Home >> Research >> Publications >> Abstract

Integrated Assessment of Oil Spill




The overall goal is to provide research results that will both provide (a) immediate contributions and (b) build a foundation for future beneficial research within NGI themes, NOAA goals, and BP research priorities.
Specific primary objectives are to:

  1. Provide early predictions of a representative hurricane ensemble so that pre-planning can be accomplished;
  2. Predict the physical distribution, dispersion and dilution of contaminants under the action of currents and storms in coastal estuaries;
  3. Determine environmental effects of the oil/dispersant system on shallow water habitats, wetlands, and beach sediments, and the science of ecosystem recovery;
  4. Improve understanding of biological degradation of the oil/dispersant systems and subsequent interaction with the marine and coastal ecosystems;
  5. Provide a Coastal and Marine Spatial Planning toolkit for displaying results useful in recovery management;
  6. Produce a sound scientific base and plans for multi-institutional research that will fully address BP's priorities within NGI's mission.

The work will be accomplished in four interrelated tasks:

  • Task A. Hurricane Effects - will examine the impact of six tropical cyclone scenarios on the Deepwater Horizon oil spill for the Northern Gulf Coast by 15 August 2010. Recommendations for further research in Phase 2 will also be made based on these results for studies after the 2010 Gulf hurricane season.
  • Task B. Fate and Transport of Oil and Dispersants - will focus on indentifying and applying available calibrated hydrologic and hydraulic models of Gulf estuaries, demonstrating how they may be used to assess the long-term impacts of the oil spill (e.g. on hypoxia, sequestration in sediments, toxicity to algae, etc.), establishing and prioritizing remedial actions, and indentifying deficiencies in the literature impacting or introducing uncertainty into those predictions, such as kinetic rates impacting fate.
  • Task C. Natural Systems - will assess early responses of intertidal habitats to oil/dispersant contamination and interaction between oil/dispersant system and soil/sediment microbial assemblages.
  • Task D. Technology and Data Integration - will make the results of Tasks A, B, and C accessible and understandable and build a foundation for future research by providing a platform and mechanism for integration of results.


Task A: Hurricane Effects - Leader, Dr. Pat Fitzpatrick, Geosystems Research Institute
Collaborators: Dr. Haldun Karan; Yee Lau; and Chris Hill, Geosystems Research Institute

The Deepwater Horizon oil spill already poses a large environmental threat to the Gulf of Mexico region. However, the imminent impact of tropical cyclones on this situation is unclear. Although destructive to coastal property, hurricanes often dilute local pollutants through ocean mixing and storm surge flushing, and occasionally with heavy rainfall. However, the Valdez oil spill was actually expanded by an extratropical cyclone which contained tropical-storm force winds. This event demands an immediate investigation of this hurricane season's possibilities.

As already demonstrated by the relatively large error margins from NOAA's oil spill trajectory forecasts, oil spill modeling currently contains a number of uncertainties. These models must capture the following processes affecting oil spill distribution: advection, spreading, evaporation, dissolution, dispersion, mixing, and emulsification. Accurate input of winds and ocean currents is also required. Both tropical cyclones and oil spills occur in a data-starved environment, further complicating determination of these storms' impact.

Hierarchies of oil spill models have been developed, differing in computer platform capability and mathematical complexity. They include "budget", 2D, and 3D models. "Budget" models request basic input such as: initial oil concentration, a time series of winds, wave, water temperature, salinity, sediment load, ocean current, type of oil release (instantaneous or continuous) to make a scalar calculation in a region. One example is the NOAA ADIOS2 model. 2D models generally assume oil concentrations travel with the ocean currents plus an additional 3% contribution from surface winds, diffused with each time step. Some also include weather processes which modify the oil substance with time. Examples include the MMS Oil-Spill Risk Analysis (OSRA) model, DHI's MIKE 2D, NOAA's GNOME model, and other models discussed in the literature. 3D models use a concentration transport equation that requires input from an ocean models and surface weather data. Most 3D models include weather processes as well, although it can be instructive to see use the transport equation alone. One example is DHI's MIKE 3 PT model. Other examples include the models discussed in Guo et al. (2009) and Chao et al. (2001).

