Investigation of turbulence dynamics is very important for the understanding of dispersion and transport of pollutants in the marine environment. Specifically, at the surface boundary layer, dispersion phenomena are governed by the interaction of different forces (i.e., currents, waves, and winds) and are characterized by a wide range of temporal and spatial scales (Gallerano et al., 2016). Estimates of turbulence parameters able to describe the interaction of these different forces are required for an accurate prediction of pollutant pathways and concentrations. At present, turbulent dispersion simulations are mainly carried out by using Lagrangian particle models, alternatively based on a Wiener process or a Langevin scheme, which require as input data such turbulence parameters as diffusivity, velocity variance, and Lagrangian time scale (Monti and Leuzzi, 2010; De Dominicis et al., 2012). Besides, a new efficient approach is represented by kinematic chaotic models (Lacorata et al., 2014; Lacorata and Vulpiani, 2017).
In this paper the dispersion analysis of drifter campaigns realized in two different domains, the Gulf of Mexico and the Mar Grande Basin (Italy), was carried out using both relative and absolute approach. Relative dispersion, diffusivity and velocity variance were computed identifying different growth regimes. An exponential growth of the dispersion process was detected on the initial phase of deployment and empirical laws for horizontal diffusivity were inferred. Velocity variance and integral time scale were obtained using different techniques to evaluate short-term dispersion phenomena and submesoscale turbulence. A spectral analysis realized in accordance with the Taylor's theory was used to estimate the Lagrangian time scale in domains strongly affected by inertial oscillations. The turbulence parameters inferred can be used as input data in particle Lagrangian models of dispersion providing useful tools for surface dispersion prediction in marine environment.
The investigation of turbulence dynamics in marine environment results to be of primary importance because they govern and determine dispersion and transport of pollutants. Specifically, at the surface boundary layer, dispersion phenomena are determined by the interaction of different forcings, as currents, waves and winds, and are characterized by a wide range of temporal and spatial scales. Estimates of turbulence parameters able to describe the interaction of these different forcings are required for an accurate prediction of pollutant pathways and concentrations.
At present, turbulent dispersion simulation is mainly carried out using Lagrangian particle models, alternatively based on a Wiener or a Langevin scheme, which require turbulence parameters as diffusivity, velocity variance and Lagrangian time scale (Monti and Leuzzi, 2010; De Dominicis et al., 2012).
Despite many previous studies, the early phase of turbulent dispersion, which is fundamental in the planning of search-and-rescue activities in case of accidental release of pollutants, remains, at present, not completely clear and resolved, and the estimates of horizontal diffusivity varies by two orders of magnitude.