Mapping of a Typhoon‐Driven Coastal Upwelling by Assimilating Coastal Acoustic Tomography Data. Issue 10 (16th October 2017)

• Record Type: Journal Article
• Authors: Chen, Minmo; Kaneko, Arata; Lin, Ju; Zhang, Chuanzheng
• Is Part Of: Journal of geophysical research. Volume 122:Issue 10(2017)
• Page Start: 7822
• Page End: 7837
• Publication Date: 2017-10-16
• Subjects: coastal upwelling -- typhoon -- coastal acoustic tomography -- data assimilation -- Princeton Ocean Model -- Hiroshima Bay; Oceanography -- Periodicals; 551.4605
• DOI: 10.1002/2017JC012812
• Languages: English
• ISSNs: 2169-9275

Abstract

Abstract: A typhoon‐driven upwelling event was observed with coastal acoustic tomography in Hiroshima Bay during September 2013. The tomography data were used to obtain state estimates from an ocean model, employing the ensemble Kalman filter (EnKF) for data assimilation. Hiroshima Bay was represented as a two‐layer system with a fresh, near‐surface layer overlaying a more saline, lower layer. Wind from the typhoon forced the surface layer southward, thus drawing the lower layer northward and causing upwelling. After the upwelling, these currents reversed for a period. During the upwelling, the total volume transports for the upper and lower layers were well balanced, showing continuity of the two layers. However, the total upper layer volume transported northward during the reverse‐flow period was significantly smaller than the total upper layer volume transported southward during the upwelling, and significantly larger than the total lower layer volume transported southward during the reverse‐flow period. The downstream reductions of transported volume indicated mixing fractions of 24%–30%, with water mixing at the interface of the upper and lower layers. Errors in state estimates were less than the changes in current and salinity associated with the upwelling. Key Points: Coastal acoustic tomography data with two separate ray paths are assimilated into the Princeton Ocean Model with tidal and wind forcing The ensemble Kalman filter method with model state vectors perturbed by pseudo random noise is applied at every model grid point Subtidal current and salinity fields associated with a typhoon‐driven coastal upwelling are mapped in the upper and lower layers