Severe Weather in a
Changing Climate
nd 2 Edition
Cindy Bruyère1, Bruce Buckley2, Andreas Prein1, Greg Holland1, Mark Leplastrier2, David Henderson2,3, Peter Chan2, James Done1, Andrew Dyer2
1 Capacity Center for Climate and Weather Extremes, National Center for Atmospheric Research, USA 2 Insurance Australia Group 3 Cyclone Testing Station at James Cook University
September 2020
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION
FOREWORD
When initially released in November 2019, this report proposed that climate change is already well underway and is considered by many to be the greatest risk currently facing humanity. Despite the economic and social upheavals wrought by the current Coronavirus-19 pandemic, this remains the case. Every year we are confronted globally with extreme weather events that become natural disasters. Our communities in Australia are exposed to just about every possible hazard, from earthquakes, severe thunderstorms and tropical cyclones, to bushfires and devastating floods. We cannot prevent these events from happening, but we can do more to prepare communities and make them more resilient. Reducing global greenhouse gas emissions will slow future changes in weather extremes and reduce the rate at which sea levels rise. Protecting communities also requires greater investment in resilience, adaptation, and mitigation planning – from governments, businesses, community organisations and individuals – to reduce the physical, economic and social recovery costs that follow a disaster. This report provides an overview of the science behind severe weather phenomena as they affect Australia. In the ten months since we published the original report, there have been numerous examples of catastrophic extreme weather events including Australia’s most destructive bushfires, major hailstorms affecting Melbourne, Canberra and Sydney, and widespread extreme rainfall and flooding down Australia’s eastern seaboard. This edition of the report is an update of the original report released in November 2019. It reflects extensive feedback from academic and government institutions across Australia, supplemented by research that has emerged since the first report was prepared. Most of the changes relate to Section 5, which deals with changes to various extremes under different warming scenarios. The report also introduces a new section on Connected Extremes. The main findings from the original report still hold, and the Executive Summary highlights the key points to consider. The report now provides a summary of the latest climate science as of July 2020, explaining how climate change is impacting the severity and frequency of extreme weather events like tropical cyclones, hailstorms and rainfall in Australia, and what is likely to happen in the future. Climate change requires broad-scale collaboration and coordination across all sectors of the community and the implementation of global initiatives without delay. Climate change is too big for one organisation and even one nation to solve, and it is our hope this updated report will continue to provide a foundation to drive even more conversation and collaboration so that the necessary changes can happen.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION
CONTENTS
1. Executive Summary ...... 2
2. Acknowledgements ...... 7
3. Introduction ...... 8
4. State of the Climate ...... 9
4.1 Natural Climate Variability and Forced Climate Change ...... 9
4.2 State of Climate Assessment ...... 9
4.3 Extreme Events Under Climate Change ...... 13
5. Changes to Extremes for Different Warming Scenarios ...... 15
5.1 Tropical Cyclones ...... 15
5.2 Extreme Precipitation and Flooding ...... 37
5.3 Damaging Hail ...... 49
5.4 East Coast Lows ...... 60
5.5 Bushfire ...... 71
5.6 Oceans: Sea Level Rise, Temperature Anomalies and Extreme Sea Levels ...... 85
5.7 Connected Extremes: Compound and Clustered Events ...... 94
6. Conclusions ...... 102
Glossary ...... 104
Bibliography ...... 111
Contacts ...... 129
This report should be cited as: Bruyère, C., Buckley, B., Prein, A., Holland, G., Leplastrier, M., Henderson, D., Chan, P., Done, J., Dyer, A. (2020). Severe weather in a changing climate, 2nd Ed. (IAG). DOI:10.5065/b64x-e729
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 1
1. EXECUTIVE SUMMARY
This report by the National Center for Atmospheric Research (NCAR) and IAG examines current and future climate change impacts on the Australian climate and the weather extremes that produce significant property, personal and economic damage and hardship. It is an updated edition of the November 2019 report of the same name. The updates reflect extensive feedback from academic and government institutions across Australia, supplemented by research that has emerged since the first report was prepared. This report reflects the understanding of climate science as of July 2020. The level of scientific knowledge and tools available now enable confident assessments on the impacts of climate change at global and national scales and over longer time frames, with an objective assessment of associated levels of confidence. But many government, business and personal decisions require information on climate and weather extremes at more local scales, such as for states, cities and towns. To meet this need, this report incorporates a review of the extensive related literature, summarised in Table ES1, together with expert judgement on potential local impacts. In some instances, the report includes tentative trends in severe weather impacts derived from insurance industry claims. The expert assessments on local impacts are indicative rather than absolute. They are intended to provide a basis for current planning, targeted research to fill knowledge gaps, and further discussion, leading to more refined and accurate future assessments.
Key assessments are: 1. Since the pre-industrial period (1850-1900), the average global mean temperature has already risen by more than 1°C due to increasing greenhouse gas emissions and deforestation. No agreement has been reached as to when global warming might reach the 1.5/2°C Paris agreement targets. There is consensus that it is extremely unlikely to remain below these targets. Estimates indicate that global warming could reach 1.5°C during the 2020 decade and 2°C by 2035. This small change in the mean will substantially increase the frequency and intensity of weather and climate extremes. The impacts will range from more warm and hot days, fewer cool and cold nights; to more droughts and extreme weather events leading to flooding.
2. The frequency of tropical cyclones (TCs) in the Australian region has declined slightly in recent decades, and this slow trend is projected to continue globally and for the Australian region. However, over the past 30 years, the proportion of the most intense TCs has increased at the expense of weaker systems, and this change is expected to continue. Over the last two decades, the number of intense TCs making landfall on the Queensland east coast has increased substantially. The frequency of TCs making landfall throughout the western South Pacific region has also increased.
There is global evidence of a poleward shift in the latitudes where TCs reach their peak intensities. New research confirmed that this shift is also evident for TCs in the Australian region. This trend is expected to continue, although in an uneven manner. Warming oceans off south-east Queensland and New South Wales will enable these cyclones to retain higher intensities further south and to penetrate further inland. TC risk is therefore predicted to increase most rapidly in the south-east Queensland / north-east New South Wales regions. There is also a potential for increased risks in the coastal districts south of Shark Bay in Western Australia.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 2 1. EXECUTIVE SUMMARY
Planning for inland penetration of TCs needs to allow for substantial increases in rainfall rate, storm total rainfall and total area affected, increasing the risk of coincident major floods in multiple river systems. Wind speeds are also likely to decay more slowly inland, so increased wind-driven rainfall ingress is to be expected. More intense storms, combined with rising sea levels, point to increasing storm surge and coastal erosion impacts.
3. Intense short duration rainfall is expected to increase almost everywhere in Australia, even in regions with ongoing drying trends, resulting in more frequent and severe flash flooding in urban areas and small, fast response river catchments. Emerging science is confirming intense rainfall rates are rising at faster rates than the Clausius-Clapeyron rainfall-temperature relationship would imply. Rainfall intensities across southern Australia have increased up to 14% per degree of increased temperature, and 21% for the tropical regions. Storm rainfall totals from both east coast lows and tropical systems are expected to increase significantly, leading to increased flood risk in the larger river catchments, particularly along the east coast. More work is required to understand and confidently assess these changes.
4. New research has supported earlier assessments that there have been increases in the frequency of large to giant hail events across south-eastern Queensland and north-eastern and eastern New South Wales in the most recent decade. Areas at greatest risk of large (2.0- 4.9cm in diameter) and giant (>=5.0cm in diameter) hail should progressively shift southwards. The most substantial increases in risk are likely to be in the region inland from the Hunter River, south through the central and southern New South Wales highlands and central to eastern Victoria. Slower increases are assessed to affect the south-west of Western Australia. At the same time, in a warmer world severe hail risk is expected to decrease in northern and central Queensland, extending to south-eastern regions later.
5. Multi-day impacts of east coast lows (ECLs) on the south-eastern seaboard of Australia are expected to increase because of wind-driven rainfall ingress, and flash and riverine flooding. This effect will be compounded by rising impacts from storm surge, waves, and coastal erosion. Tentative evidence points to an increase in impacts from the warm cored and hybrid types of ECLs, while there will be a decrease in the less damaging types of lows, notably those occurring in winter and spring. There is limited understanding of the detailed structures and frequencies of rare extreme ECLs that drive most of the widespread damaging impacts over large parts of eastern and south-eastern Australia.
6. The simplified meteorological component of bushfire risk can be measured by the McArthur Forest Fire Danger Index (FFDI). The 99th percentile or higher FFDI produce the greatest loss of life and damage to property. The upward trend in this 99th percentile FFDI is likely to increase in almost all locations nationally, leading to longer fire seasons and more frequent extreme events, many of which could be uncontrollable at the time of peak intensity. The rate of increase varies by location and will depend on weather system changes and site-specific factors, including fuel types and loads. New research investigating the recent Black Summer 2019-2020 fire season has shown that the current generation of climate models under-predicts events of this severity. Opportunities for fuel management activities are also likely to be reduced due to the earlier onset of the bushfire season. Fire-prone regions throughout the world have historically shared resources. Longer fire seasons will result in coincidences of bushfires between hemispheres, increasing the strain on limited global resources.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 3 1. EXECUTIVE SUMMARY
7. Sea level rise is expected to accelerate around the Australian coastline but at varying rates. Notably, past assessments of sea level rise are lower than current observations indicate. Sea level rise will contribute substantially to escalating impacts from storm surge, coastal erosion and the effects on coastal natural systems, buildings and infrastructure. The greenhouse gases that are already present in the atmosphere will cause sea level rises to continue for the next couple of centuries, even if there are significant global emission reductions through the coming decades. Ice sheet research is highlighting many unmodelled physical processes that have the potential for much greater sea level rises in the future that should be considered in understanding possible impacts on coastal communities.
