Environmental Science and Pollution Research

, Volume 21, Issue 11, pp 6982–6993 | Cite as

Impact of crude oil exposure on nitrogen cycling in a previously impacted Juncus roemerianus salt marsh in the northern Gulf of Mexico

  • Agota Horel
  • Rebecca J. Bernard
  • Behzad Mortazavi
Research Article


This study investigated potential nitrogen fixation, net nitrification, and denitrification responses to short-term crude oil exposure that simulated oil exposure in Juncus roemerianus salt marsh sediments previously impacted following the Deepwater Horizon accident. Temperature as well as crude oil amount and type affected the nitrogen cycling rates. Total nitrogen fixation rates increased 44 and 194 % at 30 °C in 4,000 mg kg−1 tar ball and 10,000 mg kg−1 moderately weathered crude oil treatments, respectively; however, there was no difference from the controls at 10 and 20 °C. Net nitrification rates showed production at 20 °C and consumption at 10 and 30 °C in all oil treatments and controls. Potential denitrification rates were higher than controls in the 10 and 30 ºC treatments but responded differently to the oil type and amount. The highest rates of potential denitrification (12.7 ± 1.0 nmol N g−1 wet h−1) were observed in the highly weathered 4,000 mg kg−1 oil treatment at 30 °C, suggesting increased rates of denitrification during the warmer summer months. These results indicate that the impacts on nitrogen cycling from a recurring oil spill could depend on the time of the year as well as the amount and type of oil contaminating the marsh. The study provides evidence for impact on nitrogen cycling in coastal marshes that are vulnerable to repeated hydrocarbon exposure.


Deepwater Horizon Crude oil Benthic ecology Nitrogen cycle Juncus roemerianus Salt marsh 


An estimated 7.0 × 105 m3 (±20%) of crude oil from the Deepwater Horizon oil rig explosion contaminated the Gulf of Mexico in April 2010 (Crone and Tolstoy 2010). The spilled oil polluted shorelines, beaches, and ecologically sensitive salt marshes. While portions of the shorelines were contaminated from the moderately weathered oil, in subsequent months tar balls from the sea floor washed ashore. The contaminated Alabaman coastline was only a small part of the over 1,000 km shorelines that have been affected by the spill. In June 2010, tar balls and oil slicks were observed at Point Aux Pins (PAP), Alabama, the study site for this experiment, following the Deepwater Horizon (DWH) oil spill (Horel et al. 2012b). PAP is a salt marsh with Juncus roemerianus dominating the vegetation and Spartina alterniflora (Horel et al. 2012b) at its seaward edge. Compared to S. alterniflora marsh, J. roemerianus can be more sensitive to moderate oil contamination (Lin and Mendelssohn 2012). Depending on the amount of oil spilled, J. roemerianus productivity and vitality can decrease as a result of stress-induced plant temperature increase, decreased photosynthesis, or reduced transport of atmospheric oxygen to the roots (Pezeshki et al. 2000; Hester and Mendelssohn 2000) following physical coating of oil and oil-induced oxygen deficiency (Mendelssohn et al. 2012).

Many environmental factors such as temperature, pH, nutrients, and energy source affect the biodegradation of a contaminant (Alexander 1999) and the productivity of a salt marsh ecosystem (Sousa et al. 2008; Wright et al. 1997). The amount of nutrients available for microbial consumption is an important factor controlling the abundances of microbes and hydrocarbon degraders in coastal marsh sediments. The magnitude of hydrocarbon degrader microorganisms in the sediments can provide implications of biodegradation potentials or even prior hydrocarbon contaminations at a given area. One way to estimate the number of hydrocarbon degrader microbial communities existing in samples is with the most probable number (MPN) method. Microorganisms present in marsh sediments can compete with hydrocarbon degraders for available nutrients and affect the biodegradation process. Chlorophyll-a measurements can give a good indication of microalgae biomass present in the salt marsh ecosystem that might be responsible for additional nitrogen uptake from the sediment (Anderson et al. 1997). Moreover, seasonal differences in plant growth cycles and nutrient uptake can influence the microbial communities present in the sediments and their metabolic rates can vary with temperature. The effects of different temperatures on biodegradation rates of hydrocarbon contaminations have been previously investigated by several studies (Raikos et al. 2012; Dell’Anno et al. 2012; Horel et al. 2012a), primarily concluding that higher temperature increases microbial growth rates and hydrocarbon degradation under aerobic conditions.

Changes in nitrogen cycling can occur after oil contamination, especially in highly sensitive ecosystems (Lin and Mendelssohn 2012; Mishra et al. 2012) like salt marshes. Salt marsh sediments can often be areas of high denitrification rates, a process in which nitrate (NO3) or nitrite (NO2) is dissimilatory reduced to nitrogen gas under sub-oxic conditions (Kaplan et al. 1979). Denitrification can be coupled to the products of nitrification, an aerobic process where ammonium (NH4+) is oxidized to NO2 and NO3 by nitrifying bacteria (An and Joye 2001). Nitrogen fixation by autotrophic and heterotrophic bacteria in salt marsh sediment may provide an important nitrogen source to the primary producers like S. alterniflora and J. roemerianus during marsh development (Tyler et al. 2003). Previous studies have shown significant decreases or even complete inhibition of nitrogen fixation as a result of hydrocarbon contamination (Haines et al. 1981; Atlas and Bartha 1998). Many salt marshes are nitrogen limited (Valiela and Teal 1979; Tyler et al. 2003; Elser et al. 2007) despite having high pore water NH4+ concentrations. During denitrification, nitrogen could be removed from a system that otherwise might be used for plant primary production, which could enhance nitrogen limitation.

The northern Gulf of Mexico shorelines might experience recurring hydrocarbon seeps by storm surges, especially during hurricane seasons. Thus, hydrocarbon contamination effects on the nitrogen cycle in sub-tropical salt marshes that are susceptible to repeated weathered crude oil exposure need to be further explored. The main objectives of this study were (1) to determine the response of potential nitrogen fixation, net nitrification, and denitrification rates to introduced weathered crude oil in a J. roemerianus-dominated marsh ecosystem that had been previously impacted by the DWH oil spill (Beazley et al. 2012; Horel et al. 2012b) and (2) to examine the impact of moderately and highly weathered (tar balls) crude oil on key nitrogen cycling processes to better understand the impact of different contaminant sources simulating short- and long-term weathering processes during the time it takes the oil to reach the marsh. To address these objectives, a two-factor split-plot design experiment was set up to investigate the effects of temperature and fuel amount and type on nitrogen cycling rates during short-term hydrocarbon exposure. This study presents evidence for impact on nitrogen cycling in coastal marshes that are vulnerable to repeated hydrocarbon exposure.

