Project Overview

As of June 30, 2020, C-IMAGE-III research team members have participated in more than 50 outreach related activities including school presentations, public events, invited talks, blogs, podcasts and other social media engagement, news articles and related activities. A significant part of our outreach plan was to promote the two legacy synthesis products, “Deep Oil Spills” and “Scenarios and Responses to Future Deep Oil Spills-Fighting the Next War”, a two volume book series that synthesizes the last ten years of research (C-IMAGE I, II and III) and including co-authors from other GoMRI- funded centers, the federal government and private industry. Here are a few of our key outreach products and activities:

1. Three Tampa Bay artists were commissioned to create ten pieces that served as section dividers for books, inspired by different themes in the series. These paintings were used over the course of C-IMAGE III to promote research findings through the development an art exhibit on display at the Nelson Poynter Memorial Library at USFSP in St. Petersburg, FL. The exhibit Remember the Horizon: How USF research set a standard after the 2010 Gulf oil spill combined these original pieces with deep-sea photographs taken during DEEPEND consortium cruises. The exhibit connected science and art, representing major findings through artwork from local artists and detailed photographs of mysterious deep-sea species. The artwork remains hanging in the Marine Research laboratory (MSL) at the USF College of Marine Science, in St. Petersburg, FL.The artists were also invited to the 2020 GoMOSES Conference where they sold their artwork during the poster session.

1. In reaching new audiences for GoMRI research, C-IMAGE III collaborated with the organizers of the AGU Fall Meeting to host a Story Collider event in San Francisco. Story Collider is an organization dedicated to telling and sharing true, personal stories about science on stage across the country and around the world, and in a weekly podcast. The theme of the AGU show was disaster response, and it showcased five different storytellers at the Tabletop Tap House, across the street from the Moscone Center. Laura Guertin, professor of Earth science at Penn State-Brandywine, kicked things off with a story about how she came to use quilting as a way of depicting resilience and hope in the Gulf of Mexico. Samantha (Mandy) Joye, a microbiologist, deep ocean explorer, and educator at the University of Georgia, treated the audience to a journey through important moments and turning points in her career as a researcher. Simeon Pesch, a process engineer at the Hamburg University of Technology, spoke about a life-altering and perspective-changing trip that he took to Ghana where he visited the Agbogbloshie waste dumping site, one of the top 10 most polluted places on the planet. After breaking for intermission, Paula Buchanan, a doctoral student in emergency management, talked about what it was like to experience Hurricane Katrina from a distance after having made New Orleans her second home. The evening concluded with a poignant story by Jessica Moerman, a paleoclimatologist and AAAS Science and Technology Policy Fellow, about the legacy of her grandfather’s Tennessee cabin in her family and her life.

1. Outreach efforts were direct at highlighting science diplomacy and bringing C-IMAGE research to the international audience. Our podcasts summarized our work with researchers from the University of Havana for the first joint U.S.-Cuban expedition in

over 50 years. C-IMAGE researchers also participated in the MarCuba meeting in 2018 and the RAUGM Conference in Mexico in 2019 where we unveiled highlight videos and Dispatches from the Gulf, respectively.

1. The final outreach piece we highlight was created in recognition of the Deepwater Horizon 10th anniversary. We developed a booklet that features highlights of studies funded by the Gulf of Mexico Research Initiative (GoMRI) over the past 10 years, specifically calling out key C-IMAGE findings. The booklet gives an overview of findings related to oil spill impacts on fishes, the processes involved with oil sinking to the seafloor, and sub-sea dispersant use. It’s connected to a corresponding webpage that provides additional resources, such as an interactive website exploring several Gulf of Mexico oil spills (Beneath the Horizon), audio-visual media, and graduate students profiles. [https://www.marine.usf.edu/c-image/wp-content/uploads/2020/04/C-IMAGE- BOOKLET.pdf]