8. Connected extremes – extreme weather events that are connected in time and space – exacerbate the impacts that would have occurred from the separate events. Connected extremes can, therefore, lead to multiple extreme impacts upon Australian communities. These impacts can either occur in close succession or over extended periods of time, making response and recovery activities more challenging. Emerging research is showing the damage from connected extreme weather events is exacerbated by climate change.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 4
Table ES1 Summary of the past and future impacts of climate change on metrics of key extreme events in Australia under three future global temperature scenarios (changes from pre-industrial period 1850-1900). Present climate represents the observed state for the last two decades. This is an expert assessment and includes an estimate of the confidence in the changes from the benchmark values. Climate Change Impact (confidence level) Metric Benchmark
Present Climate +1.5°C +2°C 3°C+
5-10% higher for each ~5% increase <10% increase 10%-20% increase
Peak wind speeds 2007 1°C
- (High) (Med-High) (Med-High) Variable
between (Med-High) 1973
Latitude of lifetime Possible further poleward
Poleward shift 1-6km/year Further poleward shift Further poleward shift maximum intensity shift
1989 (High) (Medium) (Medium)
(Low)
Tropical cyclone rapid Slight increase Moderate increase Moderate/high increase High increase
1990 -
intensification (Medium/High) (Medium) (Medium) (Medium/Low)
1988 ~100% Small increase Further small increase
Proportion of Australian Minimal further increase between 1975-2010 from 2010-2015 from 2010-2015 CAT 4 and 5
1975 (Low) (High) (Medium) (Low)
Intense precipitation (> ~+60% ~10% further increase Further increase ~20% further increase 600mm) within 500km of Tropical Cyclones Tropical (Medium)
TC centre (Medium) (Medium) (Medium) 1960s
Small decrease ~15% decrease Further decrease ~30% decrease for +3°C Frequency
(Medium) (Medium) (Medium) (Medium)
1960s
~50% ~100% Further increase
Area of gale force winds 2007 No info -
(Low) (Low) (Low) 1973 Potential substantial Increase Further increase Further increase Storm surge increase frequency (High) (High) (Very High)
(Very High) Since1900
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 5
Climate Change Impact (confidence level) Metric Benchmark
Present Climate +1.5°C +2°C 3°C+
0 ~22cm - close to global ~30cm* ~40cm* ~80cm but ongoing rises*
average
190 -
Sea Sea Sea level rise Rise
Level Level (High) (High) (Medium)
(Very High)
1850
Increased damaging ECL, Increased damaging ECL, Increasing frequency of Increasing frequency of Frequency of damaging decrease in other ECL decrease in other ECL intense ECL impacts intense ECL impacts
ECLs includes large natural includes large natural
East East Lows
Coast Coast (Low) (Low)
variability (Medium) variability (Medium) Since1860
Regionally variable. Further Regionally variable. Further Regionally variable. Further Annual maximum >14% increase; north of 40% or more increase
2005 10% or more increase 15% or more increase - o 1-hour rainfall intensity 23 S >21% increase. for +4°C (Med-High) (Medium)
(Medium)
1986
Regionally variable Potentially 40% increase Rainfall Annual maximum Regionally variable Regionally variable ~10%
2005 13-15% for +4°C 1-day rainfall intensity - generally slightly upward (Med-High)
(Med-High) (Medium) 1986 Variable, generally slightly Variable, generally slightly Extreme Extreme Between 15-20% dependent Between 10-60% dependent 20-year return level of 1- upward upward 2005 on the region on the region day rainfall -
(Med-High) (Med-High) (Med-High) (Medium) 1986 Increasing trend in south- Increasing trend in south-
Marked increasing trend in Potential increase in New
east Australia, decrease east Australia, decrease Frequency of hail >=2 cm east & south-east Australia South Wales & southern central, north Queensland Queensland
Hail diameter regions, decreases Large ~2000 (Med-High) (Med) (Med) elsewhere (Low)
Increases >30% in southern
and eastern Australia (High)
Increasing in almost all 15-65% increase in number Further increases in other McArthur Forest Fire Australian regions of extreme fire danger days Further increases typically regions (Medium)
Danger Index 2010 - especially in the south-east (FFDI>50) for +1°C >10% (Medium) (FFDI)
(High) 100-300% increase in Bushfire 1973 (Medium) number of extreme fire danger days (FFDI>50) for +3°C (Medium)
* Estimates reflect Intergovernmental Panel on Climate Change Fifth Assessment Report findings which could quite possibly be low-end estimates and therefore underestimate the impact.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 6
2. ACKNOWLEDGEMENTS
The revision of this report has benefitted greatly from feedback from many scientists across Australia and in the USA. Special thanks go to Hamish Clarke (University of Wollongong); Andrew Dowdy, Joshua Soderholm, Andrew Brown and Alain Protat (Bureau of Meteorology); Ian Macadam on behalf of many scientists within UNSW and CLEX1; Hamish Ramsay (CSIRO); Robert E Kopp (Rutgers University); and Janice Coen (NCAR) who provided extensive feedback and references and peer reviewed sections of this report, important for ensuring the completeness of this update. Feedback on various parts of the report has also been gratefully received from Acacia Pepler and Scott Power (Bureau of Meteorology); Kathleen McInnes, Michael Grose, Kevin Hennessy, Geoff Gooley and David Karoly (CSIRO); Owen Price (University of Wollongong); Todd Lane, Andrew King and Kevin Walsh (University of Melbourne); Andy Pitman, Jason Evans, Lisa Alexander, Nina Ridder and others at UNSW and CLEX; Robert Warren, Christian Jakob and Michael Reeder (Monash University); and Merv Lynch and Diandong Ren (Curtin University). In addition to direct feedback on various aspects of the report, numerous recent scientific papers were provided. This allowed this version of the report to reflect the latest thinking on the current and future changes in severe weather phenomena as they affect Australia. Thanks also go to Glenn Stone (IAG) who prepared some figures used in this report and Naomi Graham and Shane Grimes (IAG) who proofread the report and made useful suggestions on its structure and content. This report is also intended to complement the Climate Measurements Standards Initiative (CMSI) report Scientific guidance for climate-related physical risk to buildings and infrastructure (2020). The material covered is also based upon work partially supported by the National Center for Atmospheric Research, which is a major facility sponsored by the National Science Foundation under Cooperative Agreement No. 1852977.
1 The Centre of Excellence for Climate Extremes (CLEX) is an international research consortium of five Australian universities and a network of outstanding national and international partner organizations supported by the Australian Research Council
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 7
3. INTRODUCTION
Climate change is occurring, so it is critical that we achieve a common understanding of the increasing risk of the impacts of severe weather on the expanding, built environment across Australia. While global changes are noted, it is the regional and local interpretations that are necessary for informed discussions on the scale, frequency and potential severity of rare weather and climate events so that response, mitigation and adaptation strategies can be focused on reducing these impacts. This report summarises the current state of knowledge on climate change impacts, severe weather and climate extremes that are relevant to Australian property risk. It is based on current knowledge documented in peer-reviewed literature and highlights the significant advances that have been made since the Fifth Assessment Report (AR5) of the United Nations Intergovernmental Panel on Climate Change (IPCC 2013). This report includes evidence from:
• observed changes to the historical climate (typically from pre-industrial or the mid-19th-century levels, although for damaging natural perils the reference period tends to be late 20th century due to the lack of reliable earlier data);
• tentative trends from Australian insurance industry claims for a range of severe weather events spanning the most recent few decades;
• modelling experiments of past, present and future climate change; and
• expert assessments based on the fundamental understanding of the physics associated with the phenomena.
Where possible, changes that are broadly in line with the 1.5°C and 2°C warming targets from the Paris Agreement are addressed explicitly – acknowledging that the current level of global action makes it extremely unlikely that the 1.5°C target, or the 2°C target, will be achieved. Potential changes for scenarios that exceed these goals (>+2°C) are also discussed.
Section 4 of this report (State of the Climate) provides a brief introduction to the current state of the climate system. It includes general definitions and concepts that are important for understanding the report. Section 5 (Changes to Extremes for Different Scenarios) contains the primary information concerning changes for property-risk-relevant extreme events in Australia and is the section of the report with the most updates. Because changes to specific weather extremes (e.g. TCs) are likely to be distributed unevenly across Australia, regional interpretations are included to help improve understanding of potential community impacts. As the regional interpretation is an input for specific scenarios for insurance loss modelling, the >+2°C climate change scenario is nominally chosen to represent +3°C. Expert judgement has been used in cases where there are conflicting research findings or a lack of data for Australia, particularly concerning regional level interpretations.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 8
4. STATE OF THE CLIMATE
4.1 Natural Climate Variability and Forced Climate Change
Natural climate variability is an intrinsic characteristic of the climate system and is related to internal and external natural processes across the full range of spatial and temporal time scales. The internal processes are caused by heat exchanges between the ocean and the atmosphere (e.g. El Niño Southern Oscillation, Indian Ocean Dipole, Interdecadal Pacific Oscillation) or by chaotic behaviours that are inherent in the climate system. Natural external processes that produce variabilities include changes in the earth’s orbit around the sun, changes in solar activity, or volcanic activity. Anthropogenically forced climate change is caused by human effects on the climate system, including greenhouse gas emissions, emissions of aerosols and land use changes. These changes would not have occurred without human activities and are superimposed on top of natural climate variability. The warming that has been recorded since the 1950s cannot be explained by natural processes alone, and human activities are extremely likely2 to have been the dominant cause of the warming observed since the mid-20th century (IPCC 2013).
4.2 State of Climate Assessment
Since the pre-industrial period (1850-1900), the average global mean temperature (applying Lowess smoothing over 5-year periods to the end of 2019) has risen by 1.19°C (Figure SC1, bottom) due to increasing greenhouse gas emissions and deforestation (NOAA 2017). The largest part of the warming has occurred since 1970. The effect of internal climate variability (see Section 4.1) is also evident in Figure SC1, especially in years with strong El Niño events (e.g. 1998 and 2016) which are associated with above-average warm temperatures. Global warming reached a peak in 2016 with temperatures ~1.23°C warmer than the average temperatures in the pre-industrial area, although 2019 was only marginally below that record value (0.04°C lower). The last five years (2015-2019) have been the warmest on record (NOAA 2019). Mean maximum temperatures in almost all land areas, including Australia, are expected to increase at higher rates than the global average (IPCC 2013, Seneviratne et al. 2016, Donat et al. 2017). High- latitude regions will experience much greater temperature increases. Australian land temperatures have increased by over 1.0°C since 1910, at a slightly higher rate than the global average (Bureau of Meteorology and CSIRO 2018), and 2019 was the warmest on record at 1.52°C above the average for the period 1961-1990 (NOAA 2019).
2 IPCC definition indicating a 95-100% probability
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 9 4. STATE OF THE CLIMATE
Figure SC1 Global observed air temperatures showing trends in the seasonal cycle from 1880 to May 2020 relative to a 1980-2015 baseline (top). Trends in monthly mean global air temperature relative to 1850-1900 with the two most recent extreme El Niño years (1998 and 2016) highlighted (bottom). Source: NOAA GISS/GISTEMPv4.
There is no agreement as to when global warming might reach the Paris Agreement targets of 1.5/2°C above pre-industrial levels. However, there is consensus that it is extremely unlikely3 to remain below these targets. Raftery et al. (2017) stated that there is a 90% likelihood that temperatures will rise between 2°C and 4.9°C by 2100. They estimated that there is a 5% chance that global temperatures will remain below a 2°C increase, and only a 1% chance that it will remain below a 1.5°C increase. King and Henley (2018) estimated that under current emissions, global warming would reach 1.5°C
3 IPCC definition indicating a 0-5% probability
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 10 4. STATE OF THE CLIMATE
around 2024 and 2°C around 2035. In a special report4, IPCC stated that there is a very high risk that the global average temperature will exceed the lowest threshold agreed in the Paris Agreement, and that it will overshoot the 2.0°C target by the 2040s (IPCC 2018).
In the Paris Agreement of December 2015, signatory nations in the international community agreed on: “Holding the increase in the global average temperature to well below 2℃ above pre- industrial levels and to pursue efforts to limit the temperature increase to 1.5℃ above pre-industrial levels, recognizing that this would significantly reduce the risks and impacts of climate change.”
Limiting global temperature increases to the Paris Agreement target of 1.5°C (above pre-industrial levels) can only be achieved under ideal conditions with all the following actions achieved as a matter of priority:
• rapid and large-scale global political commitments to decarbonisation; • strongly accelerated growth in low carbon technologies; and • the development of efficient and widespread carbon capture and storage technology by mid- century (Sanderson et al. 2017).
Figure SC2 shows the rapid decrease in carbon emissions that must occur in the very near future to limit
Figure SC2 Observed and projected changes in global average temperature (right) depend on observed and projected emissions of carbon dioxide from fossil fuel combustion (left) and emissions of carbon dioxide and other heat-trapping gases from other human activities, including land use and land-use change under pathways RCP 8.5, 4.5 and 2.6. Observed emissions and temperatures are indicated in black. Source: Hayhoe et al. (2018).
