Meta-analysis of salt marsh vegetation impacts and recovery: a synthesis following the Deepwater Horizon oil spill.

Marine oil spills continue to be a global issue, heightened by spill events such as the 2010 Deepwater Horizon spill in the Gulf of Mexico, the largest marine oil spill in US waters and among the largest worldwide, affecting over 1,000 km of sensitive wetland shorelines, primarily salt marshes supporting numerous ecosystem functions. To synthesize the effects of the oil spill on foundational vegetation species in the salt marsh ecosystem, Spartina alterniflora and Juncus roemerianus, we performed a meta-analysis using data from 10 studies and 255 sampling sites over seven years post-spill. We examined the hypotheses that the oil spill reduced plant cover, stem density, vegetation height, aboveground biomass, and belowground biomass, and tracked the degree of effects temporally to estimate recovery time frames. All plant metrics indicated impacts from oiling, with 20-100% maximum reductions depending on oiling level and marsh zone. Peak reductions of ~70-90% in total plant cover, total aboveground biomass, and belowground biomass were observed for heavily oiled sites at the marsh edge. Both Spartina and Juncus were impacted, with Juncus affected to a greater degree. Most plant metrics had recovery time frames of three years or longer, including multiple metrics with incomplete recovery over the duration of our data, at least seven years post-spill. Belowground biomass was particularly concerning, because it declined over time in contrast with recovery trends in most aboveground metrics, serving as a strong indicator of ongoing impact, limited recovery, and impaired resilience. We conclude that the Deepwater Horizon spill had multiyear impacts on salt marsh vegetation, with full recovery likely to exceed 10 years, particularly in heavily oiled marshes, where erosion may preclude full recovery. Vegetation impacts and delayed recovery is likely to have exerted substantial influences on ecosystem processes and associated species, especially along heavily oiled shorelines. Our synthesis affords a greater understanding of ecosystem impacts and recovery following the Deepwater Horizon oil spill, and informs environmental impact analysis, contingency planning, emergency response, damage assessment, and restoration efforts related to oil spills.

Recovery of the salt marsh periwinkle (Littoraria irrorata) 9 years after the Deepwater Horizon oil spill: Size matters.

Prior studies indicated salt marsh periwinkles (Littoraria irrorata) were strongly impacted in heavily oiled marshes for at least 5 years following the Deepwater Horizon oil spill. Here, we detail longer-term effects and recovery over nine years. Our analysis found that neither density nor population size structure recovered at heavily oiled sites where snails were smaller and variability in size structure and density was increased. Total aboveground live plant biomass and stem density remained lower over time in heavily oiled marshes, and we speculate that the resulting more open canopy stimulated benthic microalgal production contributing to high spring periwinkle densities or that the lower stem density reduced the ability of subadults and small adults to escape predation. Our data indicate that periwinkle population recovery may take one to two decades after the oil spill at moderately oiled and heavily oiled sites, respectively.

Wetland plant litter decomposition occurring during the freeze season under disparate flooded conditions.

To investigate the heterogeneity of plant litter decomposition in the freeze and freeze-free seasons and the responses to disparate flooded conditions in seasonally frozen wetlands, in situ simulation experiments of litter decomposition were performed in the Sanjiang Plain, Northeast China. The experiments were conducted using the litter bag method for representative plants, Carex lasiocarpa and Calamagrostis angustifolia, in both non-flooded and flooded areas between November 2011 and November 2013. Heterogeneous effects of the freeze season and its interaction with hydrological regimes on the decomposition of the litter of various species and organs were observed. The litter decomposition occurred during the freeze season and made a significant contribution to the loss throughout the year. The two-year mass-loss of C. lasiocarpa and C. angustifolia and their organs were ordered differently between the freeze season and the freeze-free season. The proportion of litter mass-loss during the freeze season accounting for the whole year in the flooded area were greater than that in the non-flooded area, except for the C. angustifolia root litter. The litter mass-losses of entire C. lasiocarpa and C. angustifolia during the freeze season were greater than those during the freeze-free season in the flooded area, while the pattern was opposite in the non-flooded area. The effect of environmental factors on litter decomposition might override the effects of litter substrate quality. The total N and P of the litter of the entire C. lasiocarpa and entire C. angustifolia increased significantly relative to the initial values after two years and tended to enrich more in the litter under flooded conditions than under non-flooded conditions. The results highlighted the heterogeneous effects of the freeze season and its interaction with hydrological regimes on various species and organs, which would provide management and restoration options for degraded wetlands caused by climate change.

