- Email: [email protected]
The effects of oil, dispersant, and emulsions on the survival and behavior of an estuarine teleost and an intertidal amphipod
ENVIRONMENTAL
RESEARCH
27, 266-276 (1982)
The Effects of Oil, Dispersant, and Emulsions Survival and Behavior of an Estuarine Teleost Intertidal Amphipod R. G. BUTLER,' The Mount
Desert
Island
W. TRIVELPIECE,~
AND D. S. MILLER
Biological
Salsbury
Laboratory,
Cove.
Mairle
on the and an
04672
Received January 26, 1981 Killifish (Fundhs heteroclitus) and amphipods (Garnmartts oceanicus) were exposed separately to either a No. 2 fuel oil, AP dispersant, or emulsions of the two in a static system. Both species exhibited a concentration-dependent response to all three treatments. However, emulsification of oil with dispersant clearly increased its lethal effect on killitish survival, but did not cause a differential change in behavioral parameters such as schooling, chafing, substrate nipping, activity, or depth preference. Killifish exposed to conditions of thermal or osmotic stress were more sensitive to the lethal effects of emulsions. In contrast, emulsions caused quantitative changes in amphipod activity and precopulatory behavior, but did not increase mortality beyond that caused by exposure to oil alone. Changes in salinity had little effect on amphipod sensitivity to emulsions, but decreasing temperature did result in increased survival.
INTRODUCTION
Aquatic organisms that inhabit the estuarine and intertidal zones are frequently exposed to oil spills, as well as to the chemical dispersants used in spill cleanup operations. The latter substances are detergents, and limited data indicate that oil/dispersant emulsions are more lethal (lower LC,,,) to aquatic organisms than either pollutant alone (Swedmark et al., 1973; Anderson et al., 1974; Barnett and Toews, 1978). Exposure to sublethal levels of these contaminants may have a long-term impact on marine organisms through disruption of behavioral patterns essential to survival or reproduction. In the present laboratory study, we exposed a euryhaline teleost (killitish, Fundulus heterocfitus) and an intertidal amphipod (Gammarus oceanicus) to oil, dispersant, and oil/dispersant emulsions to determine the effects of these pollutants on survival and some aspects of behavior. METHODS
General treatments. Killifish (Fundulus heteroclitus) were captured in traps near Salsbury Cove, Maine, and amphipods (Gammarus oceanicus) were collected at low tide near Hulls Cove, Maine. Both species were maintained separately in aerated holding tanks of 100% seawater (SW; 940 mosrn/g) at 15°C. Prior to use in experiments, killifish were allowed to acclimate for 1 week, while am1 Present address: Department of Biologocial Sciences, Duquesne University, 15219. ’ Present address: Point Reyes Bird Observatory, Stinson Beach, Calif. 94970. 266 0013-9351/82/020266-11$02.00/O Copyright All rights
0 1982 by Academic Ress, Inc. of reproduction in any form reserved.
Pittsburgh,
PA.
OIL,
DISPERSANT,
AND
EMULSION
TOXICITY
267
phipods were held for at least 24 hr. Fish and amphipods were tested in water of three different temperatures (10, 15, and 20°C) and three different salinities (100, 30, and 10% SW) to assess toxicant effects on the organisms under thermal and osmoregulatory stress. In each temperature and salinity, animals were exposed to oil, dispersants, or a 20/l oil/dispersant emulsion. The oils used were obtained from the American Petroleum Institute in 1976 (analyzed reference oils). One crude from South Louisiana was weathered over SW for 36 hr (WSLC) and chemical analyses of this weathered crude are given elsewhere (Peakall et al., 1980). The dispersant used was Atlantic/Pacific (AP) oil dispersant (GFC Chemical Co., West Palm Beach, Fla.). Emulsions were made by adding 5 ml of dispersant to 100 ml of oil and vortexing for 2 min. This ratio is within the range currently employed on oil spills. Procedure for killifish. The lethality phase of this study was conducted using glass aquaria (40 x 20 x 25 cm), each containing 15 liters of water. Three males and three females (5- 10 cm in length) were placed in each aquarium. Fish were exposed to oil, dispersant, or emulsion and the number of fish surviving was counted periodically for 72 hr. Usually, two repetitions of each experiment (a total of 12 fish for each treatment) were conducted, and paired controls were included with each replication (24 fish). Recovery experiments were also carried out in which killitish were removed from contaminated tanks after 24 hr and placed in fresh seawater for an additional 72 hr. Behavioral studies were conducted utilizing four glass aquaria (50 x 26 x 29 cm), each containing 30 liters of fresh seawater (941 mosm/g) at 15°C. Three male and three female killitish (5- 10 cm in length) were placed in each tank and allowed to habituate for 2 hr prior to addition of toxicants. Duplicate experiments were conducted (total of 12 fish for each experiment) and a control tank was included in each series. Fish behavior was recorded simultaneously for all four tanks using Super-&mm movie cameras with intervalometers (one frame/2 set). A grid was placed on the back of each tank to identify the position of each fish. Observations made on individual fish through frame-by-frame analysis included: depth preference, schooling tendency, activity, substrate nips, and chafes. These data were collected for 30-min intervals at 3, 12,24, and 48 hr following addition of toxicant. Procedure for amphipods. Five pairs of amphipods, engaged in the precopulatory clasp characteristic of these invertebrates, were placed in an aluminum pan (21 x 9 X 6 cm) containing 750 ml of seawater. Amphipods were exposed by adding either No. 2 fuel oil, Atlantic/Pacific (AP) dispersant, or an emulsion of the two pollutants to the seawater. Three repetitions of each experiment (a total of 30 amphipods for each treatment) were conducted, and paired controls were included in each experiment. The following data were collected periodically for 36 hr after initial exposure: the number of animals alive, the number of animals swimming normally during a 15-set period, and the number of pairs engaged in the precopulatory clasp. Data analysis. For purposes of analysis, data from similar treatment groups were pooled, as were data from control groups. x2 tests and t tests were performed using raw data, rather than percentages.
