Numerical modeling of Hypoxia on the Washington coast
Development of a coupled interdisciplinary biophysical model of the Pacific Northwest coastal ocean
The resulting coupled biophysical model is capable of realistically reproducing currents, water properties, nutrients, primary productivity, and hypoxia along the Washington and Oregon coastlines.
The UW Coastal Modeling Group (UWCMG) is building an interdisciplinary biophysical model of the Pacific Northwest coastal ocean and inland water bodies. Numerical simulations run by physical oceanographers, Parker MacCready and Sarah Giddings (UW School of Oceanography), use realistic atmospheric forcing, river flow, tides, and oceanic boundary conditions to reproduce the coastal physical oceanography including currents, salinity and temperature. Biophysical oceanographers Neil Banas (UW Joint Institute for the Study of the Atmosphere and Oceans, JISAO) and Kristen Davis (UW Applied Physics Laboratory, now at UC Irvine) have incorporated an ecosystem model into the physical model that predicts the abundance and interaction of nitrate, phytoplankton, zooplankton, and detritus. Geochemical oceanographer Samantha Siedlecki (UW JISAO) has also added a dissolved oxygen sub-model. The resulting coupled biophysical model is capable of realistically reproducing currents, water properties, nutrients, primary productivity, and hypoxia along the Washington and Oregon coastlines.
Extensive observations of water properties (including velocity, salinity, and temperature) and biological parameters (including chlorophyll, nitrate, and dissolved oxygen) have been used to validate the model, showing that it reproduces both the physics and biology of the region well. Special thanks to B. Hickey (ECOHAB-PNW and RISE projects), E. Dever (RISE project), R. Thomson (Fisheries and Oceans Canada Institute of Ocean Sciences, IOS), and many other scientists for use of their observations for model validation. Please see the PNWTOX simulation website for more information on observations and model validation.
The movie shows results from the model over a full year simulation in 2005. The panels show (a) sea surface salinity, (b) sea surface temperature, (c) sea surface chlorophyll (a proxy for phytoplankton biomass), and (d) the bottom oxygen concentration where dangerously low hypoxic conditions are dark blue and the 100m isobath is outlined in a thin white line. (e) Along shore winds including the wind stress (black) and an 8-day weighted mean (gray) are shown over the year at the location marked by the star in panel (a). A red line indicates the time stamp of the above plots. Clear seasonal signals can be seen including strong downwelling during the winter months sending the Columbia River plume northwards, warming of the water surface during summer, and the onset of summer upwelling, which brings a band of cold deep water to the surface at the coastline. The model captures the observed late onset upwelling with a later than normal spring transition and bloom. During upwelling, nutrients brought up from depths with the cold upwelled water near the coast fuel blooms along the coastline. Hypoxic water develops on the shelf of both the Washington and Oregon coastlines over the course of the upwelling season from a combination of local respiration of organic matter and spatially variable contributions from the low-oxygen deeper source waters. The model reproduces both the spatial and temporal patterns of oxygen, primary production, temperature, salinity, and velocity fields for 2005-2007.
UWCMG scientists are using this model to investigate a variety of important interdisciplinary questions in the Pacific Northwest. It is being use as part of the NSF and NOAA funded Pacific NorthWest TOXins (PNWTOX) project to investigate the transport of harmful algal blooms from known source regions to the Washington coast. As part of the PNWTOX project, it is also being used to study the influence of freshwater sources on coastal productivity. Additional ongoing projects include investigation of spatial and temporal trends and causes of hypoxia on the shelf, and quantification of oceanic-estuarine and along-coast estuarine connectivity. Please see the PNWTOX simulation website for more information.