- Email: [email protected]
Responses of arctic marine crustaceans to crude oil and oil-tainted food
RESPONSES OF ARCTIC MARINE CRUSTACEANS TO CRUDE OIL AND OIL-TAINTED FOOD
J. A. PERCY
Arctic Biological Station, Fisheries and Marine Service, Environment Canada, P.O. Box 400, Ste. Anne de Bellevue, P.Q., Canada ABSTRACT
The responses of several arctic marine crustaceans to oil masses and oil-tainted food have been investigated. None of the species were attracted to crude oil. Amphipods tended to avoid oil masses; however, the response was significantly diminished if the oil was weathered or i f the animals were pre-exposed to light crude oil emulsions. Untainted food was preferentially selected over oil-tainted food. In contrast, an isopod was generally neutral to the presence o f oil masses and consumed oil-tainted food as readily as untainted material.
INTRODUCTION
The potential for a major oil pollution incident in the Arctic has increased dramatically over the past several years, leading Dunbar (1971) to suggest that oil pollution is the most serious and immediate threat facing the arctic ecosystem. This is particularly true for the arctic marine ecosystem now that emphasis is rapidly shifting to offshore exploratory drilling. Concern for the preservation of this largely unspoiled unique marine system has engendered much speculation as to the potential ecological impact of these developments. The glaring inadequacy of our present understanding of both short- and long-term consequences of oil-related activities in the Arctic makes it difficult realistically to assess the validity of many of the current highly speculative predictions of unmitigated disaster. As Dunbar (1971) so aptly puts it, 'we have been caught in a state of scientific near-nudity in the particular respect in which we now so urgently need protective covering.' The Beaufort Sea Project--of which this study is a part--was initiated in 1974 as a 'crash programme' to consider a broad range of problems related to offshore drilling in the Western Arctic (Logan, 1975). The study reported here is part of a broader investigation into physiological effects of northern crude oils on arctic marine invertebrates. 155 Environ. Pollut. (10) (1976)--© Applied Science Publishers Ltd, England, 1976 Printed in Great Britain
156
J . A . PERCY
Little reliable information is available concerning the behavioural responses of marine organisms to crude oils or to oil-contaminated food. A number of pertinent field observations are suggestive but of uncertain significance. Reports cited by Nelson-Smith (1973) indicate that some fish may avoid spilled oil. Blumer (1972) noted that certain hydrocarbons derived from kerosene attract lobsters and, on this basis, tentatively suggested that the massive mortality of these animals subsequent to the Buzzards Bay oil spill may in part be attributable to an attraction towards the spill site. Wilder (1970) observed that lobsters in the laboratory readily consume fish heavily contaminated with Bunker C oil. Unfortunately no observations appear to have been made on whether or not the oiled fish was consumed as readily as uncontaminated material. In connection with the spill of Bunker C by the tanker Arrow, Thomas (1970) reported that periwinkles, Littorina sp., appeared to migrate from heavily oiled areas to adjacent clean areas of beach. Furthermore, he noted that clams, Mya arenaria, generally moved out of oil-polluted burrows; however, it is not clear whether this response represents a direct aversion to the oil or merely an attempt to escape suffocation in the burrows. In the present study we have examined the behavioural responses of a number of arctic marine crustaceans to several northern crude oils, alone and in combination with food.
METHODS
Three species of benthic crustaceans were used in this study. The amphipods Onisimus a l~nis and Gammarus oceanicus were collected in modified minnow traps in the Eskimo Lakes, a complex series of marine embayments adjacent to the Mackenzie River delta in the Northwest Territories. Isopods, Mesidotea entomon, were collected with an otter trawl in Kugmallit Bay in the vicinity of Tuktoyaktuk, North West Territory. Studies were conducted at the Eskimo Lakes field laboratory of the Arctic Biological Station. Prior to use animals were maintained in a circulating seawater tank at a temperature and salinity approximating that in the natural habitat. Three oils were used in the tests; two northern oils--Atkinson Point crude and N o r m a n Wells crude--and a Venezuelan crude (Tijauna light). The behavioural responses of the animals to the presence of oil masses in their immediate vicinity was examined as follows. Test chambers consisted of shallow pans subdivided into four equal zones (designated A, B, C and D) by lines inscribed on the bottom. These were immersed in an ice bath to maintain the seawater between 2 ° and 4°C. In order to record the distribution of the animals in the various zones at intervals, a camera equipped with a flash attachment was positioned so that the entire interior of the chamber was included in the field o f view. For each run one zone was designated as an oiled zone and the remaining three as unoiled, or control, zones.
