Climate Measurement Standards Initiative Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance TECHNICAL SUMMARY Technical summary The Task Force on Climate-Related Financial Disclosures (TCFD) has developed a set of recommendations that companies can use to disclose information about climate risks. The disclosures include physical and transition risks. To determine the physical risks in a credible and comparable way, Australian companies need reliable information for scenario analyses. The Climate Measurement Standards Initiative (CMSI) is an industry-led collaboration between Australian insurers, banks, scientists, reporting standards professionals, service providers and supporting parties to assist with TCFD reporting. The CMSI is initially developing technical, business and scientific standards for climate-related physical risk to buildings and infrastructure assets. This is a technical summary of the report ‘Scenario analysis of climate-related physical risk for buildings and infrastructure: climate science guidance’. The report, which was independently authored by the Earth Systems and Climate Change Hub of the National Environmental Science Program (NESP), describes methods, scenarios for acute and chronic climate hazards, gaps and needs, and a roadmap for the future. 1 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Methods and scenarios There are 3 factors that affect uncertainty in future climate scenarios: La Niña events representing the extreme phases, typically occurring every 1 Ongoing natural climate variability 4-7 years. The IOD is characterised by changes in sea surface temperatures in the eastern Indian Ocean basin and opposite changes in the west of the 2 Globalsocio-economic development and resulting emissions of basin. SAM is the alternation of atmospheric pressure between the middle greenhouse gases and aerosols and high latitudes. 3 Regional climate responses to these emissions. Emissions of greenhouse gases Researchers use multiple lines of evidence to assess observed and Using the lowest and highest international-standard scenarios for projected climate change. Plausible ranges of change and associated atmospheric greenhouse gas concentrations, known as representative confidence ratings are consistent with guidance recommended by the concentration pathways (RCPs), provides the range of future climate Intergovernmental Panel on Climate Change (IPCC). possibilities, and meets international TCFD guidelines. Consultation with CMSI participants led to identification of a set of RCP2.6 aligns with low emissions and has a 2-in-3 chance of staying relevant emission scenarios, climate variables and timeframes. below 2°C global warming by the end of the century relative to the The report outlines two emission scenarios leading to global warming preindustrial period (1850-1900). RCP8.5 aligns with high emissions and above or below 2°C, consistent with the TCFD recommendations, along projects 3-5°C (or more) global warming by the end of the century. with the following time frames: the present; and 10, 30 and 70 years in The assumptions underlying each scenario need ongoing assessment the future, expressed as the years 2020, 2030, 2050 and 2090. over time. For example, a transition away from fossil fuels may make The ranges of projected changes for climate variables, including some of the assumptions under RCP8.5 unlikely, or sustained high changes to extreme weather, from all adequately performing climate emissions may make RCP2.6 impossible, at least in the medium term. model simulations should be considered. Expert judgment can determine whether changes outside the simulated range should also Regional climate responses be considered. The range of possibilities for a given global warming Regional climate responses to RCP2.6 and RCP8.5 are based on the scenario should be limited only if the model or the physical change it Coupled Model Inter-comparison Project phase 5 (CMIP5) set of represents is implausible. global climate model results used by the IPCC and reported in www. climatechangeinaustralia.gov.au. Researchers have supplemented this Ongoing natural variability work with regional climate modelling done for some state government Climate variability on timescales from minutes to decades affects extreme projects, and applied new insights from the emerging CMIP6 modelling. events such as floods, drought, heatwaves and fires. Large-scale drivers of This work has quantified the range of modelled responses to the natural climate variability include the El Niño – Southern Oscillation (ENSO), emissions scenarios, providing an estimate of the plausible range of Indian Ocean Dipole (IOD) and Southern Annular Mode (SAM). ENSO is a climate change. The range of model results includes uncertainty due to coupled ocean-atmosphere process in the Pacific Ocean, with El Niño and imperfect modelling, and efforts to represent deeper uncertainties. 2 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Table 1: Attributes of low and high greenhouse gas emission scenarios There may be important factors that no models incorporate. The full range of adequately performing models is not a statistical sample Sector Carbon dioxide Likely global Global warming relative Transition risk Physical risk of uncertainty, but rather researchers’ best estimate of the range of emissions SSP* to 1850-1900** plausible change. Low case Net zero by Sustainability 1.3–2.2°C by 2050 Higher Lower The climate science report presents projected changes for two around 2070 (SSP1) challenges challenges 0.9–2.4°C by 2090 categories of climate hazards that can damage buildings and (RCP2.6) infrastructure: High case High and Fossil fueled 1.8–3.0°C by 2050 Lower Higher 1 Acute—extreme weather events identified by the CMSI science accelerating development challenges challenges (RCP8.5) (SSP5) 3.2–5.4°C by 2090 committee associated with building and infrastructure damage. This includes changes in natural hazards such as tropical cyclones, east coast lows, extreme rainfall, hail, extreme sea level events and fire weather, presented in Table 2 with confidence ratings for RCP2.6 and RCP8.5 for 2030, 2050 and 2090. 