Page - 133 - in Loss and Damage from Climate Change - Concepts, Methods and Policy Options

5 Attribution:HowIs ItRelevant forLoss… 133 canbe tracedback to anthropogenic emissions.Given the challenges of attribution toanthropogenicemissions forcertainvariables,notablyprecipitationchanges, this flexibleapproachallows forevidence tobegatheredevenwhere the signal-to-noise ratio fromanthropogenicactivity is so far small. The basic premise of impacts attribution research is consistent with the atmo- spheric research(Sect.5.3.1).Onceachangeinacertainvariablehasbeendetected, potential drivers of that change are compared: the influence of regional or local climatechange iscomparedwithotherconfoundingvariables suchaspollutionand landusechange,andsometimestechnologicalinnovation,orsocialanddemographic changes.Theprecisemethodologiesvarybetweendisciplines(Stoneetal.2013),but for a causal relationship tobeestablished it is essential tounderstand theprocesses bywhichclimatechangecontributedtotheobservedimpact,whichisoftenexplored usingecological, hydrological, agricultural, or epidemiologicalmodels. Over thepast coupleofdecades, evidenceabout theobserved impactsofclimate changehas grown substantially (Hansen2015). In the IPCCreport of 2001, strong evidencewas restricted to thecryosphereand terrestrial ecosystems innorthern lat- itudes ormountainous regions (Gitay et al. 2001;Arnell et al. 2001). In theAR5, impacts of recent climate change were observed on all continents and across all oceans.There ishighconfidencethatworldwideglacial retreat,permafrostwarming and thawing, andmass bleachingof coral reefs canbemainly attributed to climate change. There is evidence that the livelihoods of indigenous people in the Arctic havebeenalteredbyclimate change, andemergingevidence for indigenouspeople inother regions (Crameret al. 2014). Asmightbeexpected,understandingcausal relationships isverychallengingfor human systems, and there is often a strong role for social and economic factors, making itdifficult to isolate the roleofclimatechange(Crameretal.2014).Hansen andCramer (2015) also highlight that the availability of evidence variesmarkedly between regions. Often there is less evidence available about impacts in regions considered to be most vulnerable to climate change: suggesting that the lack of evidencedoesnot indicate thatclimatechangeimpactshavenotoccurred,but rather than thereare fewerstudiesavailable.Forexample,between2000and2010,10,544 scientific studieswere published about climate change impacts inEurope, and just 1987aboutSouthAmerica (ibid). Increasingly, there are efforts to analyse whether impacts attributed to climate changecanalsobeattributedtoanthropogenicemissions,aswellastoextendextreme eventattributionstudiesofweathertoalsoinvestigateimpacts(i.e.linkingSects.5.3.1 and5.3.2).Attributiontoanthropogenicemissionshasbeendemonstratedforglobal scale studies of shrinking glaciers (Marzeion et al. 2014), ecological studies at a global aggregate level of ameta-analysis (Rosenzweiget al. 2008), changingwater runoff, forexampleinthewesternUnitedStates(Barnettetal.2008),andchangesin ecosystemproductivity (Sippel et al. 2018).HansenandStone (2016) analysed the roleofanthropogenicemissionsacrossallof theimpactsassessedintheIPCCWGII report(Crameretal.2014),andfoundthatapproximately65%oftheimpactsrelated to changes in atmospheric or ocean temperature could be confidently attributed to anthropogenicforcing(Fig.5.7).Thestrongestevidenceexistsforshrinkingglaciers,