Impact of Disasters in Mediterranean Regions: AnOverview in the Framework of the HYMEX Project
Olga Petrucci and Maria Carme Llasat
A review of recent articles concerning the Natural Disaster Impact Assessment (NDIA) hasbeen performed according to the HyMex (
 Hydrological cycle in the Mediterranean Exper-iment 
) framework. HyMex is an international project focused on quantifying the hydrolog-ical cycle in the Mediterranean, analyzing high-impact weather events in the context of global change. According to their approach, the articles have been sorted in three groups:(a) those that focus on short-to-medium term effects directly involving people and goodsimpacted by the disaster; (b) those that focus on medium-to-long-term socio-economiceffects; and (c) those that focus on short-to-long-term physical and physiological effects onindividuals. The aim is to highlight the approaches used to address this issue in variousscientific fields and thereby to promote the sharing of both data and methodologies andfacilitate the use of an advanced multidisciplinary approach to the NDIA.
Natural disasters
Disaster Impact from the HyMex Perspective
The population of the Mediterranean area is increasing andas a result, it is experiencing an increase in urban sprawl,even in areas threatened by weather-related hazards such asfloods and landslides. Mediterranean countries are sociallyand politically diverse, but in the context of climate change,they must confront challenging, short-time extreme events(i.e., flash floods) and long-term changes (i.e., drought).
 is an international research project intended toquantify the hydrological cycle in the Mediterranean, with anemphasisonhigh-impactweathereventsinthecontextofglobalchange ( index.php). HyMeX-WorkingGroup 5
 which focuses on societal and economical effects,aims to (a) improve knowledge of high-impact weather eventsin the Mediterranean basin, (b) monitor vulnerability factorsand adaptation strategies developed by different societies toadjust to climate change, (c) point out the lessons that can belearnedfromdifferentsocieties andindividualsseekingtocopewith climate change and hydro-meteorological extremes,and make these lessons beneficial for all Mediterraneancommunities, and (d) identify changes in the vulnerability of humans and ecosystems under future global change. In thiscontext,itisessentialtocollectdataonweather-relateddisastersand to improve the procedures for assessing their impact.The social and natural science communities working onweather-related disasters in Mediterranean countries are cur-rently fairly limited, and there is no systematic procedureused to determine the impact of the disasters. Thus, wedecided to widen the scope of our survey, selecting articlesthat address the impact of all the types of natural disasters allaround the word. The hope is that we may thereby isolateapproaches used in those contexts that would also be suitablefor the Mediterranean area.
O. Petrucci (
)CNR-IRPI, U.O.S. of Cosenza, Via Cavour, 87036 Rende,Cosenza, Italye-mail: o.petrucci@irpi.cnr.itM.C. LlasatDepartment of Astronomy and Meteorology, University of Barcelona,08028 Barcelona, SpainC. Margottini et al. (eds.),
 Landslide Science and Practice
, Vol. 7,DOI 10.1007/978-3-642-31313-4_18,
Springer-Verlag Berlin Heidelberg 2013
Natural Disaster Impact: Definitions andUsefulness
The definition of 
 Natural Disaster Impac
 (NDI) changesaccording to both the aim of the study and the scientistassessing it. It can be defined as constituting the
 losses caused on environment andsociety by a natural disaster (Swiss Re 1998).
 Direct losses
 include physical effects such as destructionand changes that reduce the functionality of an individual or structure. Damage to people (death/injury), buildings, their contents, and vehicles is included, as are clean-up and dis-posal costs.
 Indirect losses
 affect society by disrupting or damagingutility services and local businesses. Loss of revenue;increase in cost; expenses connected to the provision of assistance, lodging, and drinking water; and costs associatedwith the need to drive longer distances because of blockedroads are included.
 Intangible losses
 include psychological impairmentscaused by both direct and intangible losses that individualspersonally suffer during the disaster.The
 Natural Disaster Impact Assessment 
 (NDIA) is cru-cial in helping individuals to estimate replacement costs andto conduct cost-benefit analyses in allotting resources toprevent and mitigate the consequences of damage (UNEP-ECLAC 2000). Possible end users of NDIA include thefollowing (Lindell and Prater  2003):1.
 which have an interest in estimating directlosses to report to taxpayers and to identify segments of the community that have been (or might be) dispropor-tionately affected.2.