268
BUTLER,
TRIVELPIECE,
AND MILLER
RESULTS
All control animals (n = 120) tested survived the 3-day experimental period. Under standard conditions (100% SW at 1X>, exposure to No. 2 fuel oil caused a concentration-dependent increase in killifish mortality (Fig. 1). From these data we estimate 48-hr LCsO values to be between 5 and 10 ppt; 72-hr LC5,, values were between 1 and 5 ppt. Corresponding data for dispersant indicated greater lethality for a given set of experimental conditions, e.g., 48-hr LC,, values were roughly 0.1 ppt, and 72-hr LC5,, values were between 0.05 and 0.1 ppt. When fish were exposed to an oil/dispersant emulsion at 1.05 ppt (a level at which neither pollutant alone caused any mortality at 72 hr), the emulsion was clearly lethal, killing nearly all test animals after 48 hr (Fig. 1). Killitish exposed to 1 ppt of a South Louisiana crude (SLC) exhibited no mortality in 48 hr, but 50% mortality in 72 hr. This was a significant increase in mortality when compared to control fish (x2 = 5.042, df = 1, P < 0.05). Thus, this crude appeared to be more toxic to killifish than No. 2 fuel oil. With 1 ppt weathered SLC, no mortality was observed after 72-hr exposures. Since weathering of
12
24 HOURS
48 OF
72
EXPOSURE
FIG. 1. Effects of different levels of No. 2 fuel oil, AP dispersant, and emulsions on the survival of killitish under standard conditions. All control animals (n = 120) survived 72 hr. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
OIL,
DISPERSANT,
AND
EMULSION
TOXICITY
269
this oil resulted primarily
in the loss of lower-molecular-weight alkanes (Peakall et appear to be the most toxic to fish. As with No. 2 fuel oil, emulsification with dispersant increased the lethality of SLC to fish (100% mortality in 72 hr; x2 = 16.667, df = 1, P < O.OOl), but the toxicity of WSLC was not potentiated. In recovery experiments, fish were exposed to 10 ppt No. 2 fuel oil, 0.2 ppt dispersant, or an oil/dispersant emulsion for 24 hr and then transferred to unpolluted 100% SW. There was no mortality in any of these treatment groups after short exposure and transfer to fresh SW for 5 days. With continuous exposure, each of the pollutant levels killed at least 50% of the test animals at 48 hr. Using No. 2 oil and dispersant, we investigated the effects of varying salinity and temperature on killitish survival. Fish maintained in 30% SW (an environment that is nearly isosmotic with teleost body fluids) exhibited no mortality after 72-hr exposures to a 1.05-ppt oil + dispersant emulsion (Fig. 2). Fish maintained in pond water (100 mosm/kg) exhibited significant mortality after 48-hr exposure to the same emulsion; however, mortality was only about half of that found for SW-maintained fish. Increasing the temperature at which fish were maintained (100% SW) shortened the time period before mortality was observed. For any given time period, mortality increased with temperature (Fig. 2). Control killifish typically swam in schools of varying cohesiveness and at varying depths in the tank; they nipped and chafed substrate, and displayed pronounced startle responses to vibration or sudden movement near the aquarium. The behavior of fish exposed to lethal levels of either oil or emulsions was characterized generally by a decrease in activity, aggregation at the surface, and eventual loss of buoyancy and righting ability. However, experimentals did retain a response to vibration or sudden movement near the aquaria. Prior to death, fish exposed to lethal levels of dispersant generally exhibited a decrease in activity, aggregation on the bottom of the aquarium, and lack of pronounced startle response. Quantitative analyses of the behavior of fish exposed to sublethal pollutant levels (0.1 ppt oil, 0.005 ppt dispersant, or 0.105 ppt emulsion under standard conditions) revealed no significant differences in terms of schooling tendency, frequency of substrate nips, or frequency of chafes when compared to controls. However, for all exposure periods up to 48 hr. all three experimental groups exhibited significantly decreased activity scores (Fig. 3) and different depth preferences (Fig. 4) when compared to control groups. Similar results were also obtained at pollutant levels that were one order of magnitude lower. There were no major differences among treatment groups with regard to pollutant effects on killitish behavior. At concentrations that were two orders of magnitude lower, treatment groups did not differ from controls with regard to any of the behavioral parameters measured (Figs. 3 and 4). These findings indicate a steep doseresponse curve for behavioral effects of all three treatments at sublethal concentrations. al., 1980), these compounds
Amphipods
All control amphipods (n = 58) tested survived the 36-hr experimental period. Mortality data for No. 2 oil, AP dispersant, and oil/dispersant emulsions are
270
BUTLER,
TRIVELPIECE,
AND MILLER
. . 100
. . 4
50
100 %
SW
Pondwater 30 % SW
-
.
