Effect of crude oil on a Louisiana Spartina alterniflora salt marsh

EFFECT

OF CRUDE OIL ON A LOUISIANA ALTERNIFLORA SALT MARSH

SPARTINA

R. D. DELAUNE, W. H. PATRICK, JR. & R. J. BURESH

Lahoratory.Jor Wetland Soils and Sediments, Center./br Wetland Resources, Louisiana State UnirersiO', Baton Rouge, Louisiana 70803, USA

A BS TRA CT

Ttle q[]ect oj Louisiana crude oil on growth of Spartina alterniflora Loisel and selected anaerobic soil processes in a Louisiana salt marsh was incestigated. Greenhouse and field studies indicated that S. alterniflora ean tolerate a large amount of oil without a short-term decrease in abore-ground biomass. Neither the biological reduction o f nitrate, manganese, iron and sulphate nor the production of methane and ammonium in stirred reduced sediments were affected by additions of up to 10 '~ooil on a soi'l-weight basis. Oil placed on the water surface of unstirred sediment caused a release of iron, manganese and ammonium from the sediment to the overlying water due to the absence of oxygen in the water column.

INTRODUCTION

Louisiana coastal wetlands support a rich assortment of fisheries and wildlife that is heavily dependent upon annual production of estuarine organisms. Estuarinedependent organisms, in turn, rely on the productivity of the natural marsh. Maintenance of a viable marsh is essential for the preservation of fishery and wildlife resources. Primary production originates from marsh grasses dominated by Spartina alterniflora, phytoplankton, benthic plants and epiphytic algae on the marsh grasses (Day et al., 1973). S. alterniflora is the most important primary producer and it acts as the initial source of detrital food on which all of the estuarine animals depend either directly or indirectly. The Louisiana coastal zone is the scene of intense activity associated with oil and gas exploration, drilling and extraction that has the potential for reducing 21 Environ. Pollut. 0013-9327/79/0020-0021/$02.25 © Applied Science Publishers Ltd, England, 1979 Printed in Great Britain

22

R. D. DELAUNE, W. H. PATRICK, JR., R. J. BURESH

productivity of the area through the input of oil. Any deleterious effect upon S.

alterniflora by oil pollutants could have widespread repercussion on the food web for the entire estuarine ecosystem. Although casual observations have suggested that crude oil is toxic to marsh plants, there have been few quantitative studies dealing with the toxicity of crude oil to S. alterniflora in Gulf coast salt marshes. Crow et al. (1976) observed little damage to living S. alterniflora or reduction in regeneration of new plants after the application of 250ml/m 2 of Arabian light crude oil to marsh plots. Lytle (1975) found decreased production ofS. alterniflora immediately following an oil spill in a Mississippi pond, but the oil had no noticeable long-term effects on S. alterniflora biomass. Practically no information is available on the effect of crude oil on biological and chemical reactions in the predominantly anaerobic marsh soils that may indirectly affect productivity. Ellis & Adams (1961) demonstrated the effect of oil on aerobic processes in upland soils. Addition of oil tended to cause greater changes in aerobic processes due to increased oxygen utilisation during degradation and prevention of normal diffusion processes. Changes in oxidation-reduction potential during decomposition and assimilative processes can be expected when hydrocarbons are added to an aerated soil (Ellis & Adams, 1961). Oil degradation proceeds slowly under anaerobic conditions (Zobell, 1946; Hughes & McKenzie, 1975; Mayo et aL, 1978). This suggests that oil may have little effect as an added energy source under anaerobic conditions. Knowles & Wishart (1977) found no effect, stimulatory or inhibitory, of crude oil on nitrogen fixation and carbon dioxide evolution in reduced sediment samples. The purpose of this study was to determine the effect of crude oil on several microbiological processes in salt marsh soil and the growth of S. alterniflora.

