Dec 9, 1974 - University of Southern Mississippi, Hattiesburg, Miss. 39401. aromatic ..... land Resources, Louisiana State University, Baton. Rouge. 17.
APPLIED MICROBIOLOGY, May 1975, p. 646-652 Copyright 0 1975 American Society for Microbiology
Vol. 29, No. 5 Printed in USA.
Persistence and Biodegradation of Spilled Residual Fuel Oil on an Estuarine Beach1 R. H. PIERCE, JR.,2 A. M. CUNDELL, AND R. W. TRAXLER* Department of Plant Pathology-Entomology, University of Rhode Island, Kingston, Rhode Island 02881 Received for publication 9 December 1974
The enrichment of hydrocarbon-degrading bacteria and the persistence of petroleum hydrocarbons on an estuarine beach after a spill of residual fuel oil on 11 April 1973 in Upper Narragansett Bay, R. I. was investigated. A rapid enrichment occurred during days 4 to 16 after the oil spill and a significant population of hydrocarbon-degrading bacteria was maintained in the beach sand for at least a year. The concentration of petroleum hydrocarbons in the mid-tide area declined rapidly during the bacterial enrichment period, remained fairly constant throughout the summer, and then declined to a low concentration after 1 year. An increased concentration of branched and cyclic aliphatic hydrocarbons in the low-tide sediment 128 days after the spill suggested a migration of hydrocarbons during the summer. Hydrocarbon biodegradation was apparent during the winter months at a rate of less than 1 Ag of hydrocarbon per g of dry sediment per day.
On the night of 11 April 1973, the Liberian Tanker, Pennant, hit an underwater obstruction off Popasquash Point, upper Narragansett Bay, R. I. and an estimated 90,000 gallons (340,650 liters) (approximately 320 short tons) of no. 6 fuel oil were lost. An eyewitness reported that a portion of residual fuel oil came ashore at 7:30 a.m. the following day at Gaspee Point, Warwick, and when the tide receded, 3 to 4 inches (7.62 to 10.16 cm) of fuel oil covered the high-tide area. A local oil clean-up company removed the stranded oil from the beach on the morning of 13 April (day 2 after the spill). The cold weather at the time of the spill enabled clean-up personnel to mechanically remove the stranded congealed fuel oil with minimal amounts of sand. No detergents were used to treat the spill area. Petroleum hydrocarbons persist in coastal sediment even though large populations of hydrocarbon-degrading bacteria exist within these sediments (5, 9, 17, 21). The chemical analyses undertaken by Blumer and Sass (2-4) in an investigation of the fate of no. 2 fuel oil from a spill in Buzzards Bay, Mass. showed that the petroleum hydrocarbon disappearance was slow and appeared to be facilitated by the bacterial degradation of normal aliphatic hydrocarbons and by the partial dissolution of lower-boiling ' This paper is contribution no. 1575 of the Rhode Island
Agricultural Experiment Station. 2Present address: Institute of Environmental Science, University of Southern Mississippi, Hattiesburg, Miss. 39401.