Materials and methods

Experimental setup

Marsh sediments from 0 to 50 mm depth from the J. roemerianus marsh at PAP in the northern Mississippi Sound (between 30°22′43″ and 30°22′45″ North and between 88°18′11″ and 88°18′13″ West) were collected on May 6, 2012, homogenized thoroughly, and used in the experiment. The experiment focused on the Northern Gulf of Mexico’s seasonal temperature range and concentrated on temperatures of 10, 20, and 30 °C to signify winter, spring, and summer, respectively. Moderately weathered Louisiana sweet crude with alkanes ranging from C9 to C35 (Fig. 1a) and highly weathered tar balls (C17 to C35; Fig. 1b) collected at the West end of Dauphin Island (Aug. 4, 2010; Alabama, USA) after the DWH oil spill represented two separate stages of recently spilled oil washing up or weathered oil remaining in salt marsh areas after a spill. For each fuel type, four sediment treatments with oil amounts of 0, 500, 4,000, and 10,000 mg kg−1 were applied.
Fig. 1

Chromatographic views of the weathered fuels total ionic content (TIC). a Louisiana crude oil (LA); b tar balls collected at Katrina Cut (KC), Dauphin Island, AL. Internal standards are shown at retention time 10.96 min (naphthalene-d8) and 24.57 min (chrysene-d12)

Sediment chlorophyll-a content and microbial estimation

For analysis of sediment chlorophyll-a (chl-a) content, the top 50 mm of sediment was collected from the J. roemerianus marsh in triplicate with a 15-mm inner diameter (ID) core tube. Wet weight of the sediment was obtained from samples prior to freeze drying. Chl-a was cold extracted according to Welschmeyer (1994) in 90 % acetone for 24 h. Samples were diluted 17-fold into aliquots prior to fluorometric chlorophyll analysis on a Turner Designs 700 fluorometer. Calculations were made according to EPA Method 445.0 for “uncorrected chlorophyll α” (i.e., non-acidified) and reported on a milligram-per-square-meter basis.

The MPN technique was used to estimate the number of aerobic hydrocarbon-degrading bacteria (polycyclic aromatic hydrocarbon [PAH] degraders, alkane degraders, and total hydrocarbon degraders) originally present in the soil. Sediment samples were prepared for microbial enumeration as previously described by Horel et al. (2012b). Briefly, triplicate 1-g sediment samples were suspended in 10 ml of 1 % w/v sodium pyrophosphate (Na4P2O7) with 2 % NaCl. To each well of a 96-well microtiter plate, 180 μl of 2 % NaCl Bushnell–Haas broth growth medium, 20 μl sediment suspension and 5 μl of hexadecane, 5 μl conventional diesel fuel, or 10 μl of PAH mix as the sole carbon source for alkane, total hydrocarbon, or PAH degrader microbial enumeration, respectively, were added (Horel et al. 2012b). The calculations of aerobic hydrocarbon degraders were based on MPN 96-well plate methods described previously by Wrenn and Venosa (1996).

Physical and chemical characteristics of the sediment and water

Water column salinity was approximately 17 psu in the J. roemerianus marsh and was measured using a portable refractometer. Gravimetric water contents (GWC) were determined by oven drying sediment samples (48 h; 60 ºC). Grain size distribution was determined by sieve analysis based on ASTM C136-06 standard by using sieves #10, #60, and #230. Total nitrogen and carbon content were measured with an elemental combustion analyzer (Costech Instruments, model ECS 4010). Inorganic nutrient analyses followed a method previously described by Horel et al. (2012b), where triplicates of 10-g sediment samples were extracted with 1 M NaCl solution and supernatants were analyzed for NO3, NO2, NH4+, and inorganic phosphate concentrations with a SANplus SKALAR autoanalyzer. Concentrations are reported as milligrams per kilogram of dry weight sediment and rates as nanomoles per gram per hour of wet weight sediment.

Calibrated oxygen (O2) and hydrogen sulfide (HS) microelectrodes (Unisense OX500-UW and H2S500-UW) were used to determine the depth of the oxic layer as well as the concentration of HS in an intact sediment core (175 × 95 mm ID) collected from the J. roemerianus marsh.

Nitrogen cycling experiments

All nitrogen cycling experiments were carried out under dark conditions in a temperature-controlled environmental chamber at the Dauphin Island Sea Lab (Alabama, USA) with four oil concentration levels of 0, 500, 4,000, and 10,000 mg kg−1. Each of the slurry incubation experiments (2.4.1–2.4.3) was repeated for each temperature (10, 20, and 30 ºC), and samples were incubated for 1 to 24 h after oil addition to the top of the sediment surface before measurement of N cycle rates.

All rates and fluxes pertaining to nitrogen species are expressed on a nitrogen atom basis.

Potential nitrogen fixation

Potential nitrogen (N2) fixation was measured as ethylene (C2H4) production from acetylene (C2H2) reduction (Welsh et al. 1996). From the homogenized sediment, 20 g wet weight sediment was added to triplicate 70 ml serum vials. Control samples had no oil addition. The different fuel amounts (500, 4,000, and 10,000 mg kg−1) of Louisiana sweet crude or tar balls were added to the vials prior to the addition of 25 ml of 17 psu artificial seawater (ASW) or 25 ml of 20 mM sodium molybdate amended ASW. Sodium molybdate is a specific inhibitor of the sulfate reduction process (Capone 1993; Hardy et al. 1973). Serum vials were capped, vortexed for 20 s, and degassed for 10 min with N2 gas to ensure anoxia. The non-inhibited samples received 10 % v/v of C2H2, added to the headspace, and were dark incubated for 24 h. After 1 h of shaking incubation, sodium molybdate inhibited samples received 10 % C2H2 (v/v) and were dark incubated for 24 h. Production of C2H4 from the reduction of C2H2 as a substrate analog of N2 was measured using a Shimadzu gas chromatograph (GC-2014) with flame ionization detection (GC-FID). Production rates of C2H4 were converted to potential N2 fixation rates using a C2H2–N2 reduction ratio of 3:1 (Capone 1993). The rate of N2 fixation by sulfate reducing bacteria (SRB) was calculated based on the difference between the unamended (total N2 fixation) and the sodium molybdate (N2 fixation not by SRB) treated vials.