MATERIALS AND METHODS

Field stud), South Louisiana crude oil was added in May 1976 to a uniform stand of S.

alternifiora in a salt marsh in the Barataria Basin of Louisiana (29 °15'N, 90 °7'W). Crude oil was added at rates of 1,2, 4 and 8 litres per square metre to circular 0.25 m 2 plots enclosed by metal cylinders pushed 15cm into the marsh sediment. The retainers were 43 cm in height and coated with epoxy paint to resist corrosion. The oil was slowly poured from a beaker onto an existing 15-cm layer of water present on the marsh surface. The field layout consisted of five replications of each treatment along with four control plots to which no oil was added. Because of the low tidal range in coastal Louisiana the water level in the containers never exceeded a height of 30cm above the marsh surface. In September 1976 the plants within the retainer were harvested and weighed to

EFFECT OF CRUDE OIL ON

Spartina

23

determine total above-ground biomass. Plant samples were analysed for total nitrogen by the Kjeldahl method and for phosphorus using the molybdenum blueascorbic acid method (Murphy & Riley. 1962). In April 1977 the number of shoots regenerated from the clipped plots was measured. The above-ground biomass of the second year's growth was harvested in September 1977. Soil cores (10cm width × 15 cm depth) were taken in September 1977 for measurement of remaining crude oil. The cores were dried at 28 °C and ground to pass through a 35-mesh sieve. Fifty grammes of the soil was Soxhlet-extracted with 300ml of l:l(v/v) chloroform: methanol for 18 h. The extract was reduced to 50 ml and water-washed. Aliquots were taken for total lipids analysis and silica gel chromotagraphy which separated alkanes, aromatic and NSO fractions. Total lipids were analysed using an infrared spectrophotometer. Concentrations of the alkane, aromatic and NSO fractions were determined gravimetrically.

Greenhouse study The effect of South Louisiana crude oil on growth of S. alterniflora was also investigated in the greenhouse. Ten kilogrammes of wet sediment from Barataria Bay, Louisiana, were placed in ceramic pots 20 cm in diameter and 22 cm in depth. Ten S. alterniflora plants were then transplanted into each pot. Plants were fertilised with N and P and allowed to grow for a period of 90 days. Crude oil was then added at the rates of 0, 1,2, 4, 8, 16 and 32 litres/m z on the surface o f a 5-cm overlying water column which was maintained throughout the experiment by adding water daily to replace that lost by evapotranspiration. Plants were harvested after 75 days and the dry weight of plant material was determined. Two weeks after the first harvest the number of new tillers emerging from the pots was recorded. Above-ground plant dry weight for the regenerated second growth was obtained 60 days after the first cutting.

Laboratory stud)' The effect of South Louisiana crude oil on several soil and sediment processes was studied in the laboratory. Sediment suspensions of 4:1 water-sediment mixtures were incubated in two-litre, three-necked, flat-bottomed flasks. The suspensions were continuously stirred with a magnetic stirrer. Each flask, as shown in Fig. 1, was fitted with two platinum electrodes, a glass electrode for measuring pH, a thermometer, a serum cap, an inlet for nitrogen and an outlet tube, the end of which was submerged to prevent gaseous oxygen diffusion into the flask. The suspensions were maintained at 30 + 1 °C. Temperature was regulated by inserting thin asbestos sheets between the flask and the underlying magnetic stirrer to control the heat transfer from the stirrer motor. South Louisiana crude oil was added to the sediment suspensions at rates equivalent to 0, 0.1, 1, 5 and 10 "//oof the dry sediment weight. Redox potential was monitored to determine the rate of reduction of the sediment suspension. Samples of the sediment suspension were removed from the incubation flask through the rubber serum cap with a syringe and a 12-gauge stainless steel

24

R. D. DELAUNE, W. H. PATRICK, JR., R. J. BURESH

Salt Bridge (connected to calomel electrode) Air or Nitrogen Inlet Platinum--(

/ - Serum Cap ~ / .~ I ~-Rubber Stoppers

Ilk/_/\

Outlet

p.

,oct.o .

2 Litre-3 Necked Reaction Vessel Sediment-Water Suspension Stir Bar Magnetic StTrrer

Fig. 1.

