Researchers: (alphabetic order) - Doglioli, Andrea1 - Griffa, Annalisa2,3 - Magaldi, Marcello3 - Molcard, Anne4 - Poulain, Pierre5 - 1Univ.Marseille, Fr - 2CNR-ISMAR, Sp, Italy - 3RSMAS, Miami - 4LSEET, Toulon, Fr - 5OGS, Tr, Italy Sponsors: - CNR - INFM Publications: - Molcard A., P.M. Poulain, P. Forget, A. Griffa, Y. Barbin, J. Gaggelli, J.C. De Maistre, M. Rixen. 2008. Comparison between VHF radar observations and data from drifter clusters in the Gulf of La Spezia (Mediterranean Sea). J. Mar. Sys, in press. - Doglioli A. M., A. Griffa, M.G. Magaldi, 2004: Numerical study of a coastal current on a steep slope in presence of a cape: the case of the Promontorio di Portofino. J. Geophys. Res., 109, c12033, doi:10.1029/2004JC002422 (PDF) - Aliani S., A. Griffa, A. Molcard, 2003: Floating debris in the Ligurian Sea, North-Western Mediterranean. Marine Pollution Bulletin, 46, (9), 1142-1149.

Understanding and predicting circulation and dispersion processes in coastal flows is crucial for a number of important applications such as support of safe navigation, search and rescue operations, prediction of the spread of contaminants and of biological quantities. Numerical and experimental studies have been performed to quantify circulation and dispersion in coastal regions characterized by complex topography  1) A numerical study of coastal currents in the vicinity of a cape 2) Circulation and dispersion in the Gulf of La Spezia (Mediterranean Sea) 1. A numerical study of coastal currents in the vicinity of a cape The response of a narrow shelf current in presence of a cape has been investigated. The results have been applied to the winter circulation in the area of the Promontorio of Portofino (North-Western Mediterranean) (Fig. 1.1, Fig. 1.2). Historical current measurements in the area suggest the presence of a recirculating eddy in the lee of the Promontorio (cape), with intensity of 10% of the incoming shelf current and with extension of 15 km. A sensitivity study is first performed considering the response of a steady incoming current which propagates on an idealized shelf topography (maintaining the shallower depths on the right) and interacts with the cape. Numerical experiments are performed using both 2D (vertically integrated) and 3D versions of the POM model. From the 2D results, the main controlling parameter appears to be the equivalent Reynolds number, while the system appears quite insensitive to the values of the Rossby number, i.e. to the intensity of the incoming current. In the 3D case, the dependence on the vertical Ekman number is investigated and a significant intensification of the attached eddy with respect to the 2D solutions is obtained, while the eddy size is basically unchanged. The main difference between the 2D and 3D dynamics is the presence of a resolved bottom Ekman layer in 3D, introducing a vertical shear in the incoming current. Downstream the cape in the surface layers, a current directed inshore and associated with an upwelling vertical velocity at the shelf break is observed. This current could be explained by the formation of a ``secondary circulation'' in the case of dominant Coriolis effect and appears responsible for the observed eddy intensification. Experiments with realistic bathymetry are then performed, The 3D results, obtained for realistic values of the parameters, show a good qualitative agreement with the measurement results (Fig. 1.3). The present results are expected to have significant consequences from the point of view of biological transport, pointing out to a mechanism for eddy intensification in the lee of a cape, connected with surface inshore transport and upwelling. Examples of idealized Lagrangian particles advected by the flow are shown in (Fig. 1.4). The eddy retention and the effects of upwelling/downwelling on particle transport are clearly shown. Figures: Figure 1.1 Location of the area of interest, situated along the western Italian coast in the North-Western Mediterranean Sea. Figure 1.2 Coastline and bottom topography in the area surrounding the Promontorio di Portofino. Red squares indicate the positions of historical current meter measurements. Figure 1.3 Results for vertically integrated velocity from 2D (upper panel) and 3D (lower panel) simulations. The numbers in the small boxes indicate average velocity from observations (blue) and model (red) in 3 locations. In the 2D simulations the gyre appears significantly weaker than observations, while the 3D simulations appear in good qualitative agreement with observations. Figure 1.4 Examples of idealized Lagrangian particles advected by the 3-D flow in Fig.3. The animation refers to a 20 day simulation with 12 hour interval. Different colors indicate different release regions, while the symbol size indicate particle depth (surface particles are characterized by maximum size). 2. Circulation and dispersion in the Gulf of La Spezia (Mediterranean Sea) In the framework of the MREA (Marine Rapid Environment Assessment) exercise LASIE (Ligurian Air Sea Interaction Experiment http://geos2.nurc.nato.int/lasie07/),  in the Ligurian Sea (Mediterranean), a coastal component experiment POET (Pilot Oceanographic and Environmental Trial, http://www.iof.cnr.it/esperimento/en/index_en.htm) has been carried out in the period 15-30 June 2007 focused on the Gulf of La Spezia, also known as Gulf of Poets. The Gulf area has scales of the order of 10 km, and it can be seen as a prototype of many coastal areas especially in the Mediterranean Sea. It is adjacent to a commercial and military harbour and is characterized by a complex use of the territory and of the sea, including tourism, fishing and aquaculture activities all collocated in a region adjacent to a marine protected park. Being able to monitor surface currents and associated Lagrangian transport in an area of this type is a key factor to a correct management. Measurements include current measurements from a WERA radar system in very high frequency (VHF) mode (resolution of 250 m, University of Toulone) ,and coastal surface drifter data (OGS) (Molcard et al. 2008). The drifters have been launched in clusters of 6-3 units, aimed at investigating the significant time and space variability of flow dispersal. In particular, two clusters launched in the same location two days apart show strikingly different patterns of evolution. The first cluster is swiftly advected with little particle spreading, while the second cluster divides into two branches with significant drifter spreading (Fig.2.1) In-situ drifter trajectories are compared with synthetic trajectories computed from VHF radar fields. In all cases, the spreading patterns of the in-situ clusters are correctly riproduced by the synthetic clusters, including the dramatic difference between the two launches from the same location (Fig,2.1). The positive results are likely to be due to a combination of factors and primarily to the high resolution of the radar fields, and they confirm that radar data are well suited for the study of coastal flows in limited areas with complex patterns of velocity and transport Figure 2.1: Example of snapshot of VHF radar velocity inthe Gulf of La Spezia, with superimposed drifter trajectories (upper panel).Comparison between trajectories of drifters (red dots) and synthetic trajectories computed from VHFradar (black dots), for four clusters launched in the Gulf showing significantly different dispersion behaviour. In all cases the VHF radar based trajectories are able to capture the cluster spreading.