereis
bicanaliculata. The Vx + 1 transform of density for Corophium ascherusicum
was significantly different in oiled versus unoiled sediment. This was
the only case in which data transformation changed "significance" designation.
Substrate Recovery Indicators for the Strait of Juan de Fuca.
We have shown in the preceding section that the 13 primary species
contain a substantial part of the total number of individuals in fully
recovered experimental trays and from nearby baseline stations. Further,
while the preponderance of effects on density from the oil treatment were
reductions compared to appropriate controls, in the case of the high
density mollusk, Mysella tumida, there was, in fact, a significantly
greater density in oiled substrates as compared to controls. For the
3-month recovery data presented, there is also an indication (non-
significant) that Capitella capitata may have greater density in oiled
than in non-oiled substrates. Thus, while the total number of individuals
in the primary species follows total numbers of individuals quite well,
there is an inconsistent but conservative effect for both the total
number of individuals and total number of individuals of primary species
due to species responding differently to the oil treatment.
Two of the primary species, Exogone lourei and Leptochelia dubia,
were excellent indicators of recovery in this study and would seem to be
good region-wide indicators as well. It should be emphasized that this
is not in the sense of "indicator species" concepts where some measure of
the density for these species alone would be indicative of recovery.
Rather, the two species alone, used in a proper experimental design
implemented at the time of some future oil contamination and backed with
appropriate chemical analyses, would provide a good measure of substrate
suitability for recovery in terms of numbers of individuals. Because of
high replication needs, the limitation of interest to the two species
would result in substantial savings in sampling effort. The merits of
these species are: (1) a relatively high sensitivity to the effects of
the oil treatment; (2) a relatively high methodological sensitivity; and
(3) a rather ubiquitous distribution in the Strait of Juan de Fuca and
northern Puget Sound region. The data in this section are provided to
document these features. Before proceeding, however, we briefly digress
to two taxonomic considerations.
37
-------
Table 13. The Mean Density of Primary Species in Experiment III
(Sequim Bay, 15-Month Recovery, Sediment Type, Oil).
MEAN NUMBERS/TRAY
PRIMARY SPECIES
MOLLUSKS
My sell a tumida
Protothaca staminea
Lacuna sp.
CRUSTACEANS
Corophium ascherusicum
Photis brevipes
Leptochelia dubia
POLYCHAETES
Armandia b rev is
Exogone lourei
Platynereis bi canal iculata
Polydora social is
COARSE
CONTROL OIL
3.4 7.2
0 0.6
1.0 0.8
34.8 11.6
5.6 0.8
103.2 42.0
48.8 22.2
323.4 197.2
32.8 9.0
322.3 83.6
FINE
CONTROL
6.8
0.4
1.0
38.8
3.6
99.0
69.6
512.4
40.2
224.1
OIL
12.4
0.2
0.0
0.8
0.6
87.4
10.6
279.6
3.6
294.9
Means based on 5 replicates of groups of 7 cores per group. Hypothesis
tests shown in Table 14.
38
-------
Table 14. Hypothesis Tests of Density of Primary Species in Experiment
III (15-Month Recovery, Sequim Bay, Sediment, Oil).
PROBABILITY
PRIMARY SPECIES
MOLLUSKS
Mysella tumida
Protothaca staminea
Lacuna sp.
CRUSTACEANS
Corophium ascherusicum
Photis brevipes
Leptochelia dubia
POLYCHAETES
Armandia brevis
Exogone lourei
Platynereis bi canal iculata
Polydora social is
FOR ERROR IN REJECTING
SEDIMENT1
0.012
1.000
0.465
0.790
0.533
0.485
0.767
0.108
0.913
0.266
THE HYPOTHESIS
01 Lz
0.007*
0.514
0.278
0.0273
0.038
0.225
0.000*
0.039
0. 004*
0.349
1 The sediment hypothesis is: Density in trays with coarse sediment
is equal to density in trays with fine sediment.
2 The oil hypothesis is: Density in trays receiving oil treatment is
equal to density in trays not receiving oil treatment.
3 ANOVA on square-root transform computed "significant" according to
our criteria.
* We reject the indicated hypothesis with a maximum real probability
for error of 10%.
39
-------
The polychaete, Exogone lourei, has an abundant congener, L gemmifera,
in our region. Although they are distinctive, it may be quite important
to make that distinction (see p. 43). The crustacean, Leptochelia dubia,
has a confusion of names. It is often called L savignyi. It is beyond
the scope of this study to provide the necessary investigation of nomen-
clature to assign an appropriate name. We have followed the lead of
Nyblade (1979) who calls the species L. dubia, apparently using keys of
Light, Smith and Carlton (1975). Taxonomic keys provided by Kozloff
(1974), perhaps more widely used in the region, give the name L savignyi.
Thus, some authors who provide the comparative data may use this name.
Unusually high numbers of this species in a wide variety of situations in
our region certainly warrant studies to clarify its nomenclature.
Although neither of the above species was preselected as a primary
species for Experiment I (the site comparison 3-month fall recovery),
they were, in fact, important contributors to all segments of that experiment.
Table 15 provides the mean density information for the two species in
that experiment. In general, the pattern of density follows that discussed
in Experiments II and III. Fine sediment information is not conclusive.
Exogone lourei was less dense at Protection Island. Even there, however,
the species consistently occurred in all treatment categories. Leptochelia
dubia was even more abundant at Protection Island than at Sequim Bay, and
shows the oil treatment trends as clearly for Protection Island as those
shown for Sequim Bay.
Seasonal mean density for Mean Lower Low Water of Exogone and Leptochelia
at the adjacent baseline stations (Nyblade, 1979) is shown in Table 16.
Both the species are at least moderately abundant on a year-round basis.
The data from Nyblade (1979), Webber (1979), and Smith and Webber (1978),
indicate a predominantly lower intertidal and shallow subtidal abundance
pattern for the two species. For stations on the west shore of Whidbey
Island, Webber (1979) listed Leptochelia savignyi and Exogone sp. among
dominant species in the shallow subtidal zone. As far away as Drayton
Harbor in the north, Leptochelia savignyi was identified as an important
constituent of the low intertidal zone by Smith and Webber (1979). The
species was used by these same authors to illustrate site differences
between Cherry Point, where it was rarely found, and Shannon Point, where
it was an extremely common constitutent. Simenstad et al. (1980) found
L dubia, as a member of the epibenthic zooplankton, in high numbers at
stations west of Port Angeles but, inexplicably, not at stations east of
Port Angeles.
The relative spatial and seasonal ubiquity of these species indicates
to us that sources of these species would likely remain stable well
beyond any envisioned immediate impact zone due to oil contamination.
Given proper experimental control they should reflect very well the
degree of recovery from oil contamination.
Recovery of Other Community Members
Taken alone, the data on primary species from the two preceding
sections might be construed as an anomalous indicator of the general
community recovery in terms of numbers of individuals. Therefore, the
computed densities for other community constituents are reported here for
40
-------
Table 15. Mean Density for Recovery Indicators in Experiment I.