 Community leaders
, who may need to use loss data after adisaster strikes to determine if external assistance is nec-essary and, if so, how much.3.
 who can develop damage predictions to assessthe effects of alternative hazard adjustments. Knowingboth the expected losses and the extent to which thoselosses could be reduced makes it possible to implementcost-effective mitigation strategies.4.
 who need data on the maximum losses in their portfolios to guarantee their solvency or even to under-take additional measures to alleviate the risk that theywould face in case of a disaster (i.e., the use of catastro-phe bonds) (Noy and Nualsri 2011).Data availability and reliability represent constraints inthe NDIA context because of the following issues:1. In most countries, there are no agencies responsible for gathering damage data.2. Long-term losses must sometimes be determined over aperiod of multiple years. Slow landslides, for example,can cause damage over long periods. Intangible damagelike disaster-related stress also requires years to bedetected (Bland et al. 1996).3. Data on property damage can depreciate the value of prop-erty, thus they would not be available (Highland 2003).4. For disasters as landslides or floods, the costs of damagesto structures such as roads are often merged with mainte-nance costs and are therefore not labeled as damage. Inaddition, when heavy rains trigger both landslides andfloods (Petrucci and Polemio 2009), it is difficult to sepa-rate landslide damage from flood damage.5. Developing countries have an incentive to exaggeratedamage to receive higher amounts of international assis-tance; thus, data may not be entirely reliable (Toya andSkidmore 2007).
Review Approach
Using Google Scholar, articles and books from academicpublishers, online repositories, universities, internationalorganizations and other web sites were selected by lookingfor terms related to natural disasters (natural disaster*,cyclone*, drought, earthquake*, flood*, hurricane*, land-slide*, tsunami, volcan*) and their effects (accident*,damag*,econom*,homeless*,impact*,injur*,loss*,morbid-ity, mortality, victim*, wound*, stress). The terms followedby asterisks were truncated so that we could search for allwords commencing with those letters (Ahern et al. 2005).Because we only searched for articles whose year of publica-tion was 1990 or later, approximately 100 articles wereselected. First, papers concerning general aspects of NDIAwere separated from articles presenting specific case studies.Then, some articles were excluded because the approachesproposed therein were the same as those reported in others of theselectedarticles.Finally,50casestudieswereselectedandsorted in three groups as described in the next section.
Short-Medium-Term Effects Directly InvolvingPeople and Goods Affected by a Disaster
Group 1 includes 22 % of the sample (Table 1). In the articlesemployingthesimplestapproaches,theimpactisexpressedbythe list of damaged structures that does not include monetaryfigures or any other assessment (Ngecu and Ichang’i 1999;Whitworth et al. 2006; Bilgehan and Kilic 2008). Frequently used impact indicators include numbers of victims anddamage to buildings, roads and agriculture. In these studies,damage data are obtained by state agencies or even collectedby directly asking people involved in the disaster. Both thenumber of victims and the percentage of people affected areused to compare the impact of a disaster on various
138 O. Petrucci and M.C. Llasat
communities (Msilimba 2010) or that of disasters that haveoccurred in different time and places.Some articles focus on damage to people, analysing thecircumstances leading to loss of life and assessing them inrelation to vulnerability factors (e.g., age, race, and gender)(Jonkman et al. 2009).
Medium-Long-Term Socio-Economic Effects
Group 2 contains 46 % of the selected articles (Table 2).After individuating the affected population and the pre-disaster situation, the researchers isolated effects on socialsectors (the population, housing, health and education),service infrastructure (drinking water and sewage,communications, electricity and power), and productionsectors (agriculture, industry and trade). All of the effectswere then aggregated to measure the disaster’s impact on themacroeconomic indicators during a period of 1 to 2 yearsafter the disaster (ECLAC 1991).In such studies, natural disasters are seen as a function of a specific
 natural process
 economic activity
 (Raschky2008). The indicators used to detect the impact on nationaleconomies include (a) long-term recovery businesses (Webbet al. 2002); (b) changes in flow variables such as annualagricultural output (Patwardhan and Sharma 2005);(c) variations in fiscal pressure (Noy and Nualsri 2011);and (d) effects on the labor market (Belasen and Polachek2007; Zissimopoulos and Karoly 2010). The value of human life can be tentatively assessed usingtwo approaches that assign different values to people indifferent income groups or in countries at different stagesof development (AusAID 2005):(a). The
 human capital
 approach involves calculating theaverage expected future income that the deceased wouldhave generated assuming that he (or she) had achievednormal life expectancy.(b). In the
 willingness to pay
