Salt marsh denitrification is impacted by oiling intensity six years after the Deepwater Horizon oil spill

Accepted Manuscript Salt marsh denitrification is impacted by oiling intensity six years after the Deepwater Horizon oil spill Corianne Tatariw, Nikaela Flournoy, Alice A. Kleinhuizen, Derek Tollette, Edward B. Overton, Patricia Sobecky, Behzad Mortazavi PII:

S0269-7491(18)31934-1

DOI:

10.1016/j.envpol.2018.09.034

Reference:

ENPO 11576

To appear in:

Environmental Pollution

Received Date: 14 May 2018 Revised Date:

22 August 2018

Accepted Date: 5 September 2018

Please cite this article as: Tatariw, C., Flournoy, N., Kleinhuizen, A.A., Tollette, D., Overton, E.B., Sobecky, P., Mortazavi, B., Salt marsh denitrification is impacted by oiling intensity six years after the Deepwater Horizon oil spill, Environmental Pollution (2018), doi: https://doi.org/10.1016/ j.envpol.2018.09.034. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

1

Salt marsh denitrification is impacted by oiling intensity six years after the

2

Deepwater Horizon oil spill

3

Corianne Tatariw1,2, Nikaela Flournoy1, Alice A. Kleinhuizen1,2, Derek Tollette1,2, Edward

5

B. Overton3, Patricia Sobecky1, Behzad Mortazavi1,2*

6 7

1

8

35487, United States

University of Alabama, Department of Biological Sciences, Tuscaloosa, Alabama

10

2

11

36528, United States

SC

9

M AN U

Dauphin Island Sea Lab, 101 Bienville Blvd Dauphin Island, Dauphin Island, Alabama

12 13

3

14

70803, United States

15

*corresponding author

Department of Environmental Sciences, Louisiana State University, Baton Rouge, LA

16

21 22 23 24 25 26 27

EP

20

AC C

19

TE D

17 18

RI PT

4

28 29 30 31 1

ACCEPTED MANUSCRIPT

Abstract: Coastal salt marshes provide the valuable ecosystem service of removing

33

anthropogenic nitrogen (N) via microbially-mediated denitrification. During the 2010

34

Deepwater Horizon (DWH) spill, oil exposure killed marsh plants in some regions and

35

contributed to rapid compositional shifts in sediment microbial communities, which can

36

impact ecosystem denitrification capacity. Within 3—5 years of the spill, plant biomass

37

and microbial communities in some impacted marshes recovered to a new stable state.

38

The objective of this study was to determine whether marsh recovery 6 years after the

39

DWH oil spill results in subsequent recovery of denitrification capacity. We measured

40

denitrification capacity (isotope pairing technique), microbial 16S rRNA gene

41

composition, and denitrifier abundance (quantitative PCR) at sites subjected to light,

42

moderate, and heavy oiling during the spill that were not targeted by any clean-up

43

efforts. There were no differences in plant belowground biomass, sediment extractable

44

NH4+, inorganic nitrogen flux, 16S rRNA composition,16S rRNA diversity, or denitrifier

45

functional gene (nirS, norB, and nosZ) abundances associated with oiling status,

46

indicating that certain drivers of ecosystem denitrification capacity have recovered or

47

achieved a new stable state six years after the spill. However, on average,

48

denitrification capacities at the moderately and heavily oiled sites were less than 49% of

49

that of the lightly oiled site (27.7±14.7 and 37.2±24.5 vs 71.8±33.8 µmol N m-2 h-1,

50

respectively). The presence of heavily weathered oiled residue (matched and non-

51

matched for MC252) had no effect on process rates or microbial composition. The loss

52

of function at the moderately and heavily oiled sites compared to the lightly oiled site

53

despite the comparable microbial and environmental factors suggests that oiling

54

intensity plays a role in the long-term recovery of marsh ecosystem services.

56 57

SC

M AN U

TE D

EP

AC C

55

RI PT

32

Keywords: denitrification, Deepwater Horizon, salt marsh, 16S rRNA, recovery

58

Capsule: Six years after the Deepwater Horizon oil spill, denitrification capacities were

59

nearly 2X lower at moderately and heavily oiled salt marsh sites compared to a lightly

60

oiled site, indicating that ecosystem service recovery is occurring at a slower temporal

61

scale than for plant or microbial communities.

62 2

ACCEPTED MANUSCRIPT

63 64

Introduction The recovery of ecosystem services (i.e. biogeochemical cycling) following a disturbance is dependent on both ecosystem resilience and the resilience and/or

66

redundancy of its functional microbial community. Coastal salt marshes remove nearly

67

half of the N inputs they receive through burial, sequestration, and microbially-mediated

68

denitrification (Jordan et al., 2011; Valiela and Cole, 2002), but human-driven

69

disturbances can reduce salt marsh capacity through reduced plant health, microbial

70

community shifts, and ecosystem loss (Barbier et al., 2011; Gedan et al., 2009). The

71

2010 Deepwater Horizon (DWH) oil spill resulted in more than 3 million barrels of crude

72

oil being delivered into the Gulf of Mexico with approximately 160,000 barrels of oil

73

residue impacting over 1800 km of shoreline in the northern Gulf of Mexico (Boufadel et

74

al., 2014; Mendelssohn et al., 2012; Michel et al., 2013; Nixon et al., 2016). This

75

resulted in widespread vegetation die-off (Beland et al., 2017; Lin et al., 2016; Turner et

76

al., 2016) and rapid changes in microbial community structure (Beazley et al., 2012;

77

King et al., 2015; Kostka et al., 2011), which could potentially impact biogeochemical

78

processes such as denitrification.

M AN U

SC

RI PT

65

Denitrification is a microbially-mediated process by which inorganic N, as nitrate

80

(NO3-), is removed by reduction to nitrous oxide and dinitrogen gasses (Knowles, 1982)

81

and is of particular interest as it is a pathway of permanent N removal. Salt marsh

82

vegetation supports denitrification by promoting coupled nitrification-denitrification in the

83

rhizosphere and by providing a C source to heterotrophic denitrifiers (Alldred and

84

Baines, 2016; Koop-Jakobsen and Giblin, 2009). During the DWH spill, oil exposure

85

killed marsh plants and accelerated marsh erosion (Lin and Mendelssohn, 2012;

86

Silliman et al., 2012; Turner et al., 2016). However, within three years of the spill, some

87

marshes showed extensive recovery (Mo et al., 2017; Shapiro et al., 2016), likely driven

88

by survival of belowground plant biomass (Silliman et al., 2016). Belowground root

89

networks promote microbial oil degradation (Beyer et al., 2016), increase soil strength

90

(McClenachan et al., 2013), and provide healthy stock for aboveground recovery

91

(Delaune et al., 2003).

92 93

AC C

EP

TE D

79

Similarly, sediment microbial communities showed rapid compositional shifts following oil exposure (Beazley et al., 2012; Engel and Gupta, 2014; King et al., 2015; 3

ACCEPTED MANUSCRIPT

Kostka et al., 2011). Oiling reduces microbial diversity and promotes the growth of

95

hydrocarbon (HC)-degrading bacteria (Atlas et al., 2015; Mahmoudi et al., 2013) and

96

sulfate reducers (Boopathy et al., 2012; Kostka et al., 2002; Natter et al., 2012). As

97

more labile components of the oil residues are degraded, however, microbial

98

communities can return to a pre-oiling state or a new stable state driven by factors other

99

than oiling status (Atlas et al., 2015; Engel et al., 2017).

RI PT

94

The objective of this study was to determine whether marsh recovery following

101

the DWH oil spill results in subsequent recovery of ecosystem functional capacity (i.e.

102

denitrification capacity). We measured denitrification capacity (isotope pairing

103

technique), microbial 16S rRNA composition (amplicon sequencing), and denitrifier

104

abundance (quantitative PCR) in the Chandeleur Islands (LA, USA), a chain of barrier

105

islands subjected to a range of oiling during the spill (EMRA, 2015) that were not

106

targeted by any clean-up efforts. We sampled from three sites that experienced light,

107

moderate, and heavy oiling during the spill to evaluate the effect of oiling intensity on

108

denitrification capacity six years after the DWH spill. Two key drivers of denitrification in

109

salt marshes, plant biomass and microbial community composition, were immediately

110

impacted by oiling but showed evidence of recovery in the northern Gulf of Mexico

111

within 6 years of the spill (Engel et al., 2017; Shapiro et al., 2016). Given this evidence

112

of long-term ecosystem recovery across a range of oiling intensities following the DWH

113

oil spill, we predicted that 6 years post-spill, denitrification capacities would not differ

114

across oiling intensities.