MEAN NUMBERS/TRAY1
EXPERIMENTAL CONDITION
Protection Island
Coarse Sediment
Control
Oiled
Fine Sediment
Control
Oiled
Sequim Bay
Coarse Sediment
Control
Oiled
Fine Sediment
Control
Oiled
Exogone
1.0
0.2
0.6
0.8
51.2
21.8
12.4
5.0
Leptochelia
32.0
5.8
14.8
5.4
9.6
5.0
1.6
2.2
1 Mean density based on 5 replicates of groups of 7 cores for each mean.
41
-------
Table 16. Estimated Density for Recovery Indicator Species at
Adjacent Baseline Sites.1
NUMBERS PER SQUARE METER
LOCATION/SEASON
Beckett Point
Spring 1977
Summer 1977
Fall 1977
Winter 1977
Jamestown
Spring 1977
Summer 1977
Fall 1977
Winter 1977
Exogone lourei
720
8920
3104
1600
5220
2560
5180
3432
Leptochelia dubia
9670
1798
44632
26372
810
1492
1760
2080
1 Data from Nyblade (1979) +0' Tide Level.
42
-------
comparison in recovery. For reasons previously discussed (p. 9), we
cannot be certain of the error probabilities for these species. For this
reason we simply present the mean numbers of individuals per square
meter. To preclude perhaps needless repetition of the same point, we
include only data from Experiment I, Sequim Bay, and Experiment III in
this section. A complete data list and Analyses of Variance will be
included in the final report.
On Table 17 are the densities of polychaetes from Sequim Bay trays,
Experiment I. The trend in the data is quite clear. With the notable
exception of E. gemmifera, the congener to Exogone lourei, the species
indicate reduced densities in oiled substrates as compared to controls.
Although for a few of these species the differences were computed to be
significant (alpha = 0.05), we do not select species here and attribute
significance. The central point is that the data do not contradict the
generalizations made about effects of oil generally from the data on
primary species in the preceding sections. It appears that the effect of
the oiling was to reduce the numbers of polychaetes on a rather broad
spectrum basis.
Data for the individual crustacean species is a little less clear-cut
(Table 18). Where larger numbers of individuals are involved, e.g.,
ostracods (undetermined species), Melita dentata, amphipods (undetermined
species), the trend of reduced numbers in oiled substrates seems clear.
The ambiguity which arises when lesser numbers of individuals are involved
is undoubtedly related to methodological sensitivity. Data on the primary
species does appear to fairly represent the response of crustaceans.
The gastropod mollusks, like the primary species, do not seem to
show a pattern dependent on oil treatment (Table 19). Among bivalves,
Clinocardium sp. was consistently more dense in control trays. From
analysis of variance, that difference was computed to be significant. We
do not attribute statistical significance to this indicated effect for
previously stated reasons. However, the cockles have some importance as
recreational clams; since the result seems to be contrary to the generali-
zation that the mollusk density is not affected by oiling, some further
investigation of the species is warranted.
Also shown in Table 19 are the densities in Experiment I for the
"Other Species" category. There are no highly abundant, highly sensitive
species indicated in these data for this group.
Comparable data sets for densities of nonprimary species after
15-months recovery are on Tables 20 through 22. The striking feature of
Table 20 relating to polychaete density is the large increase, relative
to 3-month recovery, not only in the number of polychaete species but
also in the estimated densities for individual species. Overall, numerical
superiority resides with the control substrates as compared to oiled, an
observation consistent with the data on primary species. In contrast to
the 3-month recovery data (Table 17) where a preponderance of oiled
substrates had indicated densities of zero, the oiled substrates in this
case do show an appreciable amount of recovery for most species. It is
interesting, although not necessarily significant, that the congener of
our suggested recovery indicator species, Exogone gemmifera, in this
43
-------
Table 17. Mean Density of Nonprimary Species (No./m2) of Polychaetes
in Experiment I Sequim Bay Example.*
SPECIES/FAMILY
DORVILLEIDAE
Dorvillea gracilis
GLYCERIDAE
Henri podus boreal is
GONIADIDAE
Glycinde polygnatha
NEPHTYIDAE
Nepthys sp.
ONUPHIDAE
Nothria elegans
PHYLLODOCIDAE
Phyllodoce (Anaitides) maculata
POLYNOIDAE
Harmothoe imbricate
Harmothoe sp.
SYLLIDAE
Exogone gemmifera
CAPITELLIDAE
Notomastus sp.
MALDANIDAE
Axiothella rubrocincta
SPIONIDAE
Boccardia sp.
Nerine cirratulus
Polydora californica
Polydora sp.
Prionospio sp.
RhynchospiQarenicola
Spio filicornis
Spionid undet.
STERNASPIDAE
Sternaspis fossor
TOTALS
CONTROL
COARSE
60
20
20
30
20
40
20
0
0
0
280
160
40
0
0
20
740
0
20
1470
FINE
20
60
0
60
20
0
20
20
0
0
20
20
0
40
20
0
500
20
0
820
OILED
COARSE
0
0
0
0
0
20
0
0
40
0
40
0
20
0
0
0
320
0
0
440
FINE
0
0
0
0
60
0
0
0
220
20
20
0
0
0
0
0
20
60
0
400
* This experiment included: Two sites (Protection Island; Sequim Bay); Two sediment
types at each site (Coarse; Fine); One tide level (MLLW); Two oiling conditions (Oiled;
Unoiled Control). 3-Month Recovery (Started 8/78; Completed and Sampled 11/78). For
convenient comparison to Experiment III (Table 20) only Sequim Bay data are shown.
Species density is rounded to nearest 10. Each mean based on 5 replicates with 7 cores
each replicate.
44
-------
Table 18. Mean Density of Nonprimary Species (No./m2) of Crustaceans
in Experiment I Sequim Bay Example.*
CONTROL OILED
SPECIES/GROUP COARSE FINE COARSE FINE
OSTRACODA
Ostracoda undet. 360 100 140 40
COPEPODA
Copepoda undet. 120 160 20 100
AMPHIPODA
Ampithoe sp. 0 20 20 0
Caprella undet. 0 20 00
Corophiurn sp. 00 40 0
Mellta dentata 340 0 60 0
Melita sp. 60 60 100 0
Orchestoidea sp. 20 0 00
Parallorchestes sp. 60 0 00
Paraphoxus sp. 20 20 20 0
Photis sp. 00 80 20
Pontharpinia sp. 0 20 00
Pontogeneia inermis 00 20 0
Pontogeneia sp. 20 0 40 0
Amphipoda undet. 60 140 60 40
DECAPODA
Crangon sp. 0 40 20 0
Heptacarpus paludicola 20 0 00
Pandalus platyceros 0 20 00
Spirontocaris prionota 0 20 00
Spirontocaris sp. 0 20 20 0
REPTANTIA
Pagurus hirsutiusculus 20 0 00
Pinnotheres sp. 0 0 20 0
TOTALS 1100 640 660 200
* This experiment included: Two sites (Protection Island; Sequim Bay);
Two sediment types at each site (Coarse; Fine); One tide level (MLLW);
Two oiling conditions (Oiled; Unoiled Control). 3-Month Recovery (Started
8/78; Completed and Sampled 11/78). For convenient comparison to
Experiment III (Table 21) only Sequim Bay data are shown. Species density
is rounded to nearest 10. Each mean based on 5 replicates with 7 cores
each replicate.