2 Chronic—gradually emerging aspects of climate risk, including changes to annual average temperature, rainfall and sea level, time in drought and days over 35°C. Table 3 presents projected changes and confidence ratings for RCP2.6 and RCP8.5 for 2030, 2050 and 2090. * There is an emerging system of socio-economic pathways (SSPs) that can be related to different RCPs. ** CMIP5  models reported in IPCC (2013), but new generation climate modelling suggests that even greater warming late in the century than cited here can’t be ruled out (see Grose et al. 2020). 3 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Projected changes Acute and chronic climate hazards Tables 2 and 3 present projected changes in acute and chronic climate Projections for RCP2.6 and RCP8.5 are similar for 2030, but generally hazards, respectively, based on information that is limited in some cases. diverge by 2050 and beyond. The ‘central estimate’ is based on different For each climate variable, the tables list the average in the IPCC baseline lines of evidence and expert judgment, with values in square brackets period (1986-2005), and the observed change over recent decades. indicating the 80 per cent likelihood range, and a confidence rating. Where possible, the level of attribution to human influence is included Qualitative projections are provided where there is low confidence. (none, weak or strong). Projections are relative to 1986-2005, based on The tables present broad estimates for all of Australia or for large 20-year periods centred on each of 2030 (2020-2039), 2050 (2040-2059) regions. Spatial variation occurs for all projections, with models revealing and 2090 (2080-2099). possible values above or below those presented. Confidence ratings, based on IPCC guidance, provide information about whether the range of change is a reliable and complete description. Table 2: Observed and projected changes in acute climate hazards that present physical risks for Australian buildings and infrastructure. Green boxes show projections for RCP2.6, pink boxes for RCP8.5. Projections have a central estimate and a range of plausible change (based on 10th-90th percentile estimates considering multiple lines of evidence). Continued overleaf. Extreme or hazard Average in 1986-2005 Observed change 2030 2050 2090 Confidence (and attribution) Tropical cyclone (TC) frequency 10-11 per year -10% (weak) East -4% [-8% to 1%]; East -4% [-8% to 1%]; East -4% [-8% to 1%]; Medium in Australian region West -6% [-10% to -2%] West -6% [-10% to -2%] West -6% [-10% to -2%] East -8% [-15% to 2%]; East -15% [-25% to 5%]; West -12% [-20% to -4%] West -20% [-30% to -10%] Category 4-5 TC frequency 2-3 per year Little change (noting large Little change or small Little change or small Little change or small Low-Medium (relevant for damaging winds) variability) (none) increase increase increase (for examples of numbers published in previous studies, see Section 3.2 of the Science Little change or increase Little change or increase Report) 4 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Extreme or hazard Average in 1986-2005 Observed change 2030 2050 2090 Confidence (and attribution) TC location (latitude) 10-20°S common (30°S less Little change or small Little change or small Little change or small Little change or small Low with changes noted for common) poleward expansion (none) poleward expansion poleward expansion poleward expansion (for examples of numbers southern extent published in previous studies, see Section 3.2 of the Science Report) Little change or poleward Little change or poleward expansion expansion East coast low (ECL) frequency 20 per year, with 2-3 -10% (but with large -10% [-15% to -5%] -10% [-15% to -5%] -10% [-15% to -5%] Medium (Low for intense ECLs per year variability) summer and High for impacting on land (weak) -20% [-30% to -10%] -35% [-50% to -20%] winter) Extreme rainfall intensity Spatially variable intensity +10% hourly and +7% daily +10% [5% to 15%] hourly; +10% [5% to 15%] hourly; +10% [5% to 15%] hourly; High for direction of (considering 20-year return (but with large variability) +7% [4% to 10%] daily +7% [4% to 10%] daily +7% [4% to 10%] daily change and Medium for period) (weak) magnitude of change +20% [10% to 30%] hourly; +35% [15% to 55%] hourly; +15% [8% to 20%] daily +25% [15% to 35%] daily Extreme sea-level events Spatially variable Mainly driven by mean 1-in-100-year event becomes an annual event by the end of the century under RCP 2.6 High sea-level rise: 3 mm/year and by mid-century under RCP 8.5 (strong) Floods Spatially variable and No clear signal Increase more likely than a decrease for most types of floods; increases very likely for Low for large dependent on flood type coastal flooding (based on the rate of sea-level rise) and for small-scale flash floods catchments and large (based on extreme rainfall increases). floods in general (including river and surface water); High for coastal and flash floods Large hail (>2.5 cm diameter) About 5-10 per year in No information Little change, but potential As for 2030 As for 2030 Low frequency in city-scale regions eastern regions and 0-5 increase in east and per year elsewhere poleward shift in features As for 2030 As for 2030 Extreme fire weather days About 18 days per year to +15% (medium-high) +20% [+5% to +35%]; +20% [+5% to +35%]; +20% [+5% to +35%]; High; Medium