4 In a note released in January 2018, the IPCC restates that draft texts of the report can change substantially and do not necessarily represent the IPCC’s final assessment of the state of knowledge.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 11 4. STATE OF THE CLIMATE
global warming to 1.5°C or 2°C. There is little evidence that major emitters are willing to adopt measures that will achieve this level of decrease. It is becoming apparent to the authors that the observed global emissions trajectories to date mean low emissions scenarios should move away from the RCP2.6 scenario (refer to Glossary) to something more plausible.
A 2°C global warming target is therefore unlikely to be achieved and will significantly increase the risk for catastrophic events, even compared to 1.5°C warming. For example, King et al. (2017) investigated the difference between a 1.5°C and 2°C global Figure SC3 Changes in the likelihood of Australian extreme events in warming on Australian extremes. the current, 1.5°C and 2°C warmer world compared to a natural (pre- Their study showed that events industrial) world. Modified from King et al. (2017). such as the record warm summer of 2012-2013 and associated bleaching of the Great Barrier Reef in 2016 would be 87% more likely in a 2°C warmer world compared to 60% more likely for 1.5°C of warming (Figure SC3). Impacts on rainfall extremes were less clear in their study. The study should be consulted to put the changes in likelihood in context with their uncertainties. Note, King et al. (2017) is a single example pointing to the likely impact on recent events under different warming scenarios. Other studies have reported different absolute values in the expected changes due to a warming environment, but all point to a substantial increase in the likelihood that past extreme events will become more likely under these conditions. Warming above 2°C rapidly increases the risk for global-scale disruptive weather events and unforeseeable threshold changes in the climate system. These so-called tipping-point events are associated with positive feedbacks that cause accelerated and perhaps irreversible changes, regardless of human activities. One example is the collapse of the Greenland ice sheet. In recent years, this ice sheet has experienced rapid increases in summer melt, reaching an all-time record in June 2019. If the melting continues to a full melt of the ice sheet, the resulting global average sea level rise would be ~7.4m. Paleoclimatic records show that Greenland was deglaciated for extended periods during the Pleistocene epoch (2.6 million years ago to 11,700 years ago; Schaefer et al. 2016). Unfortunately, our understanding of important processes that contribute to the melting does not (so far) allow an estimate of threshold temperatures that would result in a collapse of the ice sheet (van den Broeke et al. 2017). Sea level rise and the uncertainty in some of these processes are discussed further in Section 5.6.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 12 4. STATE OF THE CLIMATE
4.3 Extreme Events Under Climate Change
Figure SC4 Climate change has shifted the likelihood of extreme heat for the Northern Hemisphere as shown by the maximum summer temperature anomalies to climate normal average (1951-1980). Modified from WXshift: http://wxshift.com/climate-change/climate-indicators/extreme-heat. (March 2018).
The frequency and intensity of extreme weather and climate events will change at a much higher amplitude than more common weather events. This is because small changes in the mean or spread of current climatic observations typically lead to dramatic changes in the extremes. An example of this is shown in Figure SC4 for Northern Hemisphere maximum summer temperature anomalies within a baseline period (1951-1980) and a recent climate period (2004-2014). Extreme heat events, defined as a temperature anomaly of three standard deviations above the mean, occurred 0.4% of the time in the baseline period (which already included a global warming component). In the recent period, the occurrence increased by a factor of 20 to 8.1%. Many climate extremes undergo similar changes (see Section 5). Not all climate extremes are equally well recorded through historical observations, and not all are well simulated in state-of-the-art climate models. Considering the current peer-reviewed literature and our expert judgement, Figure SC5 shows an assessment of our current ability to detect and understand the impact of climate change on extreme weather events (adapted from Vose et al. 2014). As a general rule, the larger the extent or time period of the extreme, the better it is recorded. For example, especially small-scale and short-period extremes related to severe convection (e.g. tornadoes and hail) are not well-observed. Similarly, our understanding of the effects of climate change on small-scale extremes is often more limited than it is for larger-scale extremes.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 13 4. STATE OF THE CLIMATE
Figure SC5 Our expert assessment of the state of knowledge regarding our ability to detect and understand the impact of climate change on extreme weather events. The horizontal axis shows how skilful we are in detecting the impact of climate change, while the vertical axis refers to our understanding of the physical processes that drive changes. Adapted from Vose et al. (2014).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 14
5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.1 Tropical Cyclones
Section Summary
Tropical cyclones (TCs) are characterised by large interannual variability. However, some trends - supported by observations, theory and modelling simulation – are emerging:
• TC frequency has declined slightly, but the proportion of intense TCs has increased markedly. For example, taking a Southern Hemisphere perspective, the seven most intense TCs in the satellite era (1979 onwards) have occurred since 2004, and the five most intense have occurred since 2015. • The latitude at which cyclones reach their maximum lifetime intensity has shifted poleward, with potentially serious consequences for south-eastern Queensland and north- eastern New South Wales. • TC rainfall is already increasing, especially in terms of inland penetration. For example, cyclone sensitivity studies have shown a near-doubling in the area experiencing >600mm during a cyclone passage over south-eastern Queensland and north-eastern New South Wales has occurred in the last decade. Further increases are expected in a warmer world. • TC translational speed appears to be slowing at higher latitudes. These slower speeds, combined with increasing intensity and rainfall, lead to a potential for substantial increases in cyclone impacts from wind, rain and water ingress into buildings. • Sea levels are rising at an accelerating pace. Combined with increasing river runoff and more intense cyclones, this points towards substantial increases in storm surge impacts and coastal erosion. Socio-economic factors are placing more people and property at risk. The effect of climate change on the impact of TCs will, therefore, act as a risk multiplier.
5.1.1 Background
Section 5.1 focuses on the impact of climate change on the damage potential of TCs and of hybrid cyclones with non-classical structures (Quinting et al. 2019, Cavicchia et al. 2020).
The world has entered a new era of global TC impacts in which economic and insured losses are doubling every 15 years (Kunreuther and Michel-Kerjan 2009, Smith and Katz 2013, Pielke Jr et al. 2008). Bhatia et al. (2019) modelled economic losses of Atlantic Ocean hurricane landfalls and looked at the potential for a US trillion-dollar season in a warming climate. He found that if the CMIP5 model predictions are realised, the 100-year event damage cost increases by US$160 billion beyond that for a static climate regime.
Changes to exposure and vulnerability dominate the trend (Weinkle et al. 2012, Höppe and Pielke 2006, Stewart et al. 2003) but changes to the TCs themselves (Walsh et al. 2016a, Knutson et al. 2020), together with sea level rise (Solomon et al. 2007) have compounding effects. Indeed, Estrada et al. (2015) quantified the climate change contribution to rising USA hurricane costs at US$136
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 15 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
million per year.
The following sections summarise our understanding of global historical and future TC changes and evaluate local changes and predictions as they apply to Australia.
5.1.2 Global Changes
Given that environmental factors influence all stages of the TC life cycle, the entire life cycle is expected to be affected by climate change. Starting with TC formation, there is no theory linking TC genesis to the mean state of the climate (Sharmila and Walsh 2017). A large empirical modelling effort has led to numerous genesis potential indices, but these have substantial issues. For example, Bruyère et al. (2012) highlighted the extreme sensitivity of these indices to the region selected. Such indices also rarely account for non-stationarity in the TC-climate relationship, such as the increasing threshold onset temperature for formation with warming (Johnson and Xie 2010), thereby limiting their application to climate change studies.
Changes in Intensity In a very recently completed analysis, Kossin et al. (2020) found there was an 18% per decade change in the major TC intensity (Saffir Simpson Categories 3-5) exceedance probability for the South Indian Ocean basin and 8% per decade for the South-West Pacific Ocean basin. They based their findings on observed changes in intensities of TCs using the Advanced Dvorak Technique to analyse a homogeneous 39-year long global satellite record from 1979 to 2017. This is the first time such an analysis has been undertaken with statistically significant changes observed at the 95% level or better for most ocean basins. An idealised modelling study by Lavender et al. (2018) suggested caution in extrapolating empirical associations with temperature from past climate. They showed that relationships between sea surface temperature (SST) and TC intensity, size, rainfall and surge might be far from linear. The theoretical basis for TC intensity is far more established. TC potential intensity is directly related to SST (Emanuel 1991, Holland 1997). Theoretical, observational and modelling studies (Strazzo et al. 2015) have all shown a 5% increase in TC maximum wind speed per degree Celsius rise in SST. Other conditions such as vertical wind shear and oceanic mixing by the cyclone act as modifiers to these theoretical findings for specific TCs. But the 5% increase per degree Celsius rule still applies on average.
Geographical and Structural Changes In addition to these advances in theoretical understanding and empirical modelling, numerical models apply physics to our understanding of TCs and climate. Global climate models have demonstrated some success in capturing the geographic distribution of TCs, their frequencies, and inter-basin differences (e.g., Strachan et al. 2013). Still, there are substantial variations between models (Shaevitz et al. 2014, Camargo et al. 2020). Therefore, when looking at potential future climate changes, it is important to use climate models that validate well against the available observational record.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 16 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Recent research highlights the need for climate models to have sufficient resolution and model physics to resolve pre-cyclone tropical disturbances, also referred to as TC seeds (Sugi et al. 2020, Bhatia et al. 2019). Camargo et al. (2020) found that climate model physics, dynamical core, and resolution all affect the TC climates produced by climate models. They (Camargo et al. 2020) completed a detailed investigation of thirty climate models used for a variety of climate change studies, including TCs and their ability to replicate the observed global TC climatologies. They evaluated 14 CMIP5 models with relatively low resolution: six models developed by the US CLIVAR Hurricane Working Group with resolutions between 0.25o and 1.25o, and ten models from the NOAA Model Diagnostics Task Force. The studies revealed substantial quantitative variations between models. It was apparent that many of the models struggle to replicate the observed climatology, and other environmental and TC specific parameters. Hence, any future investigation of TC climate must take care to ensure the models validate well in the regions of interest if their predictions of TC trends are to be believed. This sentiment is also apparent in the multi-model CMIP5 assessments of TCs in the Southern Hemisphere by Ramsay et al. (2018). Model resolution is important when trying to resolve the details of TC behaviour in a changing climate. An illustrative high-resolution study by Bacmeister et al. (2016) used a suite of 28km resolution model runs using bias-corrected SSTs for RCP scenarios 4.5 and 8.5. For both scenarios, they found an overall reduction in the frequency of global TC activity, with less reduction for RCP4.5 compared to RCP8.5. In contrast to this, they found a dramatic increase in the frequency of very intense TCs. Extreme storm-related rainfall was also found to increase in all ocean basins. Recent global models use 10-25km grid spacing, at which many key damaging TC parameters start to become resolved (e.g., Shaevitz et al. 2014, Bacmeister et al. 2016, Knutson et al. 2020), including TC clustering (sequential TC impacts in a given season) and TC rainfall (Villarini et al. 2014). The current generation of climate models still lacks the resolution necessary to capture future trends in very small, aka midget (Arakawa 1952, Walsh et al. 2007) TCs (e.g., TC Tracy, TC Larry) that frequent the Australian region. Gentry and Lackmann (2010) found a grid size in the order of 1km is needed to capture the peak wind speeds of the most intense TCs. Separating the climate influence on TCs into thermodynamic and dynamic contributions offers a useful framework for understanding future changes. Global climate model projections agree on a future climate that is warmer and more humid but disagree on changes to environmental winds and other important atmospheric parameters (Camargo et al. 2020). This disagreement worsens for changes on regional scales. Many climate models have significant biases in the Equatorial Pacific Ocean and poorly replicate the South Pacific Convergence Zone (SPCZ), a zone that extends south- east from Papua New Guinea and plays a key role in the TC climate across the South-West Pacific Ocean, including the Coral Sea region. Confidence in thermodynamically-driven changes to TCs, such as potential intensity and rainfall, is therefore far higher than for circulation-driven changes such as wind shear and steering flow. Newer generation CMIP6 models are showing improved skill in replicating the SST structure of the SPCZ (Grose et al. 2020). There is hope that downscaling the best performing of these models will lead to greater confidence in the future TC climate of this region. There is consensus on a thermodynamically-driven future increase in global and regional maximum wind speeds and on the incidence of high-intensity TCs (Knutson et al. 2020), with a 5% median projected increase in the lifetime maximum surface wind speeds by 2100. There is medium to high
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 17 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
confidence of an increase in the global proportion of major TCs (Saffir Simpson Category 4-5). Knutson et al. (2020) estimate this increase to be in the order of +13% for a 2°C warming (relative to 1986-2005). Initial analysis of CMIP6 models indicates that a warming of 2°C may be reached as early as 2035 (Grose et al. 2020). Holland and Bruyère (2014) found a relationship between anthropogenic warming and an increasing/decreasing proportion of the strongest/weakest TCs at global and ocean basin scales. Lee et al. (2016) provided evidence that changes in the incidence of rapid intensification could be driving these proportional shifts. Bathia (2019) found a detectable increase of Atlantic intensification rates with a positive contribution from anthropogenic forcing. A lack of reliable validation data hampers trends in other ocean basins – notably, the south-west of the South Indian Ocean where geostationary satellite coverage spans a short period.
