IPCC_Managing Risks of Extreme Events.pdf - Climate Access

Climate Change: New Dimensions in Disaster Risk, Exposure, Vulnerability, and ResilienceChapter 1Whether an extreme event results in extreme impacts on humans andsocial systems depends on the degree of exposure and vulnerability tothat extreme, in addition to the magnitude of the physical event (highconfidence). Extreme impacts on human systems may be associatedwith non-extreme events where vulnerability and exposure are high(Sections 1.1.2.1 and 9.2.3). A key weather parameter may cross somecritical value at that location (such as that associated with heat waveinducedmortality, or frost damage to crops), so that the distribution ofthe impact shifts in a way that is disproportionate to physical changes(see Section 4.2). A comprehensive assessment of projected impacts ofclimate changes would consider how changes in atmospheric conditions(temperature, precipitation) translate to impacts on physical (e.g.,droughts and floods, erosion of beaches and slopes, sea level rise),ecological (e.g., forest fires), and human systems (e.g., casualties,infrastructure damages). For example, an extreme event with a largespatial scale (as in an ice storm or windstorm) can have an exaggerated,disruptive impact due to the systemic societal dependence on electricitytransmission and distribution networks (Peters et al., 2006). Links betweenclimate events and physical impacts are addressed in Section 3.5, whilelinks to ecosystems and human systems impacts are addressed in 4.3.Disaster signifies extreme impacts suffered by society, which may alsobe associated with extreme impacts on the physical environment andon ecosystems. Building on the definition set out in Section 1.1.2.1,extreme impacts resulting from weather, climate, or hydrological eventscan become disasters once they surpass thresholds in at least one ofthree dimensions: spatial – so that damages cannot be easily restoredfrom neighboring capacity; temporal – so that recovery becomesfrustrated by further damages; and intensity of impact on the affectedpopulation – thereby undermining, although not necessarily eliminating,the capacity of the society or community to repair itself (Alexander,1993). However, for the purposes of tabulating occurrences, someagencies only list ‘disasters’ when they exceed certain numbers of killedor injured or total repair costs (Below et al., 2009; CRED, 2010).1.2.3.2. Complex Nature of an Extreme ‘Event’In considering the range of weather and climate extremes, along withtheir impacts, the term ‘event’ as used in the literature does notadequately capture the compounding of outcomes from successivephysical phenomena, for example, a procession of serial storms trackingacross the same region (as in January and February 1990 and December1999 across Western Europe, Ulbrich et al., 2001). In focusing on the socialcontext of disasters, Quarantelli (1986) proposed the use of the notion of‘disaster occurrences or occasions’ in place of ‘events’ due to the abruptand circumstantial nature of the connotation commonly attributed tothe word ‘event,’ which belies the complexity and temporality of disaster,in particular because social context may precondition and extend theduration over which impacts are felt.Sometimes locations affected by extremes within the ‘same’ large-scalestable atmospheric circulation can be far apart, as for example theRussian heat wave and Indus valley floods in Pakistan in the summer of2010 (Lau and Kim, 2011). Extreme events can also be interrelatedthrough the atmospheric teleconnections that characterize the principaldrivers of oceanic equatorial sea surface temperatures and winds in theEl Niño–Southern Oscillation. The relationship between modes of climatevariability and extremes is discussed in greater detail in Section 3.1.1.The aftermath of one extreme event may precondition the physicalimpact of successor events. High groundwater levels and river flows canpersist for months, increasing the probability of a later storm causingflooding, as on the Rhine in 1995 (Fink et al., 1996). A thickness reductionin Arctic sea ice preconditions more extreme reductions in the summerice extent (Holland et al., 2006). A variety of feedbacks and otherinteractions connect extreme events and physical system and ecologicalresponses in a way that may amplify physical impacts (Sections 3.1.4and 4.3.5). For example, reductions in soil moisture can intensify heatwaves (Seneviratne et al., 2006), while droughts following rainy seasonsturn vegetation into fuel that can be consumed in wildfires (Westerlingand Swetman, 2003), which in turn promote soil runoff and landslideswhen the rains return (Cannon et al., 2001). However, extremes can alsointeract to reduce disaster risk. The wind-driven waves in a hurricanebring colder waters to the surface from beneath the thermocline; for thenext month, any cyclone whose path follows too closely will have areduced potential maximum intensity (Emanuel, 2001). Intense rainfallaccompanying monsoons and hurricanes also brings great benefits tosociety and ecosystems; on many occasions it helps to fill reservoirs,sustain seasonal agriculture, and alleviate summer dry conditions in aridzones (e.g., Cavazos et al., 2008).1.2.3.3. Metrics to Quantify Social Impactsand the Management of ExtremesMetrics to quantify social and economic impacts (thus used to defineextreme impacts) may include, among others (Below et al., 2009):• Human casualties and injuries• Number of permanently or temporarily displaced people• Number of directly and indirectly affected persons• Impacts on properties, measured in terms of numbers of buildingsdamaged or destroyed• Impacts on infrastructure and lifelines• Impacts on ecosystem services• Impacts on crops and agricultural systems• Impacts on disease vectors• Impacts on psychological well being and sense of security• Financial or economic loss (including insurance loss)• Impacts on coping capacity and need for external assistance.All of these may be calibrated according to the magnitude, rate, duration,and degree of irreversibility of the effects (Schneider et al., 2007).These metrics may be quantified and implemented in the context ofprobabilistic risk analysis in order to inform policies in a variety ofcontexts (see Box 1-2).42