 (WTP) approach, surveysassess how much an individual is willing to pay toreduce the risk of death.Even environmental damage can be assessed using theWTP approach, either by asking people to state a WTPamount or by inferring this amount based on costs incurredfor environmental services (Dosi 2001).Economically, disasters can act as a barrier to develop-ment, increasing poverty and having a small but significantnegative effect on economic growth (Raschky 2008). Thiseffect can return a society to the level of human developmentit had achieved 2 years prior to the disaster (Rodriguez-Oreggia et al. 2010). Indirect societal effects such asdecreases in productivity in people affected by disaster caninfluence economic growth (Popp 2006). Human capital canbe
 affected by these disasters through death or injuryand
 affected when damage to schools decreaseshuman capital accumulation (in poor countries, decreasing
Table 1
 Group 1: articles focusing on
 short-medium-term effectsdirectly involving people and goods affected by the disaster 
N. Authors Study area Disaste1 Bilgehan and Kilic (2008) Turkey Landslides2 Brunkard et al. (2008) Louisiana Hurricanes3 FitzGerald et al. (2010) Various Floods4 Jonkman et al. (2009) Louisiana Hurricanes5 Msilimba (2010) Malawi Landslides6 Ngecu and Ichang’i (1999) Kenya Landslides7 Patwardhan and Sharma(2005)India Tropicalcyclones8 Rappaport (2000) Various Tropicalcyclones9 Shajaat Ali (2007) Bangladesh Floods10 Whitworth et al. (2006) Spain Landslides
Table 2
 Group 2: articles focusing on
 medium-long-term socio-economic effects
N. Authors Study area Disaste1 Barredo (2009) Various Flood2 Barredo (2010) Various Windstorm3 Belasen and Polachek(2007)Florida Hurricane4 Birkmann et al. (2008) Sri Lanka,IndonesiaTsunami5 Cavallo et al. (2010) Haiti Earthquake6 Guimaraes et al. (1993) South Carolina Hurricane7 Gupta and Sah (2008) India Flood8 Kellenberg and Mobarak(2008) – Nat. disasters9 Luechinger and Raschky(2009)Europe Flood10 Noy and Nualsri (2011) Various Nat. disasters11 Noy (2009) Various Nat. disasters12 Padli and Habibullah(2009)Asia Nat. disasters13 Pielke et al. (2008) United States Hurricane14 Raschky (2008) Various Nat. disasters15 Rodriguez-Oreggia et al(2010)Mexico Nat. disasters16 Schuster and Highland(2007)Various Landslide17 Smith and McCarty 1996 Florida Hurricane18 Toya and Skidmore 2007 Various Nat. disasters19 Webb et al. 2002 California,FloridaNat. disasters20 Wilhite et al. 2007 United States Drought21 Wilson et al. 2011 Chile, Argentina Volcanic ashfall22 Xiao 2011 Various Flood23 Zissimopoulos andKaroly 2010Various Hurricane
Impact of Disasters in Mediterranean Regions: An Overview in the Framework of the
. . .
school attendance rates caused by reductions in familyexpenses can occur). Even demographic effects such asmigration have been detected (Smith and McCarty 1996).Nevertheless, natural disaster can also produce positiveeffects. Disasters can create
 Schumpeterian creative destruc-tion
 (Cuaresma et al. 2004), especially if there are injectionsof funds for assistance and/or reconstruction. They can rep-resent an opportunity to update capital stock and improve aneconomy, thereby producing a long-term positive effect onthe growth of the Gross Domestic Product (GDP) (Skidmoreand Toya 2002). Activities in the construction sector mayreactivate the economy, and the demand for constructionmaterials may generate windfall profits (ECLAC 1991).Outside the disaster area, income increases can accrue for owners of commodities whose price is inflated by disaster-induced shortages (CACND 1999). For instance, in the caseof drought, when agricultural production decreases, farmersin affected areas experience the negative effects of thedisaster, and the price of agricultural products increases.Then, farmers outside affected area, who are experiencingnormal production, will reap the benefits of these higher prices (Wilhite et al. 2007). Even ways of thinking andacting can be modified by major disasters, resulting in per-sonal and community growth (Birkmann et al. 2008).Disasters are more costly for developing countries: aseconomies develop, there are fewer disaster-related deathsand damages/GDP (Toya and Skidmore 2007). Neverthe-less, increasing wealth causes relatively higher losses inhigh-income nations (Raschky 2008). Increases in incomeincrease the private demand for safety; higher incomeenables individuals (and countries) to employ additional,costly precautionary measures. Nevertheless, in countriesthat experience a concentration of assets that is larger thanthe counter-measures put in place, the income-vulnerabilityrelationship can be inverted, especially in the case of disasters related to behavioral choices such as floods andlandslides. It seems that in countries with a GDP per capitabelow $4,500–$5,500, disaster deaths increase with income;however, they start decreasing if GDP is beyond that thresh-old (Kellenberg and Mobarak 2008).Disasters in South, Southeast, and East Asia are morecostly than those occurring in the Middle East and LatinAmerica. These results might be tied to the higher popula-tion density of Asian countries. Small island developingstates are severely impacted by such events (Meheux et al.2007): the number of victims and affected individuals andthe degree of damage are twice as large on average as in anyother region (Noy 2009).Normalization procedures are used to assess what themagnitude of economic losses over time would be if a pastdisaster took place today. It seems that societal change andeconomic development are the principal factors responsiblefor the increasing losses from natural disasters to date(Pielke et al. 2008; Barredo 2009,2010). For weather-related disasters, Bouwer (2011) pointed out no trends in losses – corrected for increases in population and capital at risk – thatcould be attributed to anthropogenic climate change.