Intensification Rates, Translational Speed and Rainfall Rates In addition to these changes to frequency and intensity, other important changes to TC activity are anticipated. Emanuel (2017) suggested a future increase in the incidence of rapidly intensifying TCs just offshore that would present a challenge for future forecast and emergency preparation. Current climate models are only starting to be able to be run at the resolution necessary to capture this effect. There are emerging studies showing an increase in intensification rates in future climates. Bhatia et al. (2018) demonstrated this using GFDL’s HiFLOR model. Lee et al. (2020) found increasing rapid intensification events under future climate scenarios, using the synthetic tracks produced by their model. There is also tentative evidence of a slowdown of the median translational speed, as illustrated by Category 5 Hurricane Dorian in the North Atlantic Ocean (2019) and Severe TC Veronica in the Pilbara region of Western Australia in March 2019. There is consensus on a 5-20% thermodynamically-driven increase in large scale TC rain rates within 100km of cyclone centres by the end of this century (Christensen et al. 2013, Walsh et al. 2016a, Villarini et al. 2014, Knutson et al. 2020). As moisture content increases with warming, so does moisture convergence for a given mass convergence into the cyclone. The expected increase in wind speeds may lead to further increases in moisture convergence beyond the Clausius-Clapeyron scaling (Knutson et al. 2010).
Lifetime Maximum Intensity The global expansion of the tropics is associated with lifetime maximum wind speeds now occurring at higher latitudes than in the past (Kossin et al. 2014, IAG internal research 2020 refer Table TC1). They found that the increasing latitude of lifetime maximum intensity (LLMI) over the past 30 years applies to both the South Indian and South Pacific basins, primarily associated with a southward shift of the subtropical jet and associated reduction in wind shear. If this continues, these higher latitude locations will experience increased TC impacts. Rising sea levels will bring more damage from storm surge events with a declining recurrence interval of the more extreme events, all other factors being equal. Finally, there is a lack of climate theory for TC size, but high-resolution simulations by Sun et al. (2017) indicate a future expansion of the mean area subjected to gale force winds. TCs may be living longer, with an associated increased likelihood of reaching their maximum potential intensities, as predicted by the thermodynamic theory.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 18 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Future Climate Projections A recent consensus review of TCs in a changing climate arrived at similar findings to the above. Specifically, Knutson et al. (2020) looked at a wide range of modelling studies. This study included assessments from 11 researchers who are active in this area. Their conclusions provide a fair indication of the current state of the science and are well summarised by the abstract of their paper, the highlights of which are listed below. “Observations, theory, and models, with increasing robustness, indicate rising global TC risk for some metrics that are projected to impact multiple regions. A 2°C anthropogenic global warming is projected to impact TC activity as follows. 1) The most confident TC-related projection is that sea level rise accompanying the warming will lead to higher storm inundation levels, assuming all other factors are unchanged. 2) For TC precipitation rates, there is at least medium-to-high confidence in an increase globally, with a median projected increase of 14%, or close to the rate of tropical water vapor increase with warming (~7% per °C), at constant relative humidity. 3) For TC intensity, 10 of 11 authors had at least medium-to-high confidence that the global average will increase. The median projected increase in lifetime maximum surface wind speeds is about 5% (range: 1%–10%) in available higher-resolution studies. 4) For the global proportion (as opposed to frequency) of TCs that reach very intense (category 4– 5) levels, there is at least medium-to-high confidence in an increase, with a median projected change of +13%. Author opinion was more mixed and confidence levels lower for the following projections: 5) a further poleward expansion of the latitude of maximum TC intensity in the western North Pacific; 6) a decrease of global TC frequency, as projected in most studies; 7) an increase in global very intense TC frequency (category 4–5), seen most prominently in higher-resolution models; and 8) a slowdown in TC translation speed.” Figure TC1 (from Knutson et al. 2020) shows the impact for a 2°C warming, relative to a 1986-2005 baseline. This consensus suggests that there will be higher TC intensities and TC rain rates but fewer TCs globally. There are significant variations between ocean basins and potentially even higher differences in subregions within individual basins, the basins around Australia being prime examples. For example, the studies of Camargo et al. (2020) and Ramsay et al. (2018) showed the South-West Pacific basin is one where the CMIP5 models have difficulty replicating the observed climatology; hence future climate projections must be treated with caution.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 19 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure TC1 Consensus future projection of tropical cyclone characteristics under a 2°C global warming scenario, relative to the 1986-2005 baseline. Shown for each basin and the globe are median and percentile ranges for projected percentage changes in TC frequency, Saffir Simpson category 4–5 TC frequency, TC intensity, and TC near-storm rain rate. For TC frequency, the 5th–95th-percentile range across published estimates is shown. For category 4–5, TC frequency, TC intensity, and TC near-storm rain rates, the 10th– 90th-percentile range is shown. Note the different vertical-axis scales for the combined TC frequency and category 4–5 frequency plot vs the combined TC intensity and TC rain rate plot. Source: Knutson et al. (2020).
5.1.3 Changes in Australia
We now consider observed and predicted changes in Australian TCs. Here, published studies, as well as unpublished analyses from our own IAG/NCAR work, are incorporated. Readers should also refer to the information panel Changes in Australian TC observing practices that outlines issues with the observational record.
Inhomogeneity of the Observational Record When considering the observed changes in TCs in the Australian region (Figure TC2), it must be recognised that the official Bureau of Meteorology TC database is based on observing practices that are not homogeneous over time. A recent reanalysis of TCs in the Australian region (BoM 2018), using all available satellite imagery, recommended using only data since 1989 for intensity analyses because the quality and coverage of satellite imagery declines for the earlier years (see the informational panel on Changes in Australian TC observing practices). There is also insufficient data on the coastal crossing and inland characteristics of landfalling TCs – the region where most damage occurs.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 20 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Changes in Australian TC observing practices
The first geostationary satellite with coverage over Australia was launched in 1977 (GMS-1), and this satellite only provided analogue imagery, unsuited to modern TC analysis techniques which require digital data. There were few polar-orbiting satellites through this period, and none with high-resolution microwave channels or scatterometer instrumentation.
The Australian historical TC record is based on rare direct intensity measurements. Australia does not have an aircraft-based cyclone reconnaissance program such as the one operated by the USA for the North Atlantic Ocean, and sometimes North-East Pacific Ocean. The practice in Australia is to rely heavily on satellite intensity analysis techniques best suited to the higher resolution multi-spectral satellites that have progressively been launched since 1989.
The current requirement for the existence of gale force winds or stronger over more than half of the cyclone’s circulation was introduced in the late 1980s with the advent of more detailed satellite observations. This altered analysis procedure introduced a heterogeneity in the Australian cyclone record not found in other ocean basins where gale force winds only need to occur in less than half of the cyclone circulation. The different definitions of TCs around the world are available at: https://public.wmo.int/en/our- mandate/focus-areas/natural-hazards-and-disaster-risk-reduction/tropical-cyclones.
Low-intensity tropical lows, from non-classical tropical weather systems like the Townsville monsoon low that later formed TC Uesi 2019 or Ex-TC Oswald 2013, are more likely to be classified as monsoon lows or other hybrid lows. This is due to the observed asymmetries and non-classical structures evident in these formative cyclones, that have only become evident with the use of current day analysis techniques and improved quality and quantity of satellite data. As an illustration of this, there were three asymmetrical gale producing tropical lows in the Australian region in the 2019-20 season alone that were not classified as TCs that may have been in earlier eras.
Adopting an arbitrary “Australian region” definition can also produce potentially unrepresentative trends as the longitudes used to define the bounds of the Australian cyclone regions are not ideal when assessing regional TC trends. The 2015-16 season, for example, has a total of three cyclones, although a fourth (TC Raquel) did move into the Australian region in July 2015 but is not included in the total. Cyclone clustering approaches may also offer better insights into regional cyclone trends (Ramsay et al. 2018).
The statistics also depend upon a degree of interpretation by the best track analysts as, although satellite- based guidance has improved dramatically in recent decades, there is still a requirement for interpretation of the guidance, with the final intensities likely to have confidence limits of around +/- 10hPa for the higher intensity cyclones. As an illustration of this, TC Stan (2016), using the automated Dvorak technique, did achieve Category 3 intensity briefly but is not reflected in the best track data as other guidance indicated a lower intensity. TC Uriah (2016) was Category 2 within the Australian region and subsequently deepened to Category 4 when just west of the western boundary of the Australian region of responsibility.
Looking at earlier seasons, the 1970-71 season had 19 TCs in total, three of them severe TCs. Of the 19 TCs, four have lowest central pressures between 1004hPa and 990hPa, which raises the question: would they be included if current analysis practices were used? Of the severe TCs, two, rather than three, would be severe using lowest central pressures below 970hPa as the classifier. In the 1983-84 season, five of the TCs have lowest central pressures of 990hPa or higher. Five rather than eight of the cyclones have lowest central pressures under 970hPa. In the 1984-85 season there were 17 TCs included but two of them (Nigel, 996hPa at 160oE and Rebecca, 994hPa), may not have been included using today’s analysis techniques. The BoM 2018 reanalysis also shows nine severe TCs, fewer than the 11 that appear in the graph.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 21 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Any historical trends in TC numbers and intensities therefore need to be treated with caution, as changing observation systems and analysis techniques may produce changes that are larger than climate change alone. This applies particularly before 1989 but also to a lesser degree post-1989 (see the informational panel on Changes in Australian TC observing practices on page 21). In addition, trend analysis is further complicated by the high level of natural decadal variability in the TC record. Nevertheless, this record forms the basis of much of the industry risk assessment for TC damage potential and, as a result, these assessments include these same false trends.