in east. (exceeding 95th percentile) once every few years East +15% [+0% to +30%] East +15% [+0% to +30%] East +15% [+0% to +30%] Low confidence for lightning ignition and +40% [+10% to +70%]; +75% [+20% to +130%]; fuel load (key risk factors East +30% [+0% to +60%] East +55% [+0% to +110%] particularly in north and central Australia) 5 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Table 3: Observed and projected changes in chronic climate hazards that present physical risks for Australian buildings and infrastructure. Table details as for Table 1. Climate variable Observed change (and attribution) 2030 2050 2090 Confidence Annual average Around +1.4°C since 1910 (strong) +0.6 to 1.4°C +0.5 to 1.5°C +0.5 to 1.5°C Very high temperature +1.5 to 2.5°C +2.5 to 5.0°C Average sea level Increased by 3.1 mm/year during 1993–2009 (strong) +0.07 to 0.2 m +0.1 to 0.3 m +0.2 to 0.6 m Very high +0.1 to 0.3 m +0.4 to 1 m Average Decreased 11% in the southeast during April to October for 1999–2018 relative East: -13 to +5% Drier in the south and east, uncertain in the north High in southern annual rainfall to 1900–98, and decreased 20% in the southwest during May to July since North: -9 to +4% Australia, 1970 relative to 1900–69 (strong), with an increase of 10 mm/decade from South: -9 to +2% Drier in the south and east, uncertain in the north Low elsewhere 1900–2019 in the north (weak) Rangelands: -10 to +6% Time in drought* Insignificant (weak) Increase in many regions No data High in southern Australia, Significant increase in many regions Low elsewhere Annual days >35°C# Increase (strong) Increases Increases High Large increases * Meteorological drought (rainfall deficits) and soil moisture drought, not accounting for changes to other factors that are included in agricultural, socio-economic and other drought measures There are projections for other threshold temperatures at www.climatechangeinaustralia.gov.au/en/climate-projections/explore-data/threshold-calculator/ #  6 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Projected changes Warming over Australia is projected to be broadly similar to the global Internally consistent scenarios for multiple climate hazards average warming over land, more than the global average warming When considering projections for multiple climate hazards, including oceans, and less than the rapid warming in the Arctic and the randomly combining the ranges of change for different hazards is large inland regions of North America and Eurasia. not recommended because some combinations may be physically Rainfall is likely to continue decreasing in the south, especially in implausible. Internally consistent scenarios should be: southwest Australia and during the southern wet season of May- • Physically plausible—the correlations between the variability and October, similar to other areas in the mid-latitudes such as southwest change in climate variables should be maintained. To achieve this, Africa and southwest South America and the Mediterranean. Rainfall CSIRO’s climate futures framework can be used, supplemented by projections for northern Australia are uncertain, due to a wide range of other evidence and assessment. plausible future changes to the monsoon. Regional modelling produces more detail in projections of rainfall around mountain ranges, and for • Representative—they should sample the projected range in general Tasmania and the eastern seaboard. climate represented by average temperature and rainfall change (e.g. hotter and drier vs warmer and wetter), and the hazards and impacts Regardless of the change in average rainfall, climate change is likely of interest for the specific application. to increase rainfall variability, including an increase in the incidence of short-duration extreme rainfall. This change will affect the reliability of Researchers have identified 4 internally consistent scenarios for a water supply and lead to an increase in flash flooding. 20-year period centred on 2050 (2040–2059). (Scenarios for other timeframes can be produced if required.) Projected changes can help inform the development of scenarios of climate extremes for ‘stress testing’ industry sectors. (The terms ‘stress test’ and ‘scenario’ have different meanings to different stakeholders, so must be carefully defined for any application.) Insights from these projections, supported by other evidence, non- climate knowledge and expert judgement, can provide tests of climate hazard impacts. Specifically, for extreme events that can damage buildings and infrastructure, scenario analyses should include quantitative increases in extreme wind, rainfall, fire weather, large hail and coastal inundation. Scenario analyses can use research on compound extreme events as information emerges. 7 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Gaps and needs There are several practical limitations to consistent analysis of climate • examination of compound events (occurring at the same time or in risk for disclosures using the data described here. Some datasets are close succession) and cascading events where one hazard or impact not publicly available, some are difficult to use, and some are not for triggers another (e.g. a bushfire triggers water quality and security commercial use. issues in the months and years following the fire) Projections for some hazards have wide ranges of plausible changes • assessment of future changes in extreme weather and climate hazards and low confidence levels. The scientific needs include: • nationally consistent high-resolution projections (rather than the inconsistent high-resolution projections for regional domains currently • improved understanding of physical processes associated with available), so that regional insights from high-resolution modelling climatic hazards, especially bushfires, thunderstorms, floods, hail, are comparable across the country tropical cyclones, east coast lows, storm surges and drought • use of up-to-date climate models—projections overwhelmingly use • evaluation of present-day climate model hazard simulations with a the CMIP3/CMIP5 generation of international models, which are 7-15 focus on extreme hazards that cause damage years old, but CMIP6 models are now emerging. 