Research Highlights

As of June 30, 2020, C-IMAGE III-sponsored research resulted in 42 peer-reviewed publications, 2 books (comprised of 65 chapter titles), 229 scientific presentations and 93 datasets being submitted to the GoMRI Information and Data Cooperative (GRIIDC), which are/will be made available to the public. A bulk of the research used to inform the bulleted questions at the beginning of this report were summarized in our two-volume series that was drafted in the first year of the project. Subsequent years focused on GoMRI’s Synthesis and Legacy workshops and outcomes, and continuing to develop Gulf-wide baseline products. The project also engaged 9 PhD and 5 Masters students over its award period. Significant outcomes of this project’s research are highlighted below. Overall, C-IMAGE has generated in excess of 270 scientific publications.

Theme 2 Highlights:

Theme 2 research (tasks 1, 2, and 3) focused on completing high pressure experiments both without and with high pressure dispersant application and integrating these results in modeling efforts. The role of pressure in biodegradation has been an ongoing subject of intense investigation at the Technical University of Hamburg and at Georgia Tech. Using the experimental set up we refer to as “Deep Sea in a Can”, researchers found that both elevated pressures and the presence of dispersant had a significant impact and the rate and amount of biodegradation. The strain Rhodococcus (PC20) was shown to degrade up to 30% more oil at elevated pressures within 96 hours, especially for the monoaromatic components. When dispersants were applied at ambient pressure at a DOR of 1:100 and higher, the degradation was inhibited. However, degradation was elevated at increased pressure. Similar findings were summarized from a comparable study using a larger collection of GoM sediment microbial communities (Colwellia, Psychrobrium, Thalassoalea, and Moritella). The purpose of this more comprehensive study was not only to look at the role of microbial communities in biodegradation, but also to investigate the factors responsible for the development of these deep microbial communities during and after a deep blowout. In all treatments except for Moritella, elevated pressure increased the relative abundance of the microbes, but the impact of

dispersant was less uniform, causing increased abundances in Moritella and Thalassotalea and decreased abundances in Cycloclasticus. These results highlight the significant influence of pressure on the development of microbial communities in the presence of oil and dispersant during oil spills and related response strategies in the deep sea. The research at Hamburg also identified key phenomena including the very large pressure drop at the blowout preventer as a likely cause of explosive de-gassing of gas saturated oil droplets as simulated in laboratory experiments. This de-gassing process likely contributed to the production of deep plumes of small oil droplets even in the absence of dispersant use.

C-IMAGE researchers contributed to the National Acedemies’ study “The Use of Dispersants in Marine Oil Spill Response” by providing model results for comparisons to in situ data during the blowout. Additionally, the complicated issues associated with high pressure laboratory experiments and the integration of these results into the modeling efforts are summarized in Murawski et al. (2019), and provides some strategies on how to parse out how much droplet size is impacted by subsea dispersant application or the delicate interaction between the nozzle diameter, the pressure drop, and the presence of methane gas in the oil. The authors recommend the development of a large scale high pressure facility, an experiment at the “field scale”, similar to DeepSpill, only deeper, and making critical observations available to researchers shortly after a future “spill of opportunity”. This is a follow up publication of some of the synthesis studies included in Deep Oil Spills. In the absence of a controlled field scale experiment, C-IMAGE researchers used the comprehensive BP Gulf Science Data to quantify petroleum dynamics throughout the 87-day long blowout. Evaluating the PAH concentrations at the sea surface and at the 1000m intrusion, the data reveal higher concentrations in these two areas within a 10-km radius of the blowout source, revealing there was no significant effect of SSDI volume on PAH vertical distribution and concentration. The turbulent energy associated with the spewing of gas-saturated oil at the deep-sea blowout may have minimized the effectiveness of the SSDI response approach. Given the potential for toxic chemical dispersants to cause environmental damage by increasing oil bioavailability and toxicity while suppressing its biodegradation, unrestricted SSDI application in response to deep-sea blowout is highly questionable. More efforts are required to inform response plans in future oil spills.

