Lateral mixing and the North Atlantic Tracer Release Experiment: Observations and numerical simulations of Lagrangian particles and a passive tracer

Mixing and stirring of Lagrangian particles and a passive tracer were studied by comparison of float and tracer observations-from the North Atlantic Tracer Release Experiment (NATRE). Statistics computed from the NATRE floats were found to be similar to those estimated by Ledwell et al. [this issue] from the tracer dispersion. Mean velocities computed from the floats were ((u) over bar, (v) over bar) = (-1.2 +/- 0.3, -0.9 +/- 0.2) cm/s for the (zonal, meridional) components, and large-scale effective eddy diffusivities were (kappa(e11), kappa(e22)) = (1.5 +/- 0.7, 0.7 +/- 0.4) x 10(3) m(2)/s. The NATRE observations were used to evaluate theoretical models of tracer and particle dispersal. The tracer dispersion observed by Ledwell et al. [this issue] was consistent with an exponential growth phase for about the first 6 months and a linear growth at larger times. A numerical model of mesoscale turbulence that was calibrated with float statistics also showed an exponential growth phase of tracer and a reduced growth for longer times. Numerical results further show that Garrett's [1983] theory, relating the effective small-scale diffusivity to the rms strain rate and tracer streak width, requires a scale factor of 2 when the observed growth rate of streak length is used as a measure of the strain rate. This scale factor will be different for different measures of the strain rate and may also be affected by temporal and spatial variations in the mesoscale strain field.