Tropical cyclones presents additional modeling challenges, involving the coupling of surface winds to an ocean model to accurately represent storm-driven currents, mixing, and upwelling. The winds may be from pre-determined 2D parametric wind fields based on gradient wind equations (such as the Holland wind profile), or from an atmospheric model which responds to ocean temperature (i.e., two-way coupling) such as the Hurricane WRF (HWRF) which is coupled to the Princeton Ocean Model (POM). Wave heights may be determined based on the Bretshneider empirical equations, or using a wave model such as Wavewatch III. The storm surge can be simulated using models such as SLOSH or ADCIRC.

Task A Objectives: Due to the complexity and variety of oil spill models, as well as account for model uncertainty, the best approach is to run an ensemble of oil spill models. Since results are needed quickly, a detailed scientific assessment will be postponed until Phase 2. Phase 1 will focus on providing results to six tropical cyclone scenarios using several modeling approaches. Recommendations for further research in Phase 2 will also be made based on these results for studies after the 2010 Gulf hurricane season. This research will be coordinated with LSU via Dr. Jim Chen, since they have expertise with GNOME and MIKE. We will also coordinate with FEMA and USACE to obtain their hurricane simulation results produced for flood mapping and levee design, respectively.

Primary task objectives are to achieve primary objectives 1 and 6 by:

  • Providing a forecast ensemble using budget, 2D, and 3D models for six tropical cyclone scenarios by 15 August 2010. The scenarios are subdivided into landfall west (central Louisiana) and east (Pensacola) of the oil spill, with intensities of tropical storm, Category 2, and Category 4.
  • Developing a research strategy for Phase 2 that performs detailed analyses of hindcast oil spill simulations for the 2010 hurricane season.

    Subtask A1: Model simulations

  • Run ADIOS2 for tropical storm scenarios. Its algorithms are not valid for hurricane-force winds
  • .Run GNOME in standard mode and diagnostic mode for all six scenarios using parametric wind hurricane calculations
  • Run MIKE 3 PT for all six scenarios using output for all six scenarios using input from HWRF model
  • Run multiple versions of a 2D oil spill trajectory forecast using input from HWRF model, based on the following equations. Denoting the zonal speed component as u, meridional speed component as v, the oil drift speed can be computed aswhere the subscripts d, o, a and are for drift, ocean, and air, respectively. These values will be determined from HWRF output. The parcel distance per timestep by diffusion is: where is the random number in the interval 0-1, is the horizontal diffusion. Multiple versions will be performed by varying surface . In one version, will be determined by the POM model output from HWRF, currently computed by the Smagorinsky formulation. In other versions, it will vary from 5-15 m2s-1 . Simulations without the random number component will also be performed.

    The displacement L of each parcel per timestep will be calculated by:

    Repeat the above methodology, except include a 3D trajectory component where vertical ocean velocities from HWRF are included and vertical diffusion from the KP formulation is used. Initial oil concentrations will assume an exponentially decreasing profile to 1.5 times the ocean wave height based on the results of Delvigne and Sweeney (1989).
  • Time permitting, we will also vary oil concentrations, assuming the spill is stopped by early August and that other oil containment efforts have also been successful.
  • Particle transport calculations will be performed using a concentration transport equation already available in HWRF.

    Subtask A2: Results Transfer and Display
  • Graphics of the simulations (Task D) will be presented online at the NGI website with summaries in August and provided to Tasks B-C. A consensus plot will also be displayed. Discussion of the sensitivity to diffusion will be included. The potential shoreline impact will be examined to east and west landfall scenarios and by intensity. Natural dispersion by wave and current action will also be explored.