Short-to-Long-Term Physical and PhysiologicalEffects on People
The articles in Group 3 (32 %) focus on natural disasters andtheir effects on people’s health from either a physical or apsychological point of view (Table 3). Pre- and post-disaster conditions were compared in these studies to detect the onsetof diseases and/or the worsening of pre-existing illness, andto assess if and when disaster-related symptoms appear/ disappear. The data collection processes mainly involvedstandardized questionnaires used to collect self-reportedinformation on symptoms quantified using numerical scores(Catapano et al. 2001; Cao et al. 2003) that could measure the disaster’s impact. The risk of developing physical and/or psychiatric disorders is related to the extent of the lossessuffered (Cao et al. 2003), and it is greater in families thathave lost a family member in a disaster (Lindell and Prater 2003), have experienced evacuation, or have worse finances(Bland et al. 1996). This probability can also be increased bya lack of information on the probability that the event willre-occur (Catapano et al. 2001). Two sub-sets of articleswere isolated that focused on
effects, respectively.
 Psychological effects.
 According to the
 Conservation Resource Model
, people try to protect resources such asobjects (housing, possessions,etc.), socialroles(employment,marriages, etc.), energy (time and monetary investments), andpersonalcharacteristics(e.g.,self-confidence). Thethreatened
Table 3
 Group 3: articles focusing on
 short-to-long-term physical and  physiological effects on people
N. Authors Study area Disaste1 Bland et al. (1996) Italy Earthquake2 Cao et al. (2003) China Earthquake3 Catapano et al. (2001) Italy Landslide4 Chou et al. (2003) Taiwan Earthquake5 Fonseca et al. (2009) Louisiana Hurricane6 Hussain et al. (2011) Thailand Tsunami7 Lazarus et al. (2002) Nat. disasters8 Liao et al. (2004) Taiwan Earthquake9 Lutgendorf et al (1995) Florida Hurricane10 Montazeri et al. (2005) Iran Earthquake11 O’Neill et al. (1999) North America Flood12 Phifer (1990) Kentucky Flood13 Ramachandran et al. (2006) India Tsunami14 Seplaki et al. (2006) Taiwan Earthquake15 Suzuki et al. (1997) Japan Earthquake
140 O. Petrucci and M.C. Llasat
oractuallossofthese resources ascausedbya natural disaster leads to psychological distress (O’Neill et al. 1999). Fre-quently observed conditions such minor emotional disordersseldom come to the attention of psychiatrists but may nega-tivelyaffectsocialrelationships and workperformance. Com-monly detected symptoms are fatigue (Lutgendorf et al.1995), tics, and cognitive experiences such as confusion,impaired concentration, and attention deficit disorder.
 signs such as anxiety, depression, and grief, as well as
behavioral effects
 such as sleep and appetite changes andsubstance abuse, were also reported (Lindell and Prate2003).Eveneffectsonsuicideratesweredetected:earthquakevictims (people who had lost family members residing withthem, were injured, or experienced property loss) were 1.46times more likely than non-victims to commit suicide (Chouet al. 2003).These effects can be mild and transitory or can lead to
 Post Traumatic Stress Disorder 
 (PTSD). The mental states of victims can include three stages (Sadeghi and Ahmadi 2008):(a) an
 immediate reaction
 involving distressing symptomsaccompanying adaptive stress; (b) the
 post-immediate phase
,which includes symptoms of maladaptive stress (confusion,agitation, and occasionally neurotic or psychotic reactions);and (c) the
 long-term sequel
 phase, which involves a return tonormal health or the onset of PTSD, which can sometimesyield a
 chronic phase
 that involves personality changes.These surveys make it possible to monitor the most fragilesegments of the population, including people with preexistingmental illness, racial and ethnic minorities, and children, inwhich symptoms may differ depending on age (Lazarus et al.2002; Overstreet et al. 2011). Gender differences arise as well: for instance, after an earthquake, women report greater emo-tional distress and mental health problems than do men (Norriset al. 2002), but the occurrence of addiction disorders amongwomen is much lower (Montazeri et al. 2005).