45
-------
Table 19. Mean Density of Nonprimary Species (No./m2) of "Others"
and Mollusks in Experiment I Sequim Bay Example.*
CONTROL OILED
SPECIES/GROUP COARSE FINE COARSEFINE
OTHER SPECIES
HYDROZOA
Sertularella sp. 60 0 40 0
Sertularia sp. 20 0 00
Hydrozoa undet. 20 20 00
NEMERTEA
Tubulanus sp. 40 20 00
Emplectonema gracile 00 0 20
Paranemertes peregrina 00 0 20
Nemertea undet. 80 40 20 40
MOLLUSKS
GASTROPODA
Alvania compact a 20 0 00
Littorina sp. 0 60 40 20
Margarites sp. 00 60 0
Solariella sp. 20 20 00
BIVALVIA
Clinocardium sp. 120 20 00
Psephidia lordi _20 __0 0 0
TOTALS 400 180 160 100
* This experiment included: Two sites (Protection Island; Sequim Bay);
Two sediment types at each site (Coarse; Fine); One tide level (MLLW);
Two oiling conditions (Oiled; Unoiled Control). 3-Month Recovery (Started
8/78; Completed and Sampled 11/78). For convenient comparison to
Experiment III (Table 22) only Sequim Bay data are shown. Species density
is rounded to nearest 10. Each mean based on 5 replicates with 7 cores
each replicate.
46
-------
Table 20. Mean Density of Nonprimary Species (No./m2) of Polychaetes
in Experiment III.*
CONTROL OILED
SPECIES/GROUP
DORVILLEIDAE
Dorvillea rudolphi
Protodorvillea gracilis
GLYCERIDAE
Glycera americana
Hemipodus boreal is
GONIADIDAE
Glycinde armigera
Goniada brunnea
COARSE
60
260
0
125
400
20
FINE
160
120
0
140
400
20
COARSE
80
60
20
120
20
0
FINE
40
40
40
120
20
0
LUMBRINEREIDAE
Lumbrinereis zonata 80 460 80 360
NERIDAE
Nereis vexillosa 220 160 100 80
ONUPHIDAE
Nothria eleqans 300 460 200 500
PHYLLODOCIDAE
Eteone lonqa 40 0 60 100
Eumida bTfoTiata 1580 1540 400 620
Phyllodoce (Anaitides) maculata 500 380 480 660
POLYNOIDAE
Halosynda brevisetosa 0 100 0 600
Harmothoe imbricata 280 200 260 80
SYLLIDAE
Exogone gemmifera 1540 2180 1500 2320
TrypanosyIll's gemnripara 40 0 0 0
Trypanosyllis ingens 00 0 200
TrypanosylTTs sp. 00 20 0
CIRRATULIDAE
Cirratulus cirratus 80 3160 20 280
Tharyx multifiliis 360 120 20 520
47
-------
Table 20. (Continued)
SPECIES/GROUP
CONTROL
COARSE
FINE
OILED
COARSE
FINE
MALDANIDAE
Axiothella rubrocincta
SABELLARIDAE
Sabellid undet.
SPIONIDAE
Rhynchospio arenicola
Spio filicornis
680
40
580
380
20
160
580
40
60
240
400
420
0
380
160
TERREBELLIDAE
Eupolymnia crescentis
Thelepus crispus
Terebellid undet.
POLYCHAETES UNDET.
TOTALS
0
600
0
0
7785
40
560
20
20
11380
0
980
0
0
5160
0
2700
0
0
10240
* This experiment included: One site (Sequim Bay); Two sediment types
(Coarse; Fine); One tide level (MLLW); Two oiling conditions (Oiled;
Unoiled Control). 15-Month Recovery (Started 8/78; Completed and
Sampled 11/79). Comparable data for 3-month recovery on Table 17.
Species density is rounded to nearest 10. Each mean based on 5
replicates with 7 cores each replicate.
48
-------
Table 21. Mean Density of Nonprimary Species (No./m2) of Crustaceans
in Experiment III.*
CONTROL OILED
SPECIES/GROUP COARSE FINE COARSE FINE
OSTRACODA
Ostracoda undet. 0 120 60 40
NEBALIACEA
Nebalia pugettensis 40 200 80 120
CUHACEA
Cornella vulgarIs 60 40 20 20
AMPHIPODA
Ampithoe lacertosa 0 100 210 420
Ampithoe sp. 20 200 20 0
Aoroides columbiae 0 20 00
Caprella laeviuscula 40 60 20 0
Caprella undet. 20 60 20 0
Ischyrocerus anguipes 680 580 100 20
Orchomene pacifica 0 40 00
Paraphoxus sp. 560 360 80 60
Pontogeneia inermis 20 60 0 20
Amphipod undet. 20 0 00
DECAPODA
Heptacarpus sitchensis 0 20 00
Heptacarpus taylori 0 520 220 120
REPTANTIA
Hemigrapsus nudus
Pinmxia occidental is
Pinnixia schnritti
Pinnixla tubicola
Telemessus cheiTagpnus
Upogebia pugettensis
TOTALS 1900 2980 1550 1240
This experiment included: One site (Sequim Bay); Two sediment types
(Coarse; Fine); One tide level (MLLW); Two oiling conditions (Oiled;
Unoiled Control). 15-Month Recovery (Started 8/78; Completed and
Sampled 11/79). Comparable data for 3-month recovery on Table 18.
Species density is rounded to nearest 10. Each mean based on 5
replicates yith 7 cores each replicate.
49
-------
Table 22. Mean Density of Nonprimary Species (No./m2) of "Others"
and Mollusks in Experiment III.*
CONTROL OILED
SPECIES/GROUP COARSE FINE COARSE FINE
OTHER SPECIES
HYDROZOA
Obelia sp. 40 40 40 0
Hydrozoa undet. 0 20 00
NEMERTEA
Baseodiscus sp. 20 0 20 0
Emplectonema gracile 40 60 20 80
Nemertea undet. 160 200 160 260
ECHINODERMATA
Holothuroid undet. 0 20 60 0
MOLLUSKS
GASTROPODA
Alyania compacta 240 120 0 0
Blttlum IpT 20 0 10 20
Littorina sitkana 20 0 20 0
BIVALVIA
Clinocardium nuttallli 20 60 00
Clinpcardium sp. 40 60 20 20
Macoma inflatula 100 40 20 20
Macoma sp. 20 0 00
MvTTp". 20 0 00
Mytilus edulis 20 0 00
Saxidomus nuttallil _20 _0 _0 _0
TOTALS 780 620 370 400
* This experiment included: One site (Sequim Bay); Two sediment types
(Coarse; Fine); One tide level (MLLW); Two oiling conditions (Oiled;
Unoiled Control). 15-Month Recovery (Started 8/78; Completed and
Sampled 11/79). Comparable data for 3-month recovery on Table 19.
Species density is rounded to nearest 10. Each mean based on 5
replicates with 7 cores each replicate.
50
-------
Independent experiment again shows slight numerical advantage in oiled
substrates.
The 15-month recovery densities for crustaceans (Table 20) excluding
primary species, do not show an increase in numbers of species over 3
months (Table 18) as did the polychaetes. Overall, numerical superiority
is in favor of unoiled substrates. The difference is mainly attributable
to differences in density in fine substrates. We do not see any strong
exceptions in the data to refute the finding that crustacean numbers are
reduced in oiled substrates compared to controls based on primary species.
As for the polychaetes, the effects of oiling seem to be a rather broad
spectrum on the recovery of crustaceans.