    Subtask A3: Recommendations
  • By November, the impact of the 2010 hurricane season on the oil spill will be known. Recommendations will include details on: 1) a validation study of 2010 Gulf of Mexico tropical cyclones using the various models; 2) model deficiencies; and 3) budget calculations of the weathering processes in hurricane conditions. We may also recommend the development of a community model for oil spills, similar to the developmental approaches currently in use for several ocean, meteorology, and storm surge models

Task B: Fate and Transport of Oil and Dispersants - Leader, Dr. James Martin, Civil & Environmental Engineering

Collaborators: Drs. Vladimir Alarcon, Geosystems Research Institute, Jairo Diaz, Civil & Environmental Engineering

The oil spilled into the Gulf of Mexico is expected to have long-term residual impacts on human and environmental health, both in deep and near-shore sea waters as well as Gulf estuaries. In order to assess those impacts, quantitative cause and effect relationships are needed to relate the degree and extent of the contamination to physical, chemical, and biological processes to expected impacts. These relationships are most commonly expressed in predictive mathematical models.

The use of mathematical models is well established, such as in the regulatory environment to establish waste load allocations, estimate Total Maximum Daily Loads (TMDLs), to estimate impacts of remediation of contaminated sediments, and a variety of other purposes. (Martin and McCutcheon 1999, Lung 2001). Each of these are quantitative analyses where numeric targets or endpoints are established that equate to attainment of the water quality standard (e.g. physical, chemical or biological integrity per Clean Water Act). Predictive water quality models are typically used to develop linkages between sources and targets. The models provide a quantitative link between sources and targets, or cause-and effect relationship, in order to determine the capacity of the waterbody to assimilate contamination and to address the site-specific nature of the problem. Models of open waters and Gulf estuaries most commonly include both hydrodynamic and water quality models, due to the importance of transport on the fate of water quality constituents (Martin and McCutcheon 1999). The models may then be focused on the kinetic and transformation process impacting the specific issue of concern (organic contaminants, mercury, dissolved oxygen, nutrients, oil spills, etc.) in order to address specific concerns such as excess algal growth, hypoxia, and others.

There are a variety of models available that have been targeted to the Gulf of Mexico and/or near-shore (shelf) areas. These large scale models are the primary focus of research and development by a number of NGI institutions. Similarly, there have been a large number of models that have been developed for Gulf estuaries by and for different agencies and institutions. These models may have been developed for research purposes or to address a specific regulatory question, such as establishing TMDLs and nutrient criteria. It is these estuarine models that will be the focus of this research effort.

The research will focus on indentifying and applying available calibrated models of Gulf estuaries, demonstrating how they may be used to assess the long-term impacts of the oil spill (e.g. on hypoxia, sequestration in sediments, toxicity to algae, etc.), establishing and prioritizing remedial actions, and indentifying deficiencies in the literature impacting or introducing uncertainty into those predictions, such as kinetic rates impacting fate.

Since the models selected and applied impact their availability and use, this study will focus on two specific models:

  • EFDC: the Environmental Fluid Dynamics Code, a three-dimensional hydrodynamic and sediment transport model (Hamrick 1996, Tetra Tech 2002):
  • WASP: The Water Analysis Simulation Model, a generalized water quality model most commonly applied using the transport information for EFDC along with specific water quality algorithms (Ambrose et al. 1993, Wool et al. 2001, USEPA 2010),

WASP and EFDC are arguably the most commonly applied models to Gulf estuaries for regulatory purposes, and both models are maintained and distributed by the U.S. Environmental Protection Agency. The WASP model includes kinetic algorithms for coliform bacteria, simple and complex eutrophication (dissolved oxygen and nutrient cycling), toxic metals, toxic organics, and mercury. Although there is not a specific module for fate and transport of oils, the model can receive and incorporate loading information from other models or observations. Ongoing or recent EFDC/WASP applications include the Back Bay of Biloxi, MS (MDEQ 2002 and ongoing studies by MDEQ, pers. comm.); Bay St. Louis, MS (MDEQ 2001, Huddleston et al. 2007, and ongoing studies by MDEQ and MSU), Escatawpa and Pascagoula Rivers, MS (Rodriguez Borrelli et al. 2006); Mobile Bay, AL (Wool 2003, Wool et al. 2003, McAnally et al. 2007, Martin et al. 2008, Tetra Tech 2008, Diaz et al. 2008, Alarcon et al. 2009, Aziz et al. 2009, and ongoing studies); Weeks Bay, AL (ongoing studies by GOMA), Tampa Bay, FL(Wang et al. 1999 and ongoing studies by GOMA) and many others. The WASP model is presently planned for use in Florida to estimate site-specific nutrient criteria for all Florida estuaries (MDEQ pers. comm.). Other and inland applications include Lake Okeechobee, FL; eutrophication of the Neuse River Estuary, NC; eutrophication Coosa River and Reservoirs, AL; PCB pollution of the Great Lakes, eutrophication of the Potomac Estuary, kepone pollution of the James River Estuary, volatile organic pollution of the Delaware Estuary, and heavy metal pollution of the Deep River, North Carolina, mercury in the Savannah River, GA (USEPA 2010).