Potential nitrification

Net nitrification rates were measured in triplicate by incubating 1 g wet weight of homogenized sediment with 50 ml of ASW, 500 μM ammonium chloride (NH4+Cl) (Henriksen et al. 1981), and the different fuel amounts, under oxic and dark conditions on a shaker table. Control samples were prepared as described above without the addition of fuel. Initial (t = 0 h) and final (t = 24 h) samples were collected, filtered, and frozen immediately, and subsequently analyzed for NOx (NO3+ NO2) using standard wet chemical techniques modified for the SANplus SKALAR autoanalyzer (Pennock and Cowan 2001). Rates of potential nitrification (nanomoles of N per gram per hour, based on wet weight of sediment) were calculated from the difference between concentrations in final and initial samples.

Potential denitrification

Potential denitrification rates were measured using the acetylene block technique from triplicate slurry incubations containing 20 g wet weight of homogenized sediment, 35 ml of ASW with 100 μM KNO3, and the different fuel concentrations. Control samples in triplicates were prepared as described above but without the addition of fuel. Vials were capped with butyl rubber caps and flushed with N2 for 10 min to ensure anoxia prior to the addition of C2H2 (10 % v/v). The addition of C2H2 prevents the activity of nitrous oxide reductase (nosZ), and the denitrification pathway cannot proceed completely to N2 gas; rather, the process stops at the intermediate production of nitrous oxide (N2O). The activity of ammonia monooxygenase (AmoCAB), which facilitates the formation of hydroxylamine, is also inhibited by C2H2 (Yoshinari et al. 1977). Therefore, the present slurry incubations only account for classical denitrification and not hydroxylamine-dependent or coupled nitrification–denitrification. After a 1-h incubation, gas samples were injected into evacuated 10 ml exetainer vials and N2O production was quantified by using a Shimadzu gas chromatograph (GC-2014) with an electron capture detector (GC-ECD) (Dollhopf et al. 2005).

Hydrocarbon characteristics and analyses

An aliquot of Louisiana sweet crude oil (hereafter referred to as LA crude oil or LA) was mechanically weathered prior to the beginning of the experiment. LA crude oil (57,500 mg) was placed in an open container on a mechanical shaker. The oil was shaken at ca. 85 rpm for 5 days under a fume hood in the dark (Mortazavi et al. 2013). The moderately weathered Louisiana crude oil density prior to the start of the experiment averaged 0.94 g cm−3. The tar balls (hereafter referred to as tar balls or KC) were collected from Katrina Cut at the West side of Dauphin Island (Aug. 4, 2010; Alabama, USA) after the 2010 DWH oil spill reached the Alabama cost (collection location coordinates: 30°14′56″ North and 88°12′22″ West). The major difference between the two oil types was in the viscosity; the tar balls were in a solid phase that included an average of 80.24 % beach sand, while Louisiana crude oil was free of sediments in a liquid phase. The two types of oil were analyzed by GC-FID (Shimadzu GC-2014) as described previously by Horel et al. (2012b) and gas chromatography with mass spectrometry (GC-MS Total Ion Chromatograph, Thermo Scientific Instruments, Trace Ultra GC Gas Chromatography coupled with Triple quadruple TSQ Quantum GC Mass Spectrometry). The initial GC-FID oven temperature was held at 50 °C, increased to 300 °C at a rate of 10 °C min−1, and held at 300 °C for 20 min. The carrier gas was helium with a constant flow of 1.3 ml min−1. The GC column (Restek, Rtx-5) was 30 m in length with 0.25 mm internal diameter and 0.25 μm film thickness (Horel et al. 2012b). The initial GC-MS oven temperature was held at 60 °C for 1 min, increased to 210 °C at a rate of 12 °C min−1 and to 340 °C at a rate of 8 °C min−1, and held for 5.25 min. The carrier gas was helium with a constant flow of 1.0 ml min−1. The GC column (Restek, 5Sil-ms) was 30 m in length with 0.25 mm in internal diameter and 0.5 μm in film thickness. Injection volume used was 1.0 μL. The crude oil contamination levels in the tar balls averaged 1.65 × 105 ppm.

Statistical analyses

A two-factor split-plot design in an incomplete factorial application was used to analyze the data in SAS JMP 10. Temperature was the main plot and the fuel addition subplot had seven treatment levels (control, LA 500, LA 4,000, LA 10,000, KC 500, KC 4,000, and KC 10,000 mg kg−1), each with three replicates. The fixed effects were temperature, fuel amount, fuel types (with controls), temperature × fuel amounts, temperature × fuel types, and temperature × controls. The results of the statistical analysis are presented in Table 1 and 2. There was also a random effect for temperature × replicate (i.e., the whole plot error, data not shown). Data were tested for normality and homogeneity of variances using the Shapiro-Wilk and Levene’s tests, respectively. Since N2 fixation values were not normally distributed, the data were log-transformed prior to the analyses. When significant differences occurred, Tukey HSD post hoc tests were used to determine significant interactions. Statistical significance of the data sets was determined at p < 0.05 and p < 0.01. Error is reported as standard error.
Table 1

Overall statistical data on the effects of temperature, fuel types, and fuel amounts on the nutrient cycling

Nitrogen cycle type



DF Den

F ratio

Prob > F









Fuel types and amounts






Fuel types and amounts × Temp






N2 fixation







Fuel types and amounts






Fuel types and amounts × Temp













Fuel types and amounts






Fuel types and amounts × Temp






*p < 0.05; **p < 0.01, statistically significant differences at these levels

Table 2

Detailed statistical data on the effects of temperature, fuel types, and fuel amounts on the nutrient cycling



