Project topics and materials
Get Complete Project Material File(s) Now! »
Hydrodynamical features of the Ligurian Sea
The Ligurian Sea is an area with a wide extended bathyal plain of about 2500 m deep (its deepest point at 2746 m) associated with a quite inexistent continental shelf in the western part and a more extended shelf in the East (around 1100 m).
The major large scale hydrodynamical feature is a well-defined cyclonic circulation involving both the layers of the MAW and of the LIW (Astraldi et al., 1994). Two main currents, which are part of the general cyclonic circulation of the western Mediterranean Sea, enter into the Ligurian Sea: the Western Corsican Current (WCC) and the Eastern Corsican Current (ECC). Both advect MAW and the ECC includes also LIW. These two currents join in the North of Corsica and form the Ligurian-Provencal- Catalan current also called the Northern Current (NC). The NC flows along seafloor slope between 25 and 46 km offshore (Béthoux et al., 1988). Millot (1991) has shown that the major surface currents flowing along the coastal slope are affected by instability processes that generate mesoscale eddies, capable of inducing relatively intense currents and producing significant dynamical heterogeneity of the hydrological characteristics in this area. The Ligurian Sea region is affected by severe weather condition in winter, which are mostly caused by the periodic intrusion of energetic, cold and dry continental wind, the Tramontana. Blowing over a warmer sea, the wind induce relevant air-sea interaction processes that subtract heat and water to the involved regions. Molcard et al. (2002) have suggested that the main features of the general circulation are induced by the structure of the wind stress curl, pointing out that the magnitude and spatial variability of the wind is essential in determining characteristics of the local circulations. The Ligurian-Provencal Catalan current has a maximum flux (1.5-2 Sv) during a relatively long winter season (roughly from December to May), and its structure markedly changes seasonally (Millot, 1999). This seasonal variation is mainly due to the variation of the ECC and the WCC which form the Ligurian-Provencal-Catalan current. The variability of these currents is investigated by, e.g., Astraldi and Gasparini (1992) and Sammari et al. (1995) and shows a seasonal cycle and dependence on local atmospheric forcing. A peculiar feature of the connection between the Tyrrhenian and the Ligurian basins is that, in early winter, the ECC at all depths undergoes a sudden increase that maintains, with some oscillations, for the whole colder season. From late spring, the increase progressively vanishes (Astraldi et al., 1994). The WCC displays a seasonal variability too, though less pronounced than that on the ECC. With the lowest values in late summer and autumn, the WCC progressively increases in velocity to a maximum in early summer (May-June and early July) (Astraldi et al., 1994).
The Ligurian-Provencal-Catalan current generates frontal zones characterized by thermal, saline or density gradient. The front separates the current (dynamic zone) from the central water, characterized by a weaker stratification and by characteristics similar to those of LIW. The front exhibits numerous instable meanders and eddies driving vertical movements, discontinuous in time and space (Boucher et al., 1987). These mesoscale features are prevailing during winter and can also cause smallscale turbulences.
Main features of seasonal dynamic of primary production in the Ligurian Sea
The Mediterranean Sea is considered as oligotrophic, and characterized by low primary production (Margaleff, 1985; Minas et al., 1988). However, satellite imagery has exhibited the spatial and temporal heterogeneity of phytoplankton distribution (Morel and André, 1991). In the Ligurian Sea, spatial variations, are essentially related to the presence of the NC. In the central part of the basin the seasonal succession of hydrologic conditions induces the production systems varying from mesotrophy in spring to oligotrophy in summer and fall (Marty et al., 2002). The Ligurian Sea is subject to intense vertical convection, which mixes the cool dense surface water and the underlying saltier Levantine intermediate water during winter. The winter mixing brings nutrients into the upper layer but the short residence time of them in this layer prevents the development of biomass (Morel and André, 1991). Short periods of stabilization during winter, due to good weather conditions, can occur and are associated with short-lived diatom blooms (DYFAMED, 1995). Progressing in the year, the surface layer becomes stable over a long period, thus allowing the winter nutrient enrichment to be utilized continuously and it leads to a spring bloom. The bloom have observed between early April and mid-May (Morel and André, 1991). Oligotrophy prevails during summer while perturbations in the meteorological forcing can generate a secondary bloom in fall. As regard the vertical distribution of nutrients, in deep waters, nitrate concentrations range between 7 and 8 μmol Kg-1, phosphate between 0,4 and 0,5 μmol Kg-1, and silicate between 7 and 8 μmol Kg-1 (Marty et al., 2002). Surface nutrient concentrations follow the general pattern of thermal tratification. Nitrate, phosphate and silicate concentrations reach, respectively, 2–3 μmol Kg-1, 0.15–0.2 μmol Kg-1 and 2–3 μmol Kg-1 in surface layers during winter mixing conditions. The nutrient concentrations decrease in the surface layer with increasing stratification to reach undetectable levels (below 0.05 μmol Kg-1) from June to November. The distribution of phytoplankton biomass follows the hydrographic and nutrient structure. Chlorophyll-a is maximal in surface layers during spring (bloom) in March and April. The maximum of biomass becomes less pronounced and descends to deeper layers (around 50 m) following the nitracline during summer and fall. The Chlorophyll-a is significantly present down to 50 m during the destratification process, and can reach 150–250m during the winter convective-mixing period depending of years (Marty et al., 2002). Some works indicate that phosphate is a limiting factor in the Mediterranean Sea (Béthoux et al., 1998; Coste et al., 1988), while other study suggest that Mediterranean waters could be nitrate limited (Andersen and Nival, 1988; Owens et al., 1989). In the central Ligurian Sea, it appears that there is a shift from nitrate limitation during winter and spring to phosphate limitation during the oligotrophic period. Marty et al. indicate that during the period of well established stratification (July, August and September), the N/P ratio below 200m is always around 20, but the data for the superficial (0–200 m) layer are always greater than 20. This confirms that the P-limitation is particularly important during the period ofoligotrophy. In this respect, the potential inputs of P by rain (Migon and Sandroni, 1999) or by deposition of Saharan dusts (Bergametti et al., 1992) are very important during summer. The depth at which P concentration is significantly different from 0 is about 50 m. This is clearly consistent with the thickness of the stratified depth. In this sense, high values of N/P ratio also can be attributed to a shift between nitracline and phosphacline.
