A water-quality model for the Lagoon of Venice, Italy

Abstract

A water-quality model for the Lagoon of Venice is proposed. The model is based on the results of an existing, deterministic, hydraulic-dispersive model of the Lagoon to provide the distribution of salinity and residence time in the Lagoon of Venice. This model has been implemented by Magistrato alle Acque di Venezia and Consorzio Venezia Nuova to evaluate the environmental impact of the MOSE Project, that has the aim to defend the city of Venice from extraordinary high tides [CVN, 1997. Allegato allo studio di impatto ambientale del progetto di massima delle opere mobili per la difesa dei centri abitati lagunari dagli allagamenti, vol. 2., CVN, 2002. Studio di nuove configurazioni dei canali di bocca e del relativo adeguamento progettuale delle opere mobili alle bocche di porto]. The water-quality is simulated by statistic analysis on water-quality data, monthly collected in 30 stations. The data-set covers a period of 2 years, and has been collected in the framework of MELa1, the institutional water-quality monitoring program (Magistrato alle Acque di Venezia, Consorzio Venezia Nuova). The Spearman correlation index of salinity and residence time versus the water-quality variables (nitrogen, phosphorus, chlorophyll-a and the trophic index TRIX) has been studied on a yearly average basis and for the spring–summer periods. The spatial distribution of the water-quality variables, based on the yearly average of nutrients, is mostly driven by the dispersive processes and is well correlated to salinity [Bianchi, F., Acri, F., Alberghi, M., Bastianini, M., Boldrin, A., Cavalloni, B., Cioce, F., Comaschi, A., Rabitti, S., Socal, G., Turchetto M.M., 1999. Biologocal variability in the Venice Lagoon. In: Lasserre, P., Marzollo, A. (Eds.), The Venice Lagoon Ecosystem. Input Interaction between Land and Sea, UNESCO, Man and Biosphere Series, vol. 25, pp. 97–125]. The model has been applied to simulate the variation of nutrients and trophic index distribution in the Lagoon as a consequence of an increase of hydraulic dissipation at the Lagoon outlets. The work presented in this paper shows that, coupling a deterministic, distributed-parameters, dynamic, hydraulic-dispersive model to a statistic one that accounts for the correlation between hydraulic related forcing functions (salinity, residence time) and water-quality data is a promising and simple way to evaluate the water-quality of the Lagoon of Venice. Of course, this methodology is applicable because a very large data-set is now available. The usual limitations of the statistical model methodology are present in this application too. E.g., it cannot precisely estimate the values of the water-quality variables, but it can indicate how they react when the system hydrological features change. Besides, the outcomes depend strongly on site characteristics and on the actual ecosystem state. The model has not been validated yet, due to the short data time lag, but the aim of this work is to suggest a simple simulation tool whose reliability is at least the same of that obtained by complex, deterministic, dynamic water-quality models. These models, accounting for several processes and hence including a lot of parameters, require for calibration a much more detailed data-set not yet available. The increase of dissipation is altering nutrient concentrations in the Lagoon of an average +3.2%, while the average variation for TRIX is +0.4%, and for chlorophyll-a is +3.0%. These variations are small enough to confirm a posteriori the validity of the adopted statistical approach.