aromatic hydrocarbons. In this paper we describe chemical and bacteriological analyses of the oil-polluted beach sediment from Gaspee Point, R. I., collected at specific intervals for a period of 1 year. MATERIALS AND METHODS Collection of samples. Samples of beach sand from the top 2 to 4 inches of the sediment in the mid-tide and low-tide areas were collected during low water on days 2, 4, 8, 16, 32, 64, 128, 190, 250, 310, and 370 after the oil spill in sterile 250-ml Erlenmeyer flasks, sealed with aluminum foil, and then conveyed to the laboratory in an ice chest. The pattern of sample collection at progressively longer intervals of time made it necessary to use a geometric time scale to present the data for bacterial numbers and hydrocarbon concentration in the sediments. Larger samples were collected on the day 64 and in subsequent collections to provide more materials for chemical analysis. These additional samples were stored at -20 C before chemical analysis. Bacteriological analyses. All the glassware, dilution bottles, and media were prechilled to prevent the thermal death of psychrophilic bacteria (12). With use of the spread plate technique, nutrient seawater agar plates (NaCl, 21 g; Rila marine mix, 5 g; polypeptone [BBL], 5 g; yeast extract [Difco], 3 g; glucose, 1 g; succinic acid, 0.2 g; and granulated agar, 16 g in 1,000 ml of distilled water) were inoculated with 0.1 ml of the diluted samples (11). The inoculated plates were incubated for 2 to 14 days at the ambient water temperature and the number of colonies growing on triplicate plates were determined. Plates from each dilution series containing 30 to 300 colonies were selected and replicate transfers were 646
VOL. 29,1975
BIODEGRADATION OF SPILLED FUEL OIL
647
made onto plates of basal Rila marine salts media sediments showed no definite trend over the 370 supplemented with phosphorus and nitrogen (Rila days of this investigation. There were fluctuamarine salts, 28 g; NH4Cl, 20 mg; KH2PO4, 33 mg; tions in the bacterial population in both the low KHPO4, 67 mg; Noble agar, 16 g in 1,000 ml of and mid-tide sediments; the total number of distilled water) containing 1% (wt/vol) of naph- bacteria ranged between 3.3 x 103 and 3.2 x 105 thalene, n-dodecane, trans-decalin, or no. 1 fuel oil. organisms per g of wet sediment (Fig. 1). The proportion of the total heterotrophic bacterial Whereas there was no significant change in population capable of growing at the expense of the petroleum hydrocarbons was determined from these bacterial numbers after the introduction of replicate plates after 7 days of incubation at the additional organic carbon, namely the residual ambient water temperature (17). Colonies showing fuel oil, the proportion of hydrocarbon-degradfeeble growth or obviously hydrolyzing the agar were ing bacteria in the total population increased. discounted. Estimates of the percentage of hydrocarRapid enrichment of hydrocarbon-degrading bon degraders are not absolute and were made on the bacteria occurred during days 4 to 16. By the basis of experience gained in the screening of bacterial day 16 after the spill, the enrichment was isolates on hydrocarbons in solid media and compared complete in that almost all of the colonies to control agar plates without added hydrocarbons isolated on the nutrient seawater agar plates (15). To demonstrate the seasonal selection of popula- grew on the petroleum hydrocarbons (Fig. 2 and tions of mesophilic and psychrophilic hydrocarbon- 3). In addition to the selective pressure of the degrading bacteria, the biological oxygen demand of a hydrocarbons, the environmental temperature, residual fuel oil was measured (10). Using this tech- which increased from 6 to 25.5 C during the first nique, the relative rate of hydrocarbon biodegradation 94 days after the spill, will influence the bacteduring the winter (sediment temperature 2 C) and rial population (Fig. 4). As the temperature summer (sediment temperature 22 C) can be deter- increased to about 12 C, the predominantly mined. To each 300-ml biological oxygen demand (BOD) bottle was added 0.5 g of low-tide sedi- psychrophilic bacterial population in Narraganment and 0.1% (vol/vol) of no. 4 fuel oil, and they sett Bay was replaced by a mesophilic populawere filled with the overlying sea water. The ground tion (13). Further moblilization of petroleum glass-stoppered bottles were incubated in the dark hydrocarbons from nearby sediments induced in controlled environment incubator shakers (Psy- by the increased temperature possibly caused chrotherm, New Brunswick Scientific) at 5, 15, the secondary enrichment of a population of and 25 C and shaken at 120 rpm, and the rate of mesophilic hydrocarbon-degrading bacteria by oxygen depletion in the bottles was measured after the day 64 after the spill (Fig. 2 and 3). 1, 2, 4, and 6 days of incubation using the azide The pattern of response to incubation temmodification of the Winkler method (1). of the biological oxygen demand meaperature and Chemical analysis. Samples were air-dried passed through a 2-mm sieve to remove any small 106 pebbles. Lipoidal material was extracted by refluxing the sediment samples with a solution of benzene, 0.1 N, KOH-methanol, 60:40, (vol/vol) (6). The solution A A IL-A was washed with distilled water and the organic 105 phase, containing non-saponifiable lipoidal material, KA A was taken to dryness under vacuum at 25 C. The residue was washed onto a column of alumina over 0-0 A~~~~~~~~~" silica gel to isolate hydrocarbons from lipid extracts 04 (4). Aliphatic hydrocarbons were eluted from the 0~~~~~ column with petroleum ether and aromatic hydrocar0 bons were eluted with a solution of benzeneu petroleum ether, 50:50 (vol/vol). 1.-0 The hydrocarbon fractions were concentrated to 1-ml volume in petroleum ether and analyzed by gas -102 chromatography. The concentration of hydrocarbons the in the sediment was determined by comparing area under the chromatogram with the area of the internal standard hydrocarbons n-C22 and n-C82. Samples were analyzed using a Varian model 1800 gas chromatograph with a flame ionization detector. Each 0' ) 0 64 128 256 4 8 16 32 2 sample was analyzed by using two columns, 3% SP-2100 on Supelocoport and 12% FFAP on Days after spill Chromosorb W. The temperature program was from 100 to 275 C at 6 C per min. FIG. 1. Total plate counts determined on nutrient seawater agar plates incubated at the ambient sediRESULTS ment temperature. Symbols: low-tide sediment (0), The number of heterotropic bacteria in the mid-tide sediment (A).