Types, advantages and limitations of ocean models
Numerical ocean models can be classified in many different ways. The following is a version of Kantha and Clayson (2000).
Global or regional. The former necessarily requires high performance computing capabilities, whereas it may be possible to run the latter on powerful modern workstations. Even then, the resolution demanded (grid size in the horizontal and vertical) is critical. A doubling of the resolution in a three-dimensional model requires almost an order of magnitude increase in computing (and analysis) resources. Regional models have to contend with the problem of how to inform the model about the state of the rest of the ocean.
Deep basin or shallow coastal. The prevailing physical processes and the underlying driving mechanisms are essentially different for the two. Circulation in shallow costal regions is highly variable, driven primarily by synoptic wind and other rapidly changing surface forcing (and near river outflows, by buoyancy differences between the fresh river water and the saline ambient shelf water). Wind mixing at the surface and processes at the bottom are important, and a numerical model that has reliable mixing physics and which resolves the bottom boundary layer is therefore better suited to coastal applications. Deep basin, on the other hand, are comparatively sluggish, and the horizontal density gradients, especially below the wind-mixed upper layers, are a dominant factor in the circulation. The upper mixed layer can often be modelled less rigorously, especially for
applications that do not require consideration of air-sea interaction processes.
Table of contents :
Résumé en français
Sintesi in italiano
Abstract
1. INTRODUCTION
2. SCIENTIFIC BACKGROUND AND OBJECTIVES
2.1 The study area
2.2 Description of the Mediterranean Sea
2.2.1 Structure of the water masses
2.2.2 Driving forces of the circulation
2.2.3 Horizontal circulation of water masses
2.2.4 Deep and intermediate water formation
2.2.5 Thermohaline axes
2.3 Hydrodynamical features of the Ligurian Sea
2.4 Main feature of seasonal dynamic of primary production in the Ligurian Sea
3. NUMERICAL MODELS
3.1 Ocean models
3.1.1 Types, advantages and limitations of ocean models
3.2 ROMS, Regional Ocean Modeling System
3.2.1 Model description
3.2.2 Model formulation
3.2.2.1 Equations of motion
3.2.2.2 Vertical Terrain-Following Coordinates
3.2.2.3 Horizontal Curvilinear Coordinates
3.2.3 Parameterization of sub-grid processes
3.2.3.1 Non-local K profile
3.2.4 Numerical solution technique
3.2.4.1 Horizontal discretization
3.2.4.2 Vertical discretization
3.2.4.3 Time stepping scheme
3.3 Biogeochemical Model
3.3.1 State of the art of plankton models
3.3.1.1 NPZD model
3.3.2 Biological dynamics in NPZD model
3.3.2.1 Phytoplankton growth rate
3.3.2.2 Grazing, zooplankton growth and excretion
3.3.2.3 Regeneration and the nutrient and detritus equations
3.3.2.4 Sedimentation of phytoplankton and detritus
3.3.2.5 Phytoplankton, zooplankton, nutrient and detritus full equations
3.4 Improvement of the biogeochemistry and the regional biogeochemistry of the model
3.5 NPZD model coupled with ROMS model
4. RESULTS
4.1 Description of the main simulations
4.2 Model configuration
4.3 Initial and boundary conditions
4.4 Comparison between climatological and interannual simulations
4.5 Wind effect on mesoscale eddies structures
4.6 Downscaling from 3 km to 300 m
4.7 Biological production
4.8 Primary production and impact of
III mesoscale structures on biological production
4.9 Model evaluation 1
5. CONCLUSION AND PERSPECTIVES
Bibliography
