Permanent Meanders in the California Current System and Comparison of Near- Surface Observations with OGCM Solutions Luca Centurioni (SIO-PORD) Collaborators:

Permanent Meanders in the California Current System and Comparison of Near- Surface Observations with OGCM Solutions Luca Centurioni (SIO-PORD) Collaborators: Peter Niiler, Carter Ohlmann Acknowledgments (PI): Harley Hurlburt (NLOM), Julie McClean (POP), Jim McWilliams (ROMS), Ruth Preller (HYCOM) Outline Summary of observations from 15 depth drifters data; the bias problem: best estimate of 15 m depth geostrophic velocity field; comparison of some observation-derived quantities with OGCM solutions; Conclusions. Number of 6-hrs interval observations in a 0.5 x 0.5 bin. MEAN VELOCITY FIELD At 15 m DEPTH FROM mean field at 15 m Momentum balance (mean) at 15 m depth (dissipation is ignored): 78% 80% Vector correlation and scatter plots of geostrophic velocity residuals from drifters and AVISO UNBIASED FIELD (V C ): (Niiler et al. 2003) A running average (30 hrs) filter is applied to Lagrangian time series Ekman currents (Ralph & Niiler 1999) are removed to compute geostrophic velocities from drifters; 1.Drifter geostrophic velocities (V DG ) are binned in time (7 days) within each cell (0.5X0.5) and anomalies are computed; 2.Geostrophic velocities anomalies from AVISO (V S ) gridded maps are computed and interpolated at drifter locations and (binned) times; 3.Assume the following model: V G (t i ;x)=A(x)V S (t i ;x)+V C (x) 4.Estimate A and V C by minimizing {{(V G -V DG ) 2 }} where {{}} denotes time average over concurrent drifter and satellite velocity data, i.e Slope of the linear model V G (t,x,y)=A(x,y)V S (t,x,y)+V C (x,y) A UNBIASED GEOSTROPHIC VELOCITY VECTOR FIELD AT 15 m DEPTH Unbiased geostrophic velocity field: zonal component (cm s -1 ) MEAN GEOSTROPHIC EKE 0.5 FROM CORRECTED ALTIMETRY cm s -1 HYCOMNLOMPOPROMS spatial domainglobal ~1000 x 2000 km (USWC) vertical coordinateshybridlayerslevelssigma (ETOPO5) horizontal resolution1/12 (~7 km)1/32 (~3.5 km) 1/10 (~10 km)~5 km vertical layers/levels266 + ML4020 time step6 hour 15 minute mixed layerKPPKraus-TurnerKPP wind forcingECMWFNOGAPS/HRNOGAPSCOADS (seasonal) heat forcingECMWFNOGAPSECMWFCOADS (seasonal) buoyancy forcingCOADS (restored to Levitus) Levitus (restoring) Levitus (restoring) COADS (seasonal); parameterization for Columbia River outflow integration time years assimilationnoneSST, SSHnone otherLow computational cost open boundaries POP HYCOM NLOM ROMS MEAN SEA LEVEL (cm) EKE 0.5 FROM NUMERICAL MODELS (0-20 cm s -1 ) POP HYCOM NLOM ROMS EKE 0.5 COMPARISON (0-20 cm s-1) ROMS FROM CORRECTED ALTIMETRY Conclusions 1)Data confirm that the CCS (during the last 10 years and in the area examined) had 4 permanent meanders which are co- located with jets of zonal flow that extend nearly to Hawaii; 2)Time biases from the drifter data can be removed with the aid of satellite altimetry; Comparison of observed quantities with OGCM outputs can be addressed; 3)Preliminary comparisons show that ROMS is likely to be the model with the highest degree of realism; Number of 6-hrs interval observations in a 0.5 x 0.5 bin. Ageostrophic, non-linear velocity in ROMS and simple GFD model of cold eddy interacting with wind (Lee et al 1998) s -1 FOLLOWING THE DRIFTERS Ekman force ( ) is determined from (Ralph&Niiler 1999) MEAN EKE 0.5 at 15 m DEPTH (from drifters) cm s -1 EKMAN CURRENT AT 15m DEPTH Can we explain the jets of zonal flow? Suppose that: And use the following barotropic model to compute the stream function of volume transport per unit depth: Acceleration of a drifter: From AVISO and Unb. Vel. Field. From drifters of zonal volume transport per unit depth from barotropic model. cm s -1 The stream function of the mass transport can be computed as: Consider a one layer ocean of depth D=D 0 +D with a wind stress acting over it: Absolute sea level 27 Oct, 1993 with drifter tracks that are 21 days long Acceleration of a drifter: (horizontal velocity)