Submesoscale horizontal structure of upper-ocean velocity and density

Sina Khani1, James B Girton2, Eric L Kunze3, John Mickett1 and J. Thomas Farrar4, (1)University of Washington, Applied Physics Laboratory, Seattle, WA, United States, (2)Applied Physics Laboratory, University of Washington, Seattle, WA, United States, (3)U of Washington, Seattle, WA, United States, (4)Woods Hole Oceanographic Inst, Department of Physical Oceanography, Woods Hole, MA, United States
Upper ocean velocity and density structure results from a superposition of many processes, including internal waves, fronts, baroclinic instability, atmospheric forcing, and turbulent energy transfer to smaller or larger scales. We report on measurements from dense arrays of drifting profilers (9, 16, and 23 EM-APEX floats at a time) and contemporaneous ship surveys with ADCP and a towed CTD profiler (SWIMS) to characterize upper ocean horizontal structure on 0.5-100 km scales at 3 sites in the vicinity of the North Pacific Subtropical Front. Comparisons are made between (a) straight-line legs in the ship surveys, (b) along-trajectory estimates from the profiling floats, and (c) simultaneous differences across the float arrays. The first two approaches allow estimation of horizontal wavenumber spectra (taking into account the differences in platform speed and folding together of spatial and temporal variability), while the third provides estimates of the spatial covariance and structure functions. Furthermore, horizontal structure in narrow-band frequency peaks attributable to a single coherent tidal or near-inertial internal wave can be estimated and removed using least-squares fitting. Spectral energy estimates and real-space statistical energy distributions are compared between these methods in the regions of overlap. The results are also used to investigate the separability of frequency and wavenumber contributions and to identify the shapes and structures related to different types of forcing and dynamical processes.