ak_observes_nature.jpg
     

    Our sponsors:


    Office of Naval Research (ONR)
    National Science Foundation (NSF)
    National Oceanic and Atmospheric Administration (NOAA)
    Oregon Sea Grant


    Research directions:


    Process studies along the US West coast (Osborne, Fayman, and Yu): We have employed the comprehensive Regional Ocean Modeling System (ROMS, www.myroms.org, a community ocean circulation model, describing dynamics of three-dimensional stratified ocean flows) to run realistic simulations along the Oregon, Washington, and Northern California coasts. Using these models, we attempt to understand how the ocean works on a variety of spatial and temporal scales. Our ongoing studies have been focused on:

  • intermittency of the internal tidal motions along the Oregon shelf
  • effects of semi-diurnal and diurnal tides on cross-shelf transports
  • effects of the Columbia River fresh water discharge on the ocean surface topography, near-surface temperature, and currents along the Oregon shelf
  • connectivity of the coastal and interior and coastal ocean
  • seasonal differences in the ocean circulation on regional and coastal scales
  • effects of large scale ocean variability (including El Nino / La Nino) on the coastal ocean dynamics


  • Related recent and ongoing projects:

    2007-2012, NSF, Modeling Internal Tides in Interaction with Subinertial Wind-Forced Flows in the Coastal Ocean
    2010-2013, NSF, Modeling, Assimilation, and Analysis of the Shelf - Interior Ocean Exchange off Oregon


    Assimilation of satellite and in-situ observations in a coastal ocean model off Oregon (Yu, Kurapov): Data assimilation is a suite of methods to optimally combine available observations and models. Work in data assimilation allows us better understanding of how models represent features observed in the ocean. Assimilation has been implemented with a 3-km horizontal resolution ocean circulation model centered on the Oregon coast. We have used this model to study the effect of assimilation of various data sources on accuracy of ocean state estimates, and in particular forecasts of ocean currents, temperature fronts, and other features. Data utilized have included satellite along-track altimetry, satellite sea surface temperature (SST), surface currents from land-based high-frequency (HF) radars, temperature and salinity cross-shore sections from autonomous underwater vehicles - gliders, and velocity and temperature time series from moorings. By assimilating some data, and comparing model results to the data that are not assimilated, we learn about the impact of different data sources on fields not directly constrained by assimilation (for instance, we have demonstrated that assimilation of the slope of the sea surface height (SSH), measured by the altimeters, can help improve the geometry of the SST fronts).

    The forecast component of our data assimilation system is based on ROMS (www.myroms.org). Assimilation is done using the variational method, utilizing the tangent linear and adjoint codes AVRORA developed by our group. AVRORA is a set of stand-alone codes, which are dynamically consistent with ROMS. Utilizing our own AVRORA has allowed us flexibility studying various new aspects of data assimilation.

    Recently, our data assimilation system has been incorporated as a component of a pilot real-time coastal ocean forecast system, providing information about currents and near-surface temperature to coastal communities in Oregon. It has assimilated SSH from altimeters, SST from the geostationary GOES satellite, and surface currents from HF radars. The forecast model, constrained by data assimilation, provides daily updates of 3-day forecasts of ocean conditions (including surface currents and SST). These forecasts have been popular among Oregon fishermen community, to help planning their trips. The OSU glider group has been navigating ocean forecasts planning future glider operations. The forecast fields have also been delivered to the NOAA Office of Response and Restoration (ORR, formerly Hazmat) lab, to be coupled with the NOAA oil spill software.

    Selected related recent and ongoing projects:

    1999-2010, ONR, Data Assimilation in Shelf Circulation Models
    2011-2013, NOAA, CIOSS Support to GOES Improvement and Product Application Program
    2011-2013, NOAA-CIOSS, GOES SST assimilation for nowcasts and forecasts of coastal ocean conditions
    2007-2013, NOAA, Enhancing the Pacific Northwest Regional Coastal Observing System (RCOOS) of NANOOS


    Modeling and assimilation along the Australian coasts (S. Kim): In collaboration with Dr. Peter Oke (CSIRO, Australia), we are testing our data assimilation system in the new environment, off the Bonney Coast in southern Australia. A specific objective of this project is to compare variational and ensemble-based data assimilation methods, and to possibly improve data assimilation methodology by combining the two approaches. Experience modeling upwelling off the Bonney coast has revealed some dynamical features requiring careful modeling studies. These include confluence of water masses with different properties and non-trivial details of the atmosphere-ocean fluxes (including effects of the latent flux and evaporation).

    Related ongoing project:

    2010-2013, ONR, Combining Variational and Sequential Data Assimilation


    High-resolution modeling, analysis, and assimilation in the Bering Sea (Durski): A 2-km resolution model of the eastern Bering Sea is being developed to understand the nonlinear dynamics and then to test our ROMS-AVRORA data assimilation system in this area. Interesting dynamics include three-dimensional circulation near the shelf slope, submarine canyons, and along the ice edge, as well as tidal influences.

    Related ongoing project:

    2011-2013, NSF, Reconstruction of the eastern Bering ice-ocean system by variational assimilation of the BEST-BSIERP data


    Data assimilation with nearshore circulation models: We are applying adjoint-based and ensemble data assimilation techniques to flows along beaches and in river inlets, in particular, flows forced by breaking waves coming to the coast (including rip currents). These models feature strongly nonlinear dynamics which pose challenge to adjoint-based assimilation methods. The wave and circulation patterns strongly depend on the form of the sea bottom, which is often poorly known (and can rapidly change in response to storms). Assimilation of data on waves and currents for bathymetry inversion is one of the topics of this study.

    Related ongoing project:

    2010-2014, ONR, Remote sensing and data-assimilative modeling in littorials


    Are you looking for a PhD studentship or a post-doctoral position and would be interested to work in our group? Enthusiastic and curious researchers, with demonstrated interest to physical sciences, mathematics, and computer modeling can send their note of interest to kurapov at coas dot oregonstate dot edu