1 Community-based risk management strategies: a survey on landslide risk knowledge and perception Michele Calvello 1 , Maria N. Papa 2 , Jonathan Pratschke 3 , Maria Nacchia Crescenzo 4 1 Department of Civil Engineering, University of Salerno, Italy E-mail: michele.calvello@gmail.com 2 Department of Civil Engineering, University of Salerno, Italy E-mail: mnpapa@unisa.it 3 Department of Economics and Statistics, University of Salerno, Italy. E-mail: jpratschke@unisa.it 4 Department of Civil Engineering, University of Salerno, Italy E-mail: mary87mn@libero.it ABSTRACT: People know and perceive risk differently. The study herein presented investigates the perceptions, knowledge and opinions on landslide risk by the residents of Sarno, a small town in southern Italy significantly affected by disastrous landslides on 5-6 May 1998. The paper presents the main results of a survey conducted in the months of March, April and May 2013, using a purposefully developed questionnaire. The questionnaire was administered through individual interviews to 100 residents, chosen to include a significant percentage of people living inside (60) and outside (40) of the so-called “red zone”, a territory declared at high residual risk after the events of 1998. Keywords: risk, perception, landslide, communication, education, community, resilience. 1. INTRODUCTION The resilience of a community faced with risks deriving from natural disasters is greatest when the local population can participate in decisions about risk management systems. Perceptions of risk are one of the fundamental elements that condition the behaviour of local residents (Lupton, 2013) and thus have a decisive impact on community resilience. This topic has arguably not received sufficient attention in the scientific literature dealing with landslides, with the exception of some scattered examples (e.g. Finlay and Fell, 1997; Wagner, 2007; Nathan, 2008; Scolobig et al., 2011). Herein we present a study on the perception of landslide risk amongst residents in a community which underwent enormous damage and loss of life 15 years ago as a result of a landslide. The community in question has experienced a number of qualitatively different phases of risk management characterised by varying levels of involvement and investment (including periods of very substantial investment and intense activity) at specific moments over the course of this period. The aims of the study are multiple, starting with the need to understand and measure how risks are perceived in this area, in comparison with other contexts in which similar events have not occurred. Secondly, we aim to establish the degree of trust which exists in relation to the organisms charged with managing hydrogeological risk. Thirdly, we examine satisfaction levels in relation to various kinds of risk-reduction interventions. 2. THE SURVEY The study was carried out in Sarno, a town in the Province of Salerno which experienced disastrous landslides in May 1998, which provoked widespread damage to property as well as 137 victims amongst local residents (see, for example, Cascini, 2004). The landslides in question, known as “rapid flowslides”, are frequent phenomena in the pyroclastic terrain that covers the carbonate rock surrounding Mount Vesuvius (see, for example, Cascini et al., 2013). Numerous studies of the rainfall characteristics that trigger landslides have been carried out over the years, including the various triggers, flow propagation dynamics and interactions with urbanisation (see, for example, Cascini et al., 2008; Martino and Papa, 2008). As far as we are aware, however, there are no published studies of the economic and social acceptability of interventions carried out to reduce and manage residual risks, despite the fact that these interventions involve enormous public investments and deployment of other technical and organisational resources. The questionnaire survey was carried out in March, April and May 2013. The sample comprised two sub-samples (Figure 1a), the first of which involved residents living inside the so-called “first red zone” (the area considered to be exposed to residual risk following the events of 1998), and the second of which included residents outside this area. The first sample, which we will refer to as the “red zone”, is based on 60 people, whilst the second includes data from 40 individuals. Within each area, a combination of sampling techniques was used to identify respondents, including both random selection of addresses and snowball sampling guided by the aim of maximising the heterogeneity of the two samples. The questionnaire was designed specifically with the aims of this study in mind, and contains four distinct sections: (a) socio- demographic and socio-economic characteristics of the respondent; (b) perceptions, knowledge and opinions in relation to past events; (c) perceptions, knowledge and opinions in relation to classifications of risk; (d) perceptions, knowledge and opinions in relation to the alert system. Each of these four sections contained a series of bespoke items which were developed on the basis of