GOOS/GCOS measurements of near-surface currents Rick Lumpkin Rick Lumpkin ([email protected]) ([email protected]) National Oceanic and Atmospheric Administration (NOAA) Atlantic Oceanographic and Meteorological Laboratory (AOML) Miami, Florida USA Silvia Garzoli and Gustavo Goni Silvia Garzoli and Gustavo Goni NOAA/AOML Peter Niiler Peter Niiler NOAA/JIMO Office of Climate Observations 6 th Annual System Review, 3 September 2008
Rick Lumpkin ([email protected]) National Oceanic and Atmospheric Administration (NOAA) Atlantic Oceanographic and Meteorological Laboratory (AOML) Miami, Florida USA Silvia Garzoli and Gustavo Goni NOAA/AOML Peter Niiler NOAA/JIMO. GOOS/GCOS measurements of near-surface currents. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Pazan and Niiler (2001): uundrogued=udrogued+(7.910-3)W.
.WR
Auslip
Drag area ratio
Best fit: A=0.07
Wind speed
Holey-sock drifters: R=40.
Slip is 1.8 cm/s in 10 m/s wind.
Slip in high wind/wave stateNiiler et al. (1995) measurements of slip were in W8 m/s.
Slip may exceeds linear relationship at high wind/wave state.
Niiler, Maximenko and McWilliams (2003): absolute sea height change, 40—60°S: 2.34m, all drifters; 1.98m, only drifters in W8 m/s; 1.55 m, hydrography referenced to floats (Gille, 2003).
Discrepancies with models: problems with models or with data?
Left: mean zonal drifter speed (AOML climatology) minus mean zonal speed of ECCO-GODAE 1° state estimation, 15yr mean (figure courtesy M. Mazloff, WHOI).
Consistent offset in Southern Ocean.
How we can improve our understanding of drifter motion?
• Use high resolution scatterometer-based wind product and include ocean currents when calculating wind stress.
• Simultaneously project motion into geostrophic, wind-driven, and residual components.
slipMDT uuu
./)ˆˆ(ˆˆ)1( 5432'
1 nτu jxixjxixx av
Solve in bins using Gauss-Markov estimation.
(Lumpkin and Elipot, in preparation)
Wind and wind stress
Winds: 6 h, 25 km resolution Variational Analysis Method (VAM) product (Atlas et al., 1996; Atlas et al., 2008) derived from SSM/I, AMSR-E, TMI, QuickSCAT, SeaWinds, and in-situ observations and ECMWF analysis.
Stress: Smith (1988) algorithm as implemented in COARE 3.0 (Fairall et al., 2003) applied to VAM wind and drifter downwind speed (if drifter speed=wind speed, stress=0).
A priori errors 2/Rf
gu
Results
Globally-averaged gain: 1.13±0.06
Gain coefficient (shading) and time-mean currents (arrows)
Wind-driven motion
Comparison with Ralph & Niiler (1999)./ fAuEk .//* AfH
SummaryWithin assumed error, 5d lowpassed drifter velocities can be estimated as sum of geostrophic and wind-driven. Residual has structure related to EKE maxima.
AVISO altimetry generally underestimates observed EKE, presumably due to smoothing in OI. But some regions are overestimated with a gain of 1. Ageostrophic terms in surface momentum budget.
Wind-driven component is consistent with Ralph and Niiler (1999) in much of the tropics, subtropics. However, high wind areas have larger downwind motion. The spatial variations in the wind-driven part may be due to Stokes drift in wave field. This will be included explicitly in the next version of the model.
Stokes drift: 10—20 cm/s increase in time-mean, in some regions of the Southern Ocean.