Incubation flask for incubating Barataria Bay sediment suspension.

needle at various times and analysed for exchangeable N H ~ - N , Fe ÷ ÷ and Mn ÷ ÷ and acid-soluble sulphide. The exchangeable ions were extracted with 2N sodium acetate adjusted to pH 4.5. Iron and manganese were measured by atomic absorption spectrophotometry. Ammonium-nitrogen was determined by Nesslerisation. Sulphide was determined by the iodometric procedure for sewage by the American Public Health Association (Faber, 1960). Acid-soluble sulphide includes free hydrogen sulphide as well as metallic sulphide. Another study determined the effect of crude oil on methane production, an important biological process in coastal environments. Four hundred grammes of sediment with approximately 10 cm of overlying water were incubated at 30 °C in one-litre glass bottles. Crude oil was added at a rate equivalent to 0, 0.1, 1, 5, 10 and 20 ~ of sediment weight. Containers were shaken daily to ensure mixing of oil with sediment. Gas samples were taken with time and analysed for methane with a gas chromatograph. The effect ofoil on N O 3 reduction was measured in 150 ml bottles containing 50 g of sediment and 50 ml of water. After a 14-day pre-incubation period 200#g/g of N O 3 - N was added to sediment samples which had been treated with 0, 1,4 and 10 by weight of crude oil. Samples were taken over a period of ten days. Nitrate was measured in water extracts by the phenoldisulphonic acid method. The effect of crude oil on the release of nutrients from sediment to an overlying water column was also studied. Sediment-water columns consisting o f a 5-cm layer of sediment and 7.5 cm of overlying water were incubated at 30 °C with and without a

EFFECT OF CRUDE OIL ON

Spartina

25

0-2 cm layer of crude oil on the water surface. After 28 days redox potential of the overlying water was measured and the overlying water was analysed for Fe ÷ +, Mn + + and N H ~ .

RESULTS

Effect of oil on S. alterniflora growth under fieM conditions The average total above-ground plant biomass + 1 standard deviation harvested on 23 September 1976 is shown in Table 1. Added oil did not statistically decrease or increase the above-ground biomass ofS. alterniflora. These plant biomass values are similar to those reported by Kirby & Gosselink (1976). TABLE I I N F L U E N C E O F C R U D E OIL ON A B O V E - G R O U N D BIOMASS, N E W S H O O T S A N D STEM DENSITY OF L O U I S I A N A S A L T MA R S H

Crude oil (litres/m 2 )

0 1

2 4 8 LSD 0.05 value

Aboce-ground biomass 23 Sept. 1976 16 Sept. 1977 (g/m 2 ) 2000 + 213 1908 +_ 300 2015+_ 153 1819 ± 265 1832 ± 153

1001 ± 133 822 ± 72 1161 ± 2 7 7 935 ± 110 991 f 162

316

236

S. altern!#ora IN A

New shoots Stem density 18 April 1977 16 Sept. 1977 (number per square met re) 116 102 111 101 98

± 25 ± 13 _+ 10 _+ 10 ± 19

22

205 4-_ 30 199 ± 19 215_+21 219 ± 2 197 _+ 20 28

Second-year growth, as measured by above-ground plant biomass harvested on 16 September 1977, was also not significantly influenced by oil application. Aboveground biomass was less than that harvested the previous year. This is attributed to the fact that the first year harvest consisted of all live and dead plant material present. Crude oil had no influence on the numbers of new shoots generated the spring following the first harvest. Oil added to the marsh plots did not remain on the water surface; consequently, the emerging shoots were not exposed to a layer ofoil on the water overlying the marsh. The added oil apparently adhered to dead plant material on the marsh surface and organic matter within the marsh soil. Stem density was also not significantly different between oil treated and control plots. Uptake of nitrogen and phosphorus was not affected by crude oil. Nitrogen and phosphorus contents of above-ground plant material were 0.75 °~o and 0.072 !!,,, respectively, for treated and control plots. Apparently the levels of crude oil used were not toxic to S. alzern!/tora root or stem tissue and did not inhibit the transport of atmospheric oxygen into the roots. Thus

26

R. D . D E L A U N E ,

W . 14. P A T R I C K ,

J R . , R. J. B U R E S H

oil on the soil surface or water surface within the tidal zone apparently has little influence on mature S. alterniflora. The tidal amplitude in this area is only about 0.3 m, so oil on the surface of tidal water would not come in contact with the leaf blade. Analysis of soil cores taken from each plot on 16 September 1977 revealed that a large percentage of the oil was still present (Table 2). Approximately 60-70 ~ of the added oil could be accounted for in plots which received I and 2 litres of crude oil per square metre. A smaller percentage of the added oil was recovered at the higher rates of application. Oil had apparently moved down below the depth at which the cores were taken. Pockets o foil were observed in hollow roots and rhizomes of dead plants at depths greater than 15 cm. The metal retainers probably contributed to the high recovery of the added oil. In an actual spill tidal movement across the surface of the marsh could disperse and remove the oil. TABLE 2 INFLUENCE OF CRUDE OIL ADDITION UPON THE AMOUNT OF CRUDE OIL REMAINING IN THE SEDIMENT AND THE CONCENTRATION OF SPECIFIC HYDROCARBON FRACTIONS AFTER 16 MONTHS

