A spill of weathered Arabian Light crude oil (3.8 l.m-2) on an intertidal sand zone was simulated in the laboratory. Respirometry, chemical and microbiological methods were employed to assess the effectiveness of two fertilizers: a slow-release inorganic (Max Bac), and an oleophilic organic (Inipol EAP22). Inipol EAP22 addition caused a stimulation of CO2 evolution, increased total chemoheterotrophs, and stimulated the proportion of hydrocarbon-degraders. At the end of the experiment, residual oiled sand treated with the oleophilic fertiliser had higher pristane:C17 and phytane:C18 ratios and lower total hydrocarbon concentration than the control. Max Bac had little effect. The results suggested that Inipol EAP22 stimulated the hydrocarbon-degrading microbial population, and after a lag phase encouraged oil biodegradation in fine sandy sediments subjected to a vertical tidal cycle. In contrast Max Bac had little impact on oil biodegradation.
INTRODUCTION
The bioremediation agents Inipol EAP22 and "Customblen" were shown to stimulate oil biodegradation on cobble beaches in Alaska after the Exxon Valdez incident (Pritchard & Costa, 1991; Bragg et al. 1994). This has led to the suggestion that research in bioremediation now needs to be focused on operational aspects of the technology (Swannell & Head, 1994). The aim of our research was to compare the use of a slow-release inorganic fertiliser (Max Bac - a product derived from the "Customblen" used in Alaska by Grace-Sierra Chemicals)(Grace-Sierra, 1993) with an oleophilic organic fertiliser (Inipol EAP22, produced by Elf, France)(Sirvins & Angles, 1986) to stimulate the biodegradation of Arabian Light crude oil on columns of medium-fine sand.
The sand used was a quartz sand from a beach (Long Cove, Nova Scotia, Canada) previously used for bioremediation trials (Lee & Levy, 1987; Lee & Levy, 1989; Lee & Levy, 1991; Lee & Levy, 1992, Lee et al., 1993). Lee and Levy (1987) noted that a single application of Inipol EAP22 failed to stimulate oil biodegradation, due to the rapid removal of the oleophilic fertilizer by tidal action. The failure of Inipol EAP 22 to stimulate oil biodegradation on fine sediments was also noted by Sveum & Ladousse (1989). Safferman (1991) recorded that Inipol EAP22 was rapidly removed from oiled cobble in the laboratory. To examine the discrepancy between the results noted in Alaska and these earlier field observations of effects of fertilisers, cores of sand were subjected to a tidal cycle in the laboratory, to assess the significance of vertical water movement on the efficacy of the selected bioremediation agents.
EXPERIMENTAL PROCEDURES AND MATERIALS
Column Operation and Oil Addition.
Cores of quartz beach sand 30 cm in length were obtained from Long Cove, Nova Scotia, Canada and shipped to the UK at 4oC. On arrival the cores were sieved and washed with synthetic sea water (Instant Ocean, Aquarium Systems Inc.), and placed in 3 columns (glass, 0.32 m high, 0.1 m internal diameter, containing 1.3 l of sand). These were washed in 14 tidal cycles before addition of oil and fertiliser. Peristaltic pumps were used to simulate a tidal cycle (maximum bulk flow velocity of 33 cm.h-1 matching peak spring tides at Long Cove). Each core was covered with Arabian Light Crude oil at 3.8 l.m-2 (weathered naturally to 25% weight loss, and emulsified with 25% seawater) just as the water level was falling after high tide. Inipol EAP22 was added to one core at 984 g.m-2, and Max Bac was added to a second core at 266 g.m-2 (to provide the equivalent amount of nitrogen) 6 hours after oil-addition. The third column remained as an oiled but unfertilised control. The experiment was carried out at 21 ± 2 oC.
Sampling Strategy
Observations lasted 44 days. CO2 evolution rate was assessed daily. To avoid the effects of uneven oil penetration, the outer 1 cm and the inner 4 cm diameter area of each core were not sampled. To avoid sampling bias, 18 potential sample points were mapped out and a random number generator was used to select 3 points from each column each sampling day. The first sample was taken 1 h after the addition of the oil, before nutrient-addition.
Monitoring
CO2 evolution was monitored daily to assess respiration rate. The headspace above each column was flushed with soda lime-filtered air for 5 minutes to reduce the levels of CO2 in the headspace to ambient levels. The headspace air was then recycled through an Infra Red Gas Analyser (Servomex, UK) and readings of CO2 concentration were taken regularly over a 5 min period. The mean and average CO2 production rate were calculated by linear regression. Readings were taken whenever possible one hour after high tide, since respiration was observed to vary through the tidal cycle.