N2 fixation


t ratio

Prob > ǀtǀ

t ratio

Prob > ǀtǀ

t ratio

Prob > ǀtǀ

Temperature [10]







Temperature [20]







Temperature [30]







Fuel types and amounts [1] Control







Fuel types and amounts [2] LA 500







Fuel types and amounts [3] LA 4,000







Fuel types and amounts [4] LA 10,000







Fuel types and amounts [5] KC 500







Fuel types and amounts [6] KC 4,000







Fuel types and amounts [7] KC 10,000







Control × Temp [10]







Control × Temp [20]







Control × Temp [30]







LA 500 × Temp [10]







LA 500 × Temp [20]







LA 500 × Temp [30]







LA 4,000 × Temp [10]







LA 4,000 × Temp [20]







LA 4,000 × Temp [30]







LA 10,000 × Temp[10]







LA 10,000 × Tem [20]







LA 10,000 × Temp [30]







KC 500 × Temp [10]







KC 500 × Temp [20]







KC 500 × Temp [30]







KC 4,000 × Temp [10]







KC 4,000 × Temp[20]







KC 4,000 × Temp [30]







KC 10,000 × Temp [10]







KC 10,000 × Temp [20]







KC 10,000 × Temp [30]







*p < 0.05; **p < 0.01, statistically significant differences at these levels


Physical, chemical, and biological characteristics of the sediment

The soil properties measured at the beginning of the experiment are summarized in Table 3. The sediment was comprised mainly of sand and silt with 30.97 % water content. While the soil had limited nitrogen content, the overall C–N ratio was 16.39 , which was within the range for effective biodegradation noted by Alexander (1999) to be around 33:1 or less. The average chl-a measurement from the top 50 mm of the sediment was 58.81 mg m−2 (±6.31). The depth of the oxic layer in the J. roemerianus marsh extended to 3 mm (Fig. 2), indicating oxic conditions only for the sediment surface that is exposed to the atmosphere or lying at the water column interface. At 1 mm, O2 and HS were both present in low concentrations likely because of the spatially separated (ca. 20 mm) profiles and the heterogeneous nature of the J. roemerianus roots in the sediment core. Microelectrodes were not measuring identical vertical profiles and suggest that the presence of roots may help structure oxic conditions in the sediments. In the absence of roots, HS may be present. Below the depth of the oxic layer, a low concentration of HS (<50 uM) was observed to 6 mm (Fig. 2).
Table 3

Summary of initial soil properties. GWC stands for gravimetric water content. TDIN is the total dissolved inorganic nitrogen as the sum of NH4+, NO2-, and NO3-. Values shown are means ±SE; (n = 3)

Soil type


GWC (%)

Total nitrogen (%)

Total carbon (%)


TDIN (mg N kg−1)

PO43− (mg P kg−1)

Sand and very fine sand 50 %

6.41 (±0.02)

30.97 (±0.98)

0.34 % (±0.02)

5.62 % (±0.37)

16.39 (±0.23)

0.76 (±0.04)

0.03 (±0.00)

Silt 20 %

Fig. 2

Marsh sediment core oxygen (O2) and hydrogen sulfide ion (HS) concentrations profiles. Measurements were taken at different depths from spatially independent (ca. 20 mm apart) regions of the same core. SW seawater overlying the sediment core

Aerobic hydrocarbon-degrading microbial densities present in the sediment at the beginning of the experiment varied from 1.21 × 102 to 2.66 × 105 (data not shown). The amount of alkanes and total petroleum hydrocarbon (TPH) degraders in the sediments were similar to numbers found in a study 1 year after crude oil contamination from the DWH accident at a proximate salt marsh site (Horel et al. 2012b).

Changes in nitrogen fixation

There was strong evidence for a main effect of temperature (p = 0.0008; Table 1) on potential N2 fixation rates; however, potential N2 fixation rates were not influenced by fuel amount and type (p = 0.685; Table 1). Post hoc tests revealed potential N2 fixation rates were similar at 20 and 10 ºC (p = 0.9904). Potential N2 fixation rates were higher at 30 ºC than at 20 and 10 ºC (p = 0.0084 and p = 0.0074, respectively; Fig. 3). Low fuel amounts for both LA crude oil and tar balls seemed to have minimal effect on the overall N2 fixation rates, even at higher temperatures (p > 0.154; Table 2). However, at higher fuel amounts, especially in the 10,000 mg kg−1 LA crude oil and 4,000 mg kg−1 tar ball treatments, rates increased substantially at 30 ºC (Fig. 3). N2 fixation by SRB accounted for 95 to 98 % of total N2 fixation in the 10 ºC temperature treatment (Fig. 3), while N2 fixation by SRB ranged from 46 to 71 % and 58 to 89 % of total N2 fixation for 20 and 30 ºC, respectively (Fig 3).
Fig. 3

Potential nitrogen fixation changes in the marsh sediments at different oil concentration levels and temperatures. Temperature (p = 0.0008) but not fuel amount and type (p = 0.685) had a significant effect on potential rates of total N2 fixation. Percentage of N2 fixation by SRB was greatest at 10 °C. Values shown are means ± SE (n = 3). SRB sulfate-reducing bacteria, C control, LA Louisiana sweet crude, and KC tar balls

LA crude oil additions at low concentrations did not affect the N2 fixation rates substantially (p > 0.483; Table 2). The only treatment that affected the N2 fixation negatively at all temperatures was the medium LA crude oil addition (Fig. 3). The highest change in the N2 fixation rates were observed at 30 ºC with an average decline of 21 %. In the case of high contaminant concentration LA crude oil treatments, a substantial increase in N2 fixation was observed compared to control treatments (194 % increase at 30 ºC; Fig. 3). A similar trend was observed in the case of 4,000 mg kg−1 tar balls addition (113 % increase at 30 ºC; Fig. 3).