0 12
24
HOURS
40
OF
72
EXPOSURE
FIG. 2. Effects of varying temperature and salinity on survival of killifish exposed to a 1.05-ppt emulsion of No. 2 fuel oil and AP dispersant. All control animals (n = 12) survived to 72 hr. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
shown in Fig. 5. Based on these data, amphipods were clearly more sensitive to oil (36-hr LCSO = 0.01-0.1 ppt), but less sensitive to dispersant (36-hr LC5,, = 0.05-0.1 ppt) than killifish. In contrast to kill&h, oilldispersant emulsions appeared to be no more lethal to amphipods than oil alone. The effects of salinity and temperature on oil toxicity indicated that reducing medium salinity, or increasing the temperature from 15 to 2o”C, had no significant effect on lethality. However, reducing the test temperature to 10°C dramatically decreased amphipod mortality due to oil or emulsion exposure by 70%. Finally, amphipods that were exposed to 0.1 ppt oil or 0.105 ppt emulsion and then transferred to clean 100% SW exhibited mortality comparable to that of animals continuously exposed to pollutants. This finding indicates that, unlike the case for killifish, the initial exposure period is critical in determining the ultimate effects of these pollutants on G. oceanicus. Control amphipods generally displayed bursts of normal swimming activity, alternating with stationary periods on the bottom: at this time, their pleopods remained in continuous motion. The number of control pairs engaged in the precopulatory clasp generally decreased over the course of the time period, and all animals displayed a vigorous startle response to vibration of the holding container at all time periods. Compared with control groups, the percentage of amphipods
OIL, DISPERSANT,
AND EMULSION
TOXICITY
271
0 CONTROL
OIL
DISPERSANT
EMVLSION
FIG. 3. Activity scores (expressed as the mean number of lo-cm gridlines crossed by each fish in 15 set) for killifish exposed to No. 2 fuel oil, AP dispersant, and emulsions for 48 hr under standard conditions. Significant differences for I test comparisons between each treatment group and control animals for each time period are indicated with asterisks.
engaged in precopulatory behavior in all oil treatments, all emulsion treatments, and one dispersant treatment (0.05 ppt) decreased significantly in a time- and dose-dependent manner (Fig. 6). Animals continuously exposed to the lowest oil level (0.01 ppt) displayed significantly more precopulatory behavior than the corresponding emulsion (0.0105 ppt) group (Fig. 6). Dispersant did not appear to affect swimming activity; however, exposure to lower levels of oil and emulsions caused a biphasic response in activity. Oil levels of 0.1 and 0.01 ppt and emulsion levels of 0.105 and 0.0105 ppt resulted in hyperactivity after 1 hr, but activity was significantly reduced by 6 hr. Animals exposed to higher levels of pollutants were significantly less active than controls at 1 hr (Fig. 7); in contrast to the effects of oil exposure, all three emulsion levels reduced amphipod activity significantly by 6 hr (Fig. 7). Thus, emulsification of 0.01 ppt oil with 0.0005 ppt dispersant increased toxic effects on amphipod behavior. Behavioral patterns of animals exposed to lethal levels of oil and emulsion treatments were characterized by erratic swimming followed by gradual cessation of normal limb movement. At this stage, amphipods responded to vibration of the holding pan with only slight movement of
272
BUTLER,
S
M
B
TRIVELPIECE,
5
CONTROL
M
B
OIL
AND MILLER
S
M
DISPERSANT
B
s
M
B
EMULSION
4. Depth preference index (mean percentage of fish at the surface, S: swimming in midtank, M: or located on the bottom, B) for killifish exposed to No. 2 fuel oil, AP dispersant, and emulsions for 48 hr under standard conditions. Significant differences for x2 comparisons between each treatment group and control animals are indicated with asterisks. FIG.
the antennae or limbs. In contrast, response to the same stimulus.
controls
engaged in vigorous
swimming
in
DISCUSSION
The lethal effects of petroleum hydrocarbons on marine organisms have been the subject of numerous studies, which have been reviewed previously (e.g., see Anderson& al., 1974; Anderson 1977). However, few investigators have addressed the question of changes in toxicity due to emulsification by dispersants. Methodological, environmental, and/or species-specific differences often result in different LCSO values for the same species studied in two different laboratories. Therefore, comparative studies of two or more species often produce more useful
OIL, DISPERSANT,
I
AND EMULSION
24
6 HOURS
OF
273
TOXICITY
36
EXPOSURE
FIG. 5. Effects of different levels of No. 2 fuel oil. AP dispersant, and emulsions on the survival of amphipods under standard conditions. All control animals (n = 58) survived to 36 hr. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
insights into pollutant-organism interactions. In the present study of F. heteroclitus and G. oceanicus, our results indicate that AP dispersant potentiates the effects of No. 2 fuel oil in a static system. Mortality data indicated that killitish were clearly more sensitive to emulsions than to oil or dispersant alone (Fig. 1). In contrast, amphipods exhibited similar mortality in both oil and emulsion treatments, but demonstrated differential sensitivity to the treatments behaviorally (e.g., precopulatory clasp; Fig. 6). Both killifish and amphipods responded similarly with respect to thermal stress (i.e., generally decreased mortality in all treatment groups with decreased temperature; Fig. 2). However, only the killitish demonstrated a differential response to pollutants in salinities that were hypo- or hypertonic. Survival was significantly greater in salinities that were isosmotic with killifish body fluids. This is not surprising, since the killitish is an osmoregulator and the amphipod an osmoconformer. Effects of the interaction of temperature, salinity, and oil contamination on development of F. heterocfitus embryos have also been reported (Linden et al., 1979). The lethal effects of No. 2 fuel oil on killifish embryos were increased by temperatures above or below 25”C, especially under conditions of suboptimal salinity. Killifish in the present study also exhibited complete recovery following short-term exposures to all three pollutants, while amphipods did not recover after comparable exposures. Finally, amphipods
274
BUTLER,
1
TRIVELPIECE,
AND MILLER
6
24 HOURS
OF
36
EXPOSURE
FIG. 6. Effects of different levels of No. 2 fuel oil, AP dispersant.and emulsions on the precopulatory behavior of amphipods under standard conditions. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
were more sensitive to oil, but far less sensitive to dispersant than killifish. Results of the present study support earlier findings by Swedmark ef al. (1973) that indicated that oiYdispersant emulsions were more toxic to marine organisms than either pollutant alone, and that fish are oftentimes more sensitive to these pollutants than crustaceans. Lethality data alone are inherently conservative in their predictions of the impact of petroleum on aquatic ecosystems. Sublethal doses of oil may affect the behavior of marine organisms in a manner that may reduce survival of the individual or reproduction. Locomotion in the snail (Littorina fittorea) was significantly altered by sublethal concentrations of oil (Hargraves and Newcombe, 1973), and chemoreception, essential to mate and food location by many marine invertebrates, may be impaired by petroleum hydrocarbons (Kittridge 1973; Atema and Stein, 1974). Deterioration of feeding response and food retention has also been noted in the sea cattish (Ariusfelis) exposed to fuel oil (Wang and Nicol, 1977). Linden (1976) demonstrated that Venezuelan crude oil significantly decreased reproduction in an amphipod (Gammarus oceanicus) due to the decreased frequency with which males and females engaged in precopulatory behavior. A study of the long-term effects on an oil spill on populations of the salt marsh crab (Uca pugnax) indicated significant alterations in locomotory ability and burrow construction (Krebs and Burns, 1977).
