MESOSCALE DYNAMICS IN THE ADRIATIC SEA

Introduction

Mesoscale motions in the form of fronts, jets, meanders and eddies are ubiquitous across the Adriatic Sea, and it is presumed that they play a crucial role in several regions, especially along the Croatian coast where the topography is particularly rugged and in the region of the Po River plume where buoyancy forces are large. Our understanding of this variability, however, is rather elementary at the present time, for most of the efforts to date at understanding the physics of the Adriatic Sea have concentrated on the seasonal circulation and oscillations (tides, storm surges and seiches).

This project component of the overall ONR effort has for its main objective to arrive at a physical understanding of the various types of mesoscale motions in the Adriatic Sea. In other words, what makes them develop and what regulates their size and time scale.

The approach is primarily computational, based on two rather distinct numerical models (DieCAST and POM), each being run with high resolution (2 km) over the entire sea.

Regions of particular interest

Two regions of particularly intense mesoscale activity are selected for scrutiny:

1. The Croatian coast, at the level of the middle Adriatic basin (between 43° and 44°N). Here, the focus is on the episodic offshore jets, their relation to the local topography and wind events.
2. The Italian coast, from the Po River delta down to the Gargano Peninsula (between 42° and 45°N). Here, the mesoscale activity is suspected to be relatively pure baroclinic instability of the buoyant Western Adriatic Current.

• Spreading of river discharges
• Water exchange between coast and open sea
• Upwelling of bottom waters
• Preconditioning for dense-water formation.

Objectives of the project

To study the mesoscale variability in the Adriatic Sea, the following objectives are pursued with model simulations:

• Provide a general framework for in-depth studies of the mesoscale variability in the Adriatic,
• Uncover the dynamical reasons behind the regional variety in mesoscale structures, including their east-west contrast,
• Understand how fluctuating freshwater discharge and winds can create currents that can later be vulnerable to instabilities,
• Elucidate the role of bottom topography in generating jets and regulating their characteristics,
• Investigate the nature and role of fronts associated with mesoscale motions,
• Derive practical "rules of thumb" to predict the appearance and basic characteristics of mesoscale motions in the Adriatic, and to compose a list of ingredients necessary in any numerical model if it is designed to provide accurate simulations of mesoscale motions,
• Provide a dynamical basis for those researchers who will explore the impact of mesoscale variability on other Adriatic aspects such as acoustics and ecosystem, and
• Elucidate the cumulative contribution of mesoscale motions on the general circulation of the Adriatic.

Numerical Model

It is possible that buoyancy plays a larger role along the Italian coast and bottom topography along the Croatian Coast. In order to determine what is the dynamical nature of these instabilities and to which extend buoyancy and topography play such roles, numerical simulations need to be performed.

Demands on the computer model are:

1. High spacial resolution
2. Low dissipation
3. Ability to deal with sharp topographies.

Sample Results

Simulated current field in the Middle Adriatic - detail for Gargano Peninsula, Italian coast:

• Abundant mesoscale variability
• Preponderance of variability along the coasts (both sides)
• Propagating meanders, jets and fronts
• Length scales about 30 km
• Time scales on the order of 2-4 days
• Realistic simulations.

   
Compare vorticity at surface with that at 23 m, and note the spatial shift at the level of the Rimini Squirt, indicating baroclinic instability.

Conclusions

Simulations:

1. Seasonal circulation of the Adriatic Sea is unstable to mesoscale perturbations.
2. Mesoscale variability along the Italian coast is the result of baroclinic instability.
3. Mesoscale variability along the Croatian coast may have different dynamics (yet to be investigated).

1. The DieCAST model has a sufficiently low dissipation to allow mesoscale instabilities to develop at short length scales.
2. Simulated instabilities exhibit similarities with actual variability.
3. Simulated instabilities obey a recognizable dynamical scenario.