Changes in nitrification

There was strong evidence for a main effect of temperature (p = 0.0016) on potential net nitrification rates; however, potential nitrification rates were not influenced by fuel amount and type (p = 0.694). Post hoc tests revealed potential net nitrification rates were similar at 10 and 30 ºC (p = 0.1259) and higher at 20 ºC than at 10 and 30 ºC (p = 0.0014 and p = 0.0124, respectively; Fig. 4). The highest nitrate uptakes were measured in the 10 ºC control treatment (4.66 ± 1.51 nmol N g−1 wet h−1), 500 mg kg−1 LA crude oil (3.85 ± 2.85 nmol N g−1 wet h−1) and 4,000 mg kg−1 LA crude oil (6.32 ± 2.32 nmol N g−1 wet h−1) fuel treatments (Fig. 4).
Fig. 4

Potential net nitrification changes in the marsh sediments at different oil concentration levels and temperatures. Temperature had a significant effect on potential rates of net nitrification (p = 0.0016) but fuel amount and type did not (p = 0.694). Values shown are means ± SE (n = 3); C control, LA Louisiana sweet crude, and KC tar balls

Changes in denitrification

There was strong indication for a main temperature effect (p < 0.0001) on potential denitrification rates. With the exception of 500 mg kg−1 LA crude oil at 10 °C and 4,000 mg kg−1 at 30 °C, generally potential denitrification rates increased with fuel amount and type (p = 0.0188; Tables 1 and 2) compared to the controls. There was also a significant interaction between temperature and fuel amounts and fuel types (p = 0.0002).

Post hoc tests revealed potential denitrification rates were similar at 30 and 10 °C (p = 0.2203), were higher at 30 °C than at 20 °C (p < 0.001), and were lower at 20 °C than at 10 °C (p < 0.001). At 30 °C, potential denitrification rates in the 500 and 10,000 mg kg−1 LA crude oil, as well as the 500 and 4,000 mg kg−1tar balls treatments, were significantly higher than rates in the other fuel types and concentrations at 10 and 20 °C (p < 0.05 for all). At 30 °C, potential denitrification rates in the 4,000 mg kg−1 tar ball treatment were higher than rates in the 4,000 mg kg−1 LA crude oil (p = 0.003), the 10,000 mg kg−1 tar balls (p = 0.0022), and control (p = 0.0177) treatments. Additionally, potential denitrification rates in the 10,000 mg kg−1 LA crude oil treatment were higher than rates in the 4,000 mg kg−1 LA crude oil (p = 0.001), 10,000 mg kg−1 tar balls (p = 0.0081), and control (p = 0.0026) treatments. At 20 °C, with the exception of the 500 mg kg−1 tar ball treatment, potential denitrification rates in all treatments were lower than all potential denitrification rates at 10 °C (p < 0.05 for all). The potential denitrification rates in the 20 ºC treatments overall were lower than the other temperature treatments (Fig 5) with the lowest potential denitrification rates observed in the 500 mg kg−1 LA crude oil treatment (1.75 ± 0.26 nmol N g−1 wet h−1). The highest potential denitrification rates were found at 30 ºC in the 10,000 mg kg−1 LA crude oil (20.50 ± 2.27 nmol N g−1 wet h−1) and the 4,000 mg kg−1 tar ball (21.67 ± 3.00 nmol N g−1 wet h−1).
Fig. 5

Potential denitrification changes in the marsh sediments at different oil concentration levels and temperatures. Both temperature (p < 0.001) and fuel amount and type (p = 0.0188) had significant effects on potential denitrification rates. There was also a significant interaction between temperature and fuel amount/types (p = 0.0002). Values shown are means ± SE (n = 3). C control, LA Louisiana sweet crude, and KC tar balls


Physical, chemical, and biological characteristics of the sediment

HS has been shown to interfere with the nitrification and denitrification pathways in marine sediments (Joye and Hollibaugh 1995; Brunet and Garcia-Gil 1996); however, a previous study at PAP J. roemerianus marsh indicated that a low HS concentration in the sediments did not affect the denitrification process (data not shown). Since O2 was depleted by 3 mm and the NH4+ concentration was relatively high, any nitrate produced by nitrification in the oxic layer could support denitrification below 3 mm (Hulth et al. 2005).

Though chl-a content can vary by region, the chl-a value in this study was similar to values observed in previous studies (Anderson et al. 1997). In Mississippi, the average chl-a in S. alterniflora marsh sediment was 95 mg m−2, while in South Carolina it ranged from 73 to 102 mg m−2 (Sullivan and Currin 2002). Chl-a values ranged from 8 to 169 mg m−2 in a S. alterniflora marsh during summer months (Anderson et al. 1997), while for J. roemerianus marshes annual average chl-a measurements of 59 mg m−2 have been reported (Sullivan and Currin 2002). Sediment chl-a is an indicator of the abundance of microphytobenthos (MPB) in the sediment and high chl-a inventory in the marsh ecosystem could indicate possible competition with microbial communities for available nutrients in the sediment. However, in the present study, the J. roemerianus marsh the average chl-a value was within or below the average range reported in the literature, and considerable competition of MPB with other microorganisms for the available nutrients would not be expected because the incubations were conducted in the dark. Therefore, major differences in nitrogen cycle processes were assumed to be a result of temperature and oil treatment effects on the microbial community.

Previous hydrocarbon exposure at the study site following the DWH accident can result in an elevated hydrocarbon degrader microorganisms in the oxic layer of the marsh sediments. The different hydrocarbon chains as available carbon sources for microbial growth are shown in Fig 1. One of the main differences between the two fuel types was the missing or low amount of hydrocarbon chain lengths from C9 to C16 in the tar ball samples. The shorter chained aliphatic hydrocarbons are more degradable by microorganisms (Braddock et al. 2003); therefore, large increases in microbial populations would be expected in the LA crude oil treatments. During an event of a contamination, the hydrocarbon degrader microorganisms compete with other microbial communities and long-term changes in microbial community densities and types may be inevitable (Atlas and Bartha 1998); therefore, the relatively high average alkanes (1.59 × 105 ± 9.46 × 104 CI) and TPH (9.35 × 103 ± 4.69 × 103) degrader microorganisms at PAP were expected and are crucial for degradation processes during initial and recurring hydrocarbon contaminations.

Changes in nitrogen cycle processes

Greater N2 fixations by SRB were observed at the 10 ºC treatment, when the SRB can be more metabolically active (Praharaj and Fortin 2008), but higher total N2 fixation rates were observed in the 30 ºC treatment. Several studies reported negative correlation between hydrocarbon fuel concentrations and nitrogenese activities (Piehler et al. 1997; John et al. 2011). However, in the current experiment, reductions in N2 fixation rates by SRB were observed only at 20 ºC in the moderate fuel contamination treatments (Fig. 3). The organically rich marsh sediments provide a carbon source for the indigenous microbial communities, which might be metabolically less active at lower temperatures, therefore resulting in the relatively low total and SRB related N2 fixation values in the different treatment types.