Crude oil (litres/m 2) 0 1 2 4 8

Alkanes 0"29+0"12 3"88__+1.02 10"09+3'40 10-26__+0.87 16-72_____4.60

Hydrocarbon fractions Aromatics NSO (rng/g) 0.11 +0.12 2.82+0"90 6-81 + 1.61 7.02_+ 1.25 12.22+1.43

2.21 +0.31 4-83+0.41 7.93+ 1 - 9 9 8-71 _+0"66 11.26+1.46

Total lipids

Crude oil remaining (litres/m 2)

2.6+0.4 17.6+3.4 31.5+5.9 34.4+5.9 53-4-+7.5

0 0'66+0'13 1.18+0.22 1.30-+0"22 2.01-+0.28

There was no significant difference between the ratio of the recovered alkane, aromatic and N S O (nitrogen, sulphur and oxygen) fractions with added increments of oil. This would also indicate that only a small percentage of these excessive additions of oil had degraded. Alkanes are known to degrade faster than the more toxic aromatic fraction (Zobell, 1969; Blumer et al., 1973).

Effect of oil on S. alterniflora growth in the greenhouse The greenhouse study also demonstrated that above-ground biomass of S.

alterniflora was not reduced by the application of crude oil at rates up to 32 litres/m / (Table 3). Plant growth was indirectly affected by a reduction in number of new shoots developed after the first harvest. The number of new shoots generated was reduced to 6 and 9, respectively, for oil rates of 4 and 8 litres/m 2 as compared with 15 shoots for the check. At rates of 16 and 32 litres/m z no new shoots formed. Oil did not reduce the initiation of new shoots unless there was a visible film of oil on the water surface through which the shoot had to emerge. At 16 and 32 litres/m 2 there was a constant film of oil on the water surface. At lower rates most of the oil was adsorbed on dead plant material and the surface of the organic soil.

EFFECT OF CRUDE OIL ON S p a r l i n a

27

TABLE 3 EFFECT OF C R U D E OIL O N A B O V E - G R O U N D BIOMASS A N D N E W S H O O T F O R M A T I O N O F

S. alterniflora G R O W N

IN A

GREENHOUSE

Crude oil (litres/m 2)

Above-ground biomass First harrest Second harrest (g/m 2)

0 1 2 4 8 16 32

71.0 73.4 69.2 73.1 67.3 69-3 70-1

New shoots aJter .first harcest (number per square metre)

34-0 33"1 24.1 24.7 17.9 0 0

15 17 13 6 9 0 0

EJJect of oil on soil processes Redox potential did not vary with crude oil additions as high as 10 ~°~ o by weight of sediment. The redox potential for both oil-treated and control suspensions dropped rapidly from + 500 mv to - 2 5 0 mv after a six-day incubation. Ferric iron, Mn 4 + and S O ] - reduction rates were the same for both oil-treated and control samples (Table 4). Exchangeable Mn 2 + from Mn 4 ÷ reduction was formed at the rate of TABLE 4 Mn 2 +, Fe e +

EFFECT OF C R U D E OIL O N P R O D U C T I O N OF

Time (days)

AND S 2

Percent oil byweight in soil 0

0.1

1

5

10

Manganous manganese (#g:g)

