Increased dispersion of oil from a deep water seabed release by energetic mesoscale eddies

Highlights

• An oil spill is simulated in the deep Faroe-Shetland Channel, forced by hydrodynamic models with different resolutions.

• Explicitly resolving mesoscale eddies leads to dramatically increased pollutant dispersion at and below the surface.

• A change in stratification leads to differences in subsurface plume trapping depth and subsequent far-field transport.

Abstract

Hydrodynamics play a critical role in determining the trajectory of an oil spill. Currents, stratification and mesoscale processes all contribute to how a spill behaves. Using an industry‑leading oil spill model, we compare forecasts of oil dispersion when forced with two different hydrodynamic models of the North-West European Shelf (7 km and 1.5 km horizontal resolution). This demonstrates how the trajectory of a deep water (>1000 m) release in the central Faroe-Shetland Channel is influenced by explicitly resolving mesoscale processes. The finer resolution hydrodynamic model dramatically enhances the horizontal dispersion of oil and transports pollutant further afield. This is a consequence of higher mesoscale variability. Stratification influences the depth of subsurface plume trapping and subsequently the far-field transport of oil. These results demonstrate that the choice of hydrodynamic model resolution is crucial when designing particle tracking or tracer release experiments.

Introduction

Oil spills from seabed releases are a worldwide risk to the marine environment, and drilling is gradually pushing into deeper waters as a result of resource discovery and technological innovation (Burgherr, 2007; Gallego et al., 2018). Deepwater Horizon is the highest profile example of a deep seabed release, costing BP $145 billion (Lee et al., 2018) and spilling 7–8 × 10⁵ m³ of oil over 86 days (Camilli et al., 2010; Crone and Tolstoy, 2010; McNutt et al., 2012; Joye et al., 2016). Deep water drilling also occurs in the Faroe-Shetland Channel (FSC; Fig. 1a) in the North Atlantic, which has been a region of UK oil and gas development since the early 1990s (Smallwood and Kirk, 2005). As of August 2018, there were 162 active well heads in the FSC (source: UK Oil and Gas Authority), three-quarters of which were deeper than 200 m. A spill in the FSC lasting for a typical response time of 30 days could result in an oil release of up to 3.4 × 10⁵ m³ (Gallego et al., 2018).

The FSC is a hydrodynamically complex and energetic environment. A strong slope current (up to 1 m/s) along the West Shetland slope transports warm, saline surface water north-eastwards towards the Norwegian Sea and eventually into the Arctic Ocean. Near the seabed at >1000 m depth, a bottom current transports relatively cold, fresh deep water south-westwards, which either follows bathymetry through the Faroe Bank Channel into the open North Atlantic, or overflows the Wyville Thomson Ridge into the Rockall Trough (Turrell et al., 1999; Sherwin et al., 2008). Mesoscale eddies can extend across the width of the FSC and south of the Faroe Islands near the Faroe Bank Channel (Sherwin et al., 1999, Sherwin et al., 2006; Darelius et al., 2011). Large internal tides and non-linear internal waves have been observed in the region, which can act to increase turbulent mixing rates (Sherwin, 1991; Hosegood and van Haren, 2004; Hall et al., 2011, Hall et al., 2019). A unique stratification structure is also present, where the main thermocline typically resides at several hundred meters below the sea surface and separates the exchanging water masses (Berx et al., 2013; Fig. 1c, d, e).

The behavior of oil from a seabed release depends on ocean currents and stratification, in addition to properties of the oil such as viscosity, temperature, gas-oil ratio (GOR), flow rate and orifice diameter (Yapa and Chen, 2004). A previous plume modelling study suggested that oil will be trapped at 650–800 m depth from a 1000 m release in the FSC, depending on the release rate and ambient ocean conditions (Johansen, 2000b). Main et al. (2017) used a global ocean circulation model based on the Nucleus of European Modelling of the Oceans (NEMO; Madec, 2016) to predict the transport of oil from the FSC, and found that far-field oil transport was dependent on its depth. Oil near the surface travelled north-eastwards towards the Arctic Circle, whereas oil trapped at depth reached as far west as Greenland. However, they did not consider the influence plume dynamics might have on the vertical distribution of pollutant, or the role of surface weathering processes such as evaporation and emulsification. Additionally, the horizontal resolution of the model (1/12°) was coarser than required to explicitly resolve mesoscale eddies in the FSC region (2 km or less; Oey, 1998).

Oil Spill Contingency and Response (OSCAR) is a state-of-the-art modelling system that can be used to predict the fate and trajectory of an oil release during emergency response. OSCAR comprises of a 3-D fates model (Reed et al., 1995, Reed et al., 2000), near-field plume model (Johansen, 2000a) and droplet breakup model (Johansen et al., 2013). It is typically forced with horizontal current velocities from an operational hydrodynamic ocean model. OSCAR has been well-validated against historical emulsion observations (Abascal et al., 2010), synthetic aperture radar (SAR) satellite observations (Pan et al., 2020), and against the DeepSpill field experiment (Johansen et al., 2003).

In this study, OSCAR is used to consider how oil from a seabed release in the FSC could be transported by hydrodynamic processes, and how an increase in hydrodynamic model resolution influences the predicted dispersion. To simplify our diagnostics, biodegradation and direct wind forcing are not included. We demonstrate that enhanced mesoscale variability in fine-resolution hydrodynamic models leads to a dramatic increase in horizontal dispersion, and that stratification influences the depth of trapping and subsequent far-field transport. These results will help guide the choice of hydrodynamic forcing for emergency spill forecasting.