High values, such as 194 % N2 fixation increase at 30 °C, can indicate the presence of microbial consortia in the sediment that have the potential for both hydrocarbon degradation and N2 fixation (Al-Mailem et al. 2010). Hydrocarbon additions from tar balls at high concentrations, however, showed only moderate increase in N2 fixation values, 19, 33, and 44 % increase at 10, 20, and 30 ºC, respectively.

Hydrocarbon degraders were expected to be responsible for NO3 uptake to fuel their metabolic demand, as NO3 as an electron acceptor has been documented to enhance hydrocarbon degradation (Horel and Schiewer 2009). As nitrification is a two-step process with different groups of bacteria responsible for the conversion of NH4+ to NO2 and NO2 to NO3(Atlas and Bartha 1998), the temperature differences may have impacted these diverse bacterial communities differently. Ammonium concentrations in the sediment ranged from 0.45 to 1.66 mg kg−1 and could be converted to NO3 by nitrifying microorganisms. Low nitrification rates could be results of several reasons. Since only the 20 ºC treatment had positive net nitrification rates, the 10 and 30 ºC temperatures may have impacted the indigenous microbial community’s ability to oxidize NH4+. The negative net nitrification values could be the result of lower sediment oxygen concentrations at high temperatures (Huang and Pant 2009), which could slow or stop the nitrification process. However, oxic conditions were present during the incubation period; therefore, it was unlikely that low oxygen concentrations resulted in lower net nitrification rates. At low temperatures, the decreased microbial metabolic activities (Horel and Schiewer 2009) could have caused lower oxidation of NH4+. Low nitrification measurements also have been reported in grassland areas with sediments with high organic carbon concentrations (Strauss and Dodds 1997). When heterotrophic bacteria outcompetes nitrifying bacteria for the available ammonium, nitrate production could be reduced or eliminated, therefore resulting in minimal nitrification (Strauss and Dodds 1997). Partial NO3 uptake may be also possible by the microalgae present in the sediment (Rysgaard et al. 1995). Additionally, nitrifying microbes exposed to even low concentrations of HS may not recover and nitrification can be reduced or completely inhibited (Joye and Hollibaugh 1995). Since there were low concentrations of HS in the sediments at PAP (Fig. 2), the nitrifying community may have been inhibited and consequently accounted for the low potential net nitrification rates.

The significant interaction between fuel amounts and types indicated that the degree of weathering as well as concentrations of oil were important factors on potential denitrification rates. The lower amount (500 mg kg−1) of hydrocarbon’s effect on the potential denitrification rates was similar to another study, where Haines et al. (1981) found no effects on denitrification rates from short-term exposure to crude oil. The observed results of higher potential denitrification rates at 30 ºC were in agreement with previous studies (Raikos et al. 2012; Dell’Anno et al. 2012) that found greater hydrocarbon degradation at higher temperatures because of greater microbial metabolic activities. Carbon oxidation by denitrification has an effect on the sediment total carbon metabolism rates. Some organic carbon oxidation in marine sediments can be the result of the denitrification process, as an earlier study by Canfield et al. (1993) observed that denitrification accounted between 3 and 6 % of the total carbon oxidation.

The findings in this study imply that the efficiency of nitrogen cycle processes in marsh sediments may behave differently if an oil spill were to occur during the summer, where more diverse microbial communities can be active in the sediments (Burke et al. 2003), as opposed to winter. Competition for the available nutrients between different bacterial communities may influence the overall nitrogen fixation, nitrification, and denitrification potentials. At 10 and 30 °C temperatures in the current study, nitrate uptake was observed as well as a greater amount of N2 fixation by SRB, indicating possible nitrogen limitation. Denitrification rates, in general, were higher than nitrogen fixation values and may have been more dependent on water column NO3 diffusion rather than on coupled nitrification–denitrification. Although oil contamination resulted in significant changes in denitrification rates, N2 fixation and net nitrification rates were similar to the controls at lower contaminations levels, implying potentially some resiliency in response to contamination at lower concentrations. If recurring oil contamination was to happen during summer at PAP when temperatures are near 30 ºC, for example, the rates of denitrification may increase and more nitrogen could be removed from the marsh and leave less nitrogen available for plant primary production. In addition to the stress from physical coating by the oil, if nitrogen limitation was enhanced by greater denitrification stimulated by oil contamination, primary production could decline and the salt marsh may be negatively impacted. Weathered hydrocarbons both in solid and liquid phase can persist in the environment for a long time, endangering water sources, plant life, and wildlife health.


Potential N2 fixation, denitrification, and net nitrification rates mainly increased at 30 ºC, but only denitrification showed significant increases after exposure to hydrocarbons. The fuel contamination type and amount and season can significantly influence potential denitrification rates in marsh sediments, and hence, the amount of nitrogen being removed from the system may be higher in the summer months. If a salt marsh is already nitrogen limited, enhanced nitrogen removal by the denitrification process after hydrocarbon exposure could change the nutrient balance needed for plant primary production and productivity by marsh plants could decline. It is especially important to understand baseline conditions if a coastal marsh is vulnerable to repeated hydrocarbon exposure and has the potential for enhanced denitrification rates from oil contamination. Introduced contaminant in the subsurface can disrupt the indigenous microbial communities’ structure, densities, and types, resulting in altered benthic nitrogen cycling pathways.



This material is based upon work supported by Mississippi State University/Northern Gulf Institute (MSU/NGI) 191001-306911_01/TO 091 and Dauphin Island Sea Lab/Marine Environmental Science Consortium (DISL/MESC) 2423 Jv, T4-005UA. The authors would also like to thank Dr. Aladar Bencsath, Laura Linn for instrumentation and analyses help, and Dr. Christina Staudhammer for statistical advice.