Most Probable Number (MPN) determinations were used on 6 sampling occasions to estimate the total number of chemoheterotrophic micro-organisms and the number of organisms capable of growth on the Arabian Light crude oil (Meynell & Meynell, 1970). The former was carried out by studying growth on marine broth (Difco) and the latter was conducted using a modified 'Sheen Screen' method described by Venosa et al. (1993).
Triplicate sediment samples removed on each sampling occasion from each column were pooled, the oil extracted and the extracts analysed using gas chromatography with flame ionisation detection, according to the method given in Lee et al. (1993).
RESULTS AND DISCUSSION
CO2 Evolution
The evolution rates of CO2 from the control core increased approximately eleven-fold over the first 5 days following oil addition, thereafter achieving steady-state (Figure 1). The cores treated with Max Bac gave similar evolution rates until day 15, when a small increase in evolution rate over the control was detected (Figure 1). The pattern of CO2 evolution with oleophilic fertiliser was markedly different. By three days after Inipol EAP22 addition, CO2 evolution had increased above that noted in the other cores, peaked at day 5 and then declined, but maintained moderately high activity (approximately twice that in the control) until the end of the experiment (Figure 1). The elevated values recorded on days 15 and 41 were probably due to measurements being taken 3 h later than normal on those days.
Estimates of Microbial Numbers
Triplicate MPN determinations were performed, giving errors typically of <50%, so differences of less than half an order are not significant (Figure 2). The three columns showed numbers of chemoheterotrophic organisms of the same order at day 0 (105 to 106 per gram of sand), of which 5 x 103 to 5 x 104 (approximately 5%) were oil-degrading organisms (Figure 2). The source beach in Nova Scotia is regarded as uncontaminated by oil, so it was interesting to note the presence of a small population of natural oil-degraders.
All columns showed a rapid increase in both total chemoheterotrophic and oil-degrading organisms following addition of oil at day 0 (Figure 2). By day 5 both the control and Max Bac columns had similar populations and were approaching a plateau at new levels, achieving a mean of 5 x 107 chemoheterotrophs, of which an average of 2.5 x 107 were oil-degrading (approximately 50%). A slight increase was noted by day 13, but for the remainder of the experiment these numbers and the ratio of oil-degraders remained relatively constant, with the control and Max Bac-treated cores virtually indistinguishable (Figure 2).
The Inipol-treated column showed a different pattern (Figure 2). By day 5 total chemo-heterotrophs had achieved 8 x 108 per gram, and virtually 100% of the population were oil-degraders. This peak coincided with peak headspace CO2 generation rate (Figure 1). Thereafter numbers declined slowly, and at the end of observations (day 41) there was a mean of 2.5 x 108 chemoheterotrophs, of which, on average, 108 had oil-degrading ability (approximately 40%).
Examination of effluents at day 13 showed both the Control and the Maxbac column with 106 oil-degraders per ml, while the Inipol EAP22 column showed more than 108 per ml, two orders of magnitude higher. This was associated with a period of maximum turbidity and yellow discolouration of the Inipol-treated column effluent.
Chemical analysis
Using the changes in the pristane:C17 and phytane:C18 ratios as an indicator of oil degradation, greatest alkane biodegradation occurred in the core treated with Inipol EAP22 (Figs. 3 and 4). There appeared to be a 5-7 day lag before crude oil composition changed in this column, so changes in respiration and microbial numbers in this period are probably related largely to preferential biodegradation of organic components of the Inipol. Enhanced crude oil decomposition is noted from 13 days and the degree of overall stimulation (as measured by the percentage change in the conserved marker to alkane ratio over the 41 day period) was approximately 25 % using both markers. The control and Max Bac columns showed only approximately a 10% change in both marker:alkane ratios over the same period.
The control, Max Bac and Inipol-treated columns showed respectively 12,910, 12,860 and 7,450 µg hydrocarbon per g sand at day 41. A proportion of the Inipol column losses will have been due to physical mobilisation. There was little difference in the spectrum of residual alkanes at day 41; lighter alkanes (C12-C17) were degraded slightly more readily with Inipol, leaving the sand relatively enriched with higher hydrocarbons (C18-C23). At still higher molecular weight (C24-C35) the columns show no difference (Figure 5).
CONCLUSIONS
Inipol EAP 22 appeared to be the most effective bioremediation agent on the basis of microbiological and chemical data. The total CO2 evolution rate in the Inipol-treated column increased to approximately 5 times the control at day 5, then fell gradually to plateau at approximately twice the control from day 15 to day 44. Total chemoheterotrophs increased from 105 to 108.g-1 sand by day 5, with the HC-degraders increasing more rapidly from 5% to virtually 100%. By day 41, with a reduction in available hydrocarbon, particularly lighter oil components, and probable increase of ecological successors, total chemoheterotrophs had fallen to 107.g-1, with 40% HC-degraders. The residual oil showed depletion of C17 (relative to conserved marker pristane) and C18 (relative to phytane) at more than twice the rate of the control (approximately 25% compared to 10% change over 41 days), and disappearance (biodegradation plus wash-out) of nearly 50% more total hydrocarbon relative to the control.