~~A
D
648
PIERCE, CUNDELL, AND TRAXLER
APPL. MICROBIOL.
,00
80
0
2
c'*020 20
2
4
8
16
32
64
128
256
Days after spill
100
C
80 so 0
40.
A
A-"
Cu 4 °
20
20
-1 _
0
2
4
8
16
32
64
128
256
Days after spill
0
2
4
8
16
32
64
128
25o
Doys after sp.I1
FIG. 2. Percentage of hydrocarbon-degrading bacteria in the total plate count of the heterotrophic bacteria in the mid-tide sediment determined by a replica plating technique. Substrates: (A) n-dodecane; (B) trans-decalin; (C) naphthalene; and (D) no. 1 fuel oil.
surements, using residual fuel oil as a substrate and sediment and the overlying seawater collected during the summer and winter as an inoculum, indicated a seasonal selection of hydrocarbon-degrading bacteria with respect to temperature. During the winter months, a population of psychrophilic bacteria are contained in the sediment that are capable of degrading petroleum at 5 C (Fig. 5); however, they were absent from the sediment during the summer months (Fig. 6). A temperature limitation on hydrocarbon biodegradation was apparent. Mesophilic hydrocarbon-degrading bacteria are present in the sediment during both summer and winter. The rate of oxygen depletion was curtailed at 15 C when the BOD bottles are inoculated with sediment and sea water collected in the summer. This suggests that when the ambient temperature decreased below 15 C, the biodegradation of petroleum hydrocarbons by mesophilic bacteria was reduced, and there was an emergence of a population of psychrophilic bacteria which contribute to the biodegradation.
The seasonal selection of psychrophilic and mesophilic bacteria within the sediment led to a temporary reduction in the percentage of hydrocarbon-degrading bacteria in the total heterotrophic bacterial population. This phenomenon occurred on days 32 and 190 after the oil spill when the ambient temperature was about 12 C. It was apparent from the data that the continued presence of the petroleum hydrocarbons led to the maintenance of a significant population of hydrocarbon-degrading bacteria in the sediment. Only the percentage of naphthalene-degrading bacteria had returned to the base-line level in the mid-tide sediment after 370 days (Fig. 2). During the warm water period (days 128 and 190 after the spill), signs of anaerobic conditions were noted in the low-tide sediment. These signs consisted of a blackening of the sediment and the odor of hydrogen sulphide. Hydrocarbon analyses. The concentration of the total hydrocarbons in the mid-tide area declined rapidly during the first week after the spill (38 to 18 jig of hydrocarbons per g of
649
BIODEGRADATION OF SPILLED FUEL OIL
VOL. 29,1975
sediment), remained fairly constant through the summer, and then declined to a very low level after 1 year (3 jig of hydrocarbons per g of sediment) (Fig. 7). The low-tide sediments contained less total hydrocarbons (28 Mg of hydrocarbons per g of sediment) on day 2 after the spill than did the mid-tide beach sand (38 fg of hydrocarbons per g of sediment). The hydrocarbon concentration increased during the summer in the low-tide area (maximum, 72 Mg of hydrocarbons per g of sediment) and then decreased, after 1 year, to slightly more than day 2 after the spill (Fig. 8). The hydrocarbon concentration at day 0 represents the original oil spilled from the Pennant. The total hydrocarbon concentration was obtained by assuming that the total hydrocarbons in the mid-tide beach sand at day 2 represents a 20% loss of total hydrocarbons from the original oil. The concentrations of the aliphatic and aromatic fractions were obtained from gas chromatographic analysis of the no. 6 fuel oil directly obtained from the Pennant. The aliphatic hydrocarbon fraction of the °00