Methods

117

Site Description

118

AC C

116

EP

115

TE D

M AN U

SC

100

The Chandeleur Islands are a chain of low-lying (<2 m) barrier islands situated

119

approximately 60 km south of Biloxi, Mississippi in the Gulf of Mexico (Fig. S1). The

120

islands run north to south for 80 km and range 2 – 10 km in width. The dominant

121

vegetation is Spartina alterniflora (smooth cordgrass) and Avicennia germinans (black

122

mangrove). During the DWH oil spill, the islands were subjected to a range of very light

123

to heavy oiling (Michel et al., 2013; Nixon et al., 2016). There were no clean-up efforts

124

on the islands (Fig. S1). 4

ACCEPTED MANUSCRIPT

125 126

Experimental Design Based on the Shoreline Cleanup and Assessment Technique (SCAT) data from

128

the summer of 2010 (EMRA, 2015), we established three sampling sites subjected to a

129

range of oiling intensity. SCAT oiling intensities are determined by the distribution,

130

width, and thickness of oil bands (Michel et al., 2013). Site 1, the southernmost point

131

(29.863750º -88.841466º), was subjected to light oiling. Site 2 (29.895448º -

132

88.827780º), located 3.8 km north of Site 1, experienced light to moderate oiling. Site 3

133

(29.927599º -88.829308º) was 3.6 km north of Site 2 and was subjected to moderate to

134

heavy oiling. Marsh sediment and water column were collected from each site on five

135

dates (22 June 2016, 20 September 2016, 19 October 2016, 15 November 2016, and

136

23 February 2017) and from Sites 1 and 3 on one date (24 May 2016) from a different 2

137

m2 area on each date. A tar mat was present within 5 cm of the marsh surface at Site 2

138

in November and February, and small globules of oil residue were observed in the top

139

15 cm of sediment at Site 3 in October and February. Otherwise, no oil residues were

140

visible at the sites.

M AN U

SC

RI PT

127

142 143

TE D

141

Water Column and Sediment Analysis

Point measurements of water column temperature and salinity were taken in the field with a 556 multiprobe (YSI) within a distance <20 m offshore at a water depth of <1

145

m. Site water was filtered (0.45 µm nylon membrane filter) for DIN (NO3-, NO2-, NH4+)

146

and phosphate (PO43-). NO2+3- concentrations were determined microphotometrically via

147

vanadium(III) chloride reduction (Schnetger and Lehners, 2014). NO2- and PO43-

148

concentrations were determined photometrically (Grasshoff, 1983; Koroleff, 1983) on a

149

Genesys 10S UV-Vis spectrophotometer (Thermo Scientific). NH4+ was determined

150

fluorometrically (Holmes et al., 1999: Protocol B).

AC C

151

EP

144

Water column chlorophyll a (chl-a) samples were collected by filtering 60 ml of

152

site water through a 0.7 µm GF/F glass filter, and sediment chl-a samples were

153

determined from the top 2 cm of sediment from each habitat collected. Filters and

154

sediments were freeze-dried, sediment dry weight was recorded, and chl-a was

155

extracted with 90% acetone for 24 hours. Chl-a concentrations were determined 5

ACCEPTED MANUSCRIPT

fluorometrically (Welschmeyer, 1994). Belowground biomass was collected in triplicate

157

in each habitat (10 cm x 9.5 cm i.d.). Roots were separated from sediment by rinsing

158

with tap water through a 2 mm sieve. An additional 20 ml of sitewater was filtered into

159

ashed glass vials for non-purgeable organic carbon (NPOC) measurements on a

160

Shimadzu TOC-VS equipped with an ASI-V autosampler.

161

RI PT

156

Porewater NH4+ was extracted from sediments sampled with a syringe corer at 02 cm and 5-7 cm depths with 50% (w/v) sodium chloride (NaCl) on a shaker table. After

163

24 hours, the slurries were centrifuged and the supernatant was filtered (0.45 µm nylon

164

membrane filter) and frozen until analysis. Porewater NH4+ for samples collected in May

165

and June was determined with a Skalar San+ autoanalyzer whereas samples collected

166

on the remaining dates were analyzed fluorometrically (Holmes et al., 1999: Protocol B).

M AN U

SC

162

167 168

Oil Residue Analysis

Total Petroleum Hydrocarbon (TPH) concentrations were determined from

170

frozen, homogenized 0-10 cm (9.5 cm i.d.) sediment cores for all sampling dates except

171

February using EPA method 8015M DRO/ORO at Pace Analytical Services LLC (New

172

Orleans, LA, USA). Oil source-fingerprinting was also conducted on a subset of the

173

homogenized 0-10 cm cores collected from Site 2 in September, October, November

174

and February, and Site 3 (February only) to determine if oil residues present at Sites 2

175

and 3 were a match for Macondo-252 (MC252) source oil. Due to the risk of oil residue

176

degradation during the flow-through experiment, additional “fresh” residue samples

177

were collected in 0-5 cm cores at Site 2 in February, preserved at -80ºC, and

178

homogenized immediately prior to petroleum fingerprint analysis. Oil source-

179

fingerprinting profiles were determined from petroleum biomarker compounds as

180

measured via GC/MS analytical methods at Louisiana State University (Iqbal et al.,

181

2008; Meyer et al., 2018). Source matching was done by visually comparing hopane

182

and sterane biomarker compound concentrations in their respective m/z191 (hopanes),

183

m/z 217 (sterane) and m/z 218 (sterane) ion chromatograms. The ion chromatograms of

184

extracted field samples were compared to the equivalent data for biomarker compounds

185

in the MC252 source oil (Fig. S3).

AC C

EP

TE D

169

186 6

ACCEPTED MANUSCRIPT

187

Denitrification capacity Denitrification capacity was determined by calculating the production of 29N2 and

188 189

30

190

sediment cores from each site as described previously in Hinshaw et al. 2017. Briefly,

191

marsh sediments were collected in acrylic cores (33 cm x 9.5 cm i.d.), the headspace

192

filled with site water adjacent to the marsh platform and sealed for transport to the lab

193

on ice. On the day of collection, the cores were submerged in site water in a darkened

194

environmental chamber at in situ temperature. After a 16—18 h period, cores were

195

capped with 5 cm overlying water and set up in a continuous-flow (~2.5 ml min-1)

196

system which sent 0.7 µm filtered site water spiked to ~50 µM Na15NO3 (99 atom %;

197

Cambridge Isotope Laboratories, Inc.). Overlying water was stirred with magnetic stir

198

bars attached to the core caps. Following a 24 h equilibration period, triplicate inflow

199

and outflow samples were collected for dissolved gas and nutrient analysis in Exetainer

200

vials (Labco) that were overflowed by 2 tube volumes. Hereafter, the sum of ambient

201

denitrification (i.e., of 14NO3-) rates (D14) plus denitrification stimulated by 15NO3-

202

addition (D15), are referred to as denitrification capacity: denitrification under field

203

conditions when NO3- is not limiting (Koop-Jakobsen and Giblin, 2010).

RI PT

SC

M AN U

TE D

204

N2 according to the isotope pairing techniques (IPT, Nielsen 1992) on three intact

At the end of the sampling period, the top 10 cm of the sediment cores were homogenized and frozen at -20ºC for microbial DNA analysis. Subsamples of the

206

homogenized cores were collected in amber glass jars and stored at -20ºC for total

207

petroleum hydrocarbon (TPH) analysis. Samples for dissolved gas analysis were

208

preserved with 50% (v/w) ZnCl2 and were stored under water in the environmental

209

chamber until dissolved gas analysis by membrane inlet mass spectrometer (MIMS)

210

equipped with a copper reduction column set at 600ºC to remove oxygen (O2) (Eyre et

211

al., 2002; Kana et al., 1994). Water samples for determination of the benthic nutrient

212

fluxes were filtered (0.45 µm) and immediately frozen until analysis for DIN and PO43- as

213

previously described.

214

AC C

EP

205

N2 flux was determined by measuring the differences between the sum of 29N2

215

and 30N2 from inflow and outflow samples from the cores (Eyre et al., 2002; Kana et al.,

216

1994). Similarly, benthic nutrient fluxes were calculated from the inflow and outflow

217

concentrations (Hinshaw et al., 2017). 7

ACCEPTED MANUSCRIPT

218 219 220

DNA Isolation and 16S rRNA Gene Amplicon Sequencing Total genomic DNA was extracted in triplicate from each core with a MP FastDNATM Spin Kit for Soil (MP Biomedicals, LLC) according to the manufacturer’s

222

instructions with the addition of two-minute ice incubations following the homogenization

223

and 4oC centrifugation steps. The triplicate extractions were then pooled and purified

224

using the Zymoclean Gel DNA Recovery Kit (Zymo Research) and eluted in sterile

225

ddH2O (total volume 50 µL). Total DNA concentration was quantified

226

spectrophotmetrically using a NanoDrop ND-1000 spectrophotometer (NanoDrop

227

Technologies) and stored at -20oC.