Task B objectives: The objectives of this task are listed below, with the specific focus being on the fate and transport of oils, gases, and dispersants in Gulf estuaries and/or wetlands, achieving primary objectives 2 and 6.

Specific Objectives:

  • Develop and apply predictive tools to estimate the physical distribution, dispersion and dilution of contaminants under the action of currents and storms in coastal estuaries.
  • Incorporate fundamental scientific research from other tasks into predictive tools to aid in estimating impacts and remedial strategies protective of human and ecological health.

Secondary Specific Task Objectives are to address the following research questions:

  • What is the magnitude of the oil, gases and dispersants loaded to Gulf Estuaries?
  • How is the oil and oil/dispersant being degraded in the Gulf estuaries (surface and subsurface)?
  • What physical factors determine how the oil is being transported on the surface and subsurface in Gulf estuaries?
  • What physical, biological, and hydrogeologic factors determine how oil infiltrates wetlands once it reaches the shore?
  • What is the physical and hydrogeologic impact of increased river outflow or diversion (e.g. Mississippi River diversion) on the transport of oil and oil/dispersant into nearshore environments (e.g. wetlands, estuaries, barrier islands)?
  • What will be the impact of a hurricane storm surge on oil immediately offshore, in wetlands, and on barrier islands transported into Gulf estuaries?

Subtask B1

Identifying and assembling, wherever possible, available EFDC/WASP model applications to Gulf Estuaries. Each application will be briefly documented, along with model versions, availability of input, and test applications performed to assure the model is complete. Each model will also be assessed to determine whether updating and/or modification is required for the purposes of this study.

Subtask B2

Evaluation of existing oil spill models for providing boundary conditions for the estuarine models and/or for comparison with predictions based on tracer constituents in EFDC and predictions using WASP toxic organic routines (which includes sorption, volatilization, photolysis, hydrolysis, and biodegradation).

Subtask B3

Identifying and implementing linkages between the estuarine model applications and existing models of the Gulf of Mexico and/or Mississippi sound. This task will be conducted in cooperation with other NGI partners.

Subtask B4

Example application to two selected estuaries of how the models will be used to evaluate physical distribution, dispersion and dilution of contaminants and well as ecological impact. Processes will include hydrodynamics and sediment transport. Water quality impacts will be demonstrated using best available estimates of process coefficients and rates. The two selected estuaries include Mobile Bay and/or Weeks Bay Alabama, with final selection to be made in consultation with Task C and NGI partner personnel.

Subtask B5

Based on the model demonstrations, and evaluation of available data, focus on quantifying key processes impacting the fate and impact of oils, gases and dispersants in Gulf estuaries. Examples will include rates and processes impacting diagenesis of oils sequestered in bottom sediments and potential impacts on hypoxia, rates of degradation and oxygen consumption in the in the water column, and other key processes and rates. Results will be visualized and loaded into the Sulis toolkit of Task D.

