气候预测：超越净零死亡率预测

1February 2025 Climate forecast: Looking beyond2net-zero mortality predictionsFebruary 2025At the sametime, the green transitioncould haveits own set ofinfluences,possibly with positive outcomes:healthier diets,walking-and cycling-friendly cities; reduced air pollutionin citiesfrom lower emissions;3andmoreresilient infrastructure.Albeitsometransitions may be negative—for example,potentiallyworse diets due tothegreater difficulty and expenseofsourcingnutritious foodandagreater reliance on ultraprocessed foods.Adding to the complexity, all theseaspectswill interconnect withprevailingmacroeconomicand social factors.The National Health Service (NHS) is facingsevere capacity pressures as a result of record demand,post-COVID-19backlog,an ageing population and inflation.4Some of theseissuescould well be exacerbatedbyclimate change, namelyinflation anddemand for healthcare. At the same timeas a changing climate, the UKpopulation will continue ageingwith a projected22 millionpeopleaged 65 or over in the UK by2072;up fromaround 13million people aged65 or over in 2022.5Thenation’s general economicstatecouldalso impact theavailabilityof healthcare andgeneral health.Thishas the potential tocreate a feedback loop as poor health reducesproductivity, addingfurther economic strainto a systemthatcould simultaneously bestruggling withshocksfromclimate changeand potentially fundamental changes in the job market due toartificial intelligence (AI).Understanding how these impactsinterconnectand how theycouldaffect any specific country is a tricky task.Moreover,analysis onthe impact of climate onmortalityhas tended to focusmoreonsingle drivers, e.g.,heat,and direct impacts fromclimate,often stopping before getting to second-order impactsand interactions withmacroeconomicand socialfactors.At Milliman we haveresearchedthedevelopment ofa causal modeltolookmoreholistically atthe impactofa changingclimate on mortality.We alsoconsideredsecond-order impactsand interactions.Our research has resulted in a protype model that has allowed us toexaminethe degree towhich conclusions of modest impactson mortalitywould be robust and areas to investigate that could lead todifferent outcomes.Exploring howclimate risk drivers affect mortalityCausal modelling techniques are particularlywellsuited toexamininginterconnections and second-orderimpacts.A causal modelaims tocapture how a system operates byplacing the modelling emphasis on causallinks between different factors,assessinghow they interact and interrelate.In this, a causalmodel allows us toexplore offsetting and amplifying factors.The model we have developedis a prototypefor attempting to get to grips with the complexity ofclimate-relatedmortality drivers and second-order impacts.Itwas developed based ondesktopresearchto define the structureof the modeland determine the key climate-relatedmortalityinfluences to include within the model.Thereare ofcoursemultipleotherfactorsand influencesthat couldhave been included, but our aim for this prototype modelwas to focus on a few key variables. Future iterations can beextended to include additional drivers andinteractions.Calibration wasperformedthrough a combination of desktop research and expert judgementfor theUK market.This resulted inthe causalmodelshown in Figure 1.3.A further example oftheinterconnectedness of risksis seen instudies showingthat reducing air pollution (intheform ofaerosols) may amplifyglobal warming.See,e.g.,Hansen et al. (2023),Global Warming in the Pipeline,available athttps://academic.oup.com/oocc/article/3/1/kgad008/7335889.4.Powell, T. et al. (16 July 2024). Capacity pressures in health and social care in England. UK Parliament. Retrieved 2 February 2025 fromhttps://commonslibrary.parliament.uk/capacity-pressures-in-health-and-social-care-in-england/.5.Barton, C. et al. (16 July 2024). The UK’s changing population. UK Parliament. Retrieved 2 February 2025 fromhttps://commonslibrary.parliament.uk/the-uks-changing-population/. Climate forecast: Looking beyond4net-zero mortality predictionsFebruary 2025The structure of our model was deliberately specified to be able to exploreclimate pathwayswithfutureincreasesinglobal average temperatures ranging from +1°C to +3°Crelative tothe1986-2006period.The scenarioprojectionsunderlying thesetemperaturechangesbroadly align toa Network for Greening the Financial System(NGFS)net-zeroscenariothrough toits”hothouse”worldscenarios.These global averagetemperaturesweretranslated into plausible corresponding UK averagetemperaturesand extreme weather eventsusinginsightsfromtheClimate Analytics climate impact explorer.6It is critical to model local impacts relevant to mortalitygroupsbecauselocalclimatevariations, andparticularlyimpacts on extremetemperature,differconsiderably.TheUKclimatewill be influenced by global emissions.However,there are othernon-weatherfactorsthat couldimpact mortality. For example,air pollution,which willdepend on both temperaturesand local transition awayfrom burning f