OIL, DISPERSANT,
I
AND EMULSION
24
6
275
TOXICITY
36
z HOURS
OF
EXPOSURE
FIG. 7. Effects of different levels of No. 2 fuel oil, AP dispersant, and emulsions on the activity of amphipods under standard conditions. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
The results of the behavioral phase of the present study did reveal some quantitative, but not qualitative, changes in response to pollutant levels. At concentrations that were two orders of magnitude below lethal levels, killifish demonstrated changes in behavioral patterns (i.e., activity and depth preference), suggesting that behaviors essential for survival at both the individual and population levels (e.g., feeding, antipredator, and reproductive behavior) may be impaired with exposure to these pollutants in the wild. More importantly, the amphipod data clearly indicate that in the case of the precopulatory clasp, dispersant potentiates oil effects on the reproductive behavior of this species (Fig. 6). Recently, amphipod species diversity has been shown to be inversely correlated with the degree of marine pollution (Bellan-Santini, 1980). These results underscore the potential ecological impact that even sublethal levels of petroleum hydrocarbons may have on aquatic organisms. These data are also important in light of recent research indicating that physiological stress due to water pollution may be correlated with increased disease in fish populations (Brown et al., 1979). Finally, the present study emphasizes the need for caution in the use of commercial dispersants at the sites of major spills, especially in inshore waters.
276
BUTLER,
TRIVELPIECE,
AND MILLER
ACKNOWLEDGMENTS This research was supported by NIH Grants ES-00920 and S07-RR-05764 and NSF Grant DEB 7826821. We thank L. Baldwin and S. Dion for excellent technical assistance.
REFERENCES Anderson, J. W. (1977). Effects of petroleum hydrocarbons on the growth of marine organisms. Rapp. P. V. Rertn. Cons. Inf. Explor. Mer. 157-165. Anderson, J. W., Neff, J. M.. Cox, B. A., Tatem, H. E.. and Hightower, G. M. (1974). Characteristics of dispersions and water soluble extracts of crude and refined oils and their toxicity to estuarine crustaceans and fish. Mar. Biol. 27, 75-88. Atema, J., and Stein, L. S. (1974). Effects of crude oil on the feeding behavior of the lobster Hormms americanus.
Environ.
Pollut.
6, 77-86.
Barnett, J., and Toews, D. (1978). The effects of crude oil and the dispersant, Oilsperse 43, on respiration and coughing rates in Atlantic salmon (S&no s&u). Canad. J. Zoo/. 56, 307-310. Bellan-Santini, D. (1980). Relationship between population of amphipods and pollution. Mur. Po//ut. Bull.
11, 224-227.
Brown, E. R., Koch, E., Sinclair, T. F., Spitzer, R., and Callaghan. 0. (1979). Water pollution and diseases in fish (an epizootiologic survey). J. Emiron. P&ho/. Tuxicol. 2, 917-925. Hargraves, B. T., and Newcombe, C. P. (1973). Crawling and respiration as indices of sub-lethal effects of oil and a dispersant on an intertidal snail Littorina littorea. J. Fish. Res. Board Canad. 30, 1789- 1792. Kittridge, J. S. (1973). The effects of crude oil pollution on the behavior of marine invertebrates. Find Rep. U.S. Govt. Ad-762 047. Krebs, C. T., and Burns, K. A. (1977). Long-term effects of an oil spill on populations of the saltmarsh crab, Uca pugnax. Science 197, 484-487. Linden, 0. (1976). Effects of oil on the reproduction of the amphipod Gamnmrus oceanicus. Arnbio 5, 36-37. Linden, O., Sharp, J. R., Laughlin, R., Jr., and Neff, J. M. (1979). Interactive effects of salinity, temperature and chronic exposure to oil on the survival and development rate of embryos of the estuarine killifish Fundtrlus heteroclitus. Mar. Biol. 51, 101- 109. Peakall, D. B., Hallet, D. J., Miller, D. S., Butler, R. G., and Kinter, W. B. (1980). Effects of ingested crude oil on Black Guillemots: A combined field and laboratory study. Atnbio 9, 28-30. Swedmark, M., Granmo, A., and Kollberg, S. (1973). Effects of oil dispersants and oil emulsions on marine animals. Water Res. 7, 1649- 1672. Wang, R. T., and Nicol, J. A. C. (1977). Effects of fuel oil on sea catfish: Feeding activity and cardiac responses. Bull. Environ. Contmn. Toxicol. 18, 170- 176.