The 15-month recovery data for mollusks (Table 22) indicate a greater
number of species, especially in coarse control substrates than for
3-month recovery (Table 19). Relatively low density prevails throughout.
For the two species, Macoma inflatula and Clinocardium sp., the mean
differences due to oiling are interesting, because of the consistency in
pattern. For Clinocardium. this is a second independent experiment in
which the same trend was observed. There is likely a real effect on the
density of Clinocardium due to oiling. This experiment is the first in
which the genus Macoma has been observed. The data here reported are not
conclusive but are consistent with reports of oil sensitivity by Macoma
(Shaw et al., 1976). The overall data on mollusks suggest that quanti-
tative differences in density due to oiling may require substantially
longer recovery experiments.
The "Other Species" category, also shown in Table 22, does not
indicate a marked increase in number of species from the 3-month recovery
data (Table 19). Like the earlier experiment, there do not seem to be
any highly abundant or oil-sensitive species in this group.
Trophic Mode of Primary and Selected Other Species.
In experiments of the sort reported here we are, in effect, measuring
the suitability of the substrate habitat. A central assumption is that
the organisms for colonization are available in "normal" numbers for all
of the experimental substrates, control or oiled, on an equal basis. In
cases where there are demonstrable differences in recovery, data from the
experiment itself can tell us nothing about why organisms are at a lower
density in oiled versus unoiled substrates of like character or, indeed,
the fate of the "absent" organisms. It is a certainty that organisms
must eat to grow and survive. Therefore, trophic modes may play a central
role in determining substrate suitability. This section deals with the
trophic modes of primary species and a few examples from the nonprimary
group.
We are fortunate in having available a classification of feeding
level for each of the primary species from related MESA studies (Simenstad
et al., 1979). These data, excerpted from Simenstad et al. (1979) are
displayed in Table 23. There is one very clear trend from the data. No
organism classified as a suspension feeder was deemed to have significantly
reduced density in oiled substrates in our experiments. To the contrary,
one species, Mysella tumida. was deemed to have greater density in oiled
substrates. It is interesting that this finding includes the amphipod
51
-------
Table 23. Trophic Levels of Primary Species in Oil Recovery Experiments.
GROUPS/SPECIES
TROPHIC LEVELS
MOLLUSKS
Mysella tumida
Transennella tantilla
Lacuna sp.
CRUSTACEANS
Corophi urn ascherusicum
Phot is brevipes^
Leptochelia dubia
POLYCHAETES
Armandia brevis
Exogone lourei
Platynerels bicanaliculata
Polydora social is
Capitella capitata
Ophiodromus pugettensis
Suspension feeder on phytoplankton
Suspension feeder on phytoplankton
Herbivore on microalgae
Detritivore
Suspension feeder on detritus;
detritivore
Detritivore; carnivore on small benthic
Detritivore
Detritivore
Herbivore; macroalgae
Detritivore; carnivore on zooplankton
Detritivore
Carnivore on annelids, tanaids and
cumaceans
Data excerpted from Simenstad et al.} 1979.
52
-------
crustacean, Photis brevipes, since the amphipods as a group are usually
considered to be highly sensitive to the effects of oiling. With the
exception of Capitella capitata, each of the species classified as a
detritivore was found to have significantly reduced density in oiled
substrates in at least one of the experiments, and sometimes two (Corophium
ascherusicum, Experiments II and III; Armandia brevist Experiments I and
III). The only one of the primary species classified as a carnivore,
Ophiodromus pugettensis, was not significantly reduced in oiled substrates
in any experiment. The results on herbivores indicate some relation to
feeding mode also. Lacuna sp., an herbivore on microalgae, had consistent,
but nonsignificant, reduced densities in Experiments I and III, and
inconsistent occurrence in Experiment II. The polychaete, Platynereis
bicanaliculata, an herbivore on macroalgae, had significantly reduced
density in Experiments I and III.
The foregoing data can be considered as "hard" data with clear-cut
results. It is tempting to delve into the data on nonprimary species for
further substantiation or refutation, but the cautionary note must be made
that we do not have verifiable probability statements on the errors for
hypothesis tests of effects indicated by mean differences in density.
If we examine the mean data on polychaetes (Table 20), 13 species
are classified as carnivores. For 3 of these, there appears to be dis-
tinctly reduced density in oiled substrates as compared to controls
(Protodorvillea gracilis. Eumida bifoliata, Glycinde armigera). Only 3
species are classified as herbivores (Lumbrinereis zonata, Nereis vexi11osa
and Nothria elegans). For each of these species the mean densities are
slightly higher (H. vexi11osa distinctly higher) in unoiled as compared
to oiled substrates. Nine of the species are classified as detritivores.
Of these, three have much higher mean density in controls (Cirratulus
cirratus, Axiothella rubrocincta, Spio filicornis). Two species had
distinctly higher density in oiled substrates (Rhynchospio arenicola,
Thelepus crispus), and 4 species were inconsistent. We conclude that the
data on primary species reflects the overall relationship of feeding mode
and oil effects reasonably well.
The data on crustaceans (Table 21) is much less clear on the effects
from oil generally as earlier stated. However, there are distinctly
higher densities in controls for the detritivores, Cumella vulgaris and
Paraphoxus sp. There are also distinctly more Ischyrocerus anguipes in
controls. The group to which this latter species belongs receives various
feeding level designations by Simenstad et al. (1979), i.e., suspension
feeders on detritus, herbivores and detritivores. These data then,
however insufficient, are consistent with the generalizations on feeding
mode derived from primary species.
The data on mollusks and "other species" are even more severely
limited. It is of interest that the apparent effect on Hacoma inflatula,
a deposit feeder, fits a feeding mode picture quite well while the apparent
effect on Clinocardium ,for which we have more confidence, is in con-
tradiction to the general picture. The latter is a suspension feeder on
phytoplankton.
53
-------
Relevance of Findings to Upper Trophic Levels.
An examination of species and functional groups with major importance
to upper trophic levels (Table 4, Simenstad et al., 1979) reveals the
presence of 7 of the 13 primary species in this study. It is, therefore,
pertinent to attempt to evaluate the relevance of effects on recovery, as
elicited here, to effects on upper trophic levels, principally fishes.
Criteria for deeming major trophic importance, used by Simenstad et al.
(1979), included that the designated species or functional group: (1)
provides the majority of energy sources for consumer organisms at some
time; (2) provides important conversion or transfer of organic matter to
trophic levels where it is available to higher level consumers, or (3)
holds "keystone" roles in structuring the composition of the community
and the directions and rates of food web energy flow.
The following primary species from the present study are on the list
(Simenstad et al., 1979) so defined: Lacuna sp., Platynereis bicanaliculata,
Transennella tantilla, Photis brevipes. Ophiodromus pugettensis. Leptochelia
dubia, and a member of a family defined, Spionidae, Polydora social is.
Of these, significant effects on density were indicated for Platynereis
bicanaliculata, Leptochelia dubia, and Polydora social is. These parti-
cular species are indeed important ones in the present study since jointly
they contributed 62,980 individuals per m2 in coarse sediment controls
(15-month recovery) out of a total of 119,884 (53%).
Based on these data alone, one would expect impact on the appro-
priate upper trophic levels for at least 15 months following an oil
contamination of the magnitude used in this study.
