Comparison of the feasibility of three flood-risk extent delineation techniques using Geographic Information System: case study in Tavoliere delle Puglie, Italy

Delineation of flood-risk extent is a tedious task which normally requires the expertise of trained hydrologists. In order to provide a quicker and more automated approach for estimating flood-risk extent, three straight forward methods (Mahalanobis distance, Maximin and Equally likely decision methods) which are easily executed in Geographic Information System were investigated and applied to a study area in the so-called 'Tavoliere delle Puglie' (southern Italy). The delineated areas were then visually and statistically confronted with a methodology based on hydrologic expertise that can be considered up to date with current flood-risk zone delineation knowledge. Two major problems associated with flood-risk extent estimation are firstly the overwhelming complexity of the wide variety of spatially relevant flood-related data and the amount of information contained within these data that are not easily interpreted without often costly and time consuming hydrologic expertise, and secondly, the wide variety of criteria usually considered in flood-risk analysis, yielding a broad spectrum of different units involved. The comparative analysis has revealed that of the three alternative methods, only the Mahalanobis distance technique gave moderately comparable results with the methodology based on hydrologic expert knowledge. The results suggest and confirm that hydrologic expertise cannot be bypassed in flood-risk extent analysis.