TY - JOUR
T1 - Internal-tide spectroscopy and prediction in the Timor Sea
AU - Kelly, Sam
AU - Jones, Nicole
AU - Ivey, Gregory
AU - Lowe, Ryan
PY - 2015
Y1 - 2015
N2 - © 2015 American Meteorological Society. Spectral analyses of two 3.5-yrmooring records from the Timor Sea quantified the coherence of mode-0 (surface) and mode-1 (internal) tides with the astronomical tidal potential. The noncoherent tides had well-defined variance and were most accurately quantified for tidal species (as opposed to constituents) in long records (>6 months). On the continental slope (465m), the semidiurnal mode-0 and mode-1 velocity and mode-1 pressure variance were 95%, 68%, and 56% coherent, respectively. On the continental shelf (145 m), the semidiurnal mode-0 and mode-1 velocity and mode-1 pressure variance were 98%, 34%, and 42% coherent, respectively. The response method produced time series of the semidiurnal coherent and noncoherent tides. The spectra and decorrelation time scales of the semidiurnal tidal amplitudes were similar to those of the barotropic mean flow and mode-1 eigenspeed (∼4 days), suggesting local mesoscale variability shapes noncoherent tidal variability. Over long time scales (>30 days), mode-1 sea surface displacement amplitudes were positively correlated with mode-1 eigenspeed on the shelf. At both moorings, internal tides were likely modulated during both generation and propagation. Self-prediction using the response method enabled about 75% of semidiurnal mode-1 sea surface displacement to be predicted 2.5 days in advance. Improved prediction models will require realistic tide-topography coupling and background variability with both short and long time scales.
AB - © 2015 American Meteorological Society. Spectral analyses of two 3.5-yrmooring records from the Timor Sea quantified the coherence of mode-0 (surface) and mode-1 (internal) tides with the astronomical tidal potential. The noncoherent tides had well-defined variance and were most accurately quantified for tidal species (as opposed to constituents) in long records (>6 months). On the continental slope (465m), the semidiurnal mode-0 and mode-1 velocity and mode-1 pressure variance were 95%, 68%, and 56% coherent, respectively. On the continental shelf (145 m), the semidiurnal mode-0 and mode-1 velocity and mode-1 pressure variance were 98%, 34%, and 42% coherent, respectively. The response method produced time series of the semidiurnal coherent and noncoherent tides. The spectra and decorrelation time scales of the semidiurnal tidal amplitudes were similar to those of the barotropic mean flow and mode-1 eigenspeed (∼4 days), suggesting local mesoscale variability shapes noncoherent tidal variability. Over long time scales (>30 days), mode-1 sea surface displacement amplitudes were positively correlated with mode-1 eigenspeed on the shelf. At both moorings, internal tides were likely modulated during both generation and propagation. Self-prediction using the response method enabled about 75% of semidiurnal mode-1 sea surface displacement to be predicted 2.5 days in advance. Improved prediction models will require realistic tide-topography coupling and background variability with both short and long time scales.
U2 - 10.1175/JPO-D-14-0007.1
DO - 10.1175/JPO-D-14-0007.1
M3 - Article
SN - 0022-3670
VL - 45
SP - 64
EP - 83
JO - Journal of Physical Oceanography
JF - Journal of Physical Oceanography
IS - 1
ER -