RESPONSES OF ARCTIC MARINE CRUSTACEANS TO CRUDE OIL
157
To hold the oil, small squares of sponge measuring approximately 2 cm x 2 cm x I cm were affixed to glass microscope slides with non-toxic silicone cement. Immediately prior to the run 20 animals were placed in the chamber and distributed approximately equally among the four zones. Then 1 ml of fresh crude oil of the desired type was placed on a sponge square. As soon as most of the oil was absorbed, the sponge was rinsed rapidly in seawater to remove egcess oil. The oiled sponge was then placed in the centre of the pre-selected oiled zone. Similar sponge squares that had not been treated with oil were placed in the centre of each of the control zones. To eliminate outside disturbance and to prevent the animals from visually distinguishing between the dark-coloured oiled sponges and the paler control ones, the entire apparatus was enclosed in a light-proof hood. Each run lasted for 30 rain,. with the chamber being photographed at 2-min intervals. Four runs with a single species and a single oil type constituted a series, with each of the four zones being successively designated as the oiled zone. At the end of each run the chamber was thoroughly cleaned and refilled with fresh seawater. A freshly oiled sponge square was used for each run. The numbers o f animals within each zone at each observation period were counted from the film negative and tabulated. For each run the numbers of animals in the oiled zone at each observation were added to give an observed frequency for the oiled zone designated 0' z where z represents the particular oiled zone for a given run. The numbers o f animals in each of the three unoiled zones were similarly added and the totals for the three zones combined to yield an observed frequency for the unoiled control zones designated 0u. For each run 0'z and 0, were combined to yield the total number o f counts for that run. Statistical probability dictates that, in the absence of an attraction or repulsion response, for each run, one-quarter of the counts should occur in the oiled zone, designated e'~. The X2 value was calculated from the observed and expected frequencies for each run. The X' values for each of the four runs in a given series were added to yield a series X2 value (Spiegel, 1961) and the probability of the observed distribution being significantly different from a r a n d o m distribution was determined. In order to express the attraction or repulsion response in a clear and concise manner an affinity coefficient (AC) was defined as follows: AC --- X0'~ - Xe'z Ze'~ where 0'~ is the sum of the observed counts in the oiled zone for each of the four runs in a series (EO'z = 0'o + 0'~ + O'c + O'a) and Ze'z is the sum of the expected counts in the oiled zone for each of the four runs in a given series (Ze'~ = e'o + e'b + e'c + e'a). An AC of zero indicates a neutral response and the animal is neither attracted nor repelled by the oil. An increasingly positive value indicates an increasing
158
J.A. PERCY
degree of attraction, to a maximum of 100 which indicates a total attraction to the oiled zone. Similarly, an increasingly negative value indicates an increasing degree of repulsion, to a minimum of - 100 which indicates a total repulsion from the oiled zone. The significance of the AC for each series was determined from the Xz values calculated as indicated above. The response of animals to oil-tainted food was examined in two ways. In the first test only oil-tainted food was presented to the animals, while in the second a choice between tainted and untainted food was offered. The test food consisted of I cm blocks of fish flesh which are highly attractive to these amphipods. The food was artificially tainted by marinating the blocks in the appropriate oil for 1 h at room temperature and then vigorously rinsing them in clean seawater to remove excess oil. The test chamber was the same as that employed in the oil affinity study. In the first test, one of the sponge squares was replaced by a block of either tainted or untainted (control) fish. Ten animals were introduced into the centre of the chamber. The chamber was observed at 2-min intervals for a total of 20 min and each time the number of animals in contact with the food was noted. A relative feeding score was assigned by adding the counts for each 2-min observation and expressing the result as a percentage of the maximum possible score (all animals in contact with the food at every observation equals 100~). The tests with each oil were repeated three times using fresh animals for each run. A control run, using untainted fish, was carried out just prior to each group of experimental runs. In the choice experiments, a block of tainted fish was placed in one of the zones of the test chamber and a similar block of untainted fish in the diagonally opposite zone. Ten isopods or 20 amphipods were then introduced into the centre of the chamber. Once again, the chamber was observed at 2-min intervals for 20 min and the numbers of animals in contact with either the tainted or untainted food were recorded. For each run the total counts for contact with tainted and untainted food were each added and the sums expressed as a percentage of the total combined counts for that particular run. Each oil was tested three times, using a fresh group of animals for each test.
RESULTS AND DISCUSSION
None of the species examined were attracted to crude oil masses in their immediate vicinity; on the contrary, the results (Table l) indicate that the animals were generally repelled by the various oils. The degree of avoidance, quantified as an affinity coefficient (AC), varied considerably with the species and type of crude oil used. Both species of amphipods were significantly repelled by all three oils tested, with AC's ranging from - 17.3 to - 39.8. Atkinson Point crude evoked the greatest response in both species, although Onisimus was almost as strongly repelled by N o r m a n Wells crude. Venezuela crude repelled both species only slightly.
RESPONSES OF A R C T I C M A R I N E C R U S T A C E A N S T O C R U D E O I L
159
In marked contrast, the isopod M e s i d o t e a exhibited an essentially neutral response to both Atkinson Point and Venezuela crudes and was only slightly, if at all, repelled by the N o r m a n Wells crude. The animal appeared to be completely oblivious to the presence of the oils. The responses of the amphipods suggest that they tend to avoid fresh crude oil. The ecological importance of such a response to a species such as Onisimus is evident. Although they are normally benthic detritivores, in winter and early spring, in inshore waters, large numbers congregate in the vicinity of the ice-water interface (personal observation). Crude oil spilled under ice accumulates at the ice-water interface in the form of irregular oil lenses of varying size (Keevil & Ramseier, 1975). Preliminary studies indicate that animals coming into contact with oil, even for brief periods, are unable to clean themselves, even if transferred to fresh seawater, and ultimately succumb (Percy, 1974a).
