Ocean and Climate Group

Research

OCEAN ENERGETICS AND TROPICAL CYCLONES
Strong winds are responsible for vertical mixing in the upper ocean, that allows for entrainment of cold water into the mixed layer. This process affects both the gravitational potential energy of the water column and the heat fluxes at the air/sea interface, as it modifies the surface temperature. The upper ocean heat content, on turn, contributes to determining the intensity of tropical cyclones. For given atmospheric thermodynamic conditions, both a regime characterized by intense (with deep mixing and large upper ocean heat content) and by weak (with shallow mixing and small heat content) tropical cyclone activity can be sustained. We use a hierarchy of atmospheric and oceanic models, togheter with observational data, to study the feedback mechanisms between wind anomalies and upper ocean heat content. This project has two main goals: Understanding the sensitivity of tropical cyclone intensity to perturbations in the ocean state, and onstraining the variability in upper ocean energy content.

Zonally averaged upper ocean temperature variation in the last 50 years. Solid line indicates annual and zonal mean mixed layer depth. Isolines are every 0.15 degrees Celsius (blue negative, red positive). [World Ocean Database, 1955-2003, and NMLD].

OCEAN CIRCULATION The ocean water is characterized by non homogeneous distributions of temperature,

salinity, chemical tracers, and biological organisms. Its properties are largely affected by the large scale circulation in the ocean. The surface currents, largely driven by the winds, are relatively well known and understood. Unfortunately, a detailed and broadly accepted theory for the deep circulation is still missing, partly because of the intrinsic difficulties in observing the deep ocean. For these reason, simple models that aim at isolating specific mechanisms and quantifying their importance in the large scale circulation are useful tools. We are interested in the identification of the elements at the base of the strength of the circulation, of the heat transport associated to the water advection across thermal gradients, and of their variability. Using models of different complexity, we are currently investigating the effects that intense mixing between surface water and deep water, associated with convective events, has on the buoyancy fluxes at the air-sea interface and, eventually, on the large scale circulation.

Zonally averaged poleward salt transport in a model of the gyre circulation coupled to the overturning circulation, as function of time in an oscillatory solution. [Pasquero & Tziperman 2004].

SEAWATER EQUATION OF STATE The seawater equation of state is an empirical complicated polynomial in temperature, salinity, and pressure of the water. Several seawater properties can be explained using a simplified equation of state, that depends linearly on temperature and salinity. However, there are mechanisms that, in order to be understood, require a more complicated description of how density depends on the seawater properties.

Two wellknown examples are cabelling and thermobaricity. We are interested in studying the effects of those details on the stratification of the water column.
The term cabbeling refers to a "conspiracy" hidden in the dependence on temperature of the thermal expansion coefficient. When different water masses characterized by a different temperature and salinity but same density mix toghether, the resulting mixed water can become denser and eventually sink. Similarly, when we calculate the density of water as a function of the average temperature and salinity over a number of measurements, we do not obtain the average density but rather a somewhat larger estimate. In the figure to the right, it can be seen that in regions where water masses with different temperature meet (as between the subpolar and subtropical gyres), the density calculated from the averaged water properties and the average density differ (yellow and blue areas).

Cabelling: Error obtained when surface seawater density is computed from temperature and salinity values averaged on a 5degx5deg grid. [Levitus dataset].

Seawater thermal expansion coefficient is also dependent on pressure (thermobaricity). For this reason two water types with different salinity and temperature but same density at the surface, don't have the same density below the surface. In the Weddell Sea the surface water is usually cold and fresh, overlying over warmer and saltier water. The stable stratification is prone to an instability if the surface water is pushed down below the level of free convection (in the right figure below, the level of free convection for the surface mixed layer water is about 1600m below the surface: if the surface water reaches this level, it freely sinks downward.)

Thermobaricity: Salinity and temperature profiles in the Weddell Sea (left panel). Density difference between water from the surface mixed layer (red curve) adiabatically moved to a higher pressure, and in-situ water (right panel). [CTD data, Weddell Sea, 24Sep1989, WOCE dataset].

MARINE ECOSYSTEM DYNAMICS Phytoplankton are at the base of the food chain, with their capability of converting carbon dioxide into organic matter, directly using sunlight as the energy source. Their food include nitrate, phosphate, iron and silica, whose presence in the surface euphotic layer is therefore crucial for phytoplankton growth. Eventually, the availability of nutrients in the surface ocean, where they are consumed, depends on the supply from rivers, dust deposition, and upwelling nutrient-rich deep water. Those events can be highly intermittent in space and time.
Using simple models of the ecosystem dynamics coupled to fluid dynamics models, we investigate the importance of the small and short scales of the nutrient input event on the ecosystem dynamics. In the figure below, it can be noted that the largest phytoplankton concentrations (middle panel, red regions) are found in correspondence of relatively small zooplankton concentrations (right panel), indicating that frontier regions, where nutrients have just become available and zooplankton have not had time to grow and graze the phytoplankton are probably the most interesting area for the ecosystem productivity. A correct knowledge and representation of those frontier regions is important for understanding and modeling the marine ecosystem dynamics. Currently, we are investigating whether diffusion is a suitable way of representing the effects of small scale turbulent flows for reactive tracers.

Snapshots of the Nutrient, Phytoplankton, Zooplankton concentration fields for a numerical simulation of a N,P,Z ecosystem model coupled with a two-dimensional turbulent flow. [For details, see Pasquero, Bracco, Provenzale 2004].

TRANSPORT IN TURBULENT FLOWS The ocean transports heat, salt, momentum and vorticity, nutrients and pollutants, and many other material and dynamical quantities across its vaste space. Some of these transport properties are at the heart of mechanisms of climate variability and of marine ecosystem functioning. The task of understanding, modeling, and parametrizing oceanic transport processes can be difficult, as ocean motions include a full spectrum of different scales, from the displacement of individual molecules to the general circulation at basin and planetary scales. At each scale, structured motions appear, and eddies populate the flow. Of course, eddies at different scales behave differently: some of them are coherent structures, such as mesoscale vortices, and propagate for times as long as few years without modifying their shape and their characteristics, while others are random features that live for short times, such as whirls of three-dimensional turbulence.
We are interested in the effects that those turbulent features have on transport and on Lagrangian motion in particular, and in possible ways of parametrizing them (such as with stochastic models). The study involves both quasi-geostrophic turbulence and three-dimensional turbulence, as advecting fields for (fluid or inertial) particle motions.
We are currently working on the gravitational settling of heavy particles (e.g. plankton cells) in a three-dimensional flow.

Example of a (numerically simulated) two-dimensional turbulent vorticity field.
[from Pasquero and Falkovich, 2002].

Example of a (numerically simulated) three-dimensional turbulent energy field.