Task C. Natural Systems - Leader, Dr. Gary Ervin, Biology
Impacted coastal habitats include beaches, barrier islands, shallow water habitats (seagrass beds and other submersed vegetation), and coastal marshes and estuaries. Although the obvious negative effects of oil and dispersant contamination in coastal habitats has been broadcast worldwide, the more subtle effects of these chemicals cascade throughout coastal ecosystems, and unfortunately, little is known regarding those complex, ecosystem-level impacts (Peterson and Estes 2001). This lack of information has been attributed to the rarity and unpredictability of large-scale spills, such as the Deepwater Horizon (Peterson and Estes 2001). By comparison to the current 30,000 km2 of the Gulf covered by oil, it was estimated in 2004 that there are only approximately 14,000 km2 of coastal wetlands along the US Gulf of Mexico coast (Stedman and Dahl 2008). Thus, there presently is a clear and substantial, yet unknown and unpredictable risk to coastal ecosystems.

Task C addresses primary objectives 3, 4, and 6 and two of the research priorities set forth by BP, to improve understanding of:

  • Environmental effects of the oil/dispersant system on shallow water habitats, wetlands, and beach sediments, and the science of ecosystem recovery, and
  • Biological degradation of the oil/dispersant systems and subsequent interaction with the marine and coastal ecosystems.

We will address these research priorities by investigating two interrelated focus areas, or Subtasks:

  • Subtask C1 - Assessing early responses of intertidal habitats to oil/dispersant contamination
  • Subtask C2 - Assessing interaction between oil/dispersant system and soil/sediment microbial assemblages

Site selection for sampling will be made in consultation with Task A and B personnel and NGI partners. Findings will be provided to Task B personnel for use in model kinetics and Task D for visualization and incorporation into Sulis.

Subtask C1 - Assessing early responses of intertidal habitats to oil/dispersant contamination

PI/Coordinator: Gary Ervin, Collaborators: Deepak Mishra, Saurabh Prasad, Sal Nobrega

Overview - Coastal marshes and shallow water habitats (submersed vegetation, such as seagrasses) provide countless ecological and economic services. Coastal wetlands provide habitat for economically important wildlife species, such as coastal fisheries, and they provide other human services, including water filtration and erosion mitigation (Pennings and Bertness 2001). The Northern Gulf Coast also provides critical habitat for 100 or more species of migratory bird species who use coastal areas as refueling stopover points after making their northward migrations, and by another 100 or so species who use the coast as their wintering grounds (Moore et al. 1993).

The basis of productivity in coastal wetlands is photosynthetic production, via algae and higher plants, with the vast majority of production being contributed by the higher plants. Those plants typically occur in distinct zones, each usually dominated by one or a few species (Pennings and Bertness 2001). This aspect of coastal marshes, in particular, has made them attractive study systems for ecologists, but also makes them highly vulnerable to stresses such as those resulting from contamination by organic chemicals, wherein some species are highly susceptible to stress or death from petroleum contamination (Pezeshki et al. 2000). Variability in sensitivity exists among species, but also has been demonstrated within species, the latter attributed in part to local ecological adaptation of populations of some species (Pezeshki et al. 2000). Differences in sensitivity among species could result in large-scale shifts in species composition of coastal habitats receiving the highest influx of oil/dispersant, or in large-scale dieback of coastal marsh habitats.

Remotely sensed land cover data has been used very effectively in classifying wetland vegetation, with accuracy as high as 100% on some emergent plant species (Hirano et al. 2003). Such analyses are effective not only for emergent vegetation, but also for algae & cyanobacteria (Mishra et al. 2009), corals (Mishra et al. 2007), and mapping other types of benthic marine habitats (Mishra et al. 2006). It also has been demonstrated that in reliably classified vegetation, one can use remotely sensed reflectance data to accurately estimate vigor of those stands (Artigas & Yang 2005).

Assessment of marsh health is essential to the evaluation of the short-term coastal impact of the DwH spill and for the prioritization of future restoration actions. This study will allow the identification of 'hotspots' of early stages of marsh degradation because of the oil spill. Such areas can only be delineated by evaluating marsh biophysical characteristics including distribution of chlorophyll content (Chl), green leaf area index (GLAI) (a ratio of green foliage area vs. ground area), and green vegetation fraction (VF) (percent green canopy cover). We intend to combine field data with the Moderate Landsat 30-m datasets to retrieve the described biophysical characteristics in salt marshes before and after the spill. By examining the biophysical parameters using remotely sensed data, we can infer the overall health and productivity of coastal marshland. The question to be answered through study is: What is the impact of the oil and oil/dispersant on health and productivity of wetlands?