0

7

7

9

8

8

2 4 6 9 12 16 20

40 86 82 83 73 82 83

44 82 86 84 81 77 83

44 82 83 77 80 82 83

43 83 80 83 85 80 85

42 78 82 80 82 77 78

0 2 4 6 9 12 16 20

13 225 625 1050 1075 1025 1162 1062

10 275 512 1100 i137 1075 1050 1100

10 237 462 975 1087 1000 1162 1125

0 2

0 0

4 6 9 12 16 20

4 39 89 162 212 210

0 0 5 36 78 149 192 200

0 0 I0 52 O1 176 231 216

Ferrous iron (llg/g) 13 212 612 1975 1100 1075 1075 1000

13 285 637 1112 1162 1112 1162 1102

Sulphide (#g/g) 0 0 8 42 83 158 211 241

0 0 6 51 90 172 210 232

28

R. D. DELAUNE, W. H. PATRICK, JR., R. J. BURESH

20/~g/g/day. This reduced form of Mn reached a concentration of approximately 80 #g/g and levelled off. Exchangeable Fe 2 + was formed at rates of approximately 175/ag/g/day for all treatments to a maximum concentration near 1000/~g/g. Sulphide was produced at the rate of 13/~g/g/day and reached a total concentration near 200/~g/g. TABLE 5 EFFECT OF CRUDE OIL ON NITROGEN MINERALISATION

Time (days)

Percent oil by weight in soil 0

0" 1

1

5

10

24 30 37 46 59 73 89 118

27 32 41 47 50 74 90 111

(ltg NH~-N/g soil) 0 2 4 6 9 12 16 20

26 32 40 46 58 74 82 109

31 35 43 44 56 72 85 114

27 30 38 48 54 77 87 106

Crude oil up to 10 % by weight of soil had no noticeable influence on nitrogen mineralisation (Table 5). Ammonium-nitrogen formed from the decomposition of organic matter in the sediment was produced at the rate of 4/ag/g/day for all treatments. Two hundred microgrammes per gramme of N O 3 - N added to preincubated, reduced sediment completely disappeared after four days (Table 6). There was also no observed difference in N O 3 reduction rates between oil-treated and control samples. TABLE 6 EFFECT OF CRUDE OIL ON NITRATE REDUCTION

Time (days)

Percent oil by weight in soil 0

I

4

10

194 132 48 18 0 0 0

213 134 57 10 0 0 0

(lag NO~-N/g soil) 0 1

2 3 4 6 8

198 129 49 13 0 0 0

209 140 58 8 0 0 0

Methane production in reduced sediment also proceeded at the same rate for oiltreated and control samples (Table 7). Methane initially present in the crude oil treatments came from the added oil. Much of this CH 4 was consumed before CH 4 production began in the reduced sediment. Methane was produced at a rate of 4 nmole per gramme of sediment per day.

EFFECT OF CRUDE OIL ON S p a r t i n a

29

TABLE 7 EFFECT OF CRUDE OIL ON METHANE PRODUCTION

Time (days)

Percent oil by weight in soil 0

0.1

I

5

10

20

13.0 12.2 6-9 5-2 34.1 35- I 58.0 65.2 88.9 90.2

23.3 23.3 15.4 8.9 30.7 48.8 51-3 66-4 85-5 81-9

(nmoles CH4/g soil) 0 2 5 8 11 15 19 25 32 40

Trace 0.4 0.6 4-9 36.1 41.3 58-9 63.9 76.5 86.4

0.5 0.4 0.4 5.8 33.0 48.8 63.4 67-3 71-9 81-9

I-4 1.3 0.8 4.6 29.8 39.9 60.2 68.7 75-2 79-8

5-1 4.6 3.1 3-(1 38-2 46.2 56.0 63-4 73-2 84-2

A 0.2-cm layer ofoil placed on the water surface of a sediment water column had a drastic influence on the concentration of nutrients in the water column after 28 days (Table 8). Higher concentrations of Fe 2 +, Mn 2 + and N H ~ were present in the water column containing a surface layer of oil. Less than 0-1/~g/ml of Fe z +, Mn 2 + and N H 2 was found in the water columns containing no added oil. There was an absence of molecular oxygen in the oil-treated water column, as indicated by a low redox potential of - 250 mv. The oil layer prevented normal exchange of oxygen between the water column and the atmosphere. The redox potential of the water in the untreated column, + 4 5 0 mv, indicated an oxygenated condition. The pH of the overlying water was 7.7 in the oil-treated column compared with pH 7-2 for the water column containing no oil. TABLE 8 EFFECT OF A 0 ' 2 - c m LAYER OF CRUDE OIL ON THE OVERLYING WATER UPON THE RELEASE OF NUTRIENTS FROM THE SEDIMENT INTO THE OVERLYING WATER AFTER 2 8 DAYS

Nutrient

Fe 2+ Mn 2+ NH~-N

Concentration in orerlving water No oil layer Oil layer (#g/ml)