Figure TC2 Seasonal total numbers of severe and non-severe TCs from 1970-2017 which have occurred in the Australian region as included in the Australian Bureau of Meteorology TC database based upon inhomogeneous analysis techniques.
Hartmann et al. (2013) found that, due to the magnitude of the interdecadal variability, the significance of the trend in TC frequency is sensitive to the specific period analysed. This analysis highlights periods of very active and damaging TCs along the east coast, followed by multi-decadal quiet periods. This is partially linked to the Southern Oscillation Index (Figure TC3). Although a declining trend line is often added to the data in Figure TC3, this trend may change if a longer record was used due to the few and incomplete multi-decadal cycles. A more reliable message here is that the east coast can have periods of a decade or more with few cyclone impacts, followed by more active periods. Building codes and land planning must consider the frequency and intensity of events over a longer-term and not be based on too short a view.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 22 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure TC3 Observed decadal variability in severe landfalling TCs along Australia’s east coast and corresponding variability of the Southern Oscillation Index . Source: Power and Callaghan (2020).
Landfalling Tropical Cyclones The longest analysis of severe landfalling TCs along the east coast of Australia dating back to the 1870s was that produced by Power and Callaghan (2020) using newspaper, shipwreck and other original reports for the older cyclones and the Bureau of Meteorology’s database for the more recent period. They found a large interdecadal variability and a peak of activity in the final part of the 19th century. They identify a decline in the total numbers of landfalling cyclones, unevenly distributed in space and time with prolonged quiet intervals rapidly transitioning to active periods then declining again. The nature of this data does not allow precise quantification of intensity, but this dataset confirms the variability of extreme cyclone impacts along the east coast over the last 150 years. The observed landfalling TCs for the east coast of Queensland, since 1970, indicates that there has been a marked increase in Category 3 and 4 impacts (Figure TC4 top, adapted from Holmes 2020). This finding has important implications for the adequacy of wind loading standards for the eastern seaboard of Australia. As the map of these TC tracks shows (Figure TC4 bottom), there is considerable interdecadal variability in the latitudes affected by these damaging cyclones. Many of them also reached peak intensity very close to the coast. This highlights the need to develop a much better understanding of the risks posed by landfalling TCs, and their hybrid-versions, around the Australian coastline.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 23 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure TC4 Landfalling Australian TCs Category 3 and above (top) and their tracks and location of maximum intensity (red dots) (bottom) along the Queensland coast for the period 1970-2019. In the top figure categories 3.5/4.5 indicate TCs that just fell short of reaching intensities of categories 4/5. The shading around the regression line indicates the 95% confidence intervals. The colour coding in the bottom figure is by decade. Adapted from Holmes (2020).
Future Climate Projections In addition to the available literature, an approach to addressing the issue of limited historical data on high impact weather systems such as TCs is to use ensemble simulations. NCAR, in conjunction with IAG, is conducting a study on the variability and trends in TCs across the north-eastern Australian region. This research uses the vast quantities of data provided by the Decadal Prediction Large Ensemble (DPLE) (see insert on page 25 for more details on this dataset) suite of model runs. This study essentially attempts to quantify the changes in TC risks up to the current decade in an environment of a rapidly changing climate.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 24 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
The study region is south of the equator between 135oE and 180oE. Additionally, the Decadal Prediction Large Ensemble (DPLE) study focused on the densely populated The DPLE dataset is a set of simulations carried out south-east Queensland to north-east New to support research into near-term earth system South Wales region (25oS to 30oS, 150oE to predictions (Yeager et al. 2018). The DPLE consists 160oE – denoted subregion QLD-NSW). of 64 distinct ensembles, one for each of the 64 This is an area of immediate interest as the initialisation times (1 November 1954, 1955, …, building code standards here has a lower 2016, 2017). For each start date, a 40-member wind loading requirement, compared to the ensemble was generated by randomly perturbing the more northern coastal areas of atmospheric initial conditions at the round-off level. Queensland. Details of the analyses and The simulations were integrated forward for 122 techniques used can be found in Bruyère et months (10 years with a 2-month spin-up period) al. (2019a, 2020). after initialisation. This results in over 25,000 years’ worth of model data spanning past, current and near- future climates. Thirty of these ensemble members’ temporal resolution were high enough (six-hourly) to Frequency and Intensity Changes allow TC tracking, reducing the analysed years to South-West Pacific TC frequency appears 19,200. The atmospheric component of the DPLE is to be more sensitive to circulation changes the Community Atmosphere Model, Version 5, with a o than in other basins (Sharmila and Walsh horizontal resolution of 1 (~100km) and 30 vertical 2017). As stated earlier, circulation changes levels. The DPLE is initialised for each set of runs with observed ocean and atmosphere structures, so are not well understood, and many climate has built-in climate change influences. As such, it models have difficulty representing the does not suffer the limitations of AMIP simulations observed environmental characteristics of with fixed or prescribed SSTs and only atmospheric key TC regions, such as vertical wind and CO2 changes. shear, SSTs (particularly in the Equatorial Pacific near the dateline), and the structure of the South Pacific Convergence Zone. As stated in the global perspective section and illustrated in Figure TC1, Knutson et al. (2020) reached a consensus of higher TC intensities but fewer TCs globally. More importantly, there are significant variations between ocean basins and even higher differences in subregions within individual basins. Figure TC5 (adapted from Knutson et al. 2020) depicts the projected changes in TC Figure TC5 Projected changes in TC frequency (%) (a); frequency (a), frequency of the most projected changes in (%) frequency of the most intense intense TCs (categories 4-5) (b), and TCs (categories 4-5) (b), and; projected changes in TC changes in TC maximum intensities (c), for maximum intensities (%) (c), for the South Indian and South-West Pacific basins. Shaded boxes, whiskers, and circles denote the interquartile range, the percentiles, and maxima and minima. Horizontal lines in the shaded boxes denote the median. The percentiles used are 5th–95th (a) 10th–90th (b and c). Adapted from Knutson et al. (2020).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 25 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
the South Indian and South-West Pacific basins. This figure clearly shows that although the median changes predicted for these basins are small, the variability is substantial. In a recent study of South Indian Ocean TC trends – focused upon the Africa-Reunion region and the western half of the West Australian region – Cattiaux and Bousquet (2020) investigated potential changes in TC activity between a reference period (1965- 2014) and a future climate period (2045- 2094), for the RCP8.5 climate change scenario. They used a rotated-stretched 50km grid version of the CNRM-CM6-1 climate model and CMIP5 multi-model climate projections. In agreement with other studies, they found a 20% decrease in total TC frequency but an increase in the lifetime maximum intensity. They divided the current day TC intensity distribution into bins, focusing on 0-50%, 60-80%, and above 90%. Their findings indicated that future climate TC numbers declined by 10% in the 0-50% bin. This is offset by a 10% increase in the 60-80% bin, and a 6-7% increase in the >90% bin. They further noted a shift in the TC season with the earliest cyclone of the season shifting from the historical mid-November period to mid-December, and a redistribution of TCs more into the second half of the season. Lavender and Walsh (2011) found that the average TC lifetime increases by 12-24 hours for each 2-3°C warming. Parker et al. (2018) concluded there could be a 5-10% Figure TC6 DPLE average number of TCs per year increase in landfall wind speed for eastern (a), DPLE (solid) and observed (dotted) decadal Australia under an end-of-century RCP8.5 average annual number of TCs (b), and DPLE decadal average annual number of TCs that enter the scenario, exacerbating the impacts of TCs subregion QLD-NSW (c), for the period 1960 to 2026. on more southern parts of Australia. The dotted grey line in figure (a) indicates the number of simulations available for each modelled year. The DPLE dataset replicates significant Source: NCAR (May 2020). inter-annual variability (Figure TC6a), and it does identify the gradual decline in total numbers of TCs across the domain over
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 26 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
time (Figure TC6b). Note that the DPLE’s horizontal resolution limits the development of physically small (midget) or weak TCs and as such the simulated TC numbers are lower than observed. Figure TC6c shows the decadal trend for the QLD-NSW subregion. For this small subregion, the DPLE does not indicate a notable decrease in TC numbers. This zero trend is significant in conjunction with the declining trend in basin-wide TC numbers. The number of all DPLE TCs in the South-West Pacific basin that reach Category 3 or higher increases by 20% by the 2020s (not shown). Looking specifically at the intensities for TCs that enter the QLD-NSW subregion (Figure TC7), the change in intensities for TCs with wind speeds above 40 m/s increases rapidly after the decade centred on 2000 and particularly for the current 2020 decade. This highlights the growing risk of the more destructive TCs passing through this region and hence the rising risk of more destructive cyclone impacts across south-east Queensland and north-east New South Wales.
Figure TC7 DPLE trends in the intensity of TCs per decade that enter the subregion QLD-NSW. The right panel shows the percentage change in area under the curve with a threshold of 40m/s shown as a dashed line and 50m/s (Category 3) as a solid line. Source: NCAR (May 2020). A slight increase in the lifetime maximum wind speed appears robust across studies (Walsh et al. 2016b). Uncertainty remains, particularly on the role of tropical tropopause temperatures (Ramsay 2013). Holland and Bruyère (2014) suggested a continuation of proportional increases in the strongest TCs in the future, but also suggested an upper limit to the proportion that they referred to as ‘saturation’, imposed by basin geography.
Latitude of Lifetime Maximum Intensity Gaining an understanding of trends in the LLMI of TCs in the Australian region is instructive when considering the changing risks of TC impacts across the southern-most parts of the Australian TC regions. The available observational record is too short to develop statistically significant trends. None-the-less, the reanalysed Bureau of Meteorology best track dataset from 1989 to 2020, supplemented by IBTrACS and technical bulletins for the 2019-20 season where best track data is not yet available, were analysed for LLMI. The analysis was divided into meteorologically similar subregions (A-D, see Figure TC8) of the broader Australian region. The months from December to
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 27 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure TC8 Map showing longitudes used to delineate the subregions (A-D) used for trends in LLMI in Table TC1.
April were analysed as this time period excludes near-equatorial very late or early season TCs that do not affect the Australian mainland. All TCs that reached Category 1 intensity within the Australian region were included, even if their LMI occurred outside the subregion. Table TC1 provides a summary of the observed LLMI trends, while Figure TC9 shows a graph of LLMIs for the Coral Sea region from 1989 to 2020. There are too few TCs between 130oE and 135oE for any meaningful analysis. The LLMI trends based on this limited dataset provide consistently similar results to those from other studies. Cattiaux and Bousquet (2020) found a slight poleward extension of the TC tracks in the South Indian basin. This shift represented approximated a 1oS poleward movement over 80 years. The more robust global and ocean basin results of Kossin et al. (2014) also found results comparable with what would be expected for these subregions. It is worth noting that despite the limited data and caution needed Figure TC9 Observed trends in LLMI of TCs (blue dots) for the Coral when applying this data to future Sea region (142oE to 160oE) for the months December to April from 1989 cyclone behaviour, none of the to 2020. The shading around the regression line indicates the 95% subregions exhibits an equatorial confidence intervals. Source: IAG internal analysis (April 2020). shift in the LLMI.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 28 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Table TC1 Observed trends in the LLMI of TCs in subregions of the Australian region for the months December to April from 1989 to 2020. (IAG internal analysis April 2020).