RESEARCH
27, 266-276 (1982)
The Effects of Oil, Dispersant, and Emulsions Survival and Behavior of an Estuarine Teleost Intertidal Amphipod R. G. BUTLER,' The Mount
Desert
Island
W. TRIVELPIECE,~
AND D. S. MILLER
Biological
Salsbury
Laboratory,
Cove.
Mairle
on the and an
04672
Received January 26, 1981 Killifish (Fundhs heteroclitus) and amphipods (Garnmartts oceanicus) were exposed separately to either a No. 2 fuel oil, AP dispersant, or emulsions of the two in a static system. Both species exhibited a concentration-dependent response to all three treatments. However, emulsification of oil with dispersant clearly increased its lethal effect on killitish survival, but did not cause a differential change in behavioral parameters such as schooling, chafing, substrate nipping, activity, or depth preference. Killifish exposed to conditions of thermal or osmotic stress were more sensitive to the lethal effects of emulsions. In contrast, emulsions caused quantitative changes in amphipod activity and precopulatory behavior, but did not increase mortality beyond that caused by exposure to oil alone. Changes in salinity had little effect on amphipod sensitivity to emulsions, but decreasing temperature did result in increased survival.
INTRODUCTION
Aquatic organisms that inhabit the estuarine and intertidal zones are frequently exposed to oil spills, as well as to the chemical dispersants used in spill cleanup operations. The latter substances are detergents, and limited data indicate that oil/dispersant emulsions are more lethal (lower LC,,,) to aquatic organisms than either pollutant alone (Swedmark et al., 1973; Anderson et al., 1974; Barnett and Toews, 1978). Exposure to sublethal levels of these contaminants may have a long-term impact on marine organisms through disruption of behavioral patterns essential to survival or reproduction. In the present laboratory study, we exposed a euryhaline teleost (killitish, Fundulus heterocfitus) and an intertidal amphipod (Gammarus oceanicus) to oil, dispersant, and oil/dispersant emulsions to determine the effects of these pollutants on survival and some aspects of behavior. METHODS
General treatments. Killifish (Fundulus heteroclitus) were captured in traps near Salsbury Cove, Maine, and amphipods (Gammarus oceanicus) were collected at low tide near Hulls Cove, Maine. Both species were maintained separately in aerated holding tanks of 100% seawater (SW; 940 mosrn/g) at 15°C. Prior to use in experiments, killifish were allowed to acclimate for 1 week, while am1 Present address: Department of Biologocial Sciences, Duquesne University, 15219. ’ Present address: Point Reyes Bird Observatory, Stinson Beach, Calif. 94970. 266 0013-9351/82/020266-11$02.00/O Copyright All rights
0 1982 by Academic Ress, Inc. of reproduction in any form reserved.
Pittsburgh,
PA.
OIL,
DISPERSANT,
AND
EMULSION
TOXICITY
267
phipods were held for at least 24 hr. Fish and amphipods were tested in water of three different temperatures (10, 15, and 20°C) and three different salinities (100, 30, and 10% SW) to assess toxicant effects on the organisms under thermal and osmoregulatory stress. In each temperature and salinity, animals were exposed to oil, dispersants, or a 20/l oil/dispersant emulsion. The oils used were obtained from the American Petroleum Institute in 1976 (analyzed reference oils). One crude from South Louisiana was weathered over SW for 36 hr (WSLC) and chemical analyses of this weathered crude are given elsewhere (Peakall et al., 1980). The dispersant used was Atlantic/Pacific (AP) oil dispersant (GFC Chemical Co., West Palm Beach, Fla.). Emulsions were made by adding 5 ml of dispersant to 100 ml of oil and vortexing for 2 min. This ratio is within the range currently employed on oil spills. Procedure for killifish. The lethality phase of this study was conducted using glass aquaria (40 x 20 x 25 cm), each containing 15 liters of water. Three males and three females (5- 10 cm in length) were placed in each aquarium. Fish were exposed to oil, dispersant, or emulsion and the number of fish surviving was counted periodically for 72 hr. Usually, two repetitions of each experiment (a total of 12 fish for each treatment) were conducted, and paired controls were included with each replication (24 fish). Recovery experiments were also carried out in which killitish were removed from contaminated tanks after 24 hr and placed in fresh seawater for an additional 72 hr. Behavioral studies were conducted utilizing four glass aquaria (50 x 26 x 29 cm), each containing 30 liters of fresh seawater (941 mosm/g) at 15°C. Three male and three female killitish (5- 10 cm in length) were placed in each tank and allowed to habituate for 2 hr prior to addition of toxicants. Duplicate experiments were conducted (total of 12 fish for each experiment) and a control tank was included in each series. Fish behavior was recorded simultaneously for all four tanks using Super-&mm movie cameras with intervalometers (one frame/2 set). A grid was placed on the back of each tank to identify the position of each fish. Observations made on individual fish through frame-by-frame analysis included: depth preference, schooling tendency, activity, substrate nips, and chafes. These data were collected for 30-min intervals at 3, 12,24, and 48 hr following addition of toxicant. Procedure for amphipods. Five pairs of amphipods, engaged in the precopulatory clasp characteristic of these invertebrates, were placed in an aluminum pan (21 x 9 X 6 cm) containing 750 ml of seawater. Amphipods were exposed by adding either No. 2 fuel oil, Atlantic/Pacific (AP) dispersant, or an emulsion of the two pollutants to the seawater. Three repetitions of each experiment (a total of 30 amphipods for each treatment) were conducted, and paired controls were included in each experiment. The following data were collected periodically for 36 hr after initial exposure: the number of animals alive, the number of animals swimming normally during a 15-set period, and the number of pairs engaged in the precopulatory clasp. Data analysis. For purposes of analysis, data from similar treatment groups were pooled, as were data from control groups. x2 tests and t tests were performed using raw data, rather than percentages.