TABLE 1 BEHAVIOURAL RESPONSES OF Gammarus oceanicus, Mesidotea entomon AND Onisimus affinis IN THE PRESENCE OF VARIOUS CRUDE OILS. (RESULTS EXPRESSED AS AFFINITY COEFFICIENTS AS DESCRIBED IN THE TEXT. LE'FFERS IN PARENTHESES INDICATE STATISTICAL SIGNIFICANCE OF RESULT AT THE 5 ~o LEVEL AS ESTABLISHED BY CHI-SQUARE TEST)
Oil type Species G. oceanicus M. entomon O. affinis
(a) fresh (b) weathered (c) pre-exposed
Atkinson Point
Norman Wells
Venezuela
-- 34"2(S) --3"3 (NS)
- 17.3 (S) - 16"9(S)
- 17-4(S) --2'I (NS)
- 39'8 (S) + 6'5 (NS) -11.7(S)
- 37"9(S) - 33"9(S) - 1-2 (NS)
-23.7 (S) -- 1.7 (NS) --6'9 (NS)
Our results suggest that Onisimus would avoid oil lenses composed of fresh crude oil. However, oil released into the marine environment rapidly loses varying proportions of many of its components by a complex weathering process. The affinity tests were thus repeated with O n i s i m u s to determine if the avoidance response changes as the oil is weathered. Prior to use, the crude oil was slowly stirred in a small open beaker at room temperature for 24 h. Thus treated, the Atkinson Point and Venezuela crudes proved to be tess distasteful than the fresh oils; in fact, the animals appeared to be quite neutral to these weathered oils. In contrast, the repellent effect of N o r m a n Wells crude was reduced only slightly, if at all, by weathering. It is likely that the animals are responding to some of the more volatile components of the oils. The avoidance of crude oil masses is presumably mediated by chemoreceptor organs, although the chemical compounds involved and the mechanics of the process are unknown. Recent work indicates that certain components of the oil destroy the neuronal dentrites of crustacean chemoreceptor organs (Kittredge,
160
J . A . PERCY
1973). In an oil spill situation animals would be exposed for varying periods to both finely emulsified oil and dissolved components in addition to the oil lenses. To determine if pre-exposure to highly dispersed oil would modify the animals' response to oil masses, the affinity tests were repeated with Onisimus. The animals were first exposed at 8°C to standardised emulsions (0-5 ml oil/litre seawater) under defined conditions (Percy, 1974b) for 24 h prior to commencement of the affinity tests. This pre-exposure markedly reduces the avoidance response (Table 1) and in some cases eliminates it completely. This may represent a physiological adaptation to the oil, but more probably results from impairment of chemoreceptor organs. This would increase the likelihood of the animal accidently blundering into oil lenses and becoming entrapped or severely fouled, either of which would be ultimately lethal. To understand more fully the behavioural interactions of animals with spilled oil we require comparable information on the affinity of organisms for low level emulsions and for varying concentrations of the water-soluble components of the oils. Both Onisimus and Mesidotea are omnivorous scavangers that readily consume dead and decaying organisms. It was therefore important to establish whether or not they might expose themselves further to petroleum pollutants by consuming fish or other organisms killed and heavily contaminated by an oil spill. A number of workers have demonstrated that fish flesh becomes tainted by exposure to even relatively low concentrations of petroleum products (Nelson-Smith, 1970; Connell, 1971). The results of feeding tests involving animals presented with oil-tainted food alone and in combination with untainted food are presented in Tables 2 and 3, respectively. Onisimus clearly rejects heavily tainted fish, even when presented as the only food source. It should be noted that these animals had not been starved prior to use in the tests. It remains to be seen whether the animals will accept tainted food more readily after being starved for extended periods. As might be expected from the above, when presented with a choice of tainted or untainted food, Onisimus overwhelmingly opts for the untainted material. The greatest response occurred with Norman Wells crude which resulted in a feeding score of only 19.6 %. In contrast, the untainted food was attractive to all groups, with feeding scores ranging from 80-99 %. The variability in the controls evident in Table 2 is probably attributable to the fact that the tests with the different oils were carried out at different times. Consequently, the animals in the different groups may have differed in their nutritional state and feeding activity. Recent evidence (details to be published later) suggests that Onisimus exhibits diurnal cycles in its activity and feeding. The responses to the unoiled food determined initially are much more consistent because all three were run at the same time. The control feeding activity for each oil series was determined at the same time as the experimental feeding activities for the particular
RESPONSES OF ARCTIC MARINE CRUSTACEANS TO CRUDE OIL
161
TABLE 2 BEHAVIOURAL RESPONSE OF Onisimus affinis AND Mesidotea entomon T O F O O D T A I N T E D W I T H V A R I O U S C R U D E OILS. RESULTS EXPRESSED AS PER C E N T
FEEDING ACTIVITYAS DF~CRIBEDIN TEXT
% Feeding activity Species
Control (%)
Oil type
