Projects

Partners

Prof. P. Lionello (Scientific coordinator)	Dott. L. Santoleri	Dott. P. Malguzzi	Dott. S. Zecchetto

Purpose

This project analyzed the air-sea fluxes of heat andmomentum in the Mediterranean Sea at the medium time scale (1-10 days). The studyhas been based on data derived fromobservations and numerical simulations. Thenumerical simulation have been produced by a coupled atmosphere-ocean model, called MIAO(link to page)(Model of Interacting Atmosphere and Ocean), that describes parallelly the atmospheric circulation, the marine circulation, the wave field at the air-sea interface and theinteractions among these three components. The observations are derived from satellitedata and measurements campaigns by using the appropriate algorithms and dataprocessing.

Goals of the project

This project analyzed the air-sea fluxes of heat andmomentum in the Mediterranean Sea at the medium time scale (1-10 days). The studyhas been based on data derived fromobservations and numerical simulations. Thenumerical simulation have been produced by a coupled atmosphere-ocean model, called MIAO(link to page)(Model of Interacting Atmosphere and Ocean), that describes parallelly the atmospheric circulation, the marine circulation, the wave field at the air-sea interface and theinteractions among these three components. The observations are derived from satellitedata and measurements campaigns by using the appropriate algorithms and dataprocessing. The objectives of the project have been:

1. to evaluate the air-sea fluxes of momentum and heat (split in their different components) during events characterized by strong air-sea interaction and describe their space-time structure;

2. to evaluate the reliability of the algorithms for the computation of the air-sea fluxes in the numerical models and in the elaborations of the satellite data.

The air-sea fluxes are fundamental forcings for both the atmosphere and the ocean.The need for precise values of the fluxes of heat,momentum, and moisture is commonly accepted when describing phenomenologiescharacterized by long time scales (interseasonal and/or interannual variations and climatictrends) and large spatial scales (synoptic or global scale). Their precise specification is necessary for describing the behaviour of theatmosphere and the ocean on time scales sufficiently long not to be mostly determinedby the initial condition and on spatial scales sufficiently large to be not mostly forced by the boundary conditions. This project aims to analyze theair-sea fluxes at short time scales (the daily scale characteristic of theshort-medium term meteo-oceanic prediction) and small spatial scales (the mesoscale characteristic ofthe regional meteo-oceanic prediction). In the Mediterranean region there are, infact, interesting phenomenologies whose understanding is strongly connected to thecorrect description of the air-sea fluxes, and whose analysis allows an accurate study ofthe spatial and temporal patterns of the exchanges between atmosphere and sea and ofthe algorithms used for describing them.

The air-sea fluxes have been analyzed for a 20 day-long period during November 1996. The chosen period has been characterized by:

• marine cyclogenesis, whose development is influenced by the combined action of the orography and heat and moisture fluxes from the sea to the atmosphere;
• intense marine storms with high waves and coastal surges in the Northern Adriatic, whose evolution is determined by the transfer of momentum from the atmosphere to the sea.

The work has been split into 3 tasks:

• Numerical simulations
• Computation of momentum and latent heat fluxes using satellite data
• Comparison of the spatial and time structure of the fluxes produced by themodel and derived from the observations

The model simulations were carried out in a quasi-operational mode, as aseries of two days long simulations. In such set-up, each day of the period was been simulated twice, as first day of the present simulation, and as second day of the previous one. The initial and boundary conditions of the simulations were based on the ECMWF analysis fields for BOLAM (the atmospheric compontent of MIAO), and on climatologies for POM (the ocean circulation component of MIAO). The initialization of WAMwas based on the initial wind field. Maps of latent heat fluxs, momentum flux, surface wind and temperature, precipitation, solar radiation were computed every 6 hours.

The momentum fluxes were compared to the scatterometer data produced by NSCAT-ADEOS.
Model data and satellite observations show a minor bias, lower than the accuracyof the measurements and, therefore, not actually significant.However, the satelite data shows that the mesoscale structure isunderestimated by the model, which produces fields tendentially too smooth,though model and satellite observations have a similar (~25Km) spatialresolution.

The latent heat fluxes have been computed by the model at an approximately double resolution than the data, based on a combination of SSM/I and AVHRR. Differently than momentum fluxes, the model produces more fine scale structures than thesatellite observations. The main error in the satellite data is the substantialunderestimate of the heat flux associated the strong mistral wind episodesin the Western Mediterranean Sea.

Preliminary results have been presented at the EGS symposia

• Lionello P., E. Elvini, A. Sanna, M.G. Villani, P. Malguzzi, F.Bignami, L. Santoleri, S. Zecchetto (1999): The IAMMED Project: Air-Sea Interaction in the Mediterranean Region, Geoph. Res. Abs., E.G.S. 24th General Assembly, The Hague, NL, 1999, 1 - 449.

• Lionello P., P. Malguzzi, E. Elvini, S. Marullo, F.Bignami, R. Santoleri, S. Zecchetto (2000): Analysis of air-sea fluxes of heat and momentum over the Mediterranean Sea, Geoph. Res. Abs., E.G.S. 25th General Assembly, Nice, F, 2000, 2, Atm. Sc., 290.

Friction velocity: bias between satellite and model for the period 11 to 30 november 1996.The red areas indicate the regions where the momentum flux computed by the modelis systematically lower than the observations

 

 

Latent heat flux for the period from 11 to 30 november 1996.
Top panel: Miao model. Middle panel: satellite observations. Bottom panel: difference between observation and model latent heat flux. The bias isnegligible in the Tyrrenian and Ionian Sea, but the maximum close to theFrench coast is absent in the observations. Note that the model simulation does not include the Eastern Mediterranean Sea. The largest differences arelocated near the coasts and derive from a problematic specification of the land-sea mask in the satellite data.