The effects of Inipol EAP22 are complex. Visually the Inipol-treated column differed from the control core, showing greater depth penetration of oil and more mobilisation of oil components (free intergranular oil globules in the sand and a discoloured effluent were noted), suggesting that Inipol EAP22 has physical and chemical effects on the oil (also noted by Ladousse & Tramier, 1991). The early peak in respiration may have been due more to degradation of the oleophilic components of Inipol EAP22 than of crude oil, a phenomenon noted by other workers (Lee and Levy, 1989; Rivet et al., 1993). This hypothesis is supported by the oil composition data which suggest possible initial inhibition or lack of crude oil biodegradation, and no net positive effects until approximately day 13 (Figure 3 & 4). However, the addition of Inipol EAP22 clearly enhanced the number of hydrocarbon-degrading micro-organisms (Figure 2).
The evidence therefore suggests that organic components of Inipol EAP 22 (e.g. oleic acid) may initially be preferentially decomposed with a concommitant release of CO2, and this may result in a rapid early increase in the total number of chemoheterotrophs and the proportion of competent hydrocarbon-degrading micro-organisms. However, enhanced crude oil biodegradation is not detectable chemically until several days after the treatment, by which time in the field Inipol may have been washed away. These data concur with the hypothesis proposed by Basseres & Ladousse (1992) that Inipol EAP22 initially stimulates the hydrocarbon-degrading microbial population, subsequently facilitating hydrocarbon decomposition.
The sand core amended with inorganic slow-release fertiliser was barely distinguishable from the control over the course of the experiment. Only a very gradual rise in total respiration rate, to 1.5 times control at day 44 was noted (Fig. 1). The pristane:C17 ratio had departed slightly from the control by day 41 (Fig. 3). Otherwise Max Bac and the control were very similar; both increased gently to a plateau of total respiration rate at 5-10 days, both showed total chemoheterotrophs increasing from 105 to 107.g-1 with oil-addition, both showed an increase in HC-degraders to virtually 100% by day 5, with a slow decline in total numbers and the proportion of HC-degraders thereafter. Both showed a modest change in phytane:C18 ratio of order 10% over 41 days. Visual observations of the Max Bac-treated column confirmed that the slow-release pellets were dissolving very slowly, many persisting to termination of the experiment.
Under the simulated conditions of the experiment, with vertical flow but no horizontal flow or wash-off, Inipol EAP22 was the most effective bioremediation agent, based on residual oil concentrations at day 41 and relatively enhanced disappearance of alkanes relative to conserved isoprenoid internal markers. This result apparently contrasts with that reported by Lee & Levy (1987) in field evaluations of Inipol EAP22 with the same sand in the field at Long Cove, Nova Scotia, where Inipol had no detectable stimulatory effect on oil biodegradation. These authors noted a rapid decline in nitrogen concentrations in the sand treated with Inipol EAP22, supporting their hypothesis that the fertiliser was being washed from the sand, a finding also supported by Safferman (1991).
We propose that it is the horizontal component of the sea motion (wave energy, surface and sub-surface run-off, effects of rainfall) which may be significant factors influencing the persistence and efficacy of Inipol EAP22 in sand. The absence of horizontal (especially wave) energy may also possibly explain the relative ineffectiveness of Max Bac addition in the laboratory, as the greater energy in situ may stimulate pellet disruption and nutrient release.
Our research points to the importance of the interactions between the physical and chemical form of bioremediation agents and the chemical and physical processes occurring in the environment, and suggests that although microcosm-scale studies are a valuable method of studying the action of bioremediation agents, research must be validated in mesocosms or in the field.
ACKNOWLEDGEMENTS
This work was funded by the Marine Pollution Control Unit (Coastguard Agency) of the UK Department of Transport, and by the Canadian Department of Fisheries and Oceans.
REFERENCES
Basseres A. and A. Ladousse. 1992. "Experience in Enhanced
Bioremediation Processes on Shorelines." In CONCAWE/DGMK
Symposium "Remediation of Oil Spills" Hamburg 18-21
May 1992. CONCAWE, Brussels, Belgium.
Bragg J. R., R. C. Prince, E. J. Harner, and R. M. Atlas. 1994.
