Search

This paper performs annual runoff frequency analysis in Sicily, Italy using the index runoff method. Homogeneous regions, not necessarily defined by geographical boundaries but sharing common features from a morphologic and climatic standpoint, have been identified. For each region a single probability distribution function can be fitted to the available annual runoff data scaled by the index runoff. Starting from an initial dataset of 105 hydrometric stations, 57 stations have been selected using heuristic criteria associated with the Hosking and Wallis’s discordancy measure. The Mantel tests have been then applied to find out the physical and morphological parameters best correlated with the runoff using the distances matrix approach. These parameters have been successively used in the cluster analysis to delineate homogeneous regions and the Hosking and Wallis’s heterogeneity test has been applied to the clusters in order to verify their homogeneity. Further analysis has been carried out to test the goodness-of-fit of four different candidate distributions of the scaled annual runoff using the chi-square test. For all the identified homogeneous regions the three-parameters lognormal distribution has been proved the most suitable distribution for the dimensionless annual runoff. In order to predict the runoff quantiles at ungauged sites, the procedure finally envisages the estimation of mean annual runoff (index runoff) by multiple regression analysis, providing expressions that relate mean annual runoff to some climatic and morphologic features of the basins.

The Generalised Likelihood Uncertainty Estimation (GLUE) approach is presented here as a tool for estimating the predictive uncertainty of a rainfall–runoff model. The GLUE methodology allows to recognise the possible equifinality of different parameter sets and assesses the likelihood of a parameters set being acceptable simulator when model predictions are compared to observed field data. The results of the GLUE methodology depend greatly on the choice of the likelihood measure and on the choice of the threshold which determines if a parameters set is behavioural or not. Moreover the sampling size has a strong influence on the uncertainty assessment of the response of a rainfall–runoff model. This is one of the most controversial and criticized aspect of the GLUE methodology, because it seems that this procedure does not learn from observations. Following these premises, this paper investigated first on the effect of different likelihood measures on the uncertainty analysis in the rainfall–runoff modelling of a mediterranean catchment. Performance of individual parameter sets has been assessed using three likelihood measures with a shaping factor. The acceptability threshold influence on the uncertainty analysis has been also evaluated. Finally it can be demonstrated how, using the GLUE, the predictive uncertainty grows with the streamflow variance while remains almost the same with the sampling size. In order to overcome these inconsistencies, a new simple likelihood measure, which explicitly takes into account the sample variance and extension, is here proposed.

The feared global climate change could have important effects on various environmental variables including rainfall in many countries around the world. Changes in precipitation regime directly affect water resources management, agriculture, hydrology and ecosystems. For this reason it is important to investigate the changes in the spatial and temporal rainfall pattern in order to improve water management strategies. In this study a non-parametric statistical method (Mann–Kendall rank correlation method) is employed in order to verify the existence of trend in annual, seasonal and monthly rainfall and the distribution of the rainfall during the year. This test is applied to about 250 rain gauge stations in Sicily (Italy) after a series of procedures finalized to the estimation of missing records and to the verification of data consistency. In order to understand the regional pattern of precipitation in Sicily, the detected trends are spatially interpolated using spatial analysis techniques in a GIS environment. The results show the existence of a generalized negative trend for the entire region.

This paper compares two generators of yearly water availabilities from sources located at multiple sites with regard to their ability to reproduce the characteristics of historical critical periods and to provide reliable results in terms of the return period of critical sequences of different length. The two models are a novel multi-site Markov mixture model explicitly accounting for drought occurrences and a multivariate ARMA. In the case of the multisite Markov mixture model parameter estimation is limited to a search in the parameter space guided by the value of parameter λ to show the sensitivity of the model to this parameter. Application to two of the longest time series of streamflows available in Sicily (Italy) shows that the models can provide quite different results in terms of estimated return periods of historic droughts, although they seem to perform more uniformly when it comes to simulate drought-related statistics such as drought length, severity and intensity. The role of parameter selection for the multisite Markov mixture model and of the marginal probability of generated flows in providing results consistent with the characteristics of the observed series is discussed. Both models are applied to the system of sources supplying the city of Palermo (Sicily) and its environs showing the applicability of the newly developed multisite Markov mixture model to medium-to-large scale water resources systems.