SC

One core from each site and month was selected at random for 16S rRNA gene

M AN U

228

RI PT

221

229

sequencing for microbial community analysis. Briefly, the V3-V4 region of the 16S rRNA

230

gene from total DNA was amplified with region-specific primers that included Illumina

231

flowcell adapter sequences. Following library preparation, a 2 x 300bp amplification run

232

was conducted using Illumina MiSeq 2000 technology by the Genomics Services

233

Laboratory (Hudson Alpha; Huntsville, AL, USA).

235

Sequence Data Analysis

TE D

234

Briefly, demultiplexed paired-end reads were merged and uploaded to MG-

237

RASTv4 (Meyer et al., 2011) for adapter trimming using Skewer (Jiang et al., 2014) and

238

quality filtering. The following parameters were set in MG-RAST: i) artificial replication

239

reads were removed, ii) reads were screened for host contamination using E.coli, NCBI,

240

st. 536 reference database, iii) reads were quality filtered with a Phred score of 25, and

241

iv) a maximum number of ambiguous bases and length of homopolymers equal to 5.

242

The Blast-Like-Alignment-Tool (BLAT) was used to detect 16S rRNA genes against a

243

reduced rRNA database obtained from a 90% identity clustered version of SILVA

244

release version 128 (Quast et al., 2013). 16S rRNA gene reads were clustered at 97%

245

identity by selecting the longest read of the cluster as the representative. For taxonomic

246

classification, a BLAST similarity search was performed against the Ribosomal

247

Database Project release version 11 (Cole et al., 2014) using queries set at 80%

248

identity and an E-value of 15. MGRASTer (Braithwaite, 2014) and matR (Braithwaite

AC C

EP

236

8

ACCEPTED MANUSCRIPT

249

and Keegan, 2014) packages were used to convert biom files generated from the RDP

250

database into an operational taxonomic unit (OTU) table for further analysis.

251

253

Quantitative PCR Quantitative PCR (qPCR) was performed to assess the abundance of three

RI PT

252

genes in the denitrification pathway: nirS (nitrite reductase), norB (nitric oxide

255

reductase), and nosZ (nitrous oxide reductase) with the same reaction chemistries and

256

quality control methods as previously described in Hinshaw et al. (2017). Total

257

community abundance was determined by qPCR of 16S ribosomal RNA (rRNA).

258

Samples were run in duplicate on a 7000 Sequence Detection System (ABI Prism) with

259

the primer combinations and qPCR conditions listed in Table S1.

M AN U

SC

254

260 261 262

Statistical Analysis

Temporal changes in seawater were tested using analysis of variance (ANOVA) with sampling date as factor. Water column PO43- and chl-a were log10 transformed to

264

meet the assumptions of ANOVA. Sediment chl-a, C:N, plant belowground biomass,

265

nutrient (NO2+3-, NH4+, PO43-) fluxes, denitrification capacities, and marker gene

266

abundances were analyzed by 2-way ANOVA with date and site as factors. Plant

267

belowground biomass, 16S abundance, and norB abundance were log10-transformed to

268

meet the assumptions of ANOVA. Nutrient flux and denitrification capacity data from 15

269

November were excluded from analyses because of an incorrect 15NO3- amendment.

270

Sediment extractable NH4+ was first tested for an effect of depth (0-2 and 5-7) cm using

271

a 1-way ANOVA. Then, the effects of time and site were tested within each depth. Post-

272

hoc analysis for all ANOVAs was conducted with Tukey’s Highly Significant Differences

273

(Tukey’s HSD). Effects were considered significant at α=0.05. Sediment characteristics,

274

process rates and qPCR abundances from May are reported (Tables S2—S5) but were

275

excluded from ANOVAs, as Site 2 was inaccessible on that date.

276

AC C

EP

TE D

263

Statistical analyses including Simpsons 1-D, Shannon, and Chao diversity

277

indices were conducted in the vegan package in R (Oksanen et al., 2015). Diversity

278

indices were evaluated using order level taxa for bacteria. Principal coordinate analysis

279

(PCoA) was conducted on normalized OTU tables using weighted UniFrac to visualize 9

ACCEPTED MANUSCRIPT

changes in community structure across sites. Additionally, analysis of similarity

281

(ANOSIM) on Bray-Curtis similarity matrix was used to assess changes in community

282

structure across sites. To characterize which OTUs contributed most towards

283

differences observed between sites, SIMPER was performed. Two-way ANOVAs with

284

no interaction were used to determine changes in alpha diversity and the relative

285

abundance (%) of dominant taxa and selected HC-degrading orders across dates and

286

sites. All statistical analyses were conducted in R v.3.3.3 (R Core Team, 2015) with

287

RStudio v.1.0.136 (RStudio Team, 2015) except for PCoA and ANOSIM, which were

288

conducted in PAST 3 (Hammer et al., 2001).

Results

291

Water Column Characteristics

292

SC

290

M AN U

289

RI PT

280

Water column temperature did not vary more than 2ºC between sites for each sampling date, but varied seasonally, peaking at 30.4ºC in September and dropping to

294

19.9ºC in November and February. Salinity ranged from 21.1 ppt (May) to 33.5 ppt

295

(November) and did not vary more than 3.5 ppt between sites for each sampling date.

296

Sitewater NO2+3- was below detection for all months except October (1.0±1.0 µM) and

297

November (2.0±1.3 µM), with NO2- on average accounting for less than 5% of total

298

NO2+3-. Sitewater PO43- concentrations were low (below detection in May, 0.1±0.1 µM

299

for remaining months) and did not vary temporally (ANOVA, p=0.632). Similar to PO43-,

300

sitewater NH4+ concentrations (6.0±9.0 µM) did not vary temporally (ANOVA, p=0.814).

EP

301

TE D

293

Water column NPOC varied seasonally (ANOVA, p<0.001), with peak concentrations during the warm months (June and September) that were roughly twice

303

as high as concentrations during the cooler months (October, November, and February)

304

(Tukey’s HSD, p<0.001). Water column chl-a was higher in May (21.6±8.7 µg L-1) than

305

all months but June (Tukey’s HSD, p<0.026) and lower than all other months in

306

November (3.2±0.7 µg L-1) and February (3.7±0.3 µg L-1) (Tukey’s HSD, p<0.039).

AC C

302

307 308

Sediment Characteristics

309

Sediment molar C:N differed between sites (ANOVA, p=0.01) and was higher at

310

Site 1 (12.1±2.1) than Sites 2 (10.0±2.8) and 3 (9.5±3.2) (Tukey’s HSD, p<0.04) across 10

ACCEPTED MANUSCRIPT

all dates. Although there was a significant effect of month on sediment C:N (ANOVA,

312

p=0.03), post-hoc analysis did not identify any statistically significant seasonal trends at

313

α=0.05. Temporal patterns in sediment chl-a differed between sites (ANOVA date x site,

314

p<0.001). Site 1 showed the least temporal variation of the three sites, and while

315

October chl-a inventories at Site 1 (219±78 mg m-2) were >2.5X higher than Sites 2 and

316

3 (Tukey’s HSD, p<0.04), there were no differences between months at Site 1. Chl-a

317

inventories at Sites 2 and 3 peaked in November, with inventories at Site 2 at that time

318

being higher than all other sites and dates (Tukey’s HSD, p<0.001; 474±38 mg m-2).

319

There was no effect of oiling intensity on plant belowground biomass (ANOVA,

320

p=0.196). Rather, biomass at all three sites varied over time (ANOVA, p<0.001),

321

peaking in June then decreasing nearly 5-fold in September (Tukey’s HSD, p<0.009).

SC

Sediment extractable NH4+ concentrations did not differ between surface (0-2

M AN U

322

RI PT

311

323

cm) and subsurface (5-7) sediments (1-way ANOVA, p=0.146). Neither surface

324

sediment extractable NH4+ nor subsurface extractable NH4+ varied between months

325

(ANOVA, p=0.189, =0.110) or sites (ANOVA, p=0.303, =0.240).

326

328

Oil Residues

TE D

327

TPH was below the detection limit of 25 mg kg-1 for all but two cores from Site 3 in June (58.4 mg kg-1) and October (45 mg kg-1), and one core from Site 2 in November

330

(290 mg kg-1). Oil source-fingerprinting analyses did not detect biomarker compounds in

331

these samples (Table 1), indicating an absence of crude oil residue at Site 2 in

332

September and October, which is consistent with TPH results for these samples.