Adding to the complexity and consequences of oil and dispersant distribution, our researchers discovered that the footprint of the DWH spill was larger than the known satellite footprint, and that this oil that was invisible to satellites extended to regions like the west Florida shelf, the Texas shore, the Florida keys and along the Gulf Stream. This invisible oil contains some of the more dangerous compounds of oil, resistant to short term degradation, and can cause sublethal impacts on a wide range of wildlife.

Theme 3 Highlights

A cornerstone of the C-IMAGE III project was to synthesize the data from the comprehensive Gulf-wide surveys conducted annually since 2011, Mud & Blood and OneGulf expeditions. An overall health assessment of different fish species within the

Gulf of Mexico provided a baseline of PAH contamination in Gulf fishes, something that was not available pre-Deepwater Horizon. These baseline products will undoubtedly aid when monitoring recovery and potential long-term impacts from oil spills. A consequential publication “A First Comprehensive Baseline of Hydrocarbon Pollution in Gulf of Mexico Fishes” that was just released in early 2020 reports that highest concentrations of PAHs were found in the northern Gulf of Mexico, in the vicinity of the Deepwater Horizon and in additional “hot spots” off major population centers, suggesting that runoff from urbanized coasts may play a role in the higher concentrations of PAHs.

Other sources include chronic low-level releases from oil and gas platforms, fuel from boats and airplanes and oil seeps. These findings complement other studies of fish health coming out of C-IMAGE that are species-specific (Golden Tilefish, Hake, etc.) and extend the study to look at some pelagic species (Yellowfin Tuna). Overall, concentrations of biliary PAHs decreased in the years following the Deepwater Horizon oil spill, but the time series saw an uptick of concentrations in 2017. Biliary PAHs are an ideal indicator of recent exposures to oil and the more recent increase may be capturing new releases from leaking infrastructure, annual spills or resuspension events from an earlier depositional event.

Assessing population-level impacts due to the DWH have been difficult to quantify due to the lack of baseline studies, as previously noted. C-IMAGE paired a pre-spill opportunistic mini-ROV video survey with resources for a longer term study performed from 2009-2017 in the northern Gulf of Mexico to look at reef community changes. Reef fish communities displayed shifts in community structure, declines in species richness, diversity, evenness and density. Species richness exhibited a rebound, but species diversity has remained significantly lower than pre-DWH levels. Those reef fish that experienced high mortality had limited home ranges, making them susceptible to both contaminant exposure and resource limitations. Parallels were drawn to large red tide events along the west Florida shelf that resulted in community-wide declines. Within several years, a predictable pattern of recovery occurred. However, no such recovery has taken place in the northern Gulf where sub lethal exposures to PAHs continue, indicated by elevated PAH levels in bile and liver. Adding to the story was the competitive pressure from invasive lionfish, that hindered recovery to pre-spill populations.

End-to-end ecosystem modeling of the impacts of DWH on fish guilds using the Atlantis framework is consistent with the aforementioned findings. In simulating a dose-response scenario, we found that oil concentrations and associated toxicity had significant impacts on growth and mortality. Alternatively, fishery closures had little impact on biomass dynamics, compared to oil exposure that decreased biomass by 40%-70%. Overall, smaller fish were found to recover the fastest, within 10 years of the perturbation, but larger functional groups took as long as 20-30 years to recover. The main driver for these changes is through food web effects that communicate losses to regions far from the impact region. This finding should modify the regions of interest when looking at injury assessment.

In examining other components of the food web, C-IMAGE researchers used the MODIS nFLH product (normalized fluorescence line height, as a proxy for Chl a) to examine large temporal and spatial patters of primary productivity in the Gulf, pre and post-DWH. Environmental factors such as freshwater discharge, sea surface temperature and wind speed, are significant predictors for changes in nFLH. The MODIS nFLH overestimated the primary productivity from 2011-2014, but by 2015, productivity resumed to its pre- spill levels. However, the multi-year impacts suggest that oil residue remained in the underlying sediments. The utility of MODIS nFLH as a proxy for primary productivity is demonstrated and cane be useful for longer term spatial and temporal trends following a perturbation event.