SEPARATION OF THE SEASONAL/PERMANENT THERMOCLINE TEMPERATURE VARIATIONS

Separation of the Seasonal/Permanent Thermocline Temperature Variations
A new approach is presented for the inversion of the acoustic travel times to separate and quantify the temperature variations between the seasonal and permanent thermocline.
Two different processes are happening in the surface and subsurface layers.
	Evolution of the range-averaged sound speed anomaly (dC) profiles from the reference field on the vertical section T1-T3 (990.5 km range). It shows that the dC in the surface layer (0-100 m) had been warming up about 7 m/s during the experimental period, while the dC in subsurface layer (100-1000 m) had been cooling down, indicating two different processes happened in the surface and subsurface, respectively.

Differential travel times from a tomography ray in the surface layer compares well with the SLA.
	Top: 10-day averaged travel time series of the arrivals on the T1-T3. The red one is the steepest ray (Ray1) which can access surface layer. Bottom: Comparison of the differential TTs from Ray1 to others and the SLA.
Temperature perturbation as a function of layer
	Top: Sum of the travel time in the three layers (0-100, 101-1500, and 1501-5500 m) for the Ray1 on the T1-T3. Bottom: Assuming the temperature variation only occurs in the Layer1 and Layer2, and the percentage of the effective travel time in each layer can be calculated. The temperature perturbation in each layer can be obtained by this ratio.
The contribution of horizontal eddy advection, rather than one-dimensional local forcing, may dominant the heat budget in the recirculation region.
	Top: Estimated range-averaged dT in the two layers. The dT has been warming up about 1.0° C in Layer1, but cooling down about 0.15° C in Layer2. Bottom: Corresponding heat content in the two layers.
The warming in the surface layer derived from acoustic data is consistent with the steric height change determined from the net heat flux data from NCEP/NCAR.
	Top: Comparison of the SSHA and estimated dT in surface layer on the T1-T3. Bottom: Comparison of SSHA and the steric height derived from NECP/NCAR reanalysis net heat flux. It shows 5 cm out of a total of 12 cm of SSHA change comes from changes in steric height (one-dimensional local forcing).
The barotropic westward-flowing velocity determined from reciprocal acoustic travel times in the recirculation region is about 5 cm s^-1 and compares well with surface geostrophic current velocity determined from T/P SSHA.
	Top: Range-averaged zonal geostrophic current between 143° E and 155° E derived from SSDH. It shows that axis of the Kuroshio Extension is between 35° N and 37° N, and has been moving northward from the day 200 to day 240. Bottom: Comparison of the geostrophic current from T/P SSHA and barotropic current determined from differential TTS. The barotropic current is comparable with the geostrophic current.