Propagation of Kuroshio Extension Meanders between 143° and 149°E

A two-dimensional array of current- and pressure-recording inverted echo sounders provided synoptic measurements of the upper and deep fluctuations in the Kuroshio Extension between 143° and 149°E with mesoscale resolution. Downstream-propagating meanders with periods of 3–60 days were always present between June 2004 and September 2005. Propagation speeds were estimated by two methods: spectral analysis of path displacements and complex empirical orthogonal functions (CEOF) analysis of along-path anomalies. The two methods produced similar results. Phase speeds increased smoothly from 10 km day−1 (0.12 m s−1) for meanders with wavelengths and periods [λ, T] = [420 km, 40 days] to 35 km day−1 (0.41 m s−1) for [λ, T] = [220 km, 6 days] meanders. This empirically derived dispersion relationship is indistinguishable from that obtained for Gulf Stream meanders downstream of Cape Hatteras. The deep ocean was populated with remotely generated, upstream-propagating eddies composed of a nearly depth-independent current structure. Upper meanders and deep eddies jointly spun up when they encountered each other with the deep eddy offset about a quarter wavelength ahead of the upper meander. Subsequently, as the upper and deep features moved past each other and the vertical offset changed, intensification ceased.

Tropical Intraseasonal Variability in the MRI-20km60L AGCM*

This study documents the detailed characteristics of the tropical intraseasonal variability (TISV) in the MRI-20km60L AGCM that uses a variant of the Arakawa–Schubert cumulus parameterization. Mean states, power spectra, propagation features, leading EOF modes, horizontal and vertical structures, and seasonality associated with the TISV are analyzed. Results show that the model reproduces the mean states in winds realistically and in convection comparable to that of the observations. However, the simulated TISV is less realistic. It shows low amplitudes in convection and low-level winds in the 30–60-day band. Filtered anomalies have standing structures. Power spectra and lag correlation of the signals do not propagate dominantly either in the eastward direction during boreal winter or in the northward direction during boreal summer. A combined EOF (CEOF) analysis shows that winds and convection have a loose coupling that cannot sustain the simulated TISV as realistically as that observed. In the composited mature phase of the simulated MJO, the low-level convergence does not lead convection clearly so that the moisture anomalies do not tilt westward in the vertical, indicating that the low-level convergence does not favor the eastward propagation. The less realistic TISV suggests that the representation of cumulus convection needs to be improved in this model.

Complex EOF Analysis as a Method to Separate Barotropic and Baroclinic Velocity Structure in Shallow Water

Defining the vertical depth average of measured currents to be barotropic is a widely used method of separating barotropic and baroclinic tidal currents in the ocean. Away from the surface and bottom boundary layers, depth-averaging measured velocity is an excellent estimate of barotropic tidal flow, and internal tidal dynamics can be well represented by the difference between the measured currents and their depth average in the vertical. However, in shallow and/or energetic tidal environments such as the shelf of the South Atlantic Bight (SAB), bottom boundary layers can occupy a significant fraction of the water column, and depth averaging through the bottom boundary layer can overestimate the barotropic current by several tens of centimeters per second near bottom. The depth-averaged current fails to capture the bottom boundary layer structure associated with the barotropic tidal signal, and the resultant estimate of baroclinic tidal currents can mimic a bottom-trapped internal tide. Complex empirical orthogonal function (CEOF) analysis is proposed as a method to retain frictional effects in the estimate of the barotropic tidal currents and allow an improved determination of the baroclinic currents. The method is applied to a midshelf region of the SAB dominated by tides and friction to quantify the effectiveness of CEOF analysis to represent internal structure underlying a strong barotropic signal in shallow water. Using examples of synthesized and measured data, EOF estimates of the barotropic and baroclinic modes of motion are compared to those made using depth averaging. The estimates of barotropic tidal motion using depth-averaging and CEOF methods produce conflicting predictions of the frequencies at which there is meaningful baroclinic variability. The CEOF method preserves the frictional boundary layer as part of the barotropic tidal current structure in the gravest mode, providing a more accurate representation of internal structure in higher modes. The application of CEOF techniques to isolate internal structure co-occurring with highly energetic tidal dynamics in shallow water is a significant test of the method. Successful separation of barotropic and baroclinic modes of motion suggests that, by fully capturing the effects of friction associated with the barotropic tide, CEOF analysis is a viable technique to facilitate examination of the internal tide in similar environments.