Legacy effects of Hurricane Katrina influenced marsh shoreline erosion following the Deepwater Horizon oil spill.

Disturbance interactions occur when one perturbation influences the severity and perhaps the baseline state of succeeding disturbances. Natural and anthropogenic disturbances are frequent in dynamic coastal ecosystems and can often be linked. We evaluated potential for disturbance interactions associated with the 2010 Deepwater Horizon (DWH) oil spill, which was preceded by disturbance from Hurricane Katrina in 2005, by quantifying marsh shoreline retreat across both events. Our goal was to determine the degree to which Hurricane Katrina altered baseline rates of erosion prior to the DWH spill. We quantified erosion rate and fetch from aerial images of northern Barataria Bay, Louisiana marsh shorelines classified as reference, moderately-oiled, and heavily-oiled over three pre-spill time periods (1998-2004, prior to Hurricane Katrina; 2004-2005, during Katrina; 2005-2010, post-Katrina but pre-oil spill) and a post-spill period from 2010 to 2013. Prior to Hurricane Katrina, marsh shoreline erosion rates were low (from 0.38 to 1.10 m yr-1). In contrast during Hurricane Katrina (2004-2005), erosion increased by 661% and 756%, respectively, for shorelines that would subsequently become moderately and heavily-oiled; reference shoreline erosion increased by 59%. These high erosion rates were associated with increased fetch and higher wave action due to loss of protective geomorphic features such as small islands and spits and persisted during the post-Katrina/pre-spill period of 2005-2010 (0.62, 1.38, and 2.07 m yr-1 for reference, moderately, and heavily-oiled shorelines, respectively). Erosion rates increased modestly after the DWH event (reference = 1.13 m yr-1, moderate oiling = 1.45 m yr-1; heavy oiling = 2.77 m yr-1), but not significantly, compared to the post-Katrina period. Consequently, we could not detect a post-spill increase in marsh shoreline erosion. Rather, we concluded that Hurricane Katrina reset the erosion baseline, thereby connecting the two disturbances, and was the major driver of marsh shoreline erosion at our research sites during the study period.

Shoreline oiling effects and recovery of salt marsh macroinvertebrates from the Deepwater Horizon Oil Spill.

Salt marshes in northern Barataria Bay, Louisiana, USA were oiled, sometimes heavily, in the aftermath of the Deepwater Horizon oil spill. Previous studies indicate that fiddler crabs (in the genus Uca) and the salt marsh periwinkle (Littoraria irrorata) were negatively impacted in the short term by the spill. Here, we detail longer-term effects and recovery from moderate and heavy oiling over a 3-year span, beginning 30 months after the spill. Although neither fiddler crab burrow density nor diameter differed between oiled and reference sites when combined across all sampling events, these traits differed among some individual sampling periods consistent with a pattern of lingering oiling impacts. Periwinkle density, however, increased in all oiling categories and shell-length groups during our sampling period, and periwinkle densities were consistently highest at moderately oiled sites where Spartina alterniflora aboveground biomass was highest. Periwinkle shell length linearly increased from a mean of 16.5 to 19.2 mm over the study period at reference sites. In contrast, shell lengths at moderately oiled and heavily oiled sites increased through month 48 after the spill, but then decreased. This decrease was associated with a decline in the relative abundance of large adults (shell length 21-26 mm) at oiled sites which was likely caused by chronic hydrocarbon toxicity or oil-induced effects on habitat quality or food resources. Overall, the recovery of S. alterniflora facilitated the recovery of fiddler crabs and periwinkles. However, our long-term record not only indicates that variation in periwinkle mean shell length and length-frequency distributions are sensitive indicators of the health and recovery of the marsh, but agrees with synoptic studies of vegetation and infaunal communities that full recovery of heavily oiled sites will take longer than 66 months.