268
BUTLER,
TRIVELPIECE,
AND MILLER
RESULTS
All control animals (n = 120) tested survived the 3-day experimental period. Under standard conditions (100% SW at 1X>, exposure to No. 2 fuel oil caused a concentration-dependent increase in killifish mortality (Fig. 1). From these data we estimate 48-hr LCsO values to be between 5 and 10 ppt; 72-hr LC5,, values were between 1 and 5 ppt. Corresponding data for dispersant indicated greater lethality for a given set of experimental conditions, e.g., 48-hr LC,, values were roughly 0.1 ppt, and 72-hr LC5,, values were between 0.05 and 0.1 ppt. When fish were exposed to an oil/dispersant emulsion at 1.05 ppt (a level at which neither pollutant alone caused any mortality at 72 hr), the emulsion was clearly lethal, killing nearly all test animals after 48 hr (Fig. 1). Killitish exposed to 1 ppt of a South Louisiana crude (SLC) exhibited no mortality in 48 hr, but 50% mortality in 72 hr. This was a significant increase in mortality when compared to control fish (x2 = 5.042, df = 1, P < 0.05). Thus, this crude appeared to be more toxic to killifish than No. 2 fuel oil. With 1 ppt weathered SLC, no mortality was observed after 72-hr exposures. Since weathering of
12
24 HOURS
48 OF
72
EXPOSURE
FIG. 1. Effects of different levels of No. 2 fuel oil, AP dispersant, and emulsions on the survival of killitish under standard conditions. All control animals (n = 120) survived 72 hr. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
OIL,
DISPERSANT,
AND
EMULSION
TOXICITY
269
this oil resulted primarily
in the loss of lower-molecular-weight alkanes (Peakall et appear to be the most toxic to fish. As with No. 2 fuel oil, emulsification with dispersant increased the lethality of SLC to fish (100% mortality in 72 hr; x2 = 16.667, df = 1, P < O.OOl), but the toxicity of WSLC was not potentiated. In recovery experiments, fish were exposed to 10 ppt No. 2 fuel oil, 0.2 ppt dispersant, or an oil/dispersant emulsion for 24 hr and then transferred to unpolluted 100% SW. There was no mortality in any of these treatment groups after short exposure and transfer to fresh SW for 5 days. With continuous exposure, each of the pollutant levels killed at least 50% of the test animals at 48 hr. Using No. 2 oil and dispersant, we investigated the effects of varying salinity and temperature on killitish survival. Fish maintained in 30% SW (an environment that is nearly isosmotic with teleost body fluids) exhibited no mortality after 72-hr exposures to a 1.05-ppt oil + dispersant emulsion (Fig. 2). Fish maintained in pond water (100 mosm/kg) exhibited significant mortality after 48-hr exposure to the same emulsion; however, mortality was only about half of that found for SW-maintained fish. Increasing the temperature at which fish were maintained (100% SW) shortened the time period before mortality was observed. For any given time period, mortality increased with temperature (Fig. 2). Control killifish typically swam in schools of varying cohesiveness and at varying depths in the tank; they nipped and chafed substrate, and displayed pronounced startle responses to vibration or sudden movement near the aquarium. The behavior of fish exposed to lethal levels of either oil or emulsions was characterized generally by a decrease in activity, aggregation at the surface, and eventual loss of buoyancy and righting ability. However, experimentals did retain a response to vibration or sudden movement near the aquaria. Prior to death, fish exposed to lethal levels of dispersant generally exhibited a decrease in activity, aggregation on the bottom of the aquarium, and lack of pronounced startle response. Quantitative analyses of the behavior of fish exposed to sublethal pollutant levels (0.1 ppt oil, 0.005 ppt dispersant, or 0.105 ppt emulsion under standard conditions) revealed no significant differences in terms of schooling tendency, frequency of substrate nips, or frequency of chafes when compared to controls. However, for all exposure periods up to 48 hr. all three experimental groups exhibited significantly decreased activity scores (Fig. 3) and different depth preferences (Fig. 4) when compared to control groups. Similar results were also obtained at pollutant levels that were one order of magnitude lower. There were no major differences among treatment groups with regard to pollutant effects on killitish behavior. At concentrations that were two orders of magnitude lower, treatment groups did not differ from controls with regard to any of the behavioral parameters measured (Figs. 3 and 4). These findings indicate a steep doseresponse curve for behavioral effects of all three treatments at sublethal concentrations. al., 1980), these compounds
Amphipods
All control amphipods (n = 58) tested survived the 36-hr experimental period. Mortality data for No. 2 oil, AP dispersant, and oil/dispersant emulsions are
270
BUTLER,
TRIVELPIECE,
AND MILLER
. . 100
. . 4
50
100 %
SW
Pondwater 30 % SW
-
.
