MFSPP - Mediterranean Forecasting System Pilot Project

 

The Mediterranean Forecasting System Pilot Project (MFSPP) closed in 2002, has started to develop the science base for the implementation of a Mediterranean ocean forecasting system. The aim was the prediction of the marine ecosystem variability in the coastal areas up to the primary producers, and from the time scales of days to months.

The project has two components: observing system and numerical modelling/data assimilation able to use the past observational information to optimally initialise the forecast. The basic assumption was that both hydrodynamics and ecosystem fluctuations in the coastal/shelf areas of the Mediterranean are intimately connected to the large scale general circulation. The second assumption was that, for the physical components of the ecosystem, monitoring and numerical modelling can work almost pre-operationally. The project has shown that NRT forecasts of the large scale basin currents are possible. Components developed and implemented: automatic temperature monitoring system for the overall Mediterranean Sea (Voluntary Observing Ship-VOS system) with NRT data delivery; a pilot Mediterranean Multisensor Moored Array buoy system (M3A) to monitor temperature, salinity and currents, together with biogeochemical and optical measurements to establish the feasibility of multiparametric monitoring of the upper thermocline in the whole basin; NRT satellite data (sea surface height, sea surface temperature and colour) analysis and mapping on the numerical model grid; different data assimilation schemes in order to assimilate multivariate parameters, e.g., XBT from the VOS and satellite sea surface height and sea surface temperature; 3, 5 and 10 days forecast experiments at basin scale for three months; techniques to downscale the hydrodynamics to different shelf areas of the Mediterranean Sea with nested models of different resolution; ecosystem models in shelf areas of the basin and a strategy for validation/calibration with M3A data sets; methods for assimilating nutrient, chlorophyll and PAR into predictive ecosystem models; an overall NRT data collection and dissemination network which allows the timely release of data for the forecasting exercise.

---