Onisimus Onisimus Onisiraus Onisimus Mesidotea Mesidotea
Unoiled AP NW Ve Unoiled NW
(%)
a
(%)
(%)
c
Mean
67 3 5 3 81 39
87 6 3 5 83 26
87 2 8 3 50 21
80 4 5 4 71 29
-64 45 86 -54
b
(%)
oil. Thus the nutritional state of the animals within each oil series was similar, although it may have differed between series. Just as in the affinity studies, the isopod Mesidotea differs markedly from Onisimus in its response to oil-tainted food. The animals readily consume heavily tainted fish (Table 2), and in a choice situation show no particular preference for untainted over tainted food (Table 3), as demonstrated by the feeding scores of 47 % and 53 ~o, respectively. Norman Wells crude is the only oil that has been tested with Mesidotea thus far. Lobsters have similarly been found to consume fish very heavily contaminated with bunker C oil (Wilder, 1970). The long-term consequences of such ingestion of crude oils are at present uncertain. These studies, although highly suggestive, have barely grazed the surface of what should be an important aspect of oil pollution biology, namely, the behavioural responses of marine organisms to crude oil. The foregoing results clearly demonstrate that different species may differ markedly in their responses to oil. Thus, sweeping generalisations concerning oil effects on marine organisms will be difficult, if not impossible, to formulate. It is interesting to note that Mesidotea entomon, a species that exhibits an essentially neutral response to the presence of oil masses and to oil-tainted food, is TABLE 3 BEHAVIOURAL RE/iPONSE OF Onisimus affinis AND Mesidotea entomon PRESENTED WITH OIL-TAINTED (0~-) AND UNTAINTED ( 0 - - ) FISH SQUARES SIMULTANEOUSLY. EACH OIL TESTED IN TRIPLICATE (a, b AND C) USING DIFFERENTANIMALSEACH TIME a
b
c
O+ O--
O+ O--
O+ O--
Oil type
~.Oq-
YO--
%0+
%0--
Onisimus: AP NW V¢
3 6 0
8 34 124
17 6 2
73 82 125
1 23 0
31 28 82
21 35 2
112 144 331
15"8% 19'6% 0,6%
84'2% 80.4% 99'4%
37
24
42
21
10
33
89
78
53'3 %
46.7 %
Mesidotea: NW
162
J.A. PERCY
extremely resistant to high concentrations ofoil in the water column (Percy, 1974b). In contrast, Onisimus aj~nis, a species that tends to avoid oil masses and rejects oil-tainted food, is killed much more readily'by oil suspended in the water column. It is probable that for many species in certain types of habitats, attraction to, or repulsion from, spilled oil may be a significant factor influencing the overall impact of the pollution incident upon the population. In addition, the behavioural response to contaminated food may serve either to disrupt normal feeding activity severely or to provide an important avenue for the transfer of toxic components to animals that might not otherwise be exposed to significant concentrations of the oil. It is difficult to assess at present just how significant a role the aversion response to oil exhibited by Onisimus might have in reducing the direct impact of a major oil spill upon the population.
ACKNOWLEDGEMENTS
I wish to acknowledge the able technical assistance of Mr J. Walbridge in conducting these studies. My thanks are due to Dr E. H. Grainger for critically reviewing the manuscript and suggesting a number of improvements.
REFERENCES BLUMER, M. (1972). Oil contamination and the living resources of the sea. In Marine pollution and sea life, ed. by M. Ruivo, 476-81. London, Fishing News (Books). CONNELL,D. W. (1971). Kerosene-like tainting in Australian mullet. Mar. Pollut. Ball., 2(12), 188-90. Dur~axR, M. J. (1971). Environment and good sense. Montreal, McGill-Queens University Press. KEr.VIL, B. E. & RAMSEmR, R. O. (1975). Behavior of oil spilled under floating ice. Proceedings o f the Conference on the prevention and control o f oil pollution, San Francisco, March, 1975, 497-501. KITTREDOE,J. S. (1973). Effects o f crude oil pollution on the behavior o f marine invertebrates. Office of Naval Research, Arlington, Va., Naval Biology Program. Project no. NR 306-046 (Mimeographed). LOOAN, W. J. (1975). Oil spillcountermeasures for thc Beaufort Sea. Proceedings of the Conference on the prevention and control o f oil pollution, San Francisco, March 1975, 265-8. NELSON-SMrrn, A. (1970). The problem of oil pollution of the sea. Adv. Mar. Biol., 8, 215-306. NELSON-SMITH,A. (1973). Oilpollution and marine ecology. New York, Plenum Press. PERCY, J. A. (1974a). Effects o f crude oils on Arctic marine invertebrates. Report to EnvironmentalSocial Program Northern Pipelines, on Project No. 11, Marine Ecology of Mackenzie Delta and Tuktoyaktuk Peninsula Region, Part II. Environment Canada. (Mimeographed). PERCY, J. A. (1974b). Effects o f crude oils on Arctic marine invertebrates. Interim report of Beaufort Sea Project B5b, Environment Canada. (Mimeographed). SPmGEL, M. R. (1961). Theory and problems o f statistics. New York, Schaum. THOMAS,M. L. H. (1970). Effects o f bunker C oil on intertidal and lagoonal organisms in Chedabucto Bay, Nova Scotia. Unpublished report, Marine Ecology Laboratory, Dartmouth, Nova Scotia. WILDER, D. G. (1970). The tainting o f lobster meat by bunker C oil alone or in combination with the dispersant corexit. Unpublished report, Fisheries Research Board of Canada, St. Andrews. New Brunswick.