8 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Vision for climate science and projections Improvements to weather and climate data, information and services • extreme fire: frequency, duration and location including fire weather, are needed to support the goal of a resilient financial services sector that fuel and ignition embeds climate change understanding into rigorous institutional level • extreme heat: frequency, duration and location of days over 35, assessment and disclosure of risk. 40 and 45°C (or other sector-relevant thresholds) For climate science and projections, the ideal is a comprehensive, • drought: frequency, duration and location. fully evaluated set of high-resolution climate projections that can depict all the relevant hazards and phenomena of interest (including The vision for climate science and projections includes achievements in their extremes) with high confidence and narrow ranges of spread a range of areas, including improvements in observations (and derived in projected change, with a full description of natural variability. high-quality gridded data sets such as for wind speed), advances in Information would be available at spatial and temporal scales relevant climate process research, corresponding advances in climate models to assessing climate risks, extending from the present for 100 years. and projections, delivery of user-focussed climate services, strong Ideally, detailed data will be available for parameters including: international engagement in science and adaptation, and enhanced national coordination and funding of climate research and services.* • tropical cyclones: frequency, intensity, duration, location, peak wind speeds, rainfall intensity, latitude of maximum intensity, area of gale- force winds, storm surge intensity and frequency • extreme east coast lows: frequency, intensity, duration, location, peak wind speeds, rainfall intensity, latitude of maximum intensity, area of gale-force winds, storm surge intensity and frequency • sea-level rise: average and extreme events • extreme rainfall: frequency, intensity, duration, location, annual maximum 1-day total, 20-year maximum 1-day total, area exceeding X mm over 1 day, annual maximum 3-day total, 20-year maximum 3-day total, area exceeding Y mm over 3 days • flood: intensity, frequency, duration and area • extreme hail: frequency, duration and location of hail over 2.5 cm diameter See  *  NCSAC (2019) Climate Science for Australia’s Future. 9 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Technical summary Roadmap for achieving the vision Achieving the CMSI vision requires investment in short-term actions (next 1-3 years) and longer-term actions (next 10 years). These actions should be consistent with those in Climate Science for Australia’s Future (NCSAC 2019), which has a broader remit. Short-term actions • Collect feedback on CMSI guidelines and update them, including the use of the most recent climate observations and projections. • Compile and quality-check observations of climate hazards, including their extremes. • Evaluate present-day simulations of extreme climate hazards. • Analyse future changes in extreme climate hazards. • Analyse recent compound events. • Perform CORDEX-CMIP6 regional climate projections for Australia. • Implement climate services supporting uptake of hazard projections in impact assessment, planning and action. Longer-term actions • Improve observation systems for extreme climatic hazards. • Improve understanding and modelling of extreme climatic hazards. • Assess likely future changes in compound events. 10 Scenario analysis of climate-related physical risk for buildings and infrastructure: Climate science guidance Copyright © Climate-KIC Australia. Climate-KIC Australia is a registered trademark of Climate-KIC Australia Ltd (ABN 95616047744). The information contained herein is subject to change without notice. Climate-KIC Australia shall not be liable for technical or editorial errors or omissions contained herein. All other rights are reserved. Climate-KIC Australia as the convenor of the Climate Measurement Standards Initiative advises that the information contained in this publication comprises general statements. Readers should note that the report is not intended to provide legal advice, accounting or auditing advice, or express an opinion of any kind on the disclosures that companies may need to make to comply with their obligations under the Corporations Act 2001 and other applicable corporations, prudential and securities regulations, nor to comply with applicable accounting standards. In all cases, users of the Guidelines should form an independent view regarding the information and variables that should be considered and disclosed in order to provide a true and fair view of their company’s financial performance, position and prospects. The CMSI is being coordinated by Climate-KIC Australia, and a group of industry partners, these include: QBE, Suncorp, IAG, RACQ, NAB, Westpac, Commonwealth Bank, HSBC Australia, Munich Re, Swiss Re, Leadenhall CP, MinterEllison and Investor Group on Climate Change. The National Environmental Science Program Earth Systems and Climate Change Hub developed the Scientific Guidance. Technical support has been provided by the Institute for Sustainable Futures, UTS.