Subregion Extent Number Mean Mean Change Rate of change of TCs LLMI LLMI 1989 2020
A 142oE-160oE 94 16.4oS 18.2oS 1.8o poleward 6.4 km/year poleward (Coral Sea)
B 135oE-142oE 44 14.2oS 15.0oS 0.8o poleward 2.9 km/year poleward
C 115oE-130oE 86 16.5oS 17.6oS 1.1o poleward 3.9 km/year poleward
D 90oE-115oE 126 15.6oS 15.9oS 0.3o poleward 1.1 km/year poleward
The Coral Sea subregion (subregion A) spans the area from the Cape York Peninsula (142oE) to 160oE. LLMI analysis for this subregion was based on the 94 TCs that appear in the database from 1989 to 2020. The observed LLMI trend in this region was a 1.8o poleward shift, which equates to a mean poleward shift of 6.4km/year. Subregion B lies between the Cape York Peninsula and 135oE and encompasses the Gulf of Carpentaria. Here, the relative proximity of the Australian landmass to the south acts to limit the poleward shift of the LLMI. The detected poleward shift of around 3km/year is less than half that of the Coral Sea region and illustrates the importance of considering how landmasses may affect trend analyses. Subregion C, between 115oE and 130oE, includes the North-West Shelf, Timor, and Arafura Seas. Here, the entire region between the Indonesian Archipelago and the Australian coastline is subjected to warm waters. However, the West Australian landmass and its associated heat lows act as modifiers to the wind regimes in this region, with the LLMI still shifting poleward but at a slower rate than for the Coral Sea. In subregion D, between 90oE and 115oE, the dominant feature is the cool north then north-westward flowing West Australian Current, with the warm onshore Leeuwin Current a smaller influence. This colder water region typically pushes as far north as 15oS at 90oE and acts as an inhibitor to the poleward expansion of the tropical warm water regions. As a result, this region exhibits the smallest poleward shift in the LLMI. Figure TC10 shows the DPLE average annual LLMI for all TCs west of 160oE (i.e., the ones impacting land). On average current-day and projected future TCs reach their Lifetime Maximum Intensity (LMI) 0.5o further poleward than in the early years.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 29 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
This observed, and predicted, poleward shift in the latitudes at which TCs reach their LMI, highlights that although the risk of TC impacts in this region is low relative to regions further north, there is an increasing proportion of TCs extending further poleward over time. This is a significant finding considering the population density and lower wind loading standards in this area. Should the lifetime maximum intensities continue to migrate poleward, as suggested by Kossin et al. (2014) and Lavender and Walsh (2011) and Figure TC10 The DLPE average annual LLMI of all TCs (dots) supported by the limited internal o west of 160 E. The shading around the regression line indicates analyses of IAG discussed the 95% confidence intervals. Source: NCAR (May 2020). earlier, cities along subtropical eastern and western Australian coasts will experience greater TCs impacts than in the past.
Tropical Cyclones and Sea Surface Temperatures TCs gain their energy and moisture from the oceans, with the maximum potential intensity (MPI) of TCs highly correlated with SST. Figure TC11 shows the decadal trends in the average SST under all TCs in the domain (left) and for the QLD-NSW subregion (right). Domain-wide and in the QLD-NSW subregion there is a marked shift towards higher SSTs in the later decades, and particularly for the current decade centred on 2020. Domain-wide the time that TCs spent over water > 30°C increases by 500% for the 2020s, compared to the 1960s. This is even more evident in the QLD-NSW subregion, where the average SST under TCs increases by nearly 1°C and the time that TCs spend over water warmer than 27°C increases by over 1000%. This is consistent with observations and modelling results, which found that TCs are becoming more intense, especially further south, and contain more precipitable water, primarily imported from the warming tropical regions.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 30 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure TC11 DPLE distribution of SST under TCs. The left panel shows the decade change for all TCs in the domain west of 160oE. The right shows the TCs within the QLD-NSW subregion. The dotted lines in the right-hand panel indicate the mean SST for 1960 and 2020. Source: Bruyère (2020).
Tropical Cyclones Rainfall and Surge There is not a good understanding of historical changes in other key TC characteristics such as size and storm total rainfall for the Australia region, primarily due to a lack of data. Lavender and Abbs (2013) explored historical TC rainfall trends and found a signal of significant drying associated with TCs over the east coast of Australia. However, recent work by Bruyère et al. (2019a) found that flooding associated with TC Debbie-like cyclones was significantly enhanced as a result of oceanic warming. The likelihood of a future increase in TC rainfall rate appears robust across many studies, driven by the strong thermodynamic change processes. Parker et al. (2018) found increases of up to 27% in landfall hourly TC rain rates over eastern Australia by the end of this century. The DPLE studies find
Figure TC12 DPLE trends in the area receiving over 600mm of rainfall for the entire cyclone path and the portion over land (left) and for the QLD-NSW subregion (right). Source: Bruyère (2020).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 31 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
this increased rainfall trend is already occurring, and it is predicted to continue in the warmer climate of the future. Rainfall associated with TCs is a major contributor to the total damage they produce. Although the DPLE model resolution does not adequately resolve topographic and convective processes, there remains a substantial trend in the total rainfall within 500km of the cyclone centre. The data indicate large changes in the area exposed to rainfall accumulations over 600mm per storm (Figure TC12). Over the continent (Figure TC12, left) the area exposed to high rain volume has increased by around 60%. Focusing in on the QLD-NSW subregion, this increase is between 175 and 200% for the current decade (Figure TC12, right). The total inland area exposed to high winds is also increasing (not shown). This points to rapidly increasing risks of water ingress through wind-driven rain or flash flooding, as well as a heightened risk of major river flooding. These impacts are compounded by increasing storm surge, wave and coastal erosion impacts.
Given that Australia is recognised as holding the world record storm surge of 13m, associated with Cyclone Mahina in 1899 (Nott et al. 2014), an important question is how likely such events will be in the future. New simulation technologies are being developed to assess rare surge events for any coastal location (Bruyère et al. 2019a; Lin and Emanuel 2016). Lin and Emanuel (2016) generated 2,400 synthetic TC surge events for Cairns using wind fields from the TC model of Emanuel et al. (2006) to drive a surge model. They found that under current climate conditions the 0.01% (1 in 10,000 year) surge would be 5.7m, generated by a TC that was only slightly more intense than Cyclone Yasi but with a slightly different track. Storm surge risk associated with TCs will increase (i.e., return periods will contract by an order of magnitude) due to rising sea levels and TC intensity associated with global warming (e.g. Lin and Emanuel 2016, Woodruff et al. 2013).
An interpretation of projected regional changes applicable to Australia This section considers potential regional changes to TCs within the Australian region. It uses expert opinion to interpret the published observational and modelling studies and known limitations, as well as unpublished IAG analyses, and initial IAG/NCAR work described earlier. These changes, summarised in Figure TC13, can be used as adjustment factors to the regional TC frequencies to reflect climatologies more representative of current and future climates. These climatologies are a key input to Natural Catastrophe models used by the (re)insurance industry and provide a means to simulate past, present and future climate risk scenarios. The near coast and overland impacts are of most interest to risk assessments. Therefore, while catastrophe models use the tracks and intensities across a broader ocean basin, it focuses on deriving the landfalling intensity distributions along different sections of the coast. Importantly, the coastal sections can have significantly different intensity distributions from each other and from the broader ocean basin total. For reasons explained earlier and following, the application of climate change to these coastal areas will also be uneven and different from the expected mean ocean basin changes, such as those summarised in Figure TC1. The following sections: 1) Discuss the key overall factors influencing the expected Australian TC changes under climate change; and 2) Detail specific regional factors that are considered when deriving the regional changes.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 32 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Property damage increases non-linearly with increasing TC intensity. Therefore, the expected changes in Figure TC13 are stratified by low and high TC intensity bands to enable more refined adjustments to the base climatologies.
Overall Factors The greater proportion of intense TCs (Categories 4 and 5) in the most recent decades compared to the 1970s (Holland and Bruyère 2014) is likely to also apply across the Australian region. This has been supported by evidence from the DPLE analyses for the north-east of Australia as well as global modelling studies, a subset of which have been discussed here. The DPLE analysis found a 10% increase in the proportion of intense TCs between the decades centred on 1960 and 2010, and a further 10% increase predicted for the 2020 decade. Importantly, from a property risk perspective, the less frequent but intense TCs drive most of the damage. It is estimated that approximately 10% of all pre-industrial TCs have been Category 4 or 5. Currently, nearly 25% of all TCs reach this peak intensity. This rising trend would likely continue before the upper limit or ‘saturation point’ is reached. There has been a poleward shift of the latitudes where TCs reach their maximum lifetime intensity in the South Indian Ocean and South-West Pacific Ocean basins (Kossin et al. 2014). Tentative trends have been identified across the Australian region, pointing to different rates of poleward shift of the LLMI for different subregions. The varying trends are driven by subregional scale factors that have not yet been investigated. For the east coast of Australia, these observed and DPLE modelled trends imply a poleward expansion of the area that will be subjected to future intense TCs and damages. For the west coast, south of Shark Bay, the poleward shift was found to be smaller than off the east coast. This is due to the presence of cooler waters in the West Australian Current, which, although warming, are projected to remain relatively cool to the end of the present century. Potentially offsetting this would be variations in the warmth and structure of the Leeuwin Current. However, the oceanic component of the climate models has yet to produce conclusive results for this area. It is likely that some of the most significant impacts for the Central West and Lower West Coasts of Western Australia will be associated with the interactions of tropical lows and mid-latitude weather systems. These interactions produce complex weather systems with both tropical and mid-latitude characteristics, such as those in May 2020 and June 2012. TCs traversing this region are also likely to be undergoing extratropical transition. To date, there have been no studies of trends in extratropical transitioning systems in the Australian and New Zealand regions. TCs in the Australian and New Zealand regions are likely to last longer due to the higher heat content of the oceans over which they travel (Figure TC11), particularly in the western and Coral Sea subregions away from the constraining effects of the Australian mainland. They are also likely to spend a longer proportion of their life cycle as higher category TCs (Bruyère et al. 2019b). This is particularly true for the South-West Pacific Ocean, where a general southward shift of the subtropical jet stream could lead to a reduction in the vertical wind shear. This increases the chance of significant TC impacts on Australia, New Zealand and other South-West Pacific islands. This trend of longer- lasting and stronger TCs should continue as the world gets warmer. Under +2°C and >+2°C scenarios, TC seasons are expected to gradually favour TC development in the second half of the season. Recent late-season TCs point to this having already occurred for the
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 33 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Coral Sea. For example, TC Zane (2013) was a weak Australian Category 3 cyclone that formed in late April and weakened in early May as it crossed the north Queensland coast; TC Ann reached Australian Category 2 intensity in the Coral Sea in mid-May 2019; and TC Mangga reached Australian Category 1 intensity to the north-west of Cocos Islands in May 2020. With warmer environments, TCs are expected to produce significantly higher rainfall and runoff quantities. Warmer air has a higher moisture capacity – as indicated by the Clausius-Clapeyron relationship (Trenberth et al. 2003, Prein et al. 2017a). Therefore, at a minimum, rainfall should increase by around 7% for each degree Celsius of warming. Rainfall volumes (Bruyère et al. 2019a, 2019b) and storm runoff extremes (Yin et al., 2018) have been shown to increase at rates significantly higher than the 7% suggested by Clausius-Clapeyron scaling alone. This is due to increased intensity, longer life, expansion of the heavy rainfall area and increased inland penetration of future TCs. The potentially changing nature of the important sub-type of TCs referred to as midgets, which includes TC Tracy that devastated Darwin in 1974, remains an area of considerable uncertainty due to their small size and the inability of the current generation of climate models to resolve them adequately. For New Zealand, the poleward extension of the latitudinal band favourable for TC development and intensification is expected to lead to an increased number of TC-related impacts across the nation. Preliminary results from the DPLE dataset are confirming increased TC frequencies and intensities over waters north to north-west of the North Island of New Zealand. Most impacts are expected to be from tropical systems that are undergoing extratropical transition. The 2017-2018 season may be an early indication that this trend is already underway with a record three extratropical cyclones (ETCs) affecting the island nation. However, insufficient research means that there remains a high degree of uncertainty in ETC trends. The work of Ramsay et al. (2018) and Knutson et al. (2019, 2020) serves as a good launching pad for future, more tailored, studies on TCs and ETCs in the Australian and New Zealand region.