4 Years after the Deepwater Horizon Spill: Molecular Transformation of Macondo Well Oil in Louisiana Salt Marsh Sediments Revealed by FT-ICR Mass Spectrometry.

Gulf of Mexico saltmarsh sediments were heavily impacted by Macondo well oil (MWO) released from the 2010 Deepwater Horizon (DWH) oil spill. Detailed molecular-level characterization of sediment extracts collected over 48 months post-spill highlights the chemical complexity of highly polar, oxygen-containing compounds that remain environmentally persistent. Electrospray ionization (ESI) Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS), combined with chromatographic prefractionation, correlates bulk chemical properties to elemental compositions of oil-transformation products as a function of time. Carboxylic acid incorporation into parent MWO hydrocarbons detected in sediment extracts (corrected for mass loss relative to C30 hopane) proceeds with an increase of ∼3-fold in O2 species after 9 months to a maximum of a ∼5.5-fold increase after 36 months, compared to the parent MWO. More importantly, higher-order oxygenated compounds (O4-O6) not detected in the parent MWO increase in relative abundance with time as lower-order oxygenated species are transformed into highly polar, oxygen-containing compounds (Ox, where x > 3). Here, we present the first molecular-level characterization of temporal compositional changes that occur in Deepwater Horizon derived oil contamination deposited in a saltmarsh ecosystem from 9 to 48 months post-spill and identify highly oxidized Macondo well oil compounds that are not detectable by routine gas-chromatography-based techniques.

Response of salt marshes to oiling from the Deepwater Horizon spill: Implications for plant growth, soil surface-erosion, and shoreline stability.

We investigated the initial impacts and post spill recovery of salt marshes over a 3.5-year period along northern Barataria Bay, LA, USA exposed to varying degrees of Deepwater Horizon oiling to determine the effects on shoreline-stabilizing vegetation and soil processes. In moderately oiled marshes, surface soil total petroleum hydrocarbon concentrations were ~70mgg(-1) nine months after the spill. Though initial impacts of moderate oiling were evident, Spartina alterniflora and Juncus roemerianus aboveground biomass and total live belowground biomass were equivalent to reference marshes within 24-30months post spill. In contrast, heavily oiled marsh plants did not fully recover from oiling with surface soil total petroleum hydrocarbon concentrations that exceeded 500mgg(-1) nine months after oiling. Initially, heavy oiling resulted in near complete plant mortality, and subsequent recovery of live aboveground biomass was only 50% of reference marshes 42months after the spill. Heavy oiling also changed the vegetation structure of shoreline marshes from a mixed Spartina-Juncus community to predominantly Spartina; live Spartina aboveground biomass recovered within 2-3years, however, Juncus showed no recovery. In addition, live belowground biomass (0-12cm) in heavily oiled marshes was reduced by 76% three and a half years after the spill. Detrimental effects of heavy oiling on marsh plants also corresponded with significantly lower soil shear strength, lower sedimentation rates, and higher vertical soil-surface erosion rates, thus potentially affecting shoreline salt marsh stability.

Temporal and Spatial Changes in Black Carbon Sedimentary Processes in Wetlands of Songnen Plain, Northeast of China.

Black carbon (BC), an important component of organic carbon (OC) produced from incomplete combustion of carbon compounds, is widespread and affects the global carbon storage. The objectives of this study were to analyze the BC contents and fluxes in the last 150 years to determine the causes of differences in the three profiles of the Songnen Plain of Northeast China and to estimate the BC storage in the wetlands of the Songnen Plain. In the three sampling sites, BC fluxes in the period between 1950 and the present time increased by the ratios of 1.3, 31.1 and 1.4, respectively, compared to their own baseline between 1850 and 1900. Furthermore, the BC fluxes varying from 0.76 to 5.63 g m-2 y-1 in the three profiles had an opposite trend with the sand percentages with mean values changing from 78.9% to 19.6%, suggesting that sand desertification might additionally affect the BC processes in the region.

Effect of fire on phosphorus forms in Sphagnum moss and peat soils of ombrotrophic bogs.