J. A. PERCY
Arctic Biological Station, Fisheries and Marine Service, Environment Canada, P.O. Box 400, Ste. Anne de Bellevue, P.Q., Canada ABSTRACT
The responses of several arctic marine crustaceans to oil masses and oil-tainted food have been investigated. None of the species were attracted to crude oil. Amphipods tended to avoid oil masses; however, the response was significantly diminished if the oil was weathered or i f the animals were pre-exposed to light crude oil emulsions. Untainted food was preferentially selected over oil-tainted food. In contrast, an isopod was generally neutral to the presence o f oil masses and consumed oil-tainted food as readily as untainted material.
INTRODUCTION
The potential for a major oil pollution incident in the Arctic has increased dramatically over the past several years, leading Dunbar (1971) to suggest that oil pollution is the most serious and immediate threat facing the arctic ecosystem. This is particularly true for the arctic marine ecosystem now that emphasis is rapidly shifting to offshore exploratory drilling. Concern for the preservation of this largely unspoiled unique marine system has engendered much speculation as to the potential ecological impact of these developments. The glaring inadequacy of our present understanding of both short- and long-term consequences of oil-related activities in the Arctic makes it difficult realistically to assess the validity of many of the current highly speculative predictions of unmitigated disaster. As Dunbar (1971) so aptly puts it, 'we have been caught in a state of scientific near-nudity in the particular respect in which we now so urgently need protective covering.' The Beaufort Sea Project--of which this study is a part--was initiated in 1974 as a 'crash programme' to consider a broad range of problems related to offshore drilling in the Western Arctic (Logan, 1975). The study reported here is part of a broader investigation into physiological effects of northern crude oils on arctic marine invertebrates. 155 Environ. Pollut. (10) (1976)--© Applied Science Publishers Ltd, England, 1976 Printed in Great Britain
156
J . A . PERCY
Little reliable information is available concerning the behavioural responses of marine organisms to crude oils or to oil-contaminated food. A number of pertinent field observations are suggestive but of uncertain significance. Reports cited by Nelson-Smith (1973) indicate that some fish may avoid spilled oil. Blumer (1972) noted that certain hydrocarbons derived from kerosene attract lobsters and, on this basis, tentatively suggested that the massive mortality of these animals subsequent to the Buzzards Bay oil spill may in part be attributable to an attraction towards the spill site. Wilder (1970) observed that lobsters in the laboratory readily consume fish heavily contaminated with Bunker C oil. Unfortunately no observations appear to have been made on whether or not the oiled fish was consumed as readily as uncontaminated material. In connection with the spill of Bunker C by the tanker Arrow, Thomas (1970) reported that periwinkles, Littorina sp., appeared to migrate from heavily oiled areas to adjacent clean areas of beach. Furthermore, he noted that clams, Mya arenaria, generally moved out of oil-polluted burrows; however, it is not clear whether this response represents a direct aversion to the oil or merely an attempt to escape suffocation in the burrows. In the present study we have examined the behavioural responses of a number of arctic marine crustaceans to several northern crude oils, alone and in combination with food.
METHODS
Three species of benthic crustaceans were used in this study. The amphipods Onisimus a l~nis and Gammarus oceanicus were collected in modified minnow traps in the Eskimo Lakes, a complex series of marine embayments adjacent to the Mackenzie River delta in the Northwest Territories. Isopods, Mesidotea entomon, were collected with an otter trawl in Kugmallit Bay in the vicinity of Tuktoyaktuk, North West Territory. Studies were conducted at the Eskimo Lakes field laboratory of the Arctic Biological Station. Prior to use animals were maintained in a circulating seawater tank at a temperature and salinity approximating that in the natural habitat. Three oils were used in the tests; two northern oils--Atkinson Point crude and N o r m a n Wells crude--and a Venezuelan crude (Tijauna light). The behavioural responses of the animals to the presence of oil masses in their immediate vicinity was examined as follows. Test chambers consisted of shallow pans subdivided into four equal zones (designated A, B, C and D) by lines inscribed on the bottom. These were immersed in an ice bath to maintain the seawater between 2 ° and 4°C. In order to record the distribution of the animals in the various zones at intervals, a camera equipped with a flash attachment was positioned so that the entire interior of the chamber was included in the field o f view. For each run one zone was designated as an oiled zone and the remaining three as unoiled, or control, zones.