Specific Regional Factors Figure TC13 and the following commentary summarises the expected regional changes to both the frequency of all TCs and the frequency of low intensity (Australian Categories 1 and 2) and high intensity (Australian Categories 3, 4 and 5) TCs between the 1950s and the +3°C climate change scenario. It is important to note:
• some of the changes shown have already occurred due to the warming experienced up until 2020;
• the changes described below only relate to the wind component of TCs and do not include other important risk drivers – such as water ingress, flooding caused by more intense short duration or storm-total rainfall, storm surge, wave impacts and coastal erosion – that are all very likely to worsen.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 34 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure TC13 The potential regional changes to the frequency of all TCs, low intensity TCs (Australian Categories 1 and 2) and intense TCs (Australian Categories 3, 4 and 5) in a +3°C warming world based on all available sources of information.
Central and South-East Queensland and North-East New South Wales Figure TC14 shows the ten most significant TCs to have affected the south-east Queensland region since 1954, serving as a rough baseline for applying future changes caused by the warming climate. By applying the observed and modelled poleward shift in the LLMI and noting the warmer SSTs that will exist southwards beyond the latitude of Brisbane to the historical TC tracks, Figure TC13 serves as an illustration of what future climate TC impacts could look like across this region. In the +3°C scenario and beyond, the tentatively observed poleward shift of the LLMI, coupled with the expected thermodynamically driven support for more intense cyclones over warmer waters, broadens the highest risk area from the north coast of Queensland into the central coastal region. Due to the poleward expansion of the TC affected region, the south-east Queensland and north-east New South Wales region will experience the largest relative (not absolute) change. Although still likely to be uncommon events, the very rare high category TCs are expected to have a slower decay rate (i.e., have a higher intensity) in a warmer climate. This is due to the increased SSTs off the central and south-east coasts of Queensland, which can sustain stronger cyclones.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 35 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
21.5oS
Queensland 27.0oS
New South Wales
Figure TC14 The Tracks of the ten most significant TCs to have affected the south-east Queensland and north- east New South Wales region since 1954. The horizontal dashed lines show the mean latitude of maximum intensity for this set of TCs (21.5oS) and the latitude of Brisbane (27.0oS).
Far North Queensland, the Northern Territory and Northern Western Australia The poleward shift of the LLMI would tend to marginally reduce TC frequency in these regions. However, there will remain a relatively high risk of impacts from intense TCs as these regions are susceptible to impacts from some of the most intense TCs in the world. This reduced frequency will not reduce the impact from the most severe and destructive TCs. When rapidly increasing storm-total rainfall and storm surge-related factors are considered, it is quite likely that the total TC-related risk may increase rather than decline.
Central West and South-West Western Australia The waters off the west coast, from Shark Bay southwards, are cooler than those off the east coast. The warming effects of climate change on TCs should be slower to manifest themselves in this region. Nonetheless, there will likely be ongoing warming of the waters off the west coast of Australia and particularly in the Leeuwin Current. This warming could lead to a commensurate increase in the risk of higher intensity TCs, although the rate of increase in risk should be slower than for the east coast.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 36 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.2 Extreme Precipitation and Flooding
Section Summary
Extreme precipitation can intensify significantly with climate change, even in regions that experience drying on average. The more extreme an event is (i.e., the more intense and less frequent), the more its rainfall rate is likely to change in the future. A study of trends in Australian hourly and daily rainfall from the period 1966-1989 to 1990-2013 showed daily rainfall increased at around 7% per degree of warming. Emerging science also confirms that intense rainfall rates are increasing. These rates across southern Australia have increased nearly 14% per degree of warming, and 21% for the tropical regions. Records show observed flood severity for smaller, fast response catchments has increased. The faster the response of the catchment, the greater the increase in flood severity. As well as increased flood frequency for fast response rivers, the flood volumes and peak flow rates should also increase, leading to non-linear increases in the damage produced by changes to extreme short-duration rainfall. Housing and infrastructure currently being designed and built can be expected to still be in use around the year 2100. Traditional floodplain management decisions are based upon analyses of rainfall regimes from the historical climate and do not adequately account for current and future conditions. Wind-driven rainfall damage is expected to increase due to a combination of more intense convection, more intense TCs and possibly warm-season ECLs. Dyer et al. (2019) highlighted that catchments with historically strong land planning controls are more sensitive to increases in flood intensity or frequency due to the accumulation of properties just above the 1% AEP. Recent advances in computational resources and atmospheric model development enable regional climate model simulations in convection-permitting resolutions (≤4km horizontal grid- spacing), which explicitly resolve issues related to deep convective storms. These provide more reliable simulations of sub-daily precipitation extremes and surface-atmosphere interactions (e.g. coastlines, orography) than those where convection is not explicitly simulated.
5.2.1 Background
The IPCC states in its AR5 report (IPCC 2013) that, based on historical observations, there are likely more land regions where the number of heavy precipitation events has increased than decreased. The recent detection of increasing trends in extreme precipitation (Jakob et al. 2011, Westra and Sisson 2011, Chen et al. 2013, Guerreiro et al. 2018, Dowdy 2020) and discharge in some catchments implies a greater risk of flooding at regional scales. This trend of increasing frequency
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 37 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
and intensity of extreme precipitation events will very likely continue and intensify with rising temperatures. From a risk to life and property perspective, consideration of extremes in rainfall should focus on the rainfall events which lead to damaging flood events and which form the basis of land use planning and building standards, with annual exceedance probabilities (AEP) in the order of 2% to 0.001%. Such extreme rainfall events are sparse in the historical data and therefore poorly represented in the scientific literature, with the majority of published research focusing on daily or multi-day totals around the 99th to 99.9th percentile (equivalent to AEP of ~30% to ~3%). Events close to the 99th percentile typically do not exceed design criteria and therefore produce only limited levels of damage to the built environment. The rate at which extreme precipitation will intensify is theoretically related to the water-holding capacity of the atmosphere (i.e. the Clausius-Clapeyron relationship). Air can hold approximately 7% more water vapour for each degree Celsius of warming, which translates to a theoretical increase in short-term (hourly to daily accumulations) extreme precipitation by approximately the same rate (Trenberth et al. 2003, Prein et al. 2017a, BoM and CSIRO 2018). These rates can, however, be strongly modified by changes in storm characteristics, such as size, translation speed or non-linear storm dynamics (Westra et al. 2014, Prein et al. 2017b, Prein and Heymsfield 2020). Extreme precipitation increases of higher than 7% per degree Celsius of warming can occur for cold season extremes due to a transition from stratiform to convective rainfall for temperatures above ~10°C (Berg et al. 2013). Other changes in storm dynamics, such as an increase in latent heat release or changes in rainfall efficiency (e.g., Prein and Heymsfield 2020), might also result in larger than 7% per degree Celsius increases in extreme precipitation. However, further research is needed to understand their contribution better (Fowler et al. 2020). Furthermore, the atmospheric circulation patterns that trigger heavy precipitation events are also likely to change, thereby either enhancing or partially offsetting thermodynamic effects. The changes in atmospheric circulation patterns are a major source of uncertainty, while thermodynamic changes (e.g., rising temperatures and increasing atmospheric moisture) are more certain (Shepherd 2014). It is important to realise that changes in extreme precipitation rates are decoupled from changes in mean precipitation. This means that extreme precipitation can intensify significantly with climate change in regions that experience drying on average (Giorgi et al. 2011, Ban et al. 2015, Prein et al. 2017a). This results in global extreme precipitation increases being more homogeneous than changes in mean precipitation (e.g., Fischer et al. 2014). Frequency changes in extreme precipitation are expected to change at even higher rates than intensity changes, meaning higher return level events will experience a bigger increase (see Section 4.2). Figure EP1 shows the return periods for the 20th century 100-year flood at the end of the 21st century under a >+2°C scenario. In the Murray Basin, for example, the 100-year current climate flood is projected to occur every 5-25 years by the end of the century, which is a 400-2000% increase in frequency. Similar large projected increases have been reported for other regions of the world (e.g., Te Linde et al. 2011). There is an increasing body of literature that indicates that the increase in frequency and intensity of future extreme precipitation will be larger for rare events (Figure EP2: e.g., Pendergrass 2018, Li et al. 2019, Lopez‐Cantu et al. 2020). This means that the rarest and most costly extreme precipitation events might increase the fastest under climate change.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 38 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure EP1 Projected return period of the 20th century 100-year flood at the end of the 21st century under the RCP8.5 scenario at the outlets of 29 selected river basins. The colour of each basin indicates the multi- model median return period at basin outlets. Source: Hirabayashi et al. (2013).
Figure EP2 Simulated global mean daily extreme precipitation change as a function of the rarity of an event. The more extreme an event is (i.e., the more intense and less frequent), the more its rainfall rate changes in the future. Source: Pendergrass (2018).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 39 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.2.2 Changes in Australia
Observed changes Assessing the impact of climate change on extreme precipitation in Australia and its regions is more uncertain than the discussed global estimates. This is due to the large natural variability in Australian rainfall on annual to decadal timescales. Observed changes in heavy precipitation in Australia are consistent with global studies. Still, the existing climatology of severe convective storms across Australia is poor and complicates the detection and attribution of historical changes in extreme rainfall. The fraction of Australia (Figure EP3) that receives a high proportion of its annual rainfall (greater than 90% of daily precipitation) from extreme events (greater than 90th percentile) has increased since the 1970s (Gallant et al. 2013). The exception is the east coast which has generally experienced a significant decrease in extreme daily rainfall events since 1950 (Gallant et al. 2014).
Figure EP3 Trend in maximum one-day precipitation from 1920 to 2019. Source: Australian Bureau of Meteorology http://www.bom.gov.au/cgi-bin/climate/change/extremes/trendmaps.cgi?map=RX1d&period=1920.