original no. 6 fuel oil was subjected to urea adduction (8) to separate branched and cyclic alkanes from normal alkanes. With the use of internal standard hydrocarbons, it was determined that the branched and cyclic hydrocarbons were the major constituents of the unresolved complex and the normal alkanes comprised the resolved peaks of the gas chromatogram. Thus, it was possible to estimate changes in branched and cyclic alkanes versus normal alkanes by reference to the unresolved and resolved portions, respectively, of the gas chromatogram for the aliphatic fraction. Interpretation of hydrocarbon data in the low-tide area is complicated by the constant influx of hydrocarbons from other sources (7). The concentration of the resolved aliphatic fraction remained fairly constant, whereas the unresolved aliphatic and aromatic components increased and then decreased again (Fig. 8).
DISCUSSION This investigation demonstrates the complexity of environmental factors regulating the bi-
A
80
80
A
,
B
'00
0
60
IoO
A
/\A/ \ /
0
40
A
e20o _-A
20
20
2
4
16 8 32 Days after spill
64
128
256
0
2
8 32 16 Days ofte- sp,11
4
28
6
250
AiA10
°00 . 60
i
0
_2 40
vI AA
201 A A-A
C
2
4
8
IC
32
Doys offe, spod
04
128
256
2
4
32 8 lo Days afte, sp.i1
64
128
256
FIG. 3. Percentage of hydrocarbon-degrading bacteria in the total plate count of the heterotrophic bacteria in the low-tide sediment determined by a replica plating technique. Substrates: (A) n-Dodecane; (B) trans-decalin; (C) naphthalene; and (D) no. 1 fuel oil.
650
PIERCE, CUNDELL, AND TRAXLER j
20
AAAA
AA A
A A
A
A
A
AA
0
AA
a
oil, the cleanup procedure, and the physical and biological features of the area polluted by oil (14). In the mid-tide area the rapid decrease in all fractions of the residual fuel oil may be indicative of rapid physical migration. However, petroleum hydrocarbons percolating from the mid-tide area did not appear in the same
A
A A A~~~~~~~~~~~~~~~~~ A~~~~
o 15
APPL. MICROBIOL.
A A
&10 E
A
A
*A
'A
A A A
5
A
A A
A
A
AAA
A
AA
0
100 300 200 Days after Spill FIG. 4. The ambient water temperature in Upper
c x
Narragansett Bay after the oil spill. E i2
10 w
\
8
2
0
Time
in
3
4
5
6
Days
FIG. 6. Oxygen depletion by beach sediment collected at an ambient temperature of 22 C. BOD bottles containing 0.5 g of sediment, 0.1% (vol/vol) of no. 4 fuel oil and filled with seawater from the sampling site. Key as in Fig. 5.
6
0
A
50 2
40
9 0
3
2
Time
in
4
5
6
Days
FIG. 5. Oxygen depletion by beach sediment collected at an ambient temperature of 2 C. BOD bottles containing 0.5 g of sediment, 0.1% (vollvol) of no. 4 fuel oil and filled with seawater from the sampling site. Incubated 5 C, 120 rpm, no. 4 fuel oil (0); incubated 5 C, 120 rpm, no added substrate (a); incubated 15 C, 120 rpm, no. 4 fuel oil (0); incubated 15 C, 120 rpm, no added substrate (0); incubated 25 C, 120 rpm, no. 4 fuel oil (A); incubated 25 C, 120 rpm, no added substrate (A).
c
ci
E , -0
330
05
0
0
-D2
AA"
au
>1 ti
A-A~~~~
10
0)
odegradation and persistance of spilled oil and the difficulties of examining the parameters under field conditions. The bacterial population in the beach sands shifted to compensate for the stressed environment immediately upon influx of the oil. Obviously, the environmental changes after an oil spill will depend on the type of oil spilled, the weather conditions at the time of the spill, the previous exposure of the area to
0
o ~ ~ 2 4
8
~
16
~ 32 - 64
128
256
Days after spill
FIG. 7. Hydrocarbon content in the mid-tide sediment after the oil spill. Symbols: total petroleum hydrocarbons (A), normal aliphatic hydrocarbons (y), branched and cyclic aliphatic hydrocarbons (a), and aromatic hydrocarbons (a).