The effect of burning Sphagnum moss and peat on phosphorus forms was studied with controlled combustion in the laboratory. Two fire treatments, a light fire (250 °C) and a severe fire (600 °C), were performed in a muffle furnace with 1-h residence time to simulate the effects of different forest fire conditions. The results showed that fire burning Sphagnum moss and peat soils resulted in losses of organic phosphorus (Po), while inorganic phosphorus (Pi) concentrations increased. Burning significantly changed detailed phosphorus composition and availability, with severe fires destroying over 90% of organic phosphorus and increasing the availability of inorganic P by more than twofold. Our study suggest that, while decomposition processes in ombrotrophic bogs occur very slowly, rapid changes in the form and availability of phosphorus in vegetation and litter may occur as the result of forest fires on peat soils.

Historical trends of atmospheric black carbon on Sanjiang Plain as reconstructed from a 150-year peat record.

Black carbon (BC), one of the major components of atmosphere aerosol, could be the second dominant driver of climate change. We reconstructed historical trend of BC fluxes in Sanjiang Plain (Northeast China) through peat record to better understand its long-term trend and relationship of this atmosphere aerosol with intensity of human activities. The BC fluxes in peatland were higher than other sedimentary archives. Although global biomass burning decreased in last 150 years, regional large scale reclaiming caused BC fluxes of the Sanjiang Plain increased dramatically between 1950s' and 1980s', most likely resulting from using fire to clearing dense pastures and forests for reclaiming. The BC fluxes have increased since 1900s with increasing of the population and the area of farmland; the increase trend has been more clearly since 1980s. Based on Generalized additive models (GAM), the proportional influence of regional anthropogenic impacts have increased and became dominant factors on BC deposition.

Impacts of Macondo oil from Deepwater Horizon spill on the growth response of the common reed Phragmites australis: a mesocosm study.

We investigated impacts of Macondo MC252 oil from the Deepwater Horizon (DWH) spill on the common reed Phragmites australis (Cav.) Trin. ex Steud., a dominant species of the Mississippi River Delta. In greenhouse experiments, we simulated the most common DWH oiling scenarios by applying weathered and emulsified Macondo oil to aboveground shoots at varying degrees of coverage (0-100%) or directly to marsh soil at different dosages (0-16 Lm(-)(2)). P. australis exhibited strong resistance to negative impacts when oil was applied to shoots alone, while reductions in above- and belowground plant growth were apparent when oil was applied to the soil or with repeated shoot-oiling. Although soil-oiling compromised plant function, mortality of P. australis did not occur. Our results demonstrate that P. australis has a high tolerance to weathered and emulsified Macondo oil, and that mode of exposure (aboveground versus belowground) was a primary determinant of impact severity.

Impacts and recovery of the Deepwater Horizon oil spill on vegetation structure and function of coastal salt marshes in the northern Gulf of Mexico.

We investigated the impacts of the Deepwater Horizon (DWH) oil spill on two dominant coastal saltmarsh plants, Spartina alterniflora and Juncus roemerianus, in the northern Gulf of Mexico and the processes controlling differential species-effects and recovery. Seven months after the Macondo MC 252 oil made landfall along the shoreline salt marshes of northern Barataria Bay, Louisiana, concentrations of total petroleum hydrocarbons in the surface 2 cm of heavily oiled marsh soils were as high as 510 mg g(-1). Heavy oiling caused almost complete mortality of both species. However, moderate oiling impacted Spartina less severely than Juncus and, relative to the reference marshes, had no significant effect on Spartina while significantly lowering live aboveground biomass and stem density of Juncus. A greenhouse mesocosm study supported field results and indicated that S. alterniflora was much more tolerant to shoot oil coverage than J. roemerianus. Spartina recovered from as much as 100% oil coverage of shoots in 7 months; however, Juncus recovered to a much lesser extent. Soil-oiling significantly affected both species. Severe impacts of the Macondo oil to coastal marsh vegetation most likely resulted from oil exposure of the shoots and oil contact on/in the marsh soil, as well as repeated oiling events.