333

Heavily weathered oil residue was detected at Site 2 in November and was a possible

334

match to MC252 (Table 1). The source-fingerprinting results from Site 2 in February

335

were conflicting. The quantitative composition of the hopane and sterane biomarker

336

compounds in surficial (0-5 cm) sediments were a very close match to the composition

337

in MC252 source oil, while the respective biomarker composition in the flow-through

338

core (0-10 cm) was not similar, and thus classifies as a non-match with the MC252

339

composition (Fig. S2, S3). At Site 3 in February, a small amount of oil residue was

340

detected but was a non-match for MC252. The non-matching biomarker compositions in

341

these oily residues indicate that they came from a different crude oil than was spilled

AC C

EP

329

11

ACCEPTED MANUSCRIPT

342

during the Deepwater Horizon accident and this no-match classification was not the

343

result of MC252 biomarker weathering.

344 345

Denitrification capacities and Nutrient Fluxes Denitrification capacities varied both spatially (ANOVA, p<0.001) and temporally

347

(ANOVA, p=0.001). On average, denitrification capacity was nearly 2X higher at Site 1,

348

the lightly oiled site (71.8±33.8 µmol N m-2 h-1), compared to Sites 2 and 3 (Tukey’s

349

HSD, p<0.003; 27.7±14.7 and 37.2±24.5 µmol N m-2 h-1, respectively) (Fig. 1A). The

350

lowest rates occurred in June (27.7±17.8 µmol N m-2 h-1), which were less than half of

351

the highest rates, which occurred in September (63.9±27.4 µmol N m-2 h-1). Less than

352

7% of total denitrification was from D14, indicating low in situ rates.

SC

M AN U

353

RI PT

346

Unlike denitrification capacities, DIN fluxes varied temporally (ANOVA, p<0.001) but not spatially ANOVA, p>0.225). NH4+ flux was positive (i.e. net production) for all

355

months (Fig. 1B), with the highest rates in June more than 5X greater than the lowest

356

rates in February (Tukey’s HSD, p<0.001). Unlike NH4+ flux, NO2+3- flux was negative (i.e.

357

net consumption) for all months but February (Tukey’s HSD, p<0.001), with the highest

358

consumption occurring in October (Tukey’s HSD, p<0.001; Fig. 1C).

359

TE D

354

In general, PO43- flux was positive but lower (<20 µmol PO43- m-2 h-1) in magnitude than the other nutrient fluxes, with different temporal patterns depending on the site

361

(ANOVA, date x site p<0.001; Fig. 1D). PO43- flux at Site 1 was highest in June (Tukey’s

362

HSD, p<0.008) and was >14X greater at than Sites 2 or 3 (Tukey’s HSD, p<0.001) for

363

that date. At Sites 2 and 3, PO43-

365 366

flux

was highest in September (Tukey’s HSD, p<0.021).

AC C

364

EP

360

16S rRNA Community Structure Of the 10,163,297 quality filtered sequences, 79% (8,032,957) were assigned

367

OTU IDs from the Ribosomal Database Project in order to evaluate bacterial community

368

composition. There were no changes in alpha diversity (Simpson, Shannon, and Chao)

369

over sites or months (Table S5). There were no differences in community structure

370

between sites (ANOSIM, R=0.159 p=0.013; Fig. 2A). Proteobacteria, Firmicutes,

371

Bacteroidetes, and Planctomycetes accounted for 62% of the total OTUs overall (Fig.

372

S4). More than a quarter (27%) of OTUs were unclassified. The most abundant phyla 12

ACCEPTED MANUSCRIPT

had different spatial and temporal patterns (Figure S4). Neither Proteobacteria nor

374

Bacteroidetes relative abundance differed between sites (ANOVA, p=0.806 and 0.741).

375

On average, Firmicutes was more abundant at Site 1 than the other two sites (Tukey’s

376

HSD, p<0.003); in contrast Planctomycetes was more abundant at Site 3 compared to

377

Site 1 (Tukey’s HSD, p=0.002). Firmicutes was the only phylum to change over time,

378

with lower relative abundance in February than May and October (Tukey’s HSD,

379

p=0.007 and 0.006).

RI PT

373

Twelve hydrocarbon (HC)-degrading classes were detected in the 16S

381

sequences (Alteromonadales, Chromatiales, Desulfobacterales, Desulfovibrionales,

382

Desulfuromonadales, Enterobacteriales, Methylococcales, Oceanospirillales,

383

Pseudomonadales, Rhizobiales, Rhodobacterales, and Thiotrichales group) (Engel and

384

Gupta, 2014; Gao et al., 2015; Harwati et al., 2007; Hazen et al., 2010; Head et al.,

385

2006; Joye et al., 2011; Kostka et al., 2011; Prince et al., 2003; Valentine et al., 2012).

386

The relative abundance (%) within the total community was calculated for each. In total,

387

the HC-degrading classes contributed to 4% of microbial abundance on average (Fig.

388

2B). The presence of a tar mat at Site 2 did not affect relative HC-degrader relative

389

abundance; there were no differences between sites (ANOVA, p=0.656; Fig. 2B) or

390

dates (ANOVA, p=0.285). Only one taxa (Alteromonadales) contributed to at least 1%

391

relative abundance (Fig. S5). Only Desulfovibrionales differed between sites (ANOVA,

392

p=0.033), with a trend of lower relative abundance at Sites 2 sand 3 compared to Site 1,

393

though the effect was marginal (Tukey’s HSD, p=0.072 and 0.038). There was no

394

change in the relative abundance of Desulfovibrionales over time (ANOVA, p=0.388)

395

(Fig S5). No other HC-degrading taxa differed between sites and months (Fig S5).

397 398

M AN U

TE D

EP

AC C

396

SC

380

QPCR

16S abundance (i.e. bacterial abundance) did not differ between sites (ANOVA,

399

p=0.402; Fig. 3A). 16S abundance was the same across all dates except for February,

400

when it was >5X lower compared to all other sampling dates (ANOVA, p<0.001; Tukey’s

401

HSD, p <0.001).

402 403

The effect of site on nirS-type denitrifier (nitrite reductase) abundance depended on sampling date (ANOVA, date x site p=0.001). The highest nirS-type denitrifier 13

ACCEPTED MANUSCRIPT

abundance occurred at Site 3 in November, at which point abundance was >4X greater

405

than Sites 1 and 2 (Tukey’s HSD, p<0.001; Fig. 3B). In contrast, norB (nitric oxide

406

reductase) abundance did not differ between sites (ANOVA, p=0.94; Fig. 3C). Rather,

407

norB abundance followed a similar pattern to 16S, with the lowest abundances

408

occurring in February compared to all other sampling dates (ANOVA, p<0.001; Tukey’s

409

HSD, p<0.001). Unlike the other functional markers, there was no effect of site (ANOVA

410

p=0.425) or date (ANOVA p=0.265) on nosZ (nitrous oxide reductase) abundance (Fig.

411

3D).

RI PT

404

413

SC

412

Discussion

There were no differences in plant belowground biomass, sediment extractable

415

NH4+, DINflux, 16S rRNA composition,16S rRNA diversity, or denitrifier functional gene

416

abundance associated with oiling status, indicating that certain drivers of ecosystem

417

denitrification capacity have recovered or achieved a new stable state six years after the

418

DWH oil spill. However, while denitrification capacities at all sites were comparable to

419

denitrification capacities in Gulf of Mexico salt marshes prior to and following the DWH

420

spill (Hinshaw et al., 2017; Rivera-Monroy et al., 2013, 2010), the lower capacities at

421

moderately to heavily oiled sites compared to the lightly oiled site suggest that

422

ecosystem service recovery is occurring at a slower temporal scale than plant or

423

microbial communities. Although vegetation in impacted marshes typically recovered

424

within a few years of oiling (Beland et al., 2017; Mo et al., 2017; Shapiro et al., 2016),

425

we expect that as in restored salt marshes, the recovery of ecosystem services at more

426

heavily oiled sites occurs at a longer timescale (e.g. Moreno-Mateos et al., 2012).

TE D

EP

The sediment C:N pools and PO43-flux rates were characteristic of younger (i.e.