RESPONSES OF ARCTIC MARINE CRUSTACEANS TO CRUDE OIL
157
To hold the oil, small squares of sponge measuring approximately 2 cm x 2 cm x I cm were affixed to glass microscope slides with non-toxic silicone cement. Immediately prior to the run 20 animals were placed in the chamber and distributed approximately equally among the four zones. Then 1 ml of fresh crude oil of the desired type was placed on a sponge square. As soon as most of the oil was absorbed, the sponge was rinsed rapidly in seawater to remove egcess oil. The oiled sponge was then placed in the centre of the pre-selected oiled zone. Similar sponge squares that had not been treated with oil were placed in the centre of each of the control zones. To eliminate outside disturbance and to prevent the animals from visually distinguishing between the dark-coloured oiled sponges and the paler control ones, the entire apparatus was enclosed in a light-proof hood. Each run lasted for 30 rain,. with the chamber being photographed at 2-min intervals. Four runs with a single species and a single oil type constituted a series, with each of the four zones being successively designated as the oiled zone. At the end of each run the chamber was thoroughly cleaned and refilled with fresh seawater. A freshly oiled sponge square was used for each run. The numbers o f animals within each zone at each observation period were counted from the film negative and tabulated. For each run the numbers of animals in the oiled zone at each observation were added to give an observed frequency for the oiled zone designated 0' z where z represents the particular oiled zone for a given run. The numbers o f animals in each of the three unoiled zones were similarly added and the totals for the three zones combined to yield an observed frequency for the unoiled control zones designated 0u. For each run 0'z and 0, were combined to yield the total number o f counts for that run. Statistical probability dictates that, in the absence of an attraction or repulsion response, for each run, one-quarter of the counts should occur in the oiled zone, designated e'~. The X2 value was calculated from the observed and expected frequencies for each run. The X' values for each of the four runs in a given series were added to yield a series X2 value (Spiegel, 1961) and the probability of the observed distribution being significantly different from a r a n d o m distribution was determined. In order to express the attraction or repulsion response in a clear and concise manner an affinity coefficient (AC) was defined as follows: AC --- X0'~ - Xe'z Ze'~ where 0'~ is the sum of the observed counts in the oiled zone for each of the four runs in a series (EO'z = 0'o + 0'~ + O'c + O'a) and Ze'z is the sum of the expected counts in the oiled zone for each of the four runs in a given series (Ze'~ = e'o + e'b + e'c + e'a). An AC of zero indicates a neutral response and the animal is neither attracted nor repelled by the oil. An increasingly positive value indicates an increasing
158
J.A. PERCY
degree of attraction, to a maximum of 100 which indicates a total attraction to the oiled zone. Similarly, an increasingly negative value indicates an increasing degree of repulsion, to a minimum of - 100 which indicates a total repulsion from the oiled zone. The significance of the AC for each series was determined from the Xz values calculated as indicated above. The response of animals to oil-tainted food was examined in two ways. In the first test only oil-tainted food was presented to the animals, while in the second a choice between tainted and untainted food was offered. The test food consisted of I cm blocks of fish flesh which are highly attractive to these amphipods. The food was artificially tainted by marinating the blocks in the appropriate oil for 1 h at room temperature and then vigorously rinsing them in clean seawater to remove excess oil. The test chamber was the same as that employed in the oil affinity study. In the first test, one of the sponge squares was replaced by a block of either tainted or untainted (control) fish. Ten animals were introduced into the centre of the chamber. The chamber was observed at 2-min intervals for a total of 20 min and each time the number of animals in contact with the food was noted. A relative feeding score was assigned by adding the counts for each 2-min observation and expressing the result as a percentage of the maximum possible score (all animals in contact with the food at every observation equals 100~). The tests with each oil were repeated three times using fresh animals for each run. A control run, using untainted fish, was carried out just prior to each group of experimental runs. In the choice experiments, a block of tainted fish was placed in one of the zones of the test chamber and a similar block of untainted fish in the diagonally opposite zone. Ten isopods or 20 amphipods were then introduced into the centre of the chamber. Once again, the chamber was observed at 2-min intervals for 20 min and the numbers of animals in contact with either the tainted or untainted food were recorded. For each run the total counts for contact with tainted and untainted food were each added and the sums expressed as a percentage of the total combined counts for that particular run. Each oil was tested three times, using a fresh group of animals for each test.
RESULTS AND DISCUSSION
None of the species examined were attracted to crude oil masses in their immediate vicinity; on the contrary, the results (Table l) indicate that the animals were generally repelled by the various oils. The degree of avoidance, quantified as an affinity coefficient (AC), varied considerably with the species and type of crude oil used. Both species of amphipods were significantly repelled by all three oils tested, with AC's ranging from - 17.3 to - 39.8. Atkinson Point crude evoked the greatest response in both species, although Onisimus was almost as strongly repelled by N o r m a n Wells crude. Venezuela crude repelled both species only slightly.
RESPONSES OF A R C T I C M A R I N E C R U S T A C E A N S T O C R U D E O I L
159
In marked contrast, the isopod M e s i d o t e a exhibited an essentially neutral response to both Atkinson Point and Venezuela crudes and was only slightly, if at all, repelled by the N o r m a n Wells crude. The animal appeared to be completely oblivious to the presence of the oils. The responses of the amphipods suggest that they tend to avoid fresh crude oil. The ecological importance of such a response to a species such as Onisimus is evident. Although they are normally benthic detritivores, in winter and early spring, in inshore waters, large numbers congregate in the vicinity of the ice-water interface (personal observation). Crude oil spilled under ice accumulates at the ice-water interface in the form of irregular oil lenses of varying size (Keevil & Ramseier, 1975). Preliminary studies indicate that animals coming into contact with oil, even for brief periods, are unable to clean themselves, even if transferred to fresh seawater, and ultimately succumb (Percy, 1974a).