The proposed research develops scientific products that can be used to assess marsh physiological characteristics at large spatial scales, which will directly inform restoration and conservation decision-making. Furthermore, we expect this project will have the potential to integrate with hydrodynamic transport models as a means of correlating marsh and shallow water habitat stress with estimates of oil/dispersant loading rates along the Northern Gulf Coast. We also anticipate that results from this research will enable estimation of wildlife habitat loss and degradation for coastal fish, shellfish, birds, and other economically and culturally important species.

Approach - We plan to employ cutting edge hyperspectral and/or multispectral imagery products to visualize and detect stress induced in sensitive estuarine vegetation. We intend to combine field data with the Moderate Landsat 30-m datasets to retrieve the biophysical characteristics (Chl, GLAI, VF) in salt marshes before and after the DwH spill. By examining the biophysical parameters using remotely sensed data, we can infer the overall health and productivity of coastal marshland. By analyzing multiple images over time, we can evaluate changes in vegetation health and potentially correlate those changes with known spatiotemporal patterns in oil movements within coastal habitats. We also anticipate that these efforts will be able to interface with hydrodynamic transport models developed through the "Amount, Fate, and Transport of Oil and Dispersants" task area.

Publicly available AVIRIS imagery also will be employed for these Phase I studies. Similarly to analyses of Landsat data, we will attempt analysis of the AVIRIS data from multiple aspects, such as simple vegetation indices, broad spectral as well as fine-scale spectral characteristics, and time-series analyses. These studies will provide valuable insight into the affects of the oil spill on sensitive and delicate vegetation along the gulf coast. This study will enable us to identify or develop more advanced algorithms for a broader-scale Phase II study.

To facilitate the coordination among the research areas in the Natural Systems task area, we will develop a pilot version of an interactive tool to geographically support decision making based on the hierarchical multi-scale research undertaken by the NGI/Natural Systems team. This will be designed for collaborative involvement, scenario evaluation and comparative analysis. Given the objective of NGI Phase-I quick response, this project will serve as a pilot study and may be expanded to other NGI teams in the future. The core-solution will be added to Sulis (Task D) and will be capable to couple to existing applications, such as the NOAA's Environmental Response Management Application (ERMA) and integrate other factors such as socio-economic research results and data collection.

Subtask C2 - Assessing interaction between oil/dispersant system and soil/sediment microbial assemblages

Findings from studies of the aftermath of the Exxon Valdez spill have indicated that (Peterson et al. 2003):

  • Microenvironment influences rates of oil degradation, with chronic effects persisting for many years,
  • Long-term exposure, even at concentrations of parts per billion, can negatively impact wildlife populations, and
  • Effects of oil contamination persist over time, in part through trophic interactions among ecosystem compartments.

This subtask will investigate impacts on exposed soil microbial assemblages (e.g., algae, bacteria, fungi), as well as the potential for those assemblages to degrade the oil/dispersant system washing ashore in Northern Gulf coastal habitats, including marshes, shores, and barrier islands. Three specific areas of investigation will be employed:

  • Subtask C2A - Response of microbial assemblages to oil/dispersant contamination, including assimilation of petroleum-derived carbon into the food web
  • Subtask C2B - Impact on microbial assemblages, including potential enrichment of bioremediation species
  • Subtask C2C - Parameters for successful biological degradation in sand, sediment and water

Subtask C2A - Response of microbial assemblages to oil/dispersant contamination, including assimilation of petroleum-derived carbon into the food web. PI/Coordinator: Mark Williams

Overview - Microorganisms, especially bacteria, form the base of many terrestrial and aquatic food webs. As the primary decomposers of organic materials, it is likely that particular taxa of bacteria will respond positively to the influx of organic rich material derived from oil and dispersants associated with the DwH oil release of April 2010. If there is evidence that bacteria are utilizing the organic materials, this could be interpreted as a positive sign that petroleum breakdown is taking place. However, this also would provide a warning that petroleum, dispersants, and their degradation products are moving through the coastal ecosystem food webs.