AC C

427

M AN U

414

428

recovering) salt marshes. Increased PO43- demand at the moderately and heavily oiled

429

sites was concomitant with peak belowground biomass in June, possibly due to the

430

higher plant P demand of younger marshes (Murphy et al., 2017; van Wijnen and

431

Bakker, 1999). We also measured lower sediment C pools at the more intensely oiled

432

sites, which could reflect lower C pools such as observed in developing or recovering

433

wetlands (Craft et al., 2003; Moreno-Mateos et al., 2012; Tyler and Zieman, 1999). This

434

loss of C may account for the lower denitrification rates that occurred when NO3- was 14

ACCEPTED MANUSCRIPT

435

not limiting, as the lower C pools in newer marshes can limit heterotrophic denitrification

436

(He et al., 2016). While denitrifier marker abundance varied little over space and time,

437

nirS-type denitrifier abundance at the heavily oiled site coincided with higher sediment

438

chl-a inventories, a proxy for labile C availability. Oil residue can increase sediment NO3- uptake and denitrifier functional group

RI PT

439

abundance (Brian M. Levine et al., 2017; Scott et al., 2014) by providing an organic C

441

source to the heterotrophic denitrifying community, but we did not a detect a response

442

when oil residues were present in our samples. It is important to note that the oily

443

residues that were detected in 2016 and 2017 were very heavily weathered, and thus

444

would degrade extremely slowly, limiting the available C source from these residues. In

445

most cases, crude oil that reached salt marsh sediments following the 2010 spill was

446

rapidly weathered by oil degrading microbial communities, leaving recalcitrant

447

hydrocarbon components behind (Aeppli et al., 2012; Mahmoudi et al., 2013; Natter et

448

al., 2012). Source-fingerprinting analysis showed that residues (whether MC252

449

matches or otherwise) from the study sites were all heavily weathered, suggesting that

450

they are not a readily available C-source to heterotrophic denitrifiers. Given that oil

451

residues in the environment are not evenly distributed, it is not surprising that the results

452

of oil residue analyses in 2016 and 2017 do not exactly line up with the SCAT oiling

453

designations from 2010 surveys. By the time these samples were collected six years

454

post-spill, only very heavily weathered oily residues would have remained in any areas

455

impacted by the 2010 spill (Bagby et al., 2017).

EP

TE D

M AN U

SC

440

Despite the different denitrification rates, we measured comparable NO2+3-

457

demand across the oiling gradient, suggesting an alternate pathway of nitrate reduction

458

such as dissimilatory nitrate reduction to ammonium (DNRA) at the moderately and

459

heavily oiled sites. Oiling inhibits sediment aeration (B.M. Levine et al., 2017) and

460

promotes the accumulation of reduced sulfur in sediment as hydrocarbons favor the

461

growth of sulfate-reducing bacteria (Boopathy et al., 2012; Natter et al., 2012), which

462

inhibits coupled nitrification-denitrification (Dollhopf et al., 2005; Kostka et al., 2002).

463

Anaerobic conditions in sediment following marsh die-back can also increase sulfide

464

accumulation in sediment (sensu King et al., 2008). DNRA can account for more than

465

30% of NO3- reduction in coastal salt marshes (Giblin et al., 2013), and while direct

AC C

456

15

ACCEPTED MANUSCRIPT

466

measurements are necessary to confirm this hypothesis, it is important to consider that

467

oiling may have shifted the dominant NO3- reduction pathway from removal to recycling. Oiling promotes the growth of hydrocarbon degrading bacteria (Bae et al., 2018;

469

Beazley et al., 2012; Engel et al., 2017; Kostka et al., 2011), and we expected to see a

470

greater relative abundance of HC-degraders at the more oiled sites. However, only one

471

of the targeted hydrocarbon degrading taxa (Desulfovibrionales) showed a consistent

472

trend of higher relative abundance in response to moderate to heavy oiling. We expect

473

that the spatial distribution of HC-degraders is associated with the distribution of

474

residues in the sediment such as the higher relative abundance Pseudomonadales,

475

Rhizobiales, and Rhodobacteriales at Site 2 when residues were present (Fig. S3),

476

though it should be noted that their contribution to the total microbial community was

477

less than 1%. That there were no differences between oiling intensities in community

478

structure or alpha-diversity suggests that microbial communities at the study site have, if

479

not recovered to a pre-oiling state, achieved a new or different stable state shaped by

480

factors other than oiling intensity (e.g. Engel et al., 2017).

M AN U

SC

RI PT

468

We found that 6 years after DWH oil spill, unremediated marsh sites subjected to

482

moderate and heavy oiling had reduced denitrification capacity by nearly half compared

483

to a lightly oiled site. This loss in function is likely due to a loss in sediment C, which

484

limits heterotrophic denitrifier activity, similar to what has been observed in restored

485

marshes. Our findings also indicated that the highly weathered oil residue at the

486

moderately and heavily oiled sites did not impact denitrification capacity or microbial

487

community structure, likely due to its patchy distribution and recalcitrant composition.

488

Rather, it appears that the microbial communities at all three levels of oiling intensity

489

have shifted to comparable post-oiling stable states. The loss of function at the

490

moderately and heavily oiled compared to the lightly oiled site in spite of the comparable

491

rate of plant and microbial recovery suggests that oiling intensity plays a role in the

492

long-term recovery of marsh ecosystem services.

AC C

EP

TE D

481

493 494

Acknowledgements. We thank Patrice Crawford, Taylor Ledford, Patrick Chanton,

495

Peter Whitehurst, and Caitlyn Taylor for their help in the field and lab. This research was

496

made possible by a grant from The Gulf of Mexico Research Initiative awarded to the 16

ACCEPTED MANUSCRIPT

Alabama Center for Ecological Resilience consortium. Data are publicly available

498

through the Gulf of Mexico Research Initiative Information & Data

499

Cooperative (GRIIDC) at https://data.gulfresearchinitiative.org (DOI:10.7266/N7251GJ3,

500

10.7266/N7V69H36, 10.7266/N7GQ6WCZ). Illumina MiSeq raw reads are available

501

from NCBI Sequence Read Archive (SRA) database (Bioproject PRJNA390775). We

502

thank the three anonymous reviewers for their insightful comments.

RI PT

497

503

Literature Cited

505

Aeppli, C., Carmichael, C.A., Nelson, R.K., Lemkau, K.L., Graham, W.M., Redmond,

SC

504

M.C., Valentine, D.L., Reddy, C.M., 2012. Oil weathering after the Deepwater

507

Horizon disaster led to the formation of oxygenated residues. Environ. Sci. Technol.

508

46, 8799–8807. https://doi.org/10.1021/es3015138

509

M AN U

506

Alldred, M., Baines, S.B., 2016. Effects of wetland plants on denitrification rates: a

510

meta-analysis. Ecol. Appl. 26, 676–685. https://doi.org/https://doi.org/10.1890/14-

511

1525

Atlas, R.M., Stoeckel, D.M., Faith, S.A., Minard-Smith, A., Thorn, J.R., Benotti, M.J.,

513

2015. Oil Biodegradation and Oil-Degrading Microbial Populations in Marsh

514

Sediments Impacted by Oil from the Deepwater Horizon Well Blowout. Environ. Sci.

515

Technol. 49, 8356–8366. https://doi.org/10.1021/acs.est.5b00413

516

TE D

512

Bae, H.S., Huang, L., White, J.R., Wang, J., DeLaune, R.D., Ogram, A., 2018. Response of microbial populations regulating nutrient biogeochemical cycles to

518

oiling of coastal saltmarshes from the Deepwater Horizon oil spill. Environ. Pollut.

519

241, 136–147. https://doi.org/10.1016/j.envpol.2018.05.033

AC C

520

EP

517

Bagby, S.C., Reddy, C.M., Aeppli, C., Fisher, G.B., Valentine, D.L., 2017. Persistence

521

and biodegradation of oil at the ocean floor following Deepwater Horizon. Proc.

522

Natl. Acad. Sci. 114, E9–E18. https://doi.org/10.1073/pnas.1610110114

523

Barbier, E.B., Hacker, S.D., Kennedy, C., Koch, E.W., Stier, A.C., Silliman, B.R., 2011.

524

The value of estuarine and coastal ecosystem services. Ecol. Monogr. 81, 169–

525

193. https://doi.org/10.1890/10-1510.1

526

Beazley, M.J., Martinez, R.J., Rajan, S., Powell, J., Piceno, Y.M., Tom, L.M., Andersen,

527

G.L., Hazen, T.C., Van Nostrand, J.D., Zhou, J., Mortazavi, B., Sobecky, P.A., 17

ACCEPTED MANUSCRIPT

528

2012. Microbial community analysis of a coastal salt marsh affected by the

529

Deepwater Horizon oil spill. PLoS One 7, e41305.

530

https://doi.org/10.1371/journal.pone.0041305

531

Beland, M., Biggs, T.W., Roberts, D.A., Peterson, S.H., Kokaly, R.F., Piazza, S., 2017. Oiling accelerates loss of salt marshes, southeastern Louisiana. PLoS One 12,

533

e0181197. https://doi.org/https://doi. org/10.1371/journal.pone.0181197 Editor:

534

RI PT

532

Beyer, J., Trannum, H.C., Bakke, T., Hodson, P. V., Collier, T.K., 2016. Environmental effects of the Deepwater Horizon oil spill: A review. Mar. Pollut. Bull. 110, 28–51.