Although one could pose the specific counter arguments: (1) that
reductions in these particularly high density species will be replaced by
other species with similar proclivities toward high density; or (2) these
are small species and in spite of high density comprise small biomass, we
see no evidence in any of the experimental data to support such arguments.
Rather, we see a broad spectrum reduction in density across polychaetes,
crustaceans, and even mollusks with detritiyorous and herbivorous feeding
modes. In the case of small versus larger individuals, the data for
Nereis vexillosa (Table 20), a rather large polychaete, seems to indicate
that biomass as well as density may be substantially reduced.
Based on an extensive amount of stomach analysis and synthesis of
published information, Simenstad et al. (1979) have constructed composite
food webs for several shallow sublittoral habitats of the northern Puget
Sound region. It is our judgment that the Sequim Bay experimental site
in the present study is best represented by the composite constructed for
protected sand/eelgrass shallow sublittoral habitat. In this, as in all
the composite food webs constructed by Simenstad et al. (1979), with the
exception of the neritic web, detritus forms the base of the web. We
have included (Figure 2) a copy of the composite web for this habitat.
The trophic groups containing primary species for which significant
effects due to oiling were demonstrated are indicated in Figure 2 by
solid fill boxes. Trophic groups containing nonprimary species for which
the mean density data suggest a detrimental effect on density are indicated
in Figure 2 by boxes with cross-hatch marking. The reader should keep in
54
-------
01
CH
Significant effects due to oiling in Recovery Experiments.
Effects due to oiling indicated by mean differences between control and
oiled substrates.
PROTECTED SANO-GBAVEL/EELGRASS SHALLOW SUBLITTORAL FOOD WEB
""""^/SSJ-J --'-- ^7/-
• ri»m-.Ki' I , / f I
r+ js n»wt
•/ f- VOILUICI
' ~"--— I
FIGURE 2.
COMPOSITE FOOD WEB CHARACTERISTIC OF PROTECTED SAND/
EELGRASS, SHALLOW SUBLITTORAL HABITATS IN NORTHERN
PUGET SOUND AND THE STRAIT OF JUAN DE FUCA.
-------
mind that other groups in the food web may contain species susceptible to
oiling which were not sampled in the present study.
The clearest and most widespread effects on density from the oiling
in the present study were on detritivorus tanaid crustaceans, amphipod
crustaceans, and polychaete annelids. One would expect reductions in
density for the constituent species to result in a shortage of food and
reduced growth for bottom feeding fishes principally using these groups.
Contributions to the food web are expressed in terms of an index of
relative importance (IRI, Simenstad et al., 1974). From the composite
web, the gammarid amphipods contribute, at least on a tertiary basis, to
a wide variety of fishes (25-49% of IRI). They are a primary food source
for the crescent gunnel (75-100% of IRI). The principal users of the
polychaete annelids are among the flatfishes and include the English
sole, rock sole, sand sole, and C-0 sole. For the C-0 sole they are
indicated as a secondary source (50-75% of total IRI). The tanaids are
principally tertiary contributors (25-49% of IRI) to snake prickleback
and shiner perch and contribute incidentally to carnivorous polychaetes.
There is now some evidence that oil spillage may, in fact, result in
reduced growth for young flat fish in a real-world situation (Desaunay,
1980). This study, conducted following spillage of the AMOCO CADIZ,
indicated a marked decrease in growth for a young year class of a bottom-
feeding flatfish.
Severity of Treatment in Infauna Experiments.
Characterization of the oil treatment was in terms of large numbers
of samples analyzed by infrared spectrophotometry (IR) and capillary gas
chromatography (CGC) (see Table 1 for schedule).
The oil treatments applied in Experiments I and III were equivalent.
A target concentration of 2000 ppm total oil was sought. Calculated
total oil (IR) in sediments was determined by substracting mean concen-
trations measured in control substrates from mean concentrations in
oil-treated substrates in the appropriate treatment categories. Table 24
shows total oil data for substrates sampled prior to placement in Experi-
ments I and III (preliminary), Experiment I (3-month fall recovery), and
Experiment III (15-month fall recovery). Total oil concentrations slightly
in excess of 1700 ppm in both fine and coarse sediments were reasonably
close to the target concentration. For Experiment I (3-months) initial
concentrations were reduced by 34 and 35% for fine and coarse sediment,
respectively, leaving total oil concentrations of slightly in excess of
1100 ppm at the termination of this experiment. By the end of Experiment
III (15-months), total oil concentrations were reduced by 87 and 85% for
fine and coarse sediments, respectively, leaving total oil concentrations
of between 200 and 300 ppm at the termination of this experiment. Thus,
there are no differences indicated for retention of oil by fine and
coarse sediments using the IR methods. The marked increases in control
substrate IR of CC14 extractable organics by the end of 15 months are of
interest. Initially, control substrates gave IR readings of 1 or 2% of
oil-treated substrates. By the end of 3 months the controls gave readings
of 2 and 3% (primarily derived from reduction in oil-treated concentrations).
At 15 months, the controls represented 35 and 36% of oil-treated concen-
trations. This increase is due primarily to increases in the controls
themselves. The increase in control IR measurements probably relates
56
-------
Table 24. Comparisons of Total Oil (IR) in Experiments I and III,
at Sequim Bay.1
CONCENTRATION (ppm)
SEDIMENT/TREATMENT
COARSE
Treated
Control
Total Oil
FINE
Treated
Control
Total Oil
PRELIMINARY
1758 ± 252
21 ± 8
1737
1794 ± 66
37 ± 10
1757
3-Month
EXPERIMENT I
1154 ± 260
32 ± 15
1122
1183 ± 153
24 ± 7
1159
15-Month
EXPERIMENT III
391 ± 43
136 ± 52
258
366 ± 44
132 ± 54
234
1 Each mean presented based on 9 core analyses. Standard deviations
shown are between trays (n = 3 per mean).
57
-------
to an increase in organic matter over the longer time period. The fact
that the relative variability in control IR data is not greatly increased
at 15-months versus 3-months (Table 24) indicates to us that cross con-
tamination is not a likely source for these increased readings. Sub-
sampling analyses and Protection Island data have been previously reported.
The total oil concentrations at Protection Island did not significantly
(p = 0.05) differ from the Sequim Bay data.
A time series plot of total oil IR (controls subtracted) is on
Figure 3. A loss of oil from substrates from the initial concentration
to the 3-month concentration appears to have taken place early (perhaps
in the first month) and, thereafter, the loss of total oil appears to
follow a linear course.
The comparable data for Experiment II, spring (3-month recovery in
coarse substrates) are shown in Table 25. In this case the target concen-
tration of 1000 ppm was achieved. Three month losses of total oil were
43 and 53% for Mean Lower Low Water and minus two feet tide levels,
respectively. From these data the losses at both tide levels appear a
little more rapid than for the comparable fall experiment. Also, the
lower tide level appears to have lost oil a little more quickly than the
higher. For the tide level differences indicated, mean differences are
not significant. It is inappropriate to statistically compare the apparent
seasonal difference.