 Physical effects
 encompass symptoms affecting peoplewho have not been directly involved in a disaster. Thedeterioration of hygiene, housing, and basic services caninduce the outbreak of diseases such as
(AusAID 2005) or increase the risk of morbidity and mortal-ity caused by communicable diseases (Waring and Brown2005). In developing countries, for instance, contagious andnon-contagious diseases are reported during the first weeksafter floods. Moreover, in some environments, even theincidence of snake bites can increase (Shajaat Ali 2007).Disaster-related stress can have several secondary impactsonhumanhealth,suchaseffectsonthehumanimmunesystem(Solomon et al. 1997), diabetes (Ramachandran et al. 2006; Fonseca et al. 2009), and gastro duodenal ulcers (Suzuki et al.1997). Also increases in serum
 levels have beendetected in subjects with PTSD, which explains the hyper-vigilance of people who have faced danger and uncertainty(Liao et al. 2004). In addition, after major earthquakes, thenumber of patients with
 Acute Myocardial Infarction
 (AMI)has been reported to increase 3.5-fold, and the part of womenwithAMIseemssignificantlygreaterthanintheyearspreced-ing the disaster (Suzuki et al. 1997).
Concluding Remarks
Direct economic cost is not a sufficient indicator of disaster seriousness: estimating indirect and intangible losses is cru-cial to assessing the effects of disasters for welfare.Restricting research to the effects in a single sector can resultin fragmented coverage of the impacts (Meheux et al. 2007).This review highlighted techniques and findings in theeconomics and medical literature, which are not wellknown by those researchers working on physical aspects of natural disasters. A general NDIA procedure has not yetbeen developed; the applicability of the availableapproaches depends on the data accessibility. Only 12 % of the selected articles concern landslides impact: four dealingwith
 short-medium-term effects directly involving peopleand goods
, one focusing on
 medium-long-term socio-economic effects
 and one on
 short-to-long-term physicaland physiological effects on people
. This low attention tolandslides impact depends on two factors: (a) landslidescould be classified as
 disasters if compared toearthquakes or hurricanes; (b) landslides can be secondaryconsequence of 
 disasters such as earthquakes. Thus itis basic to develop impact assessment procedures plannedfor landslides, more spatial-oriented, and even taking intoaccount that exposure can be related to behavioral choices.
Ahern M, Kovats RS, Wilkinson P, Few R, Matthies F (2005) Globalhealth impacts of floods: epidemiologic evidence. Epidemiol Rev27:36–46AusAID (Australian Agency for Inter. Development) (2005) Economicimpact of natural disasters on development in the Pacific. http:// Last accessed June 2010Barredo JI (2009) Normalised flood losses in Europe: 1970–2006. NatHazard Earth Syst Sci 9:97–104Barredo JI (2010) No upward trend in normalised windstorm losses inEurope: 1970–2008. Nat Hazards Earth Syst Sci 10:97–104Belasen AR, Polachek SW (2007) How disasters affect local labor markets: the effects of hurricanes in Florida. IZA DP 2976, 38pBilgehan R, Kilic RK (2008) The landslides threatening Tasova town(Central Anatolia, Turkey) and their environmental impacts. Envi-ron Geol 55:179–190Birkmann J, Buckle P, Jaeger J, Pelling M, Setiadi N, Garschagen M,Fernando N, Kropp J (2008) Extreme events and disasters: a win-dow of opportunity for change? Analysis of organizational, institu-tional and political changes, formal and informal responses after mega-disasters. Nat Hazards 55(3):637–655Bland SH,O’Leary E,FarinaroE,Jossa F,Trevisan M(1996)Long-termpsychological effects of natural disasters. Psychosom Med 58:18–24Bouwer LM (2011) Have disaster losses increased due to anthropo-genic climate change? Am Meteorol Soc 92:39–46
Impact of Disasters in Mediterranean Regions: An Overview in the Framework of the
. . .
of 7