536

https://doi.org/10.1016/j.marpolbul.2016.06.027

537

SC

535

Boopathy, R., Shields, S., Nunna, S., 2012. Biodegradation of crude oil from the BP oil spill in the marsh sediments of southeast Louisiana, USA. Appl. Biochem.

539

Biotechnol. 167, 1560–1568. https://doi.org/10.1007/s12010-012-9603-1

540

M AN U

538

Boufadel, M.C., Abdollahi-Nasab, A., Geng, X., Galt, J., Torlapati, J., 2014. Simulation

541

of the landfall of the Deepwater Horizon oil on the shorelines of the Gulf of Mexico.

542

Environ. Sci. Technol. 48, 9496–9505. https://doi.org/10.1021/es5012862 Braithwaite, D.T., 2014. MGRASTer.

544

Braithwaite, D.T., Keegan, K., 2014. matR.

545

Cole, J., Wang, Q., Fish, J., Chai, B., McGarrell, D., Sun, Y., Brown, C., Porras-Alfaro,

546

A., Kuske, C., Tiedje, J.M., 2014. Ribosomal Database Project: data and tools for

547

high throughput rRNA analysis. Nucleic Acids Res. 42, D633–D642.

548

https://doi.org/10.1093/nar/gkt1244

EP

TE D

543

Craft, C., Megonigal, P., Broome, S., Stevenson, J., Freese, R., Cornell, J., Zheng, L.,

550

Sacco, J., 2003. The pace of ecosystem development of constructed Spartina

551

alterniflora marshes. Ecol. Appl. 13, 1417–1432.

552

AC C

549

Delaune, R.D., Pezeshki, S.R., Jugsujinda, A., Lindau, C.W., 2003. Sensitivity of US

553

Gulf of Mexico coastal marsh vegetation to crude oil: Comparison of greenhouse

554

and field responses. Aquat. Ecol. 37, 351–360.

555

https://doi.org/10.1023/B:AECO.0000007047.42636.96

556

Dollhopf, S.L., Hyun, J., Smith, A.C., Adams, H.J., O’Brien, S., Kostka, J.E., 2005.

557

Quantification of Ammonia-Oxidizing Bacteria and Factors Controlling Nitrification

558

in Salt Marsh Sediments. Appl. Environ. Microbiol. 71, 240–246. 18

ACCEPTED MANUSCRIPT

559 560

https://doi.org/10.1128/AEM.71.1.240 EMRA, 2015. Web Application: Gulf of Mexico Environmental Response Management

561

Application, National Oceanic and Atmospheric Administration [WWW Document].

562

URL https://erma.noaa.gov/gulfofmexico/erma.html (accessed 2.21.18). Engel, A.S., Gupta, A.A., 2014. Regime shift in sandy beach microbial communities

RI PT

563 564

following Deepwater Horizon oil spill remediation efforts. PLoS One 9, e102934.

565

https://doi.org/10.1371/journal.pone.0102934

Engel, A.S., Liu, C., Paterson, A.T., Anderson, L.C., Turner, R.E., Overton, E.B., 2017.

567

Salt marsh bacterial communities before and after the Deepwater Horizon oil spill.

568

Appl. Environ. Microbiol. 83, e00784-17. https://doi.org/10.1128/AEM.00784-17

SC

566

Eyre, B.D., Rysgaard, S., Dalsgaard, T., Christensen, P.B., 2002. Comparison of

570

Isotope Pairing and N2:Ar Methods for Measuring Sediment Denitrification—

571

Assumptions, Modifications, and Implications. Estuaries 25, 1077–1087.

572

https://doi.org/https://doi.org/10.1007/BF02692205

M AN U

569

Gao, X., Gao, W., Cui, Z., Han, B., Yang, P., Sun, C., Zheng, L., 2015. Biodiversity and

574

degradation potential of oil-degrading bacteria isolated from deep-sea sediments of

575

South Mid-Atlantic Ridge. Mar. Pollut. Bull. 97, 373–380.

576

https://doi.org/10.1016/j.marpolbul.2015.05.065

577

TE D

573

Gedan, K.B., Silliman, B.R., Bertness, M.D., 2009. Centuries of human-driven change in salt marsh ecosystems. Ann. Rev. Mar. Sci. 1, 117–141.

579

https://doi.org/10.1146/annurev.marine.010908.163930

EP

578

Giblin, A., Tobias, C., Song, B., Weston, N., Banta, G., Rivera-Monroy, V., 2013. The

581

Importance of Dissimilatory Nitrate Reduction to Ammonium (DNRA) in the

582

Nitrogen Cycle of Coastal Ecosystems. Oceanography 26, 124–131.

583

https://doi.org/10.5670/oceanog.2013.54

584

AC C

580

Grasshoff, K., 1983. Determination of nitrite, in: Grasshoff, K., Ehrhardt, M., Kremling,

585

K. (Eds.), Methods of Seawater Analysis. verlag chemie, pp. 139–142.

586

Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: Paleontological statistics

587 588 589

software package for education and data analysis. Palaeontol. Electron. 4. Harwati, T.U., Kasai, Y., Kodama, Y., Susilaningsih, D., Watanabe, K., 2007. Characterization of Diverse Hydrocarbon-Degrading Bacteria Isolated from 19

ACCEPTED MANUSCRIPT

590

Indonesian Seawater. Microbes Environ. 22, 412–415.

591

https://doi.org/10.1264/jsme2.22.412

592

Hazen, T.C., Dubinsky, E.A., DeSantis, T.Z., Andersen, G.L., Piceno, Y.M., Singh, N., Jansson, J.K., Probst, A., Borglin, S.E., Fortney, J.L., Stringfellow, W.T., Bill, M.,

594

Conrad, M.E., Tom, L.M., Chavarria, K.L., Alusi, T.R., Lamendella, R., Joyner,

595

D.C., Spier, C., Baelum, J., Auer, M., Zemla, M.L., Chakraborty, R., Sonnenthal,

596

E.L., D’haeseleer, P., Holman, H.-Y.N., Osman, S., Lu, Z., Van Nostrand, J.D.,

597

Deng, Y., Zhou, J., Mason, O.U., 2010. Deep-Sea Oil Plume Enriches Indigenous

598

Oil-Degrading Bacteria. Science (80-. ). 330, 204–208.

599

https://doi.org/10.1126/science.1195979

SC

He, Y., Widney, S., Ruan, M., Herbert, E., Li, X., Craft, C., 2016. Accumulation of soil

M AN U

600

RI PT

593

601

carbon drives denitrification potential and lab-incubated gas production along a

602

chronosequence of salt marsh development. Estuar. Coast. Shelf Sci. 172, 72–80.

603

https://doi.org/10.1016/j.ecss.2016.02.002

605 606

Head, I.M., Jones, D.M., Röling, W.F.M., 2006. Marine microorganisms make a meal of oil. Nat. Rev. Microbiol. 4, 173–182. https://doi.org/10.1038/nrmicro1348 Hinshaw, S.E., Tatariw, C., Flournoy, N., Kleinhuizen, A., Taylor, C.B., Sobecky, P.,

TE D

604

607

Mortazavi, B., 2017. Vegetation loss decreases salt marsh denitrification capacity:

608

Implications for marsh erosion. Environ. Sci. Technol. 51, 8245–8253.

609

https://doi.org/10.1021/acs.est.7b00618 Holmes, R.M., Aminot, A., Kérouel, R., Hooker, B.A., Peterson, B.J., 1999. A simple and

EP

610

precise method for measuring ammonium in marine and freshwater ecosystems.

612

Can. J. Fish. Aquat. Sci. 56, 1801–1808.

613

https://doi.org/doi/pdf/10.4319/lo.1992.37.1.0105

614

AC C

611

Iqbal, J., Overton, E.B., Gisclair, D., 2008. Polycyclic aromatic hydrocarbons in

615

Louisiana rivers and coastal environments: Source fingerprinting and forensic

616

analysis. Environ. Forensics 9, 63–74. https://doi.org/10.1080/15275920801888301

617

Jiang, H., Lei, R., Ding, S.W., Zhu, S., 2014. Skewer: A fast and accurate adapter

618

trimmer for next-generation sequencing paired-end reads. BMC Bioinformatics 15,

619

1–12. https://doi.org/10.1186/1471-2105-15-182

620

Jordan, S.J., Stoffer, J., Nestlerode, J.A., 2011. Wetlands as Sinks for Reactive 20

ACCEPTED MANUSCRIPT

621

Nitrogen at Continental and Global Scales: A Meta-Analysis. Ecosystems 14, 144–

622

155. https://doi.org/10.1007/s10021-010-9400-z

623

Joye, S.B., MacDonald, I.R., Leifer, I., Asper, V., 2011. Magnitude and oxidation potential of hydrocarbon gases released from the BP oil well blowout. Nat. Geosci.

