Forecasting Hypoxia to Support the Dungeness Crab Fishery

Samantha Siedlecki, Marine Sciences, University of Connecticut, Groton, CT, USA, Jan Newton, Applied Physics laboratory, University of Washington, Seattle, WA, USA, Parker MacCready, School of Oceanography, University of Washington, Seattle, WA, USA, Simone Alin, Pacific Marine Environmental Laboratory, NOAA, Seattle, WA, USA, Dan Ayres, Washington Department of Fish and Wildlife, Olympia, WA, USA, and Joe Schumacker, Quinault Marine Resources Program, Taholah, WA, USA. Extracted from OOI Science Plan, 2021.

Seasonally, the upwelling region of Washington, British Columbia, and Oregon coastal waters experiences a decline in oxygen levels on the shelf that is well observed and simulated historically (Hales et al., 2006; Connolly et al., 2010; Peterson et al., 2013; Adams et al., 2013; Siedlecki et al., 2015). This seasonal decline is primarily driven by respiration of locally produced organic matter, that results from high productivity fueled by source waters rich in nutrients and influenced by transport. The same processes that enrich the source waters with nutrients cause them to be lower in oxygen relative to other regions, as well. Hypoxia is regularly experienced in the region and is expected to increase in frequency and severity with deoxygenation and climate change (Siedlecki et al., in review). Increases in hypoxia will lead to a decrease in biodiversity in the affected habitats (Levin et al., 2009), challenging managers in the region who manage species sensitive to these changes.

Hypoxia has already been linked to mass mortality events of hypoxia-intolerant species of invertebrates and fish, and in particular crab, off the coast of Oregon (Grantham et al ., 2004; Chan et al ., 2008; Barth et al., 2018). The Dungeness crab fishery is the most valuable single-species fishery on the U.S. West Coast, with landed values up to $250 million per year (Pacific States Marine Fisheries Commission, 2019) and plays an enormous cultural role in the lives of tribal communities in the region. While Dungeness crabs can reposition themselves out of hypoxic waters (Bernatis et al., 2007; Froehlich et al., 2014), mass mortality events have been recorded for crabs exposed to hypoxia for more than a few days within fishery pots in Washington and Oregon waters (Grantham et al., 2004; Barth et al., 2018).

Seasonal and short-term forecasts of hypoxia and other ocean conditions have been made in the region by JISAO’s Seasonal Coastal Ocean Prediction of the Ecosystem (J-SCOPE) in Washington and Oregon outer coast waters since 2013. J-SCOPE forecasts have significant skill in forecasting ocean conditions, including bottom oxygen on seasonal timescales (Siedlecki et al., 2016; Kaplan et al., 2016; Norton et al., 2020; Malick et al., in review). The skill from the forecasts is thought to emerge from El Nino and Southern Oscillation (ENSO) teleconnections (Jacox et al., 2017), but subsurface oceanic teleconnections likely also contribute (Jacox et al., 2020; Ray et al., 2020). January forecasts have out-performed the April-initialized forecasts historically. The onset of hypoxia has been successfully forecasted at mooring locations (Siedlecki et al., 2016).

LiveOcean, supported by the Washington State Ocean Acidification Center, has been providing 72-hour forecasts of Washington and Oregon waters, including coastal estuaries and the Salish Sea, since 2015. A comparator is available in real-time for this system, which allows direct comparison of the forecast with real-time observations. This kind of transparency in model performance is essential to building trust with stakeholders.

Both forecasts are hosted through the regional IOOS portal for the Northwest Association of Networked Ocean Observing Systems, called NANOOS, which provides a connection to regional stakeholders through existing long-term relationships. NANOOS has established working partnerships with local user communities since its Implementation Charter in 2003. Its Governing Council, now with over 70 member institutions, has provided direction, but much of the work comes from individual connections that NANOOS has fostered for years. An example is the need by state and tribal managers for understanding hypoxia effects on crab. The inclusion of J-SCOPE has enabled managers to have easy and direct access to data and forecasts. But the partnership extends beyond that. These managers also provide input into development of the products, including extensive input within J-SCOPE’s development of crab habitats and oxygen forecast products. Regular calls and webinars with the forecast scientists and managers help to assure that the products meet their needs. Together with real-time observations, these forecasts empower the region’s community with advance knowledge about the upcoming season’s ocean conditions to use in their decision- making process.

For example, in late June of 2018, emails were sent around to the J-SCOPE team initiated by the managers and NOAA scientists, relaying fishers’ experience in the region pulling up dead crabs in pots without knowing the cause. Scientists on the email chain pulled up real-time OOI observations through the NANOOS data portal, and found that the Washington Inshore Surface Mooring of the Endurance Array (CE06ISSM) had measured hypoxia from June 7th onwards (Recovered, Fig. to the right). While retrospectively there were QA/QC concerns for the oxygen data from this deployment, the “recovered” data stream is plotted here as an example of real-time conditions, with less focus on the specific value. The oxygen concentration threshold below which crabs perish is elusive, but there has been some discussion of it falling around the “severe” hypoxia threshold—22 μmol/kg or 0.65 mg/L, which is lower than the traditional hypoxia definition of 65 μmol/kg or 2 mg/L (Barth et al., 2018). J-SCOPE forecasts had forecasted onset of hypoxia earlier than usual, and LiveOcean forecasts indicated the spatial extent of the event was widespread nearshore (Fig. 2.8). Managers suspect the widespread low oxygen waters impacted the distribution of crabs that year, forcing them out of typically productive regions. The Quinault Indian Nation did take management action based on observations and J-SCOPE forecasts to close the 2018 fishery early due to recurring hypoxic conditions in the summer. A similar event occurred in 2017, but the NOAA-funded project had not yet begun at that time. The 2017 event is documented in Barth et al. (2018).

Ocean forecast systems can be relied on to help manage these events sustainably by providing guidance as to regions that will likely require soak time limitations to ensure crabs are captured alive, and aid in spatial management of the fishery itself. Observing systems like the OOI can continue to aid forecast system development in this region by extending observations into the poorly monitored winter months, helping to identify thresholds for crabs by ensuring the historical data are both available and quality controlled, and continuing to stream the observed fields in real-time. Future projections under the most severe emissions scenario explored predict that the region will continue to experience hypoxic events of greater duration and severity in the future (Dussin et al., 2019; Siedlecki et al., in review], making forecast tools on short timescales critical for the effective management into the future of the West Coast’s most valuable fishery.