0 12
24
HOURS
40
OF
72
EXPOSURE
FIG. 2. Effects of varying temperature and salinity on survival of killifish exposed to a 1.05-ppt emulsion of No. 2 fuel oil and AP dispersant. All control animals (n = 12) survived to 72 hr. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
shown in Fig. 5. Based on these data, amphipods were clearly more sensitive to oil (36-hr LCSO = 0.01-0.1 ppt), but less sensitive to dispersant (36-hr LC5,, = 0.05-0.1 ppt) than killifish. In contrast to kill&h, oilldispersant emulsions appeared to be no more lethal to amphipods than oil alone. The effects of salinity and temperature on oil toxicity indicated that reducing medium salinity, or increasing the temperature from 15 to 2o”C, had no significant effect on lethality. However, reducing the test temperature to 10°C dramatically decreased amphipod mortality due to oil or emulsion exposure by 70%. Finally, amphipods that were exposed to 0.1 ppt oil or 0.105 ppt emulsion and then transferred to clean 100% SW exhibited mortality comparable to that of animals continuously exposed to pollutants. This finding indicates that, unlike the case for killifish, the initial exposure period is critical in determining the ultimate effects of these pollutants on G. oceanicus. Control amphipods generally displayed bursts of normal swimming activity, alternating with stationary periods on the bottom: at this time, their pleopods remained in continuous motion. The number of control pairs engaged in the precopulatory clasp generally decreased over the course of the time period, and all animals displayed a vigorous startle response to vibration of the holding container at all time periods. Compared with control groups, the percentage of amphipods
OIL, DISPERSANT,
AND EMULSION
TOXICITY
271
0 CONTROL
OIL
DISPERSANT
EMVLSION
FIG. 3. Activity scores (expressed as the mean number of lo-cm gridlines crossed by each fish in 15 set) for killifish exposed to No. 2 fuel oil, AP dispersant, and emulsions for 48 hr under standard conditions. Significant differences for I test comparisons between each treatment group and control animals for each time period are indicated with asterisks.
engaged in precopulatory behavior in all oil treatments, all emulsion treatments, and one dispersant treatment (0.05 ppt) decreased significantly in a time- and dose-dependent manner (Fig. 6). Animals continuously exposed to the lowest oil level (0.01 ppt) displayed significantly more precopulatory behavior than the corresponding emulsion (0.0105 ppt) group (Fig. 6). Dispersant did not appear to affect swimming activity; however, exposure to lower levels of oil and emulsions caused a biphasic response in activity. Oil levels of 0.1 and 0.01 ppt and emulsion levels of 0.105 and 0.0105 ppt resulted in hyperactivity after 1 hr, but activity was significantly reduced by 6 hr. Animals exposed to higher levels of pollutants were significantly less active than controls at 1 hr (Fig. 7); in contrast to the effects of oil exposure, all three emulsion levels reduced amphipod activity significantly by 6 hr (Fig. 7). Thus, emulsification of 0.01 ppt oil with 0.0005 ppt dispersant increased toxic effects on amphipod behavior. Behavioral patterns of animals exposed to lethal levels of oil and emulsion treatments were characterized by erratic swimming followed by gradual cessation of normal limb movement. At this stage, amphipods responded to vibration of the holding pan with only slight movement of
272
BUTLER,
S
M
B
TRIVELPIECE,
5
CONTROL
M
B
OIL
AND MILLER
S
M
DISPERSANT
B
s
M
B
EMULSION
4. Depth preference index (mean percentage of fish at the surface, S: swimming in midtank, M: or located on the bottom, B) for killifish exposed to No. 2 fuel oil, AP dispersant, and emulsions for 48 hr under standard conditions. Significant differences for x2 comparisons between each treatment group and control animals are indicated with asterisks. FIG.
the antennae or limbs. In contrast, response to the same stimulus.
controls
engaged in vigorous
swimming
in
DISCUSSION
The lethal effects of petroleum hydrocarbons on marine organisms have been the subject of numerous studies, which have been reviewed previously (e.g., see Anderson& al., 1974; Anderson 1977). However, few investigators have addressed the question of changes in toxicity due to emulsification by dispersants. Methodological, environmental, and/or species-specific differences often result in different LCSO values for the same species studied in two different laboratories. Therefore, comparative studies of two or more species often produce more useful
OIL, DISPERSANT,
I
AND EMULSION
24
6 HOURS
OF
273
TOXICITY
36
EXPOSURE
FIG. 5. Effects of different levels of No. 2 fuel oil. AP dispersant, and emulsions on the survival of amphipods under standard conditions. All control animals (n = 58) survived to 36 hr. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
insights into pollutant-organism interactions. In the present study of F. heteroclitus and G. oceanicus, our results indicate that AP dispersant potentiates the effects of No. 2 fuel oil in a static system. Mortality data indicated that killitish were clearly more sensitive to emulsions than to oil or dispersant alone (Fig. 1). In contrast, amphipods exhibited similar mortality in both oil and emulsion treatments, but demonstrated differential sensitivity to the treatments behaviorally (e.g., precopulatory clasp; Fig. 6). Both killifish and amphipods responded similarly with respect to thermal stress (i.e., generally decreased mortality in all treatment groups with decreased temperature; Fig. 2). However, only the killitish demonstrated a differential response to pollutants in salinities that were hypo- or hypertonic. Survival was significantly greater in salinities that were isosmotic with killifish body fluids. This is not surprising, since the killitish is an osmoregulator and the amphipod an osmoconformer. Effects of the interaction of temperature, salinity, and oil contamination on development of F. heterocfitus embryos have also been reported (Linden et al., 1979). The lethal effects of No. 2 fuel oil on killifish embryos were increased by temperatures above or below 25”C, especially under conditions of suboptimal salinity. Killifish in the present study also exhibited complete recovery following short-term exposures to all three pollutants, while amphipods did not recover after comparable exposures. Finally, amphipods
274
BUTLER,
1
TRIVELPIECE,
AND MILLER
6
24 HOURS
OF
36
EXPOSURE
FIG. 6. Effects of different levels of No. 2 fuel oil, AP dispersant.and emulsions on the precopulatory behavior of amphipods under standard conditions. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
were more sensitive to oil, but far less sensitive to dispersant than killifish. Results of the present study support earlier findings by Swedmark ef al. (1973) that indicated that oiYdispersant emulsions were more toxic to marine organisms than either pollutant alone, and that fish are oftentimes more sensitive to these pollutants than crustaceans. Lethality data alone are inherently conservative in their predictions of the impact of petroleum on aquatic ecosystems. Sublethal doses of oil may affect the behavior of marine organisms in a manner that may reduce survival of the individual or reproduction. Locomotion in the snail (Littorina fittorea) was significantly altered by sublethal concentrations of oil (Hargraves and Newcombe, 1973), and chemoreception, essential to mate and food location by many marine invertebrates, may be impaired by petroleum hydrocarbons (Kittridge 1973; Atema and Stein, 1974). Deterioration of feeding response and food retention has also been noted in the sea cattish (Ariusfelis) exposed to fuel oil (Wang and Nicol, 1977). Linden (1976) demonstrated that Venezuelan crude oil significantly decreased reproduction in an amphipod (Gammarus oceanicus) due to the decreased frequency with which males and females engaged in precopulatory behavior. A study of the long-term effects on an oil spill on populations of the salt marsh crab (Uca pugnax) indicated significant alterations in locomotory ability and burrow construction (Krebs and Burns, 1977).
