Progetto: Research project

Dettagli progetto


The program and tasks of the Palermo Research Unit are here described with reference to the same "Research Issues" used in the previous section ("Descrizione(Obiettivi e Metodi)).

1) Research Issue “Water”: Costruction of quality assessment grids for the Sicilian coastal marine waters

Over the two years of this research program, the activity will focus on the use of information on the level and mechanisms of growth of Posidonia oceanica meadows, along with information on the chemical-physical characteristics of the water column and on sediments, to build up quality assessment grids for the Sicilian coastal marine waters.
The database used for this analysis, which will be carried out as a collaborative effort with the marine ecologists of the Department of Botanical Sciences at Palermo University, is being collected within the framework of an agrrement with the Sicily Environmental Protection Agency (ARPA Sicilia) and is one of the richest in the whole Mediterranean basin: at each of three stations of different depth (superficial, intermediate, deep) along 90 transects located at different sites of the Sicilian coast a sample of 30 P. oceanica shoots are extracted by a systematic random sampling scheme.
By means of lepidochronology, in the laboratory several variables concerning the annual growth of each shoot are measured. At each transect, physical-chemical variables concerning the water column and variables concernig the sediments are also collected.
The use of information on P. oceanica growth for the quality assessment of coastal marine waters requires, preliminarily, the ability to model the relationships between potentially explanatory
variables and response variables measuring growth performance. As mentioned in the Scientific background, this will be done through Generalised Linear Mixed Models. This choice requires also the development of appropriate methods for inference and diagnostics, since many problems are still open in this area, concerning on one hand the definition of goodness-of-fit tests and on the other hand the definition of appropriate residuals and their use in graphical diagnostic tools (for some preliminary results, see Lovison, 2005 and Sciandra, Lovison, 2005). In a second stage, the research activity will address the methodological problems involved in the construction of synthetic indicators of water quality, with special attention to both their scientific rigour and their interpretability for decision makers.

2) Scientific Issues Area “Air”: Analysis of the influence of meteorological variables on the concentrations of atmospheric pollutants in the Sicilian provinces

The research activity will aim at the extension to the Sicilian regional level of the study about the effect of meteorological variables on pollutant concentration already carried out in the city of Palermo (Bondì,Plaia, 2005). The available database contains the concentrations of the main pollutants and the measurements of some meteorological variables, recorded hourly/bihourly since 2001 by the monitoring networks of the Sicilian municipalities and/or provinces.
Tha data therefore display a high time frequency and a variable space density, locally high within municipalities but globally low in the region; as a consequence, special attention will be devoted to modelling the spatial correlation.
Two problems of special concern will be the acceptability of the assumption of separability of time and space components (see, for example, Shaddick, Wakefield, 2002), and the problem of "misalignement" between meteorological and pollution data, mentioned in the Scientific Background. In particular, considering the first problem, the use of (classical) hierarchical linear models seems to be particularly reasonable, as they allow to distinguish between small/medium scale and large scale variability. This is possible thanks to the possibility of considering units at three level
Data di inizio/fine effettiva2/9/072/9/09


Esplora i temi di ricerca toccati da questo progetto. Queste etichette sono generate sulla base dei riconoscimenti/sovvenzioni sottostanti. Insieme formano una fingerprint unica.