Approach - Bacterial, fungal, and algal community biomass and their δ 13C content (a measure of the isotopic composition of carbon within biomass) will be determined, to assess the flow of petroleum carbon into this part of the coastal food web. Because petroleum contains a bulk δ 13C signature of approximately -40, and most autotrophic forms of life overwhelmingly contain a δ 13C value between -12 and -27, it will be possible to assess microbial utilization of petroleum derived from the DwH oil spill. The total abundance and isotope signatures of microbial biomass, phospholipid fatty acids (PLFA), and microbial produced CO2 will be assessed using both laboratory batch methods and by monitoring and sampling field sites at 2 to 3 intervals between July and Dec 2010. Understanding whether and how the microbial community is utilizing petroleum is an important step to assessing how petroleum may impact ecosystem communities and their associated nutrient dynamics.

Subtask C2B - Impact on microbial assemblages, including potential enrichment of bioremediation species. PI/Coordinator: Mark Lawrence

Overview - The marine bacterial community is a foundational part of the coastal food web and is critical for bioremediation. In this subtask, the effects of the oil/dispersant system on the microbial community will be assessed by examining bacterial species diversity and composition. The deliverable for this project would be a determination of impact on microbial diversity and quantities, including whether there is a "natural" enrichment of potential bioremediation species, as a result of the influx of oil and dispersant into coastal habitats. From this data we could draw conclusions regarding the impact on metabolic capabilities of the microbial community.

Approach - To assess bacterial community structure, DNA will be isolated from affected and unaffected soil and sediment samples, from sites used in Subtask C2A, above. PCR amplification of 16S rDNA genes will be conducted using eubacterial and archaebacterial primers. Amplicons will be cloned, and population diversity assessed by restriction mapping. Unique clones will be sequenced to provide more detailed information on microbial assemblage composition.

Subtask C2C - Parameters for successful biological degradation in sand, sediment and water, PI/Coordinator: Susan Diehl. Collaborators: Hamid Borazjani, Richard Baird, Thomas Cathcart, Pete Melby

Overview - Numerous studies on bioremediation of crude oil clearly show that the characteristics of every site and each oil source are unique, and before any large scale bioremediation effort begins, preliminary studies are needed to determine optimal parameters in each case. The best parameters for successful bioremediation of the oil from the DwH spill, need to be established by evaluating factors that influence biodegradation capabilities of sand, sediment, and aquatic microbial assemblages from the affected areas.

Approach - The impact of nitrogen and phosphorus addition, dispersant, and oil concentration on successful biodegradation of the oil in sand, sediment, and water samples will be determined. A vast majority of field bioremediation studies have shown that the major limiting factors in successful bioremediation are nutrient limitations (nitrogen and phosphorus), contaminant concentration, bioavailability and aeration (Atlas 1995). Addition of exogenous microbes is rarely if ever truly successful or beneficial. The native microbes are capable of degrading most contaminants. To determine what factors need to be amended in order for bioremediation to be successful in the DwH spill, a series of replicated microcosms will be set up to evaluate nitrogen and phosphorus addition, bioavailability and oil concentration. Water, sand and sediment containing crude oil in the concentrations encountered on the Gulf will be evaluated for biodegradation with and without nitrogen and phosphorus addition. An additional set of replicates will be amended with nitrogen and phosphorus and dispersant (to manipulate bioavailability of the oil). Some dispersants are actually toxic to bacteria and fungi, and instead of stimulating biodegradation by making the contaminant more bioavailable, they reduce biodegradation by negatively impacting the microbial community. Companion studies will first absorb the oil into a 100% natural absorbent, such as uncontaminated plant material from coastal marshes, and then measure biodegradation of the remaining oil with and without nitrogen addition plus/minus dispersant. The oil-soaked plant material will be composted and the oil break down measured in the compost. Thus, there would be no remaining wastes for landfills. This should help determine what the limiting factors are for successful field bioremediation in different affected habitats. Oil breakdown would be measured by total TPH or gravimetric oil and grease analysis.