TABLE 1 BEHAVIOURAL RESPONSES OF Gammarus oceanicus, Mesidotea entomon AND Onisimus affinis IN THE PRESENCE OF VARIOUS CRUDE OILS. (RESULTS EXPRESSED AS AFFINITY COEFFICIENTS AS DESCRIBED IN THE TEXT. LE'FFERS IN PARENTHESES INDICATE STATISTICAL SIGNIFICANCE OF RESULT AT THE 5 ~o LEVEL AS ESTABLISHED BY CHI-SQUARE TEST)
Oil type Species G. oceanicus M. entomon O. affinis
(a) fresh (b) weathered (c) pre-exposed
Atkinson Point
Norman Wells
Venezuela
-- 34"2(S) --3"3 (NS)
- 17.3 (S) - 16"9(S)
- 17-4(S) --2'I (NS)
- 39'8 (S) + 6'5 (NS) -11.7(S)
- 37"9(S) - 33"9(S) - 1-2 (NS)
-23.7 (S) -- 1.7 (NS) --6'9 (NS)
Our results suggest that Onisimus would avoid oil lenses composed of fresh crude oil. However, oil released into the marine environment rapidly loses varying proportions of many of its components by a complex weathering process. The affinity tests were thus repeated with O n i s i m u s to determine if the avoidance response changes as the oil is weathered. Prior to use, the crude oil was slowly stirred in a small open beaker at room temperature for 24 h. Thus treated, the Atkinson Point and Venezuela crudes proved to be tess distasteful than the fresh oils; in fact, the animals appeared to be quite neutral to these weathered oils. In contrast, the repellent effect of N o r m a n Wells crude was reduced only slightly, if at all, by weathering. It is likely that the animals are responding to some of the more volatile components of the oils. The avoidance of crude oil masses is presumably mediated by chemoreceptor organs, although the chemical compounds involved and the mechanics of the process are unknown. Recent work indicates that certain components of the oil destroy the neuronal dentrites of crustacean chemoreceptor organs (Kittredge,
160
J . A . PERCY
1973). In an oil spill situation animals would be exposed for varying periods to both finely emulsified oil and dissolved components in addition to the oil lenses. To determine if pre-exposure to highly dispersed oil would modify the animals' response to oil masses, the affinity tests were repeated with Onisimus. The animals were first exposed at 8°C to standardised emulsions (0-5 ml oil/litre seawater) under defined conditions (Percy, 1974b) for 24 h prior to commencement of the affinity tests. This pre-exposure markedly reduces the avoidance response (Table 1) and in some cases eliminates it completely. This may represent a physiological adaptation to the oil, but more probably results from impairment of chemoreceptor organs. This would increase the likelihood of the animal accidently blundering into oil lenses and becoming entrapped or severely fouled, either of which would be ultimately lethal. To understand more fully the behavioural interactions of animals with spilled oil we require comparable information on the affinity of organisms for low level emulsions and for varying concentrations of the water-soluble components of the oils. Both Onisimus and Mesidotea are omnivorous scavangers that readily consume dead and decaying organisms. It was therefore important to establish whether or not they might expose themselves further to petroleum pollutants by consuming fish or other organisms killed and heavily contaminated by an oil spill. A number of workers have demonstrated that fish flesh becomes tainted by exposure to even relatively low concentrations of petroleum products (Nelson-Smith, 1970; Connell, 1971). The results of feeding tests involving animals presented with oil-tainted food alone and in combination with untainted food are presented in Tables 2 and 3, respectively. Onisimus clearly rejects heavily tainted fish, even when presented as the only food source. It should be noted that these animals had not been starved prior to use in the tests. It remains to be seen whether the animals will accept tainted food more readily after being starved for extended periods. As might be expected from the above, when presented with a choice of tainted or untainted food, Onisimus overwhelmingly opts for the untainted material. The greatest response occurred with Norman Wells crude which resulted in a feeding score of only 19.6 %. In contrast, the untainted food was attractive to all groups, with feeding scores ranging from 80-99 %. The variability in the controls evident in Table 2 is probably attributable to the fact that the tests with the different oils were carried out at different times. Consequently, the animals in the different groups may have differed in their nutritional state and feeding activity. Recent evidence (details to be published later) suggests that Onisimus exhibits diurnal cycles in its activity and feeding. The responses to the unoiled food determined initially are much more consistent because all three were run at the same time. The control feeding activity for each oil series was determined at the same time as the experimental feeding activities for the particular
RESPONSES OF ARCTIC MARINE CRUSTACEANS TO CRUDE OIL
161
TABLE 2 BEHAVIOURAL RESPONSE OF Onisimus affinis AND Mesidotea entomon T O F O O D T A I N T E D W I T H V A R I O U S C R U D E OILS. RESULTS EXPRESSED AS PER C E N T
FEEDING ACTIVITYAS DF~CRIBEDIN TEXT
% Feeding activity Species
Control (%)
Oil type