The time series course of total oil concentration for Experiment II
is shown in Figure 4. The intermediate data, as did the final mean data,
indicate a more rapid loss of oil at the lower tide level. From the time
pattern in total hydrocarbon concentrations in Experiments I and III
(Figure 3), one can derive a rate of loss over the 15-month period. The
loss of total hydrocarbons was more rapid at the beginning of the experiment,
averaging 12% per month of the initial concentration. For the entire
15-month period, the average was 5.6% per month. For the period between
3 and 15 months, the indicated loss appears to be relatively stable at a
rate of 4.2% per month. In terms of concentration data, this amounts to
roughly 71 ppm per month. At that rate of loss, sediments might be
expected to reach background levels in terms of IR measurements in a
further 3.5 months, i.e., a total of 18.5 months. Obviously, because of
a greater lack of homogeneity in substrate in a real-world situation, and
because oil would be distributed over a wider area, the data are not
directly translatable. They do, however, provide a conservative estimate
and give a framework for assessing the time to full recovery for environ-
mental conditions similar to our experiment.
Means of summed analyzed saturate and aromatic compounds in Experi-
ments I and III are shown in Table 26. Losses of both of these compound
groups were somewhat more rapid than were losses of total hydrocarbons.
Again, we took the approach of substracting out "background" as indicated
by concentrations in unoiled control units. Initial concentrations of
the analyzed saturate and aromatic compounds were somewhat higher in
coarse substrates as compared to fine. Reductions of 81 and 86% are seen
in coarse substrate in 3 months. At 15 months the comparable losses are
95 and 98% of initial concentrations, or nearly complete. In the case of
fine substrates, losses were slower during the first three months, amounting
58
-------
1900
1700
1500
10
1300
1100
900
700
500
A
A
A
A
A
300
100
A
A
S 0 N D
1978
M
A
M
J J
•1979
A
0
FIGURE 3.
CONCENTRATIONS OF TOTAL OIL IR (CCli* EXTRACTABLE ORGANICS IN OIL
TREATED SUBSTRATES MINUS CCli* EXTRACTABLE ORGANICS IN CONTROL
SUBSTRATES) .
-------
Table 25. Comparisons of Total Oil (IR) in Experiment II at
Sequim Bay.1
TREATMENT
Treated
Control
Total Oil
CONCENTRATION (ppm)
PRELIMINARY MLLW MINUS 2'
1069 ± 144 612 ± 99 503 ± 16
21 ± 82 17 ± 6 15 ± 2
1048 595 488
1 Each mean presented based on 9 core analyses. Standard deviations
shown are between trays (n = 3 per mean).
2 These control data are from Experiment I.
60
-------
1100
1000
900
800
700
600
500
400
300
200
100
O--O -2 Below MLLW
MLLW
M
J
-2979-
I IGURE4, CONCENTRATIONS OP TOTAL OIL IR (CCl^ EXTRACTABLE
ORGANICS IN OIL TREATED SUBSTRATES MINUS CCltt
EXTRACTABLE ORGANICS IN CONTROL SUBSTRATES),
EXPERIMENT II.
61
-------
Table 26. Means of Summed Analyzed Saturate and Aromatic Compounds (Capillary GC) in Experiments I and III
at Sequim Bay.
en
MEAN CONCENTRATIONS (pfl/g)
SEDIMENT/TREATMENT
Coarse Sediment
Oil Treated
Control
Total Due to Oiling
Fine Sediment
Oil Treated
Control
Total Due to Oiling
PRELIMINARY
SATURATES
157
0.58
156.52
121.6
0.53
121.07
AROMATICS
19.5
0.17
19.33
17.6
0.14
17.46
3-MONTH (I)
SATURATES
30.7
0.73
29.97
52.1
0.43
51.57
AROMATICS
2.8
0.09
2.72
5.1
0.08
5.02
15-MONTH (III)
SATURATES
8.1
0.42
7.68
3.3
1.9
1.40
AROMATICS
0.4
0.08
0.32
0.1
0.1
0.0
Each mean presented based on analysis of one core from each of three experimental trays receiving the
treatment indicated. Analysis of variance and measurement sensitivity in Vanderhorst et al. (1979).
-------
to 57 and 71% for saturates and aromatics, respectively. By 15 months,
the losses were virtually complete in fine substrates (98% for saturates
and 100% for aromatics).
In Experiment II, the spring-summer tide level experiment (Table 27)
commensurate with the reduced target and total oil concentrations, the
saturates were a little less than one-half of that measured in preliminary
data for Experiments I and III. Inexplicably, the analyzed aromatic
concentrations were considerably lower than expected. In this experiment,
aromatic concentration was about 4% of saturate concentration as compared
to 12% (Table 26) of saturates in Experiments I and III. The losses of
measured saturate compounds for this spring-summer, 3-month period, are
comparable to rates of loss indicated in the fall experiment (I), amounting
to 83% of initial concentration at MLLW and 87% at the -2' tide level.
Measured aromatic compounds were at background concentrations in 3 months
at each of the tide levels.
RECOVERY ON HARD SUBSTRATES
There is a clear difference in status between the infaunal studies
just reported and the recovery on hard substrates reported here. The
former studies are complete and conclusions are warranted. The hard
substrate studies are under way as of this writing. The points addressed
in this section relate to Task F, a testing of attachment method for hard
substrates in the exposed rocky intertidal, and Task G, a series of
one-month duration, independent experiments on the effects of oil on
recovery of communities on hard substrates. Task H, commercial clam bed
recovery, and effects of key species removal and oil on recovery will be
reported in a subsequent report.
For Task F, polyvinylchloride (PVC), asbestos, and concrete plates
were attached to conglomerate rock at Observatory Point on the Strait of
Juan de Fuca to evaluate plate survival and colonizing characteristics.
Attachments were made in the spring following the severest winter storm
activity with the intent of following gross colonization of the differing
materials. Triplicate plates of each type were attached to individual
PVC backing (60 cm x 30 cm, .25" thickness). The backing was bolted to
individual wooden frames. The wooden frames were attached to a northern
exposed face, approximately 45° inclination, of intertidal rock approximately
2' above MLLW.
Types of attachment within the plate systems included: (1) wooden
frames attached to intertidal rock using masonry nails (2.0") driven with
a powder actuated impact gun; (2) PVC backing bolted to wooden frames;
(3) concrete plates attached to PVC backing by gluing PVC strips (.25"
thick, 5" wide) on edge to form a 30 cm x 30 cm border on the PVC backing.
These strips were glued with PVC cement. Holes (.25") were drilled in
the border strips to form "key" ways, and concrete was poured directly
into the bordered area. (4) Asbestos plates, cut (30 cm x 30 cm) from
^.25" thick asbestos sheet were glued directly to the PVC backing with
contact cement.
63
-------
Table 27. Means of Summed Analyzed Saturate and Aromatic Compounds (Capillary GC) in Experiment II at
Sequim Bay.1
MEAN CONCENTRATIONS (ug/g)
SEDIMENT/TREATMENT
Coarse Sediment
Oil Treated
Control
Total Due to Oiling
PRELIMINARY
SATURATES
70.5
0.582
69.92
AROMATICS
2.6
0.172
2.43
MLLW
SATURATES
12.4
0.41
11.99
AROMATICS
0.11
0.10
0.01
-2" BELOW MLLW
SATURATES
9.2
0.38
8.82
AROMATICS
0.12
0.11
0.01
1 Each mean presented based on analysis of one core from each of three experimental trays receiving the
treatment indicated.
2 These control values from Experiment I preliminary analysis.
-------
All attachments had failed at the first observation, with the
exception of one wooden frame with bolted on PVC backing.
