Gulf Oil Spill


Understanding the ecotoxicology of the Deepwater Horizon blowout


Please note the primary repository for our Gulf Oil Spill work is our NCEAS website.  Please check there for current information


The Deepwater Horizon deep sea drilling platform blew out on April 20, 2010, killing 11 people outright and ushering in the largest discrete point source pollution release in modern American history.  By early summer it was apparent that we were far into uncharted waters.  The magnitude of the disaster and our inability to deal promptly with the unfolding crises sent shockwaves throughout the Gulf Coast and our nation.  With the realization that those of our outside the Gulf Coast had valuable experience and expertise that might be of help in this event, I assembled a national team of ecologists, toxicologists, geochemists, and surfactant experts under the auspices of the National Center for Ecological Analyses and Synthesis to help produce scoping documents about the likely long-term effects this release of hydrocarbons might have on the ecological communities of the Gulf of Mexico.  Most of us have deep experience with the ecotoxicology of the Exxon Valdez, Cosco Busan, and a multitude of other spills over recent decades.  Our group is now creating a framework of potential ecotoxicological impacts upon Gulf populations and communities with the primary goal of providing this information to decision-making and funding entities in an expedited manner.

Our Challenge

On April 20, 2010, the ultra-deepwater ocean drilling rig Deepwater Horizon 76 km southeast of the Louisiana coastline experienced a catastrophic wellhead failure, explosion, and fire while drilling an exploratory well.  Eleven rig operators were killed outright, with the entire rig subsequently lost on April 22, coming to rest on the seafloor some 400 m from the wellhead.  Shortly after the rig sank, the U.S. Coast Guard formally announced the leaking of oil from the broken riser pipe on the seafloor, approximately 1,500 m below the surface of the Gulf of Mexico. Additional leaks from kinks in the now-broken pipe were observed over the next several days by a suite of remotely operated vehicles.

The extreme depth, lax safety protocols, and poor pre-spill contingency planning created a situation in which the oil and gaseous hydrocarbons ultimately flowed unchecked for nearly three months. The inability of industry and government managers to stop the spill rapidly motivated an unprecedented use of dispersants at the spill’s source and across the region.  Large-scale aerial application of dispersants from heavy lift aerial tankers began on May 1. Three small-scale in situ efficacy trials at the wellhead itself followed and paved the way for the large-scale and continuous use of direct injection of dispersant into the oil plume at depth proximate to the riser break beginning on May 14.

Ultimately 1.8 million gallons of dispersants (roughly one third of the global supply) were released into the northern Gulf of Mexico.  The surface extent of the spill exceeded 50,000 km2 and ranged from Texas Louisiana to the Florida Panhandle. Subsurface spill extent is equivocal, but reports include vast subsea plumes extending perhaps as far as 142 kilometers, apparently often highly emulsified or fractionated.

Much remains unknown regarding the amount, character, fate, and ultimate impacts of the oil-dispersant-methane brew.  Only a subset of these toxicants (i.e. drilling mud and asphaltenes) are likely to be highly localized.  Characterizing the ecotoxicological endpoints of this aggregation of toxicants is exceedingly difficult given the inadequate data collection/reporting, lack of transparency driven by the NRDA process, and logistical difficulties inherent with a disaster of this magnitude.  Non-human toxicological endpoints are often poorly characterized.  Extrapolating such impact to population and community levels is made even more complicated in regions such as the northern Gulf of Mexico with poor or absent long-term, sentinel monitoring programs.

Most efforts to date have focused on stopping the spill or on documenting acute potential toxicity (e.g. EPA’s air quality monitoring in the coastal zone, BP’s dispersant toxicity trials).  Initially only the National Science Foundation (via their RAPID program) funded any work that begins to look beyond immediate toxicity, with most of these funded projects being extensions or modifications of existing projects/capacities or descriptions of the abiotic characteristics of the spill.  All indicators suggest an initial lack of a broad-based overview of the potential ecotoxicological effects of the disaster.  The EPA even lacked threshold lab-based maximum toxicity “cut off” levels for the 14 dispersants at BP’s disposal.  Despite this, the EPA has released data on the toxicity of Corexit 9500A and 7 other oil dispersants in early summer, finding that short term acute mortality upon adult mysid shrimp (Americamysis bahia) and estuarine atherinid fish (Menidia beryllina) was roughly equivalent to reference crude oil toxicity (with the widely used Corexit 9500A’s LC50 occurring at 42 ppm for Americamysis and 130 ppm for Menidia).  Fortunately, Corexit 9500A and other dispersants apparently demonstrate limited cytotoxicity or endocrine disruption potential in adult mammalian cells, but similar data do not exist for larval, juvenile, or non-mammalian Gulf species.  Even more critical to ecosystem impacts, no data are available to examine the potential for synergistic or antagonistic effects of simultaneous Gulf crude and dispersant exposure.  In other coastal systems we have seen significant sub-lethal effects of oil, produced water, and dispersant upon coastal and pelagic species with direct implications for local and regional population and community dynamics.  As this blowout spanned the critical spring and summer reproductive seasons for birds, marine mammals, fish, and invertebrates, we see great potential for negative impacts to wide ranges of coastal and marine assemblages, particularly in trophic cascades.  The massive scale of this event necessitates considering spatially and taxonomically broad impacts.

Into this mix of uncertainty and uncharted waters, federal and private funders are preparing to fund large-scale impact assessment efforts that may well last for years or decades.  Our working group comprised of ecotoxicologists with experience in coastal oil spills, ecologists with experience in benthic and pelagic coastal ecology, and biologists and oceanographers currently collecting in situ spill data in the Gulf will attempt to begin to tackle these vexing questions.  Our group will synthesize existing knowledge of general oil and dispersant toxicology, knowledge from a range of spills from Ixtoc I to Exxon Valdez to smaller spills, and the ecology of the Gulf of Mexico to create a holistic framework of potential ecotoxicological endpoints to help guide subsequent monitoring and assessment efforts for the Gulf.  We will house our resulting framework and amassed data at NCEAS and, to the extent funding allows, provide access to them using advanced IT techniques previously developed for the Gulf in the wake of Hurricanes Katrina and Rita.