Vegetative ecological characteristics of restored reed (Phragmites australis) wetlands in the Yellow River Delta, China.

In this study, we compared ecological characteristics of wetland vegetation in a series of restoration projects that were carried out in the wetlands of Yellow River Delta. The investigated characteristics include plant composition structure, species diversity and community similarity in three kinds of Phragmites australis wetlands, i.e. restored P. australis wetlands (R1, R2, R3 and R4: restored in 2002, 2005, 2007 and 2009, respectively), natural P. australis wetland (N) and degraded P. australis wetland (D) to assess the process of wetlands restoration. The coverage of the R1 was 99%, which was similar to natural wetland. Among all studied wetlands, the highest and lowest stem density was observed in R1 and R2, respectively, Plant height and stem diameter show the same trend as N > R2 > R1 > R3 > D > R4. Species diversity of restored P. australis wetlands became closed to natural wetland. Both species richness and Shannon-Wiener index had similar tendency: increased first and then decreased with restored time. The highest species richness and species diversity were observed in R2, while the lowest values of those parameters were found in natural P. australis wetland. Similarity indexes between restored wetlands and natural wetland increased with the restoration time, but they were still less than 50%. The results indicate that the vegetation of P. australis wetlands has experienced a great improvement after several years' restoration, and it is feasible to restored degraded P. australis wetlands by pouring fresh water into those wetlands in the Yellow River Delta. However, it is notable that costal degraded P. australis wetland in this region may take years to decades to reach the status of natural wetland.

In situ burning restores the ecological function and structure of an oil-impacted coastal marsh.

As the use of in situ burning for oil spill remediation in coastal wetlands accelerates, the capacity of this procedure to restore the ecological structure and function of oil-impacted wetlands becomes increasingly important. Thus, our research focused on evaluating the functional and structural recovery of a coastal marsh in South Louisiana to an in situ burn following a Hurricane Katrina-induced oil spill. Permanent sampling plots were set up to monitor marsh recovery in the oiled and burned areas as well as non-oiled and non-burned (reference) marshes. Plots were monitored for species composition, stem density, above- and belowground productivity, marsh resiliency, soil chemistry, soil residual oil, and organic matter decomposition. The burn removed the majority of the oil from the marsh, and structurally the marsh recovered rapidly. Plant biomass and species composition returned to control levels within 9 months; however, species richness remained somewhat lower in the oiled and burned areas compared to the reference areas. Recovery of ecological function was also rapid following the in situ burn. Aboveground and belowground plant productivity recovered within one growing season, and although decomposition rates were initially higher in the oiled areas, over time they became equivalent to those in reference sites. Also, marsh resiliency, i.e., the rate of recovery from our applied disturbances, was not affected by the in situ burn. We conclude that in situ burning is an effective way to remove oil and allow ecosystem recovery in coastal marshes.

In situ burning of oil in coastal marshes. 1. Vegetation recovery and soil temperature as a function of water depth, oil type, and marsh type.

In-situ burning of oiled wetlands potentially provides a cleanup technique that is generally consistent with present wetland management procedures. The effects of water depth (+10, +2, and -2 cm), oil type (crude and diesel), and oil penetration of sediment before the burn on the relationship between vegetation recovery and soil temperature for three coastal marsh types were investigated. The water depth over the soil surface during in-situ burning was a key factor controlling marsh plant recovery. Both the 10- and 2-cm water depths were sufficient to protect marsh vegetation from burning impacts, with surface soil temperatures of <35 and 48 degrees C, respectively. Plant survival rate and growth responses at these water depth burns were not significantly different from the unburned control. In contrast, a water table 2 cm below the soil surface during the burn resulted in high soil temperatures, with 90-200 degrees C at 0-0.5 cm soil depth and 55-75 degrees C at 1-2 cm soil depth. The 2-cm soil exposure to fire significantly impeded the post-burn recovery of Spartina alterniflora and Sagittaria lancifolia but did not detrimentally affect the recovery of Spartina patens and Distichlis spicata. Oil type (crude vs diesel) and oil applied to the marsh soil surface (0.5 L x m(-2)) before the burn did not significantly affect plant recovery. Thus, recovery is species-specific when no surface water exists. Even water at the soil surface will most likely protect wetland plants from burning impact.