625

4, 160–164. https://doi.org/10.1038/ngeo1067

626

RI PT

624

Kana, T.M., Darkangelo, C., Hunt, M.D., Oldham, J.B., Bennett, G.E., Cornwell, J.C., 1994. Membrane inlet mass spectrometer for rapid high-precision determination of

628

N2, O2, and Ar in environmental water samples. Anal. Chem. 66, 4166–4170.

629

https://doi.org/10.1021/ac00095a009

630

SC

627

King, G.M., Klug, M.J., Wiegert, R.G., Chalmers, A.G., 2008. Relation of Soil Water Movement and Sulfide Concentration to Spartina alterniflora Production in a

632

Georgia Salt Marsh. Science (80-. ). 218, 61–63.

633

M AN U

631

King, G.M., Kostka, J.E., Hazen, T.C., Sobecky, P.A., 2015. Microbial Responses to the

634

Deepwater Horizon Oil Spill: From Coastal Wetlands to the Deep Sea. Ann. Rev.

635

Mar. Sci. 7, 377–401. https://doi.org/10.1146/annurev-marine-010814-015543 Knowles, R., 1982. Denitrification. Microbiol. Rev. 46, 43–70.

637

Koop-Jakobsen, K., Giblin, A.E., 2010. The effect of increased nitrate loading on nitrate

638

reduction via denitrification and DNRA in salt marsh sediments. Limnol. Oceanogr.

639

55, 789–802. https://doi.org/doi/10.4319/lo.2010.55.2.0789

640

TE D

636

Koop-Jakobsen, K., Giblin, A.E., 2009. New approach for measuring denitrification in the rhizosphere of vegetated marsh sediments. Limnol. Oceanogr. Methods 7,

642

626–637. https://doi.org/10.4319/lom.2009.7.626

644 645

Koroleff, F., 1983. Determination of phosphorus, in: Grasshoff, K., Ehrhardt, M.,

AC C

643

EP

641

Kremling, K. (Eds.), Methods of Seawater Analysis. verlag chemie, pp. 125–131. Kostka, J.E., Gribsholt, B., Petrie, E., Dalton, D., Skelton, H., Kristensen, E., 2002. The

646

rates and pathways of carbon oxidation in bioturbated saltmarsh sediments. Limnol.

647

Oceanogr. 47, 230–240. https://doi.org/10.4319/lo.2002.47.1.0230

648

Kostka, J.E., Prakash, O., Overholt, W.A., Green, S.J., Freyer, G., Canion, A.,

649

Delgardio, J., Norton, N., Hazen, T.C., Huettel, M., 2011. Hydrocarbon-degrading

650

bacteria and the bacterial community response in Gulf of Mexico beach sands

651

impacted by the Deepwater Horizon oil spill. Appl. Environ. Microbiol. 77, 7962– 21

ACCEPTED MANUSCRIPT

652 653

7974. https://doi.org/10.1128/AEM.05402-11 Levine, B.M., White, J.R., DeLaune, R.D., 2017. Impacts of the long-term presence of

654

buried crude oil on salt marsh soil denitrification in Barataria Bay, Louisiana. Ecol.

655

Eng. 99, 454–461. https://doi.org/10.1016/j.ecoleng.2016.11.017 Levine, B.M., White, J.R., DeLaune, R.D., Maiti, K., 2017. Crude Oil Effects on Redox

RI PT

656 657

Status of Salt Marsh Soil in Louisiana. Soil Sci. Soc. Am. J. 81, 647.

658

https://doi.org/10.2136/sssaj2016.12.0398

Lin, Q., Mendelssohn, I.A., 2012. Impacts and recovery of the Deepwater Horizon oil

660

spill on vegetation structure and function of coastal salt marshes in the northern

661

Gulf of Mexico. Env. Sci Technol 46, 3737–3743.

662

https://doi.org/10.1021/es203552p

M AN U

663

SC

659

Lin, Q., Mendelssohn, I.A., Graham, S.A., Hou, A., Fleeger, J.W., Deis, D.R., 2016.

664

Response of salt marshes to oiling from the Deepwater Horizon spill: Implications

665

for plant growth, soil surface-erosion, and shoreline stability. Sci. Total Environ.

666

557–558, 369–377. https://doi.org/10.1016/j.scitotenv.2016.03.049 Mahmoudi, N., Porter, T.M., Zimmerman, A.R., Fulthorpe, R.R., Kasozi, G.N., Silliman,

668

B.R., Slater, G.F., 2013. Rapid degradation of Deepwater Horizon spilled oil by

669

indigenous microbial communities in Louisiana saltmarsh sediments. Environ. Sci.

670

Technol. 47, 13303–13312. https://doi.org/10.1021/es4036072

671

TE D

667

McClenachan, G., Turner, R.E., Tweel, A.W., 2013. Effects of oil on the rate and trajectory of Louisiana marsh shoreline erosion. Environ. Res. Lett. 8, 044030.

673

https://doi.org/10.1088/1748-9326/8/4/044030 Mendelssohn, I.A., Andersen, G.L., Baltz, D.M., Caffey, R.H., Carman, K.R., Fleeger,

AC C

674

EP

672

675

J.W., Joye, S.B., Lin, Q., Maltby, E., Overton, E.B., Rozas, L.P., 2012. Oil Impacts

676

on Coastal Wetlands: Implications for the Mississippi River Delta Ecosystem after

677

the Deepwater Horizon Oil Spill. Bioscience 62, 562–574.

678

https://doi.org/10.1525/bio.2012.62.6.7

679

Meyer, B.M., Adhikari, P.L., Olson, G.M., Overton, E.., Miles, M.S., 2018. Louisiana

680

coastal marsh environments and MC252 oil biomarker chemistry, in: Wang, Z.,

681

Stout, S.A. (Eds.), Oil Spill Environmental Forensics Case Studies. Butterworth-

682

Heinemann, Oxford, UK. 22

ACCEPTED MANUSCRIPT

683

Meyer, F., Paarmann, D., D’Souza, M., Olson, R., Glass, E., Kubal, M., Paczian, T., Rodriguez, A., Stevens, R., Wilke, A., Wilkening, J., Edwards, R., 2011. The

685

Metagenomics RAST Server: A Public Resource for the Automatic Phylogenetic

686

and Functional Analysis of Metagenomes. BMC Bioinformatics 8, 325–331.

687

https://doi.org/10.1002/9781118010518.ch37

688

RI PT

684

Michel, J., Owens, E.H., Zengel, S., Graham, A., Nixon, Z., Allard, T., Holton, W.,

Reimer, P.D., Lamarche, A., White, M., Rutherford, N., Childs, C., Mauseth, G.,

690

Challenger, G., Taylor, E., 2013. Extent and degree of shoreline oiling: Deepwater

691

Horizon Oil Spill, Gulf of Mexico, USA. PLoS One 8, e65087.

692

https://doi.org/10.1371/journal.pone.0065087

Mo, Y., Kearney, M.S., Riter, J.C.A., 2017. Post-Deepwater Horizon Oil Spill Monitoring

M AN U

693

SC

689

694

of Louisiana Salt Marshes Using Landsat Imagery. Remote Sens. 9, 1–12.

695

https://doi.org/10.3390/rs9060547

696

Moreno-Mateos, D., Power, M.E., Comín, F.A., Yockteng, R., 2012. Structural and

697

functional loss in restored wetland ecosystems. PLoS Biol. 10, e1001247.

698

https://doi.org/10.1371/journal.pbio.1001247

Murphy, J.L., Boyer, K.E., Carpenter, E.J., 2017. Restoration of Cordgrass Salt

700

Marshes: Limited Effects of Organic Matter Additions on Nitrogen Fixation.

701

Wetlands 1–11. https://doi.org/10.1007/s13157-017-0973-6

702

TE D

699

Natter, M., Keevan, J., Wang, Y., Keimowitz, A.R., Okeke, B.C., Son, A., Lee, M.K., 2012. Level and degradation of Deepwater Horizon spilled oil in coastal marsh

704

sediments and pore-water. Environ. Sci. Technol. 46, 5744–5755.

705

https://doi.org/10.1021/es300058w

AC C

706

EP

703

Nielsen, L.P., 1992. Denitrification in sediment determined from nitrogen isotope pairing.