"Effectiveness of bioremediation for the Exxon Valdez oil
spill." Nature 368 (No.6470): 413-418.
Grace-Sierra International B.V. 1993. Max Bac. Controlled
Release Nutrient Package. Rijnzathe 6, 3454 PV De Meern,
Netherlands.
Ladousse A., and B. Tramier. 1991. Results of 12 years of research
in spilled oil bioremediation: INIPOL EAP 22. Proceedings of
1991 International Oil Spill Conference. pp. 577-581. American
Petroleum Institute Pub. No. 4529, Washington DC, USA.
Lee K. and E. M. Levy. 1987. "Enhanced biodegradation of
a light crude oil in sandy beaches." Proceedings of 1987
International Oil Spill Conference. pp. 411-416. American
Petroleum Institute Pub. No. 4452., Washington DC, USA.
Lee K., and E. M. Levy. 1989. "Enhancement of the natural
biodegradation of condensate and crude oil on beaches of Atlantic
Canada." Proceedings of 1989 International Oil Spill
Conference. pp. 479- 486. American Petroleum Institute
Pub. No. 4479., Washington DC, USA.
Lee K., and E. M. Levy. 1991. "Bioremediation: Waxy Crude
Oils stranded on Low-Energy Shorelines." Proceedings
of 1991 International Oil Spill Conference. pp. 541-547. American
Petroleum Institute Pub. No. 4529, Washington DC, USA.
Lee K., and E. M. Levy. 1992. "Microbial degradation of
petroleum in an intertidal beach environment - in situ
sediment enclosures study. In Marine Ecosystem Enclosed Experiments.
pp. 140-155. International Development Centre, Ottawa, Canada.
Lee K., G. H. Tremblay, and E. M. Levy. 1993. "Bioremediation:
Application of slow release fertilisers on low-energy shorelines."
Proceedings of 1993 International Oil Spill Conference (Prevention,
Preparedness, Response). pp. 449-454. American Petroleum
Institute Pub. No. 4580. Washington DC, USA.
Meynell, G. G. and E. Meynell. 1970. Theory and practice
in experimental bacteriology. 2nd ed., Cambridge University
Press, Cambridge, UK.
Pritchard P. H., and C. F. Costa. 1991. "EPA's Alaska Oil
Spill Bioremediation Project." Environ. Sci. Technol.
25: 372-379.
Rivet, L., G. Mille, A. Basseres, A. Ladousse, C. Gerin, M. Acquavia,
and J-C Bertrand. 1993. "n-Alkane biodegradation by a marine
bacterium in the presence of an oleophilic nutriment." Biotechnol.
Lett. 15 (6): 637-640.
Sirvins, A., and M. Angles. 1986. "Development and effects
on the marine environment of a nutrient formula to control pollution
by petroleum hydrocarbons." In C. S. Giam and H. J. M. Dou
(Eds), Strategies and Advanced Techniques for Marine Pollution
Studies: Mediterranean Sea, pp.357-404. Springer-Verlag, Berlin,
Germany, NATO ASI Series, Vol. G9.
Sufferman, S.I. 1991. "Selection of nutrients to enhance
biodegradation for the remediation of oil spilled on beaches."
Proceedings of the 1991 Oil Spill conference: 571-576.
American Petroleum Institute publication no: 4529; Washington
D.C., USA.
Sveum P., and A. Ladousse. 1989. "Biodegradation of the
oil in the Arctic: Enhancement by Oil-Soluble fertiliser application."
Proceedings of 1989 International Oil Spill Conference.
pp. 436-446. American Petroleum Institute Pub. No. 4529., Washington
DC, USA.
Swannell R. P. J., and I. M. Head. 1994. "Bioremediation
comes of age." Nature 368 (No.6470): 396-397.
Venosa A. D., M. Kadkhodayan, D. W. King, B. A. Wrenn, J. R. Haines,
T. Herrington, K. Strohmeier, and M. T. Suidan. 1993. "Testing
the Efficacy of Oil Spill Bioremediation Products." In Proceedings
of 1993 International Oil Spill Conference (Prevention, Preparedness,
Response, pp. 487-493. American Petroleum Institute Pub. No.
4580., Washington DC, USA.
FIGURES
Figure 1. Absolute carbon dioxide evolution rates over the course of the Experiment.
Figure 2. Heterotrophic(H) and Oil-degrading(O) organisms measured using the MPN technique.
Figure 3. The Change in Pristane:C17 of Arabian Light Crude Oil during the experiment.
Figure 4. The Change in Phytane:C18 of Arabian Light Crude Oil during the experiment.
Figure 5. Comparison of the alkane spectra of crude oil and control, Max Bac and Inipol-treated columns at day 41.