Onisimus Onisimus Onisiraus Onisimus Mesidotea Mesidotea
Unoiled AP NW Ve Unoiled NW
(%)
a
(%)
(%)
c
Mean
67 3 5 3 81 39
87 6 3 5 83 26
87 2 8 3 50 21
80 4 5 4 71 29
-64 45 86 -54
b
(%)
oil. Thus the nutritional state of the animals within each oil series was similar, although it may have differed between series. Just as in the affinity studies, the isopod Mesidotea differs markedly from Onisimus in its response to oil-tainted food. The animals readily consume heavily tainted fish (Table 2), and in a choice situation show no particular preference for untainted over tainted food (Table 3), as demonstrated by the feeding scores of 47 % and 53 ~o, respectively. Norman Wells crude is the only oil that has been tested with Mesidotea thus far. Lobsters have similarly been found to consume fish very heavily contaminated with bunker C oil (Wilder, 1970). The long-term consequences of such ingestion of crude oils are at present uncertain. These studies, although highly suggestive, have barely grazed the surface of what should be an important aspect of oil pollution biology, namely, the behavioural responses of marine organisms to crude oil. The foregoing results clearly demonstrate that different species may differ markedly in their responses to oil. Thus, sweeping generalisations concerning oil effects on marine organisms will be difficult, if not impossible, to formulate. It is interesting to note that Mesidotea entomon, a species that exhibits an essentially neutral response to the presence of oil masses and to oil-tainted food, is TABLE 3 BEHAVIOURAL RE/iPONSE OF Onisimus affinis AND Mesidotea entomon PRESENTED WITH OIL-TAINTED (0~-) AND UNTAINTED ( 0 - - ) FISH SQUARES SIMULTANEOUSLY. EACH OIL TESTED IN TRIPLICATE (a, b AND C) USING DIFFERENTANIMALSEACH TIME a
b
c
O+ O--
O+ O--
O+ O--
Oil type
~.Oq-
YO--
%0+
%0--
Onisimus: AP NW V¢
3 6 0
8 34 124
17 6 2
73 82 125
1 23 0
31 28 82
21 35 2
112 144 331
15"8% 19'6% 0,6%
84'2% 80.4% 99'4%
37
24
42
21
10
33
89
78
53'3 %
46.7 %
Mesidotea: NW
162
J.A. PERCY
extremely resistant to high concentrations ofoil in the water column (Percy, 1974b). In contrast, Onisimus aj~nis, a species that tends to avoid oil masses and rejects oil-tainted food, is killed much more readily'by oil suspended in the water column. It is probable that for many species in certain types of habitats, attraction to, or repulsion from, spilled oil may be a significant factor influencing the overall impact of the pollution incident upon the population. In addition, the behavioural response to contaminated food may serve either to disrupt normal feeding activity severely or to provide an important avenue for the transfer of toxic components to animals that might not otherwise be exposed to significant concentrations of the oil. It is difficult to assess at present just how significant a role the aversion response to oil exhibited by Onisimus might have in reducing the direct impact of a major oil spill upon the population.
ACKNOWLEDGEMENTS
I wish to acknowledge the able technical assistance of Mr J. Walbridge in conducting these studies. My thanks are due to Dr E. H. Grainger for critically reviewing the manuscript and suggesting a number of improvements.
REFERENCES BLUMER, M. (1972). Oil contamination and the living resources of the sea. In Marine pollution and sea life, ed. by M. Ruivo, 476-81. London, Fishing News (Books). CONNELL,D. W. (1971). Kerosene-like tainting in Australian mullet. Mar. Pollut. Ball., 2(12), 188-90. Dur~axR, M. J. (1971). Environment and good sense. Montreal, McGill-Queens University Press. KEr.VIL, B. E. & RAMSEmR, R. O. (1975). Behavior of oil spilled under floating ice. Proceedings o f the Conference on the prevention and control o f oil pollution, San Francisco, March, 1975, 497-501. KITTREDOE,J. S. (1973). Effects o f crude oil pollution on the behavior o f marine invertebrates. Office of Naval Research, Arlington, Va., Naval Biology Program. Project no. NR 306-046 (Mimeographed). LOOAN, W. J. (1975). Oil spillcountermeasures for thc Beaufort Sea. Proceedings of the Conference on the prevention and control o f oil pollution, San Francisco, March 1975, 265-8. NELSON-SMrrn, A. (1970). The problem of oil pollution of the sea. Adv. Mar. Biol., 8, 215-306. NELSON-SMITH,A. (1973). Oilpollution and marine ecology. New York, Plenum Press. PERCY, J. A. (1974a). Effects o f crude oils on Arctic marine invertebrates. Report to EnvironmentalSocial Program Northern Pipelines, on Project No. 11, Marine Ecology of Mackenzie Delta and Tuktoyaktuk Peninsula Region, Part II. Environment Canada. (Mimeographed). PERCY, J. A. (1974b). Effects o f crude oils on Arctic marine invertebrates. Interim report of Beaufort Sea Project B5b, Environment Canada. (Mimeographed). SPmGEL, M. R. (1961). Theory and problems o f statistics. New York, Schaum. THOMAS,M. L. H. (1970). Effects o f bunker C oil on intertidal and lagoonal organisms in Chedabucto Bay, Nova Scotia. Unpublished report, Marine Ecology Laboratory, Dartmouth, Nova Scotia. WILDER, D. G. (1970). The tainting o f lobster meat by bunker C oil alone or in combination with the dispersant corexit. Unpublished report, Fisheries Research Board of Canada, St. Andrews. New Brunswick.