VOL. 29,1975
651
BIODEGRADATION OF SPILLED FUEL OIL
proportions or expected place in time in the low-tide area. The ability of the bacterial population to utilize all fractions of the oil suggests a bacterial role in oil degradation. The role of bacteria may be particularly important during the winter when the total hydrocarbon content of the mid-tide sediment was steadily reduced from more than 20 to 3 Mg/g of dry sediment. If the assumption is made that bacteria are responsible for the disappearance of the petroleum, the rate of biodegradation is in the order of less than 1 usg of hydrocarbon per g of dry sediment per day. The situation in the low-tide area is complicated by an influx of oil during the midsummer. The large increase in branched and cyclic aliphatic hydrocarbons in the sediment indicates an influx of weathered oil. Since no investigation of the vertical and horizontal distribution of the spilled oil in upper Narragansett Bay was undertaken the nature and extent of post-spill migrations of the residual fuel oil are unknown. The constant concentration of the n-alkane fraction in the low-tide sediment (1.8 to 3.2 Mg/ g) suggests a steady state where the hydrocarbon biodegradation equals the hydrocarbon influx into the sediment. Microbial emulsification of the residual fuel oil has been reported in pure culture experiments (20), and may cause the migration of the oil in the sediment during the summer months. Another factor responsible for the accumulation of weathered oil in the lowtide sediment may be the transportation of the hydrocarbons out of anaerobic sediments by evolved gases such as methane, carbon dioxide, and hydrogen sulphide (19). The organic carbon content in beach sediments is the lowest in coarse-grained, wellsorted mid-tide sediments and the highest in fine-grained, low-tide sediments. Heavy estuarine pollution coupled with the high summer temperatures led to anaerobic conditions in the low-tide sediments during the summer months, which may have curtailed the biodegradation of petroleum hydrocarbons. However, limited information is available on the role of anaerobic bacteria in the degradation of petroleum hydrocarbons (16). Another enrichment experiment conducted by Williams et al. (18) clearly demonstrated a rapid response by marine heterotrophic bacteria to sudden increases in amino acid concentration. Within a 2-day period, the bacteria in estuarine water were able to accommodate 10to 100-fold substrate concentration increases, thereby increasing a normal mineralization rate of 10 to 50 Mg of amino acids degraded per liter per day to 1,000 to 2,000 Mg/liter per day.
8OA
.E 0
~~~~~~I \kA
*
0 "I _0 c
-a
-
-
o0
-.o
-A-
0
-U/,
~-
C
0
22
8
4
8
4
166
3322
.
6644
128 256 256
128
Days after spill
FIG. 8. Hydrocarbon content in the low-tide sediment after the oil spill. Symbols: total petroleum hydrocarbons (A), normal aliphatic hydrocarbons (y), branched and cyclic aliphatic hydrocarbons (a) and aromatic hydrocarbons (U).