Species Composition and Density in Monthly Experiments
Monthly experiments for Task G were completed through May 1980.
Data from the first four monthly experiments (October-January) concerning
recovery of epifauna on hard substrates are presented. The counts of
organisms at daily intervals during the first five days of each month
resulted only in random occurrences of "crawl-on" type organisms.
Enumeration of organisms after 30 days of colonization resulted in some
specific patterns of abundance within the major taxomonic categories.
The same major taxonomic categories dominated the hard substrates as
were previously seen in the infauna experiments (polychaetes, crustaceans,
mollusks). Data for mean number of individual polychaetes for the first
four monthly experiments are in Table 28. Polychaetes occurred principally
in the October experiment and at the Mean Lower Low Water tide level. On
the average, slightly more polychaetes were observed on control substrates
as compared to oiled, but the standard deviations are quite large and no
significance can be attributed to the difference. Without exception, the
polychaete species were ones observed commonly in the infauna experiments.
Two of the seven species were ones which were designated as primary
species in the infaunal studies. One of these, Platynereis bicanaliculata,
was the predominate polychaete observed in this hard substrate experiment,
and dominates the density pattern shown in Table 28.
Table 29 shows the mean numbers of individual crustaceans found in
the first four experiments. Both the numbers of individuals and species
(10) were higher than for polychaetes. While the highest numbers of
individuals follow the same patterns as observed for polychaetes (principal
occurrence during October and at Mean Lower Low Water), there were measurable
densities at the higher tide level and during the winter months. These
densities relate to the contribution of one species, the isopod, Exosphaeroma
sp. and may account for the apparently anomalous result of higher densities
on control substrates at Mean Lower Low Water, and lower densities on
control substrates at the +21 tide level.
Density of individual mollusks is shown in Table 30. The mollusks
were represented by 7 species, perhaps a preliminary indication that this
group will have more importance here than was seen in the infauna. The
individual numbers, however, are quite low throughout the experiments and
the brick to brick variation is high.
These preliminary data indicate to us that even the relatively high
replication used (15 substrate units per treatment category) is not
sufficient to allow us to distinguish treatment effects of the magnitude
which may be indicated by mean differences between oiled and unoiled
substrates. The amount of variation seen is high but somewhat less than
that reported for baseline studies at rocky sites (Nyblade, 1979). The
general patterns of distribution in densities are consistent with what is
known about annual recruitment patterns and tide height distribution for
the involved species. It may be that the spring and early summer experiments
will prove more useful in distinguishing effects from the oiling, if
these exist.
65
-------
Table 28. Mean Numbers of Individual Polychaetes in Hard Substrate Oil Recovery Experiments.
(One-month exposure periods, oil administered at the start of each monthly experiment)
MEAN NUMBERS OF
EXPERIMENTAL CONDITIONS October
MEAN (S
Mean
Plus
.D.)
INDIVIDUALS/BRICK (N = 15 BRICKS)
November
MEAN (S.D.)
December January
MEAN (S.D.) MEAN (S.D.)
Lower Low Water
Oiled
Control
2' above MLLW
Oiled
Control
Species Included:
5.
7.
0.
0.
Platynereis
8
0
1
9
(5
(6
(0
(1
bicanal
Nothria elegans;
.1)
.3)
.3)
-8)
iculata
Halosynda
0.
0.
0.
0.
5
5
3
2
; Harmothoe
brevisetosa
(0.8)
(0.5)
(0.5)
(0.4)
imbricata
; Family
0.0
0.0
0.1
0.1
; Thelepus
Spionidae.
(0.0) 0.0
(0.0) 0.1
(0.3) 0.0
(0.3) 0.0
sp. ; Armandia brevis ;
(0.0)
(0.3)
(0.0)
(0.0)
-------
er»
Table 29. Mean Numbers of Individual Crustaceans in Hard Substrate Oil Recovery Experiments.
(One-month exposure periods, oil administered at the start of each monthly experiir
monthly experiment).
EXPERIMENTAL CONDITIONS
ME/I
MEAN NUMBERS OF INDIVIDUALS/BRICK (N
October
iN (S.D.)
November
MEAN (S.D.)
= 15 BRICKS)
December January
MEAN (S.D.) MEAN (S.D.)
Mean Lower Low Water
Oiled
Control
Plus 2' above MLLW
Oiled
Control
Species Included:
58.
86.
20.
6.
Exosphaeroma
ochotensis;
6 (38.0)
7 (75.0)
2 (13.7)
7 ( 5.5)
sp. ; Melita sp. ;
34.1 (21.6)
21.9 (22.4)
27.3 (18.6)
10.8 ( 9.64)
Corophium sp. ; Ampithoe
Caprella laeviuscula; Leptochelia dubia;
8.3
9.9
4.9
4.3
sp.;
(3.6) 2.8
(4.9) 1.1
(3.0) 0.1
(3.0) 0.5
Jassa sp. ; Paral lorchestes
(2.2)
(1.1)
(0.3)
(0.6)
Cancer sp. ; Pinnixia sp.
-------
en
CO
Table 30. Mean numbers of individual molluscs in hard substrate oil recovery experiments.
(One-month exposure periods, oil administered at the start of each monthly experiment).
EXPERIMENTAL CONDITIONS
Mean Lower Low Water
Oiled
Control
Plus 2' above MLLW
Oiled
Control
Species Included: Lacuna
tantil
MEAN NUMBERS OF INDIVIDUALS/BRICK (N = 15 BRICKS)
October
MEAN (S.D.)
0.4 (0.7)
0.9 (1.2)
0.7 (0.9)
0.5 (0.5)
November
MEAN (S.D.)
0.3 (0.5)
0.1 (0.3)
0.3 (0.5)
0.3 (0.6)
sp. ; Clinocardium nuttallii; Mytilus edulis;
1 a; Protothaca stami nea ;
Odostomia sp.
December
MEAN (S.D.)
0.0 (0.0)
0.0 (0.0)
0.3 (0.6)
0.3 (0.5)
Cooperella subdiaphana;
January
MEAN (S.D.)
0.0
0.0
0.1
0.0
(0.0)
(0.0)
(0.3)
(0.0)
Transennella
-------
Severity of Treatment on Hard Substrates.
The conventional expression of treatment severity in terms of weight/
weight or volume/weight ratios commonly used for sediment or water exposures
has little meaning in regard to hard substrates. The methods used in
this study result in measurements of total hydrocarbon weight on a per
brick basis or, in the case of top surface extraction, on surface area
basis. The data for whole brick extractions are on Table 31. Since the
extraction methods differed during the first two monthly experiments,
overall comparisons cannot be directly made. For the first experiment
(October), preliminary samples taken immediately post treatment were
appreciably higher in the total amount of oil per brick than were 30-day
MLLW samples. The indicated loss of oil during this period is 86%. For
the November and January experiments, however, the total amount of oil
present after 30 days represented a much higher percentage of the initial
concentration with 30-day reductions being 16 to 55%. Although variations
related to method and brick-to-brick variability in treatment are high
enough to preclude conclusions regarding the effect of tide level, or in
fact, 5-day versus 30-day losses of oil, these data, nevertheless, indicate
that the whole brick retains substantial amounts of oil for the entire
30-day periods of these experiments. Recovery of known applied amounts
of oil to bricks indicated that the whole brick extraction method recovered
66% of applied oil.