  1. Alexander M (1999) Biodegradation and bioremediation, 2nd edn. Academic Press, San DiegoGoogle Scholar
  2. Al-Mailem DM, Sorkhoh NA, Salamah S, Eliyas M, Radwan SS (2010) Oil-bioremediation potential of Arabian Gulf mud flats rich in diazotrophic hydrocarbon-utilizing bacteria. Int Biodeterior Biodegrad 64:218–225CrossRefGoogle Scholar
  3. An S, Joye SB (2001) Enhancement of coupled nitrification-denitrification by benthic photosynthesis in shallow estuarine sediments. Limnol Oceanogr 46:62–74CrossRefGoogle Scholar
  4. Anderson IC, Tobias CR, Neikirk BB, Wetzel RL (1997) Development of a process-based nitrogen mass balance model for a Virginia (USA) Spartina alterniflora salt marsh: implications for net DIN flux. Mar Ecol Prog Ser 159:13–27CrossRefGoogle Scholar
  5. Atlas RM, Bartha R (1998) Microbial ecology: fundamentals and applications, 3rd edn. Benjamin/Cummings, Redwood CityGoogle Scholar
  6. Beazley MJ, Martinez RJ, Rajan S, Powell J, Piceno YM, Tom LM, Andersen GL, Hazen TC, Van Nostrand JD, Zhou J, Mortazavi B, Sobecky PA (2012) Microbial Community Analysis of a Coastal Salt Marsh Affected by the Deepwater Horizon Oil Spill. Plos One. doi:10.1371/journal.pone.0041305 Google Scholar
  7. Braddock JF, Lindstrom JE, Prince RC (2003) Weathering of a subarctic oil spill over 25 years: the Caribou-Poker Creeks Research Watershed experiment. Cold Reg Sci Technol 36:11–23CrossRefGoogle Scholar
  8. Brunet RC, Garcia-Gil LJ (1996) Sulfide-induced dissimilatory nitrate reduction to ammonia anaerobic freshwater sediments. FEMS Microbiol Ecol 21:131–138CrossRefGoogle Scholar
  9. Burke DJ, Hamerlynck EP, Hahn D (2003) Interactions between the salt marsh grass Spartina patens, arbuscular mycorrhizal fungi and sediment bacteria during the growing season. Soil Biol Biochem 35:501–511CrossRefGoogle Scholar
  10. Canfield DE, Jorgensen BB, Fossing H, Glud R, Gundersen J, Rasming NB, Thamdrup B, Hansen JW, Nielsen LP, Hall POJ (1993) Pathways of organic carbon oxidation in three continental margin sediments. Mar Geol 113:27–40CrossRefGoogle Scholar
  11. Capone DG (1993) Determination of nitrogenase activity in aquatic samples using the acetylene reduction procedure. In: Kemp PF, Sherr BF, Sherr EB, Cole JJ (eds) Handbook of methods in aquatic microbiology. Lewis, pp 621–631.Google Scholar
  12. Crone TJ, Tolstoy M (2010) Magnitude of the 2010 Gulf of Mexico oil leak. Science 330(6004):634–634CrossRefGoogle Scholar
  13. Dell’Anno A, Beolchini F, Rocchetti L, Luna GM, Danovaro R (2012) High bacterial biodiversity increases degradation performance of hydrocarbons during bioremediation of contaminated harbor marine sediments. Environ Pollut 167:85–92CrossRefGoogle Scholar
  14. Dollhopf SL, Hyun JH, Smith AC, Adams HJ, O’Brien S, Kostka JE (2005) Quantification of ammonia-oxidizing bacteria and factors controlling nitrification in salt marsh sediments. Appl Environ Microbiol 71(1):240–246. doi:10.1128/AEM.71.1.240-246.2005 CrossRefGoogle Scholar
  15. Elser JJ, Bracken MES, Ckekand EE, Gruner DS, Harpole WS, Hillebrand H, Ngai JT, Seabloom EW, Shurin JB, Smith JE (2007) Global analysis of nitrogen and phosphorus limitation of primary producers in freshwater, marine and terrestrial ecosystems. Ecol Lett 10:1135–1142CrossRefGoogle Scholar
  16. Haines JR, Atlas RM, Griffiths RP, Morita RY (1981) Denitrification and nitrogen fixation in Alaskan continental shelf sediments. Appl Environ Microbiol 41(2):412–421Google Scholar
  17. Hardy RWF, Burns RC, Holsten RD (1973) Applications of the acetylene-ethylene assay for measurement of nitrogen fixation. Soil Biol Biochem 5:47–81CrossRefGoogle Scholar
  18. Henriksen K, Hansen JI, Blackburn TH (1981) Rates of nitrification, distribution of nitrifying bacteria and nitrate fluxes in different types of sediment from Danish waters. Mar Biol 61:299–304CrossRefGoogle Scholar
  19. Hester MW, Mendelssohn IA (2000) Long-term recovery of a Louisiana brackish marsh plant community from oil-spill impact: vegetation response and mitigating effects of marsh surface elevation. Mar Environ Res 49(3):233–254CrossRefGoogle Scholar
  20. Horel A, Schiewer S (2009) Investigation of the physical and chemical parameters affecting biodegradation of diesel and synthetic diesel fuel contaminating Alaskan soils. Cold Reg Sci Technol 58(3):113–119CrossRefGoogle Scholar
  21. Horel A, Mortazavi B, Sobecky P (2012a) Responses of microbial community from Northern Gulf of Mexico sandy sediments following exposure to Deepwater Horizon crude oil. Environ Toxicol Chem 31(5):1004–1011CrossRefGoogle Scholar
  22. Horel A, Mortazavi B, Sobecky P (2012b) Seasonal monitoring of hydrocarbon degraders in Alabama marine ecosystems following the Deepwater Horizon oil spill. Water Air Soil Pollut 223(6):3145–3154CrossRefGoogle Scholar
  23. Huang S, Pant HK (2009) Nitrogen transformation in wetlands and marshes. J Food Agric Environ 7(3&4):946–954Google Scholar
  24. Hulth S, Aller RC, Canfield DE, Dalsgaard T, Engstrom P, Gilbert F, Sundback K, Thamdrup B (2005) Nitrogen removal in marine environments: recent findings and future research challenges. Mar Chem 94:125–145CrossRefGoogle Scholar