Task D. Technology and Data Integration - Leader Dr. William H. McAnally, NGI
Collaborators: Phil Amburn, John Cartwright, Rita Jackson, Geosystems Research Institute

The objectives of Task D are to:

  • Make the results of Tasks A, B, and C accessible and understandable, satisfying primary objective 5 - Provide a Coastal and Marine Spatial Planning toolkit for displaying results useful in recovery management.
  • Build a foundation for future research by providing a platform and mechanism for integration of results.

It will be accomplished through 3 subtasks.

Subtask D1. Data and Metadata

Data collected and generated by Tasks A-C will be submitted to the National Coastal Data Development Center (NCDDC) databases along with metadata prepared using NOAA's MERMAid web software to be consistent with Federal Geographic Data Committee (FGDC) metadata standards.


  • Phase 1 data and metadata
  • Tech note on data and access

Subtask D2. Extend Sulis Toolkit

We will extend the Sulis toolkit (See Figure D1 below) to encompass products of Tasks A, B, and C. Sulis is a decision support system and toolkit with the purpose of providing users ready access to environmental and natural resources information in a useful form to better understand aquascapes (a complete hydrologic footprint, including a watershed - an area of the earth's surface from which water flows downhill to a single outflow point - plus the water-spread - the coastal and ocean area over which the watershed's flow spreads and ocean forcings affect coastal and upstream waters) and their processes, to evaluate the probable consequences of coastal hazards and management decisions, and to make informed decisions with a holistic perspective. Sulis employs a modular technology, in which individual models can be replaced by others as models evolve and user needs change. As such, it is not a model itself, but a collection of data sources - observations and models (presently hydrologic and hydraulic models) - from which results can be extracted, viewed, and analyzed (H3O Team 2010).

Tasks A-C include measured data from collection efforts and modeled data generated by numerical models. These data will be collected and cataloged in a single location for further review and analysis. Additionally, much of these data require 2D and 3D visualization for thorough analysis. Task D2 will provide 2D visualization through the use of GIS technology and 3D visualization (see examples Figure D2 below) through the use of scientific visualization technology, delivered through Sulis.


  • Sulis data repository with metadata
  • Scientific visualization displays of both collected and modeled data associated with the Deep Water Horizon event: Still images and animated sequences.

Subtask D3. Workshop on Cleanup and Remediation Technology

A workshop will be convened to examine innovative technologies for identification, cleanup, and remediation of oil spills. Experts will be invited to present their experimental technologies, evidence of applicability, and testing plans to develop and prove the technologies. For example, MSU researchers have been testing small satellites' ability to deliver rapid, complete coverage of the Gulf and absorbent pine shavings that sequester oil and toxins until microbial action can decompose them. Workshop proceedings will be summarized with recommendations for Phase 2 research experiments.

Deliverable: Workshop Report


  • Monthly status reports by teleconference
  • Task A: Hurricane Effects: August 2010
  • Task B: Simulation Results: October 2010
  • Task C: Data & Metadata: November 2010
  • Task D: Workshop Report; November
  • All Tasks: Sulis database & tools: December 2010
  • Final Project report: February 2011

Principal Investigators

William H. McAnally, Northern Gulf Institute, Mississippi State University

Pat Fitzpatrick, Geosystems Research Institute, Mississippi State University

Gary Ervin, Department of Biological Science, Mississippi State University

Collaborators and Partners

Northern Gulf Institute, Mississippi State University
Geosystems Research Institute, Mississippi State University
Department of Biological Science, Mississippi State University
Department of Civil & Environmental Engineering, Mississippi State University

For more information on this project, please contact the Program Office. A more detailed review of this project can be found in the NGI Documents.

For internal use only: Project No. 10-BP_GRI_MSU-01. Expiration Date: 12/31/2010.