As part of attempts to make the analyzed oil relate to surface
exposure conditions, top surface extractions were used in addition to the
whole brick extractions in the November, December, and January experiments,
as well as those which have followed. The dsta on these extractions are
on Table 32. These data indicate much greater percentage losses for the
30-day period, ranging from 76 to 96% of the initial concentrations. The
5-day data from December and January, taken after extraction procedures
were standardized, indicate that most of this loss probably takes place
during the first 5 days. These data have the added advantage that they
may be transformed into surface area estimates of oil coverage by multi-
plication with an appropriate factor.
The two types of data presented here give diametrically opposed
views of "best" approaches to study recovery in the rocky intertidal
zone. In natural situations, if oil is retained in pores or cracks for a
significant period of time, as would seem to be indicated by these early
data on whole brick extractions, then future studies had best emphasize
highly controlled treatment and development of standardized extraction
and analytical methods. If, on the other hand, the very short retention
times indicated by the top surface extractions are typical, a greater
emphasis needs to be placed on understanding the natural recovery rates.
This is an extremely important distinction since the feasibility of the
former types of studies appears to be quite limited with presently available
methods, while studies of the latter type are not only quite feasible but
well under way in our region. It is anticipated that the studies reported
here will aid in making that distinction.
69
-------
o
Five- Day
Table 31. Mean Total CCl4-Extractable Organics (Grams/Brick) for Whole Brick Extractions in Recovery
Experiments.
TOTAL HYDROCARBONS (GRAMS/BRICK)
EXPERIMENTAL CONDITIONS October November December January
MEAN (S.D.) Method1 MEAN (S.D.) Method MEAN (S.D.) Method MEAN (S.D.) Method
Preliminary
Post Treatment 3.47 (1.24) W 4.06 (1.19) W 3.09 (1.02) D 2.91 (0.90) D
MUW - 1.76 (0.57) W 1.97 (1.07) D 2.86 (1.63) D
+2' - 4.80 (3.38) M 3.64 (2.00) D 2.01 (1.15 D
Thirty-Day
MLLW 0.49 (0.16) W 3.41 (1.56) D 1.92 (0.83) D
+2' - 2.43 (1.17) D 1.32 (0.51) D
1 Methods differed in some experimental conditions; symbols indicate: W = extraction using "wet" brick
directly from sea water; D = extraction from dry brick, air-dried 48 hours before extraction;
M = mixture of methods, part wet-part dry, means indicate total amount of material extracted (bricks/mean).
-------
Table 32. Total Hydrocarbons (Grams/Brick) Extracted From Top Surface of'Hard Substrates.1
EXPERIMENTAL CONDITIONS
Preliminary
Post Treatment
Five-Day
MLLW
+2'
Thirty- Day
TOTAL
November
MEAN (S.D.)
0.82
0.49
0.89
(0.33)
(0.10)
(0.72)
HYDROCARBONS (GRAMS/BRICK)
December
MEAN (S.D.)
1.04
0.07
0.07
(0.47)
(0.05)
(0.03)
MEAN
1.60
0.08
0.32
January
(S.D.)
(0.51)
(0.05)
(0.60)
MLLW 0.03 (0.02) - 0.39 (0.18)
+2' 0.07 (0.03) - 0.32 (0.12)
1 All extractions from top surface made using "wet" bricks (N = 5 bricks/mean).
-------
REFERENCES
Clark, R.C., Jr., J.S. Finley, B.C. Patten, D.F. Stefani, and E.E.
De Nike. 1973. Interagency investigations of a persistent oil
spill on the Washington coast. Proc. J. Conf. Prevent, and Control
Oil Spills. (EPA, API, USCG). Washington, D.C., March 13-15, 1973,
pp. 793-808.
Kozloff, E.N. 1974. Keys to the Marine Invertebrates of Puget Sound,
the San Juan Archipelago, and Adjacent Regions. Univ. Washington
Press (Seattle and London) 226 pp.
Michael, A.D., C.R. Van Raalte, and L.S. Brown. 1975. Long-term effects
of an oil spill at West Falmouth, Massachusetts. Proc. J. Conf.
Prevent, and Control Oil Pollution. (EPA, API, USCG). San Francisco,
Calif. March 25-27, 1975, pp. 573-582.
Nyblade, C.F. 1978. The intertidal and shallow subtidal benthos of the
Strait of Juan de Fuca, Spring 1976-Winter 1977. NOAA Tech. Memo.
ERL MESA-26. Marine Ecosystems Analysis Program, Boulder, Colo., 156
pp.
Nyblade, C.F. 1979. The Strait of Juan de Fuca Intertidal and subtidal
benthos. DOC/EPA Interagency Engery/Environment R&D Program Report
(EPA-600/7-79-213). Nat. Tech. Info. Serv., Washington, D.C. ,
129 pp.
Shaw, D.G., A.S. Paul, L.M. Cheek, and H.M. Feder. 1976. Macoma
balthica: An indicator of oil pollution. Mar. Poll. Bull. 7(2):29-31.
Simenstad, C.A., B.S. Miller, C.F. Nyblade, K. Thornburg, and L.J.
Bledsoe. 1979. Food web relationships of Northern Puget Sound and
the Strait of Juan de Fuca. DOC/EPA Interagency Energy/Environment
R&D Program Report. (EPA-60018-79-259) Nat. Tech. Info. Serv.,
Washington, D.C., 335 pp.
Simenstad, C.A., W.J. Kinney, and B.S. Miller. 1980. Epibenthic
zooplankton assemblages of selected sites along the Strait of Juan
de Fuca. NOAA Tech. Memo MESA-46. Marine Ecosystems Analysis
Program, Boulder, Colo., 73 pp.
Smith, G.F. and H. Webber. 1978. A biological sampling program of
intertidal habitats of Northern Puget Sound. Washington State
Department of Ecology Baseline Study Program. Olympia, Wash.,
311 pp.
Smith, R.I. and J.T. Carl ton (eds.). 1975. Light's Manual: Intertidal
Invertebrates of the Central California Coast (3rd EdTT!Univ.
Calif. Press (Berkeley, Los Angeles, London) 716 pp.
72
-------
Straughan, D. 1978. Biological studies of the Metula Oil Spill. Proc.
Conf. on Assessment of Ecol. Impacts of Oil Spills. (AIBS). Keystone,
Colo., June 14-17, 1978, pp. 364-377.
Vanderhorst, J.R., J.W. Anderson, P. Wilkinson, and D. L. Woodruff.
1978. Estimation of effects from oil on intertidal populations:
experimental perturbations versus natural variation. Proc. Conf. on
Assessment of Ecol. Impacts of Oil Spills. (AIBS). Keystone, Colo.,
June 14-17, 1978, pp. 807-820.
Vanderhorst, J.R., J.W. Blaylock, and P. Wilkinson. 1979. Research to
investigate effects from Prudhoe Bay crude oil on intertidal infauna
of the Strait of Juan de Fuca. NOAA Tech. Memo. ERL MESA-45.
Marine Ecosystems Analysis Program, Boulder, Colo., 39 pp.
Webber, H.H. 1979. The intertidal and shallow subtidal benthos of the
west coast of Whidbey Island, Spring 1977-Winter 1978. NOAA Tech.
Memo. ERL MESA-37. Marine Ecosystems Analysis Program, Boulder,
Colo., 108 pp.
73
------- |