This investigation has demonstrated a similar enrichment of a population of hydrocarbondegrading bacteria in beach sediments. This population would apparently play a significant role in the degradation and perhaps translocation of petroleum hydrocarbons in response to a spill of residual fuel oil. After 1 year, the original fuel oil was still recognizable in the low-tide sediment and although the normal aliphatic and aromatic hydrocarbon fractions were considerably reduced, the branched and cyclic aliphatic hydrocarbons persisted presumably as a part of the organic material sequestered in the sediment. ACKNOWLEDGMENTS This investigation was supported by contract no. N00014-63-A-0215-0013 from the Office of Naval Research and the Rhode Island Agricultural Experiment Station. We are indebted to the Department of Food and Resource Chemistry for the use of their Varian gas chromatograph. LITERATURE CITED 1. American Public Health Association. 1972. Standard methods, p. 477-481. In Methods for the examination of water and waste water, 13th ed. American Public Health Association, Inc., Washington, D.C. 2. Blumer, M., and J. Sass. 1972. Indigenous and petroleum-derived hydrocarbons in a polluted sediment. Mar. Pollut. Bull. 3:92.94. 3. Blumer, M., and J. Sass. 1972. Oil pollution: persistence and degradation of spilled fuel oil. Science
176:1120-1122. 4. Blumer, M., and J. Sass. 1972. The West Falmouth oil spill: chemistry technical report no. 72-19. Woods Hole Oceanographic Institute, Woods Hole, Mass. 5. Cundell, A. M., and R. W. Traxler. 1973. The isolation and characterization of hydrocarbon-degrading bacteria from Chedabucto Bay, Nova Scotia, p. 421-
652 6.
7. 8.
9.
10. 11. 12. 13.
PIERCE, CUNDELL, AND TRAXLER
426. In Proc. Joint Conf. Prevention Control Oil Spills. American Petroleum Institute, New York. Farrington, J. W., and J. G. Quinn. 1973. Petroleum hydrocarbons in Narragansett Bay I survey of hydrocarbons in sediments and clams. Estuarine Coastal Mar. Sci. 1:71-79. Farrington, J. W., and J. G. Quinn. 1974. Petroleum hydrocarbons and fatty acids in sewage effluents. J. Water Pollut. Control. Fed. 45:706-711. Felbeck, G. T. 1966. Normal alkanes in muck soil organic-matter hydrogenolysis products, p. 11-17. Trans. Comm. II and IV, Int. Soc. Sci. Aberdeen, Scotland. Gunkel, W. 1968. Bacteriological investigations of oil-polluted sediments from the Cornish Coast following the "Torrey Canyon" disaster, p. 151-158. In The biological effects of oil pollution on littoral communities. Field Studies Supplement no. 2. Field Studies Council, London. Haines, J. R., and M. Alexander. 1974. Microbial degradation of high-molecular-weight alkanes. Appl. Microbiol. 28:1084-1085. Jones, G. E., and H. W. Jannash. 1959. Bacterial populations in seawater as determined by different methods of enumeration. Linmol. Oceanogr. 4:128-139. Morita, R. Y. 1966. Marine psychrophilic bacteria. Oceanogr. Mar. Biol. Annu. Rev. 4:105-121. Sieburth, J. McN. 1967. Seasonal selection of estuarine bacteria by water temperature. J. Exp. Biol. Ecol. 1:98-121.
14.
APPL. MICROBIOL.
Straughan, D. 1972. Factors causing environmental changes after an oil spill. J. Pet. Technol. March,
1972:250-254. 15. Traxler, R. W. 1973. Bacterial degradation of petroleum materials in low temperature marine environments, p. 163-170. In D. G. Ahearn and S. P. Meyer (ed.), The microbial degradation of oil pollutants. Sea Grant publ. no. LSU-SG-73-01, Center for Wetland Resources, Louisiana State University, Baton
Rouge.
17. Walker, J. D., and R. R. Colwell, 1973. Microbial ecology of petroleum utilization in Chesapeake Bay, p.
18.
19. 20.
21.
65-690. In Proc. Joint Conf. Prev. Contr. Oil Spills. American Petroleum Institute, New York. Williams, P. J. Leb., and R. W. Gray. 1970. Heterotrophic utilization of dissolved organic compounds in the sea. II. Observations on the responses of heterotrophic marine populations to abrupt increases in amino acid concentration. J. Mar. Biol. Assoc. U.K. 50:871-881. Voroshilora, A. A., and E. A. Dianora. 1950. Bacterial oxidation of oil and its migration in natural waters. Mikrobiologiya 19:203-210. Zajic, J. E., B. Supplisson, and B. Volesky. 1974. Bacterial degradation and emulsification of no. 6 fuel oil. Environ. Sci. Technol. 8:664-668. ZoBell, C. E., and J. F. Prokop. 1966. Microbial oxidation of mineral oil in Barataria Bay. Z. Allg. Microbiol. 6:143-162.
