A GIS tool for historical instability processes data entry: An approach to hazard management in two Italian Alpine river basins Chiara Audisio à , Guido Nigrelli, Giorgio Lollino CNR–IRPI Sede di Torino, Strada delle Cacce 73,10135 Torino (TO), Piemonte, Italy article info Article history: Received 2 July 2007 Received in revised form 23 December 2008 Accepted 3 January 2009 Keywords: Instability processes Historical research GIS tool Natural hazard Italy abstract Analysis of instability processes requires historical data over a range of temporal and spatial scales. While historical data offer a wealth of information about when, where and how a flood or a landslide happened or may recur, managing the data remains problematic. Before the data can be entered into historical and geographical databases, they need to be extracted from a vast variety of paper documents and transformed into a standard format. To do this, we developed a Geographical Information System (GIS)-based tool that permits easy data entry for comparing information on different temporal and spatial scales. The GIS tool was combined with a methodology for spatial data analysis to identify main hazardous areas. The historical and geographical databases were then queried with this tool to obtain the frequency of catastrophic events and their spatial recurrence. The GIS tool allowed accurate and rapid data management for establishing a connection between textual and spatial information for new data generation. This paper illustrates a methodology that utilizes the GIS tool for analyzing instability processes in two Italian river basins in the Western Alps. & 2009 Elsevier Ltd. All rights reserved. 1. Introduction According to a report by Munich Re (2002), floods throughout Europe in August 2002 caused damage of over 15 billion euros. Since 1990, over 30 floods have cost billions of US dollars and claimed numerous human lives (Kundzewicz et al., 2005) in Southeast Asia, Europe and North America (Munich Re, 2005 and 2006). Catastrophic floods have also struck the northwest Mediterranean basin (Spain and France in Dolz, 1993 and Llasat et al., 2005) and northern and central Europe (Germany, Poland, Hungary and Romania) (Benito et al., 2004; Dubicki et al., 2005; Kundzewicz et al., 2005; Masaryk Institute, 2004; Varga, 2000 1 ). In recent years, instability processes in Italy have caused calamities and considerable damage to property and infrastruc- tures. In the last 80 years alone 11,000 landslides and 5400 floods have occurred (Luino, 2005). Since 1980, the Italian State has spent 42.4 billion euros or about 5.7 million euros per day (Luino, 2005) in compensation and remedial works for natural disasters. These studies underscore a decline in the number of disaster victims in Western countries and a significant increase in material damage with mounting costs to national governments. The rise in instability processes and related costs is beyond the scope of this paper. Nonetheless, a knowledge of historical events and their causes is useful for studying them on a long-term scale to understand where instability processes may happen and what preventive measures should be undertaken to counteract them. Floods and landslides are complex phenomena with different hazard levels. Meteorology and hydrology play an influential role in triggering them: heavy rains, long rainy periods and snowmelt are major factors but are insufficient alone to explain why calamitous phenomena occur. Other conditions such as previous precipitation, terrain and surface run-off need to be taken into account (Llasat et al., 2005). In addition, these natural phenomena interact with human activities. Land use and its evolution, as well as civil and hydraulic infrastructures, can influence the natural response of floods and landslides (Arnaud-Fassetta et al., 2005). In brief, each event is the result of a complex interaction between meteorological, environmental and human factors. In-depth historical research has established that nearly all areas currently affected by instability processes were so in the past; understanding this continuity may aid in hazard investiga- tion and zonation (Benito et al., 2004; Tropeano and Turconi, 2004). Furthermore, historical research can play an integral part in reconstructing instability processes, describing different trigger- ing factors particular to a type of phenomenon, and constructing local and regional scenarios. Flood frequency estimates derived from historical data are based on hydrological measurements such as peak water levels recorded by gauges and marks on buildings (Bardsley, 1989; Bayliss and Reed, 2001; Cheng-Zheng, 1987; ARTICLE IN PRESS Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/cageo Computers & Geosciences 0098-3004/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.cageo.2009.01.012 à Corresponding author. Tel.: +390113977831; fax: +390113977820. E-mail addresses: chiara.audisio@irpi.cnr.it (C. Audisio), guido.nigrelli@irpi.cnr.it (G. Nigrelli), giorgio.lollino@irpi.cnr.it (G. Lollino). 1 Varga M. (2000). An Extraordinary Challenge for the Hungarian Basin Organizations: The Historic Flood in the Tisza River Basin. http://www.riob.org/ ag2000/varga_english.htm. Computers & Geosciences 35 (2009) 1735–1747