In-situ burning of oil in coastal marshes. 2. Oil spill cleanup efficiency as a function of oil type, marsh type, and water depth.

In-situ burning of spilled oil, which receives considerable attention in marine conditions, could be an effective way to cleanup wetland oil spills. An experimental in-situ burn was conducted to study the effects of oil type, marsh type, and water depth on oil chemistry and oil removal efficiency from the water surface and sediment. In-situ burning decreased the totaltargeted alkanes and total targeted polycyclic aromatic hydrocarbons (PAHs) in the burn residues as compared to the pre-burn diesel and crude oils. Removal was even more effective for short-chain alkanes and low ring-number PAHs. Removal efficiencies for alkanes and PAHs were >98% in terms of mass balance although concentrations of some long-chain alkanes and high ring-number PAHs increased in the burn residue as compared to the pre-burn oils. Thus, in-situ burning potentially prevents floating oil from drifting into and contaminating adjacent habitats and penetrating the sediment. In addition, in-situ burning significantly removed diesel oil that had penetrated the sediment for all water depths. Furthermore, in-situ burning at a water depth 2 cm below the soil surface significantly removed crude oil that had penetrated the sediment. As a result, in-situ burning may reduce the long-term impacts of oil on benthic organisms.

The dose-response relationship between No. 2 fuel oil and the growth of the salt marsh grass, Spartina alterniflora.

The effect of No. 2 fuel oil on the biomass production of the salt marsh plant, Spartina alterniflora, was studied in a greenhouse dose-response experiment. S. alterniflora were transplanted into soil with 10 dosage levels of No. 2 fuel oil ranging from 0 to 456 mg g(-1) dry soil. Three months after transplantation, values for plant biomass, stem density, and shoot height decreased significantly with increasing fuel oil level in a dose-response fashion. Evapo-transpiration rates were correlated with the total biomass response. Relative to the control, a significant decrease in total (above- plus below-ground) plant biomass was observed at concentrations above 57 mg g(-1) dry soil. Within the 3-month experimental period, detrimental effects on below-ground biomass accumulation and bioluminescence of the marine bacterium Viberio fisheri in the Microtox Solid Phase Test were observed at oil concentrations >29 mg g(-1) dry soil, suggesting that biological effects of oil within the sediment matrix may be more pronounced than on above-ground biomass, requiring a dosage 228 mg g(-1) dry soil to elicit a significant detrimental effect. Hence, measurements of oil effects with biological end-points based solely on above-ground responses may underestimate the potential impacts of petroleum hydrocarbon spills, especially when the oil has penetrated the soil. While S. alterniflora was proved to be relatively tolerant to the No. 2 fuel oil spills, its effectiveness in phytoremediation operations may be limited at fuel oil levels 228 mg g(-1) dry soil, as both plant growth and microbial activity may be constrained.

Salt marsh recovery and oil spill remediation after in-situ burning: effects of water depth and burn duration.

Effects of water depth, burn duration, and diesel fuel concentration on the relationship between recovery of marsh vegetation, soil temperature, and oil remediation during in-situ burning of oiled mesocosms were investigated. The water depth over the soil surface during in-situ burning was a major factor controlling recovery of the salt marsh grass, Spartina alterniflora. Ten centimeters of water overlying the soil surface was sufficient to protect the marsh soil from burn impacts with soil temperatures <37 degrees C and high plant survival rate. In contrast, a water table 10 cm below the soil surface resulted in mean soil temperatures > 100 degrees C at the 2-cm soil depth, which completely inhibited the post-burn recovery of S. alterniflora. Although poor plant recovery was also apparent in the treatments with 0 and 2 cm of water over the soil surface, this result was likely due to the chemical stress of the diesel fuel used to create the fire rather than the heat, per se, which never reached the estimated lethal temperature of 60 degrees C. In-situ burning effectively removed more than 95% of floating oil from the water surface. Thus, in-situ burning prevented the oil from potentially contaminating adjacent habitats. However, in-situ burning did not effectively remediate the oil that had penetrated the soil.