  25. John RC, Itah AY, Essien JP, Ikpe DI (2011) Fate of nitrogen-fixing bacteria in crude oil contaminated wetland ultisol. Bull Environ Contam Toxicol 87:343–353CrossRefGoogle Scholar
  26. Joye SB, Hollibaugh JT (1995) Influence of sulfide inhibition of nitrification on nitrogen regeneration in sediments. Science 270:623–625CrossRefGoogle Scholar
  27. Kaplan W, Valiela I, Teal JM (1979) Denitrification in a salt marsh ecosystem. Limnol Oceanogr 24:726–734CrossRefGoogle Scholar
  28. Lin Q, Mendelssohn IA (2012) Impacts and recovery of the Deepwater Horizon oil spill on vegetation structure and function of coastal salt marshes in the Northern Gulf of Mexico. Environ Sci Technol 46(7):3737–3743CrossRefGoogle Scholar
  29. Mendelssohn IA, Andersen GL, Baltz DM, Caffey RH, Carman KR, Fleeger JW, Joye SB, Lin Q, Maltby E, Overton EB, Rozas LP (2012) Oil impacts on coastal wetlands: implications for the Mississippi river delta ecosystem after the Deepwater Horizon oil spill. Bioscience 62(6):562–574CrossRefGoogle Scholar
  30. Mishra DR, Cho HJ, Ghosh S, Fox A, Downs C, Merani PBT, Kirui P, Jackson N, Mishra S (2012) Post-spill state of the marsh: Remote estimation of the ecological impact of the Gulf of Mexico oil spill on Louisiana salt marshes. Remote Sens Environ 118:176–185CrossRefGoogle Scholar
  31. Mortazavi B, Horel A, Anders JS, Mirjafari A, Beazley MJ, Sobecky PA (2013) Enhancing the biodegradation of oil in sandy sediments with choline: A naturally methylated nitrogen compound. Environ Pollut 182:53–62CrossRefGoogle Scholar
  32. Pennock JR, Cowan JLW (2001) Analytical instrumentation methods manual. Mar Environ Sci Consortium Tech Rep 98–003Google Scholar
  33. Pezeshki SR, Hester MW, Lin Q, Nyman JA (2000) The effects of oil spill and clean-up on dominant US Gulf coast marsh macrophytes: a review. Environ Pollut 108:129–139CrossRefGoogle Scholar
  34. Piehler MF, Swistak JG, Pinckney JL, Paerl HW (1997) Sub-lethal effects of coastal petroleum pollution on Spartina alterniflora stem epiphytes. Chemosphere 35(11):2665–2674CrossRefGoogle Scholar
  35. Praharaj T, Fortin D (2008) Seasonal variations of microbial sulfate and iron reduction in alkaline Pb–Zn mine tailings (Ontario, Canada). Appl Geochem 23:3728–3740CrossRefGoogle Scholar
  36. Raikos V, Vamvakas SS, Sevastos D, Kapolos J, Karaiskakis G, Koliadima A (2012) Water content, temperature and biocide effects on the growth kinetics of bacteria isolated from JP-8 aviation fuel storage tanks. Fuel 93:559–566CrossRefGoogle Scholar
  37. Rysgaard S, Christensen PB, Nielsen LP (1995) Seasonal variation in nitrification and denitrification in estuaine sediment colonized by benthis maicroalgae and bioturbating infauna. Mar Ecol Prog Ser 126:11–121CrossRefGoogle Scholar
  38. Sousa AI, Lillebo AI, Cacador I, Pardal MA (2008) Contribution of Spartina maritima to the reduction of eutrophication in estuarine systems. Environ Pollut 156(3):628–635CrossRefGoogle Scholar
  39. Strauss EA, Dodds WK (1997) Influence of protozoa and nutrient availability on nitrification rates in subsurface sediments. Microbil Ecol 34:155–165CrossRefGoogle Scholar
  40. Sullivan MJ, Currin CA (2002) Community structure and functional dynamics of benthic microalgae in salt marshes. In: Concepts and Controversies in Tidal Marsh Ecology, vol 2. Springer, pp 81–106. doi:10.1007/0-306-47534-0_6
  41. Tyler AC, Mastronicola TA, McGlathery KJ (2003) Nitrogen fixation and nitrogen limitation of primary production along a natural marsh chronosequence. Oecologia 136:431–438CrossRefGoogle Scholar
  42. Valiela I, Teal JM (1979) The nitrogen budget of a salt marsh ecosystem. Nature 280:652–656CrossRefGoogle Scholar
  43. Welschmeyer NA (1994) Fluorometric analysis of chlorphyll a in the presence of chlorophyll b and pheopigments. Limnol Oceanogr 39:1985–1992CrossRefGoogle Scholar
  44. Welsh DT, Bourgues S, de Wit R, Herbert RA (1996) Seasonal variations in nitrogen-fixation (acetylene reduction) and sulphate-reduction rates in the rihizosphere of Zostera noltii: nitrogen fixation by sulphate-reducing bacteria. Mar Biol 125:619–628CrossRefGoogle Scholar
  45. Wrenn BA, Venosa AD (1996) Selective enumeration of aromatic and aliphatic hydrocarbon degrading bacteria by a most-probable-number procedure. Can J Microbiol 42(3):252–258CrossRefGoogle Scholar
  46. Wright AL, Weaver RW, Webb JW (1997) Oil bioremediation in salt marsh mesocosms as influenced by N and P fertilization, flooding, and season. Water Air Soil Pollut 95:179–191Google Scholar
  47. Yoshinari T, Hynes R, Knowles R (1977) Acetylene inhibition of nitrous oxide reduction and measurement of denitrification and nitrogen fixation in soil. Soil Biol Biochem 9(3):177–183CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Agota Horel
    • 1
    • 2
    • 3
  • Rebecca J. Bernard
    • 1
    • 2
  • Behzad Mortazavi
    • 1
    • 2
  1. 1.Department of Biological SciencesUniversity of AlabamaTuscaloosaUSA
  2. 2.Dauphin Island Sea LabDauphin IslandUSA
  3. 3.Institute of Soil Sciences and Agricultural Chemistry, Centre for Agricultural ResearchHungarian Academy of SciencesBudapestHungary

Personalised recommendations