707

FEMS Microbiol. Ecol. 86, 357–362. https://doi.org/https://doi.org/10.1111/j.1574-

708

6968.1992.tb04828.x

709

Nixon, Z., Zengel, S., Baker, M., Steinhoff, M., Fricano, G., Rouhani, S., Michel, J.,

710

2016. Shoreline oiling from the Deepwater Horizon oil spill. Mar. Pollut. Bull. 107,

711

170–178. https://doi.org/10.1016/j.marpolbul.2016.04.003

712

Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B.,

713

Simpson, G.L., Solymos, P., Stevens, M.H.M., Wagner, H., 2015. vegan: 23

ACCEPTED MANUSCRIPT

714 715

Community Ecology Package. R package version 2. 3-1. Prince, R.C., Garrett, R.M., Bare, R.E., Grossman, M.J., Townsend, T., Suflita, J.M., Lee, K., Owens, E.H., Sergy, G.A., Braddock, J.F., Lindstrom, J.E., Lessard, R.R.,

717

2003. The roles of photooxidation and biodegradation in long-term weathering of

718

crude and heavy fuel oils. Spill Sci. Technol. Bull. 8, 145–156.

719

https://doi.org/10.1016/S1353-2561(03)00017-3

720

RI PT

716

Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Peplies, J.,

Glöckner, F.O., 2013. The SILVA ribosomal RNA gene database project: Improved

722

data processing and web-based tools. Nucleic Acids Res. 41, D590–D596.

723

https://doi.org/10.1093/nar/gks1219

SC

721

R Core Team, 2015. R: A language and environment for statistical computing.

725

Rivera-Monroy, V.H., Branoff, B., Meselhe, E., McCorquodale, A., Dortch, M., Steyer,

M AN U

724

726

G.D., Visser, J., Wang, H., 2013. Landscape-level estimation of nitrogen removal in

727

coastal Louisiana wetlands: Potential sinks under different restoration scenarios. J.

728

Coast. Res. 67, 75–87. https://doi.org/10.2112/SI

729

Rivera-Monroy, V.H., Lenaker, P., Twilley, R.R., Delaune, R.D., Lindau, C.W., Nuttle, W., Habib, E., Fulweiler, R.W., Castañeda-Moya, E., 2010. Denitrification in coastal

731

Louisiana: A spatial assessment and research needs. J. Sea Res. 63, 157–172.

732

https://doi.org/10.1016/j.seares.2009.12.004

TE D

730

RStudio Team, 2015. RStudio: Integrated Development for R.

734

Schnetger, B., Lehners, C., 2014. Determination of nitrate plus nitrite in small volume

735

marine water samples using vanadium(III)chloride as a reduction agent. Mar.

736

Chem. 160, 91–98. https://doi.org/10.1016/j.marchem.2014.01.010

AC C

737

EP

733

Scott, N.M., Hess, M., Bouskill, N.J., Mason, O.U., Jansson, J.K., Gilbert, J.A., 2014.

738

The microbial nitrogen cycling potential is impacted by polyaromatic hydrocarbon

739

pollution of marine sediments. Front. Microbiol. 5, 1–8.

740

https://doi.org/10.3389/fmicb.2014.00108

741

Shapiro, K., Khanna, S., Ustin, S.L., 2016. Vegetation Impact and Recovery from Oil-

742

Induced Stress on Three Ecologically Distinct Wetland Sites in the Gulf of Mexico.

743

J. Mar. Sci. Eng. 4, 1–19. https://doi.org/10.3390/jmse4020033

744

Silliman, B.R., Dixon, P.M., Wobus, C., He, Q., Daleo, P., Hughes, B.B., Rissing, M., 24

ACCEPTED MANUSCRIPT

745

Willis, J.M., Hester, M.W., 2016. Thresholds in marsh resilience to the Deepwater

746

Horizon oil spill. Sci. Rep. 6, 1–7. https://doi.org/10.1038/srep32520

747

Silliman, B.R., van de Koppel, J., McCoy, M.W., Diller, J., Kasozi, G.N., Earl, K., Adams, P.N., Zimmerman, A.R., 2012. Degradation and resilience in Louisiana salt

749

marshes after the BP— Deepwater Horizon oil spill. PNAS 109, 11234–

750

11239. https://doi.org/10.1073/pnas.1204922109/-

751

/DCSupplemental.www.pnas.org/cgi/doi/10.1073/pnas.1204922109

752

RI PT

748

Turner, R.E., McClenachan, G., Tweel, A.W., 2016. Islands in the oil: Quantifying salt marsh shoreline erosion after the Deepwater Horizon oiling. Mar. Pollut. Bull. 110,

754

316–323. https://doi.org/10.1016/j.marpolbul.2016.06.046

SC

753

Tyler, A.C., Zieman, J.C., 1999. Patterns of development in the creekbank region of a

756

barrier island Spartina alterniflora marsh. Mar. Ecol. Prog. Ser. 180, 161–177.

757

Valentine, D.L., Mezić, I., Maćešić, S., Črnjaric-Žic, N., Ivić, S., Hogan, P.J., Fonoberov,

758

V.A., Loire, S., 2012. Dynamic autoinoculation and the microbial ecology of a deep

759

water hydrocarbon irruption. Proc. Natl. Acad. Sci. 109, 20286–20291.

760

https://doi.org/10.1073/pnas.1108820109

M AN U

755

Valiela, I., Cole, M.L., 2002. Comparative evidence that salt marshes and mangroves

762

may protect seagrass meadows from land-derived nitrogen loads. Ecosystems 5,

763

92–102. https://doi.org/10.1007/s10021-001-0058-4

764

TE D

761

van Wijnen, H.J., Bakker, J.P., 1999. Nitrogen and phosphorous limitation in a coastal barrier salt marsh: the implications for vegetation succession. J. Ecol. 87, 265–272.

766

https://doi.org/https://doi.org/10.1046/j.1365-2745.1999.00349.x Welschmeyer, N.A., 1994. Fluorometric analysis of chlorophyll a in the presence of

AC C

767

EP

765

768

chlorophyll b and pheopigments. Limnol. Oceanogr. 39, 1985–1992.

769

https://doi.org/10.4319/lo.1994.39.8.1985

770 771 772 773 774 775

25

ACCEPTED MANUSCRIPT

776

Tables

777

Table 1. Sediment core samples subjected to forensic biomarker GCMS analyses and

779

their results when matched to MC252 source oil. 0-10 cm samples are from

780

homogenized flow-through cores. Samples that were matched to MC252 are italicized. Date

Site

Core

Depth (cm)

20 Sep. 2016

2

A

0-10

No residue detected

18 Oct. 2016

2

A

0-10

No residue detected

15 Nov. 2016

2

A

0-10

Heavily weathered residue, possible MC252 match

2

A B C

784 785 786 787 788 789 790

Heavily weathered residue

23 Feb. 2017

2

NA

0-5

Heavily weathered MC252 match

23 Feb. 2017

3

A

0-10

No residue detected

B

0-10

No residue detected

C

0-10

Little residue detected, non-match

TE D

783

Heavily weathered residue

EP

782

Heavily weathered residue

AC C

781

0-10

M AN U

23 Feb. 2017

SC

Result

RI PT

778

791 792 793 794 26

ACCEPTED MANUSCRIPT

795

Figure Captions

796

Figure. 1. Flow-through core flux rates when NO3- is not limiting. A) Denitrification

798

capacity, B) NH4+flux, C) NO2+3-flux, and D) PO43- flux. For nutrient fluxes, positive =

799

production, negative = consumption. Note the difference in y-axes. Bars are standard

800

deviation.

801

Figure 2. A. Principal coordinates analysis of 16S microbial orders by site. B. Relative

802

abundance (%) of all hydrocarbon degrading taxa by site and month.

803

SC

RI PT

797

Figure 3. Q-PCR marker abundances for A) 16S, B) nirS, C) norB, and D) nosZ. Note

805

the difference in y-axes. Bars are standard error.

M AN U

804

806 807

AC C

EP

TE D

808

27

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Salt marsh denitrification is impacted by oiling intensity six years after the Deepwater Horizon oil spill

B. Overton3, Patricia Sobecky1, Behzad Mortazavi1,2* 1

RI PT

Corianne Tatariw1,2, Nikaela Flournoy1, Alice A. Kleinhuizen1,2, Derek Tollette1,2, Edward

University of Alabama, Department of Biological Sciences, Tuscaloosa, Alabama

SC

35487, United States 2

Dauphin Island Sea Lab, 101 Bienville Blvd Dauphin Island, Dauphin Island, Alabama

M AN U

36528, United States 3

Department of Environmental Sciences, Louisiana State University, Baton Rouge, LA

70803, United States *corresponding author

TE D

Highlights •

Oiling intensity impacted salt marsh denitrification capacity 6 years post-spill

•

Neither 16S diversity nor denitrifier functional markers were affected by oiling intensity

function

EP

Oiling intensity plays a role in the long-term recovery of marsh biogeochemical

AC C

•