OIL, DISPERSANT,
I
AND EMULSION
24
6
275
TOXICITY
36
z HOURS
OF
EXPOSURE
FIG. 7. Effects of different levels of No. 2 fuel oil, AP dispersant, and emulsions on the activity of amphipods under standard conditions. Significant differences for x2 comparisons between each treatment group and control animals for each time period are indicated with asterisks.
The results of the behavioral phase of the present study did reveal some quantitative, but not qualitative, changes in response to pollutant levels. At concentrations that were two orders of magnitude below lethal levels, killifish demonstrated changes in behavioral patterns (i.e., activity and depth preference), suggesting that behaviors essential for survival at both the individual and population levels (e.g., feeding, antipredator, and reproductive behavior) may be impaired with exposure to these pollutants in the wild. More importantly, the amphipod data clearly indicate that in the case of the precopulatory clasp, dispersant potentiates oil effects on the reproductive behavior of this species (Fig. 6). Recently, amphipod species diversity has been shown to be inversely correlated with the degree of marine pollution (Bellan-Santini, 1980). These results underscore the potential ecological impact that even sublethal levels of petroleum hydrocarbons may have on aquatic organisms. These data are also important in light of recent research indicating that physiological stress due to water pollution may be correlated with increased disease in fish populations (Brown et al., 1979). Finally, the present study emphasizes the need for caution in the use of commercial dispersants at the sites of major spills, especially in inshore waters.
276
BUTLER,
TRIVELPIECE,
AND MILLER
ACKNOWLEDGMENTS This research was supported by NIH Grants ES-00920 and S07-RR-05764 and NSF Grant DEB 7826821. We thank L. Baldwin and S. Dion for excellent technical assistance.
REFERENCES Anderson, J. W. (1977). Effects of petroleum hydrocarbons on the growth of marine organisms. Rapp. P. V. Rertn. Cons. Inf. Explor. Mer. 157-165. Anderson, J. W., Neff, J. M.. Cox, B. A., Tatem, H. E.. and Hightower, G. M. (1974). Characteristics of dispersions and water soluble extracts of crude and refined oils and their toxicity to estuarine crustaceans and fish. Mar. Biol. 27, 75-88. Atema, J., and Stein, L. S. (1974). Effects of crude oil on the feeding behavior of the lobster Hormms americanus.
Environ.
Pollut.
6, 77-86.
Barnett, J., and Toews, D. (1978). The effects of crude oil and the dispersant, Oilsperse 43, on respiration and coughing rates in Atlantic salmon (S&no s&u). Canad. J. Zoo/. 56, 307-310. Bellan-Santini, D. (1980). Relationship between population of amphipods and pollution. Mur. Po//ut. Bull.
11, 224-227.
Brown, E. R., Koch, E., Sinclair, T. F., Spitzer, R., and Callaghan. 0. (1979). Water pollution and diseases in fish (an epizootiologic survey). J. Emiron. P&ho/. Tuxicol. 2, 917-925. Hargraves, B. T., and Newcombe, C. P. (1973). Crawling and respiration as indices of sub-lethal effects of oil and a dispersant on an intertidal snail Littorina littorea. J. Fish. Res. Board Canad. 30, 1789- 1792. Kittridge, J. S. (1973). The effects of crude oil pollution on the behavior of marine invertebrates. Find Rep. U.S. Govt. Ad-762 047. Krebs, C. T., and Burns, K. A. (1977). Long-term effects of an oil spill on populations of the saltmarsh crab, Uca pugnax. Science 197, 484-487. Linden, 0. (1976). Effects of oil on the reproduction of the amphipod Gamnmrus oceanicus. Arnbio 5, 36-37. Linden, O., Sharp, J. R., Laughlin, R., Jr., and Neff, J. M. (1979). Interactive effects of salinity, temperature and chronic exposure to oil on the survival and development rate of embryos of the estuarine killifish Fundtrlus heteroclitus. Mar. Biol. 51, 101- 109. Peakall, D. B., Hallet, D. J., Miller, D. S., Butler, R. G., and Kinter, W. B. (1980). Effects of ingested crude oil on Black Guillemots: A combined field and laboratory study. Atnbio 9, 28-30. Swedmark, M., Granmo, A., and Kollberg, S. (1973). Effects of oil dispersants and oil emulsions on marine animals. Water Res. 7, 1649- 1672. Wang, R. T., and Nicol, J. A. C. (1977). Effects of fuel oil on sea catfish: Feeding activity and cardiac responses. Bull. Environ. Contmn. Toxicol. 18, 170- 176.