Using the Australian observational record, Wasko et al. (2016) showed that total precipitation and the maximum precipitation intensity increases with temperature across Australia, while the storm’s spatial extent decreases. Guerreiro et al. (2018) studied trends in hourly and daily rainfall across Australia
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 40 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
from the period 1966-1989 to 1990-2013. Daily rainfall was found to follow Clausius-Clapeyron scaling. In contrast, hourly rainfall exceeded this scaling by at least a factor of two and, for tropical regions, by a factor of three. This indicates that over the observational record, flood severity increased for smaller, fast response catchments since extreme precipitation has become more intense and spatially concentrated (Fowler et al. 2020). The shorter and faster the response time of the catchment, the greater the increase in flood severity. For large catchments such as the Murray-Darling Basin, the joint probabilities of El Niño–Southern Oscillation (ENSO) and the Indian Ocean Dipole and other large-scale teleconnections need to be factored in. This increased complexity of the widespread very heavy rainfall producing mechanisms means the changes in the return periods for the rare extreme floods are far less certain. Additionally, changes in antecedent conditions (e.g., soil moisture) are increasing in importance with increasing catchment size and can mitigate or amplify future flooding (Sharma et al. 2018). Observed changes in precipitation are consistent with changes in annual maximum streamflow (Ishak et al. 2013). Ishak et al. (2013) attributed most of the decreasing flood magnitudes to natural climate variability, indicating that forced climate change impacts on extreme precipitation are small compared to internal variability in the observational records. This is in line with other studies showing that extreme rainfall time series are strongly affected by climate variability, with ENSO being the dominant driver (King et al. 2013, King et al. 2014). Changes in ENSO due to climate change are uncertain and generally within the range of natural variability (Chen et al. 2017). Prein and Heymsfield (2020) combined analyses of radiosonde data and two reanalyses (ERA- Interim: Dee and Uppala 2009, Simmons et al. 2007 and ERA-20C: Stickler et al. 2014) to evaluate precipitation differences between 1979 and 2010. They found that there were significant trends in the height of the melting level (Figure EP4) and the warm cloud layer (WCL) depth (Figure EP5) across Australia. A change in melting level will affect the distribution of snow, damaging hail and heavy rainfall. In general, a higher melting level will lead to decreases in large hail and less snowfall. Increases in melting level across Australia for the warm season (December to May) were typically 30- 60m/decade, while the increases in the deep tropical areas exceeded 140m/decade. Another consequence of the increasing height of the melting level is the increase in the WCL depth. The WCL depth influences the liquid water content in the cloud and surface precipitation rates. This is because large, fast-falling drops form very efficiently in clouds with deep WCLs. These findings are supported by observational studies, which show that precipitation intensity responds non-linearly with a deepening of clouds and precipitation efficiency increases with mid-tropospheric humidity. As Figure EP5 shows, large parts of New South Wales have experienced increasing rates of WCL of over 60m/decade. Sizeable increases have also been observed across large parts of Western Australia. The rate of extreme rainfall increases four-fold in environments with WCLs deeper than ~3.5km, corresponding to double the Clausius-Clapeyron rate. Where WCLs more frequently exceed 3.3km (a threshold used by forecasters as an indication of flash flood potential), the potential for flash flood events escalates.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 41 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure EP4 Linear trend estimates of changes in the melting level height above the surface. Shown is the warm season (December to May) average trend based on ERA-Interim and ERA-20C within the period 1979-2010. Hatching from top left to bottom right/bottom left to top right shows significant trends (alpha is 0.1) in ERA- 20C/ERA-Interim and dotted areas show regions with potential breaking points in their record. Source: Prein and Heymsfield. (2020).
Figure EP5 Linear trend estimates of changes in the WCL depth (cloud base to melting level depth). Shown is the warm season (December to May) average trend based on ERA-Interim and ERA-20C within the period 1979-2010. Hatching from top left to bottom right/bottom left to top right shows significant trends (alpha is 0.1) in ERA-20C/ERA-Interim and white areas show regions with non-homogeneous records. Source: Prein and Heymsfield (2020).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 42 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
The most robust change signal is an intensification of eastward-propagating warm sea surface anomalies that are a characteristic of very strong El Niño events (Cai et al. 2015).A new study by Freund et al. (2020) – using simulations from both CMIP5 and CMIP6 models – found no overall model agreement on the projected sign or intensity changes of East Pacific and Central Pacific El Niño events. However, the small number of available CMIP6 models did show increasing proportions of Central Pacific El Niños. These results have considerable uncertainties and studies that only include the best performing climate models are needed to gain a better understanding of future trends in ENSO events. A typical assumption in engineering is that floods occur randomly over time. This assumption is very likely invalid as has been shown by McMahon and Kiem (2018) for flood events in south-east Queensland. They showed that ~80% of historic large floods in this region occurred within sets of 5- year periods followed by 35-year periods of lower flood occurrence. This cyclic behaviour is at least in part related to the Interdecadal Pacific Oscillation (IPO: Power et al. 1999), which highlights the importance of including natural variability in flood risk assessments.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 43 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Projected changes CMIP5 global climate models (GCM) simulate a consistent increase in various extreme precipitation indices over Australia with varying confidence (see Figure EP6: CSIRO and Bureau of Meteorology 2015). These results are consistent with global estimates (see Section 5.2) and show that: 1. Changes in extreme precipitation are increasing in all regions independent of changes in mean precipitation, even in regions with strong drying trends (e.g., south-west Western Australia);
2. Changes increase in magnitude for more intense extreme events and for higher emission scenarios and are less evident under RCP2.6; and
3. Tendencies for these increases are already detectable. The confidence for increases in daily extreme rainfall should only be reduced to ‘medium confidence’ in south-west Western Australia, where the strong drying trend could offset some of the increases Figure EP6 Changes to extreme rainfall in East Australia (EA), North in extremes. Australia (NA), Rangelands (R), and South Australia (SA). Each panel shows projected change in 2080–99 for annual mean precipitation, annual maximum 1-day rainfall, and 20-year return level of annual Regional and sub-daily estimates maximum 1-day rainfall relative to the 1986-2005 average. The in extreme precipitation changes horizontal tick denotes the median and the bar denotes the 10th to 90th from GCMs should be treated with percentiles of the CMIP5 results. Scenarios are shown in grey (natural care due to the coarse grid variability), blue (RCP4.5) and red (RCP8.5). Source CSIRO and spacing, which does not resolve Bureau of Meteorology (2015). the spatial scales that cause
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 44 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
extreme rainfall events in the real world. Damaging flooding often occurs from severe local storms that are not resolved by the relatively coarse climate models used in the above studies. Higher-resolution regional models can help to close this gap. Within the NARCliM project (Evans et al. 2014), several GCMs were dynamically downscaled with regional climate models. Evans et al. (2014) used a two-way nested approach with a 50km grid-spacing outer domain over Australia and Oceania, and a 10km grid-spacing nest over south-east Australia. The 10km NARCliM simulations use a relatively small domain. Therefore, for some parts of this domain, the effects of weather systems moving in from the outer domain may not be fully captured at the local scale. Bao et al. (2017) used the 50km NARCliM ensemble to study the scaling between temperature increases and daily precipitation increases in Australia. Consistent with Evans et al. (2014), they showed a systematic increase in extreme rainfall over Australia. The scaling rates generally increase for higher extreme percentiles. The 99th percentile of daily precipitation increases close to ~7% per degree Celsius of warming while the 99.9th percentile shows higher rates (Figure EP7).
Figure EP7 Projected future climate scaling rates for the ensemble mean daily precipitation 99th (a), 99.5th (b), and 99.9th (c) percentile climate scaling rates (%/°C) from NARCliM between 2060–2079 and 1990– 2009 under the A2 scenario. Source Bao et al. (2017).
Recent advances in computational resources and atmospheric model development enable regional climate models simulations in convection-permitting resolutions (≤4km horizontal grid-spacing), which explicitly resolve issues related to deep convective storms (Prein et al. 2015). These models have been shown to provide more reliable simulations of sub-daily precipitation extremes and surface- atmosphere interactions (e.g. coastlines, orography) than those where convection is not explicitly simulated. Since no convection-permitting climate simulations have been published for Australia (so far), a summary of relevant convective extreme studies from other areas is presented. Sub-daily precipitation extremes are likely to intensify at faster rates than daily and multi-day extremes due to dynamical and thermodynamical changes in convective storms, as recent observational studies have shown (Kendon et al. 2014, Ban et al. 2015, Prein et al. 2017b, Prein and Heymsfield 2020). Dowdy (2020) has provided the first look at changes in the distribution of severe thunderstorms across Australia in a warming climate that serves as a pointer to the changing risk of flash flooding (see Section 5.5: Bushfire).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 45 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Prein et al. (2017b) analysed these kinds of changes for mesoscale convective systems (e.g., squall- lines, TCs), which are the leading cause of warm-season flooding in North America. They showed an increase in peak hourly precipitation that is approximately 7% per degree Celsius of warming and an increase in heavy precipitating storm frequency (more than 90 mm h-1) of at least 400%. The largest changes occur in storm-total precipitation volume, which can increase by up to 20% per degree Celsius of warming. This results in an almost doubling of storm-total rainfall volume at the end of the century under an RCP8.5 scenario. Wind-driven rain – that is rain that is given a horizontal velocity component by the wind – can be particularly damaging to buildings due to its entrainment into building facades (Blocken and Carmeliet 2004, Cyclone Testing Station 2018, Henderson et al. 2018, Boughton et al. 2017). While there is a growing understanding of increases in rainfall extremes currently being experienced, and expected to continue into coming decades, changes in wind hazards are less certain. Very little research has been done regarding climate change effects on wind-driven rain. In general, confidence in future changes in wind hazards depends on the storm type. Much of the wind-driven rain damage in Australia comes from either TCs, ECLs, frontal systems or severe convective storms. Wind-driven rain events might therefore intensify due to an increase in TC wind speed. Walsh et al. (2016b) showed that large-scale frontal system-related events are projected to decrease in northern and southern Australia and increase along the east coast. Projections for wind-driven rain events associated with convective storms and summertime ECLs are highly uncertain.
5.2.3 Regional Interpretations for Risk Assessment for Australia
This section only refers to riverine flooding caused by the changes to rainfall regimes. It is important to note that for most urban developments around Australia the so-called 1 in 100-year flood event (1% AEP) is used to delineate areas that can be developed against those that cannot, or at least have significant restrictions on the types of structures allowed. In Australia, the publication Australian Rainfall and Runoff (ARR) was first published in 1958 and updated in 1987. This release used historical rainfall records to determine rainfall distributions around Australia. The historical observations roughly spanned the period from around 1945 to 1985, which is a period approximately 35 -75 years ago. The climate has changed significantly since those decades and is likely to continue to do so for the next century or more. A new version of ARR was released in 2019 (http://arr.ga.gov.au/). It highlights the fact that most land planning decisions around Australia for river flood are being based upon analyses of rainfall regimes from the climate of the past, and not for the climate extremes which the built environment will experience within the lifetime of the structures. The result is that sub-optimal information is currently being used when making planning decisions for river flood. Changes to rainfall regimes are discussed in more detail in the individual sub-sections within this report dedicated to the phenomena that produce them, such as TCs and ECLs. However, many extreme rainfall events are produced by other weather systems ranging from cut-off lows, monsoon lows, hybrid systems, coastal troughs and convergence zones in the trade winds. Antecedent conditions like clustering of events; hydrological processes (such as response to vegetation, mitigation measures and changes to watercourse characteristics); water management strategies; and land use changes, are complex and are not considered in this discussion.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 46 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
As mentioned above, from a property damage perspective, community flood risk is driven by very rare events that exceed the typical land planning threshold of 1% AEP. A paper by Dyer et al. (2019), highlighted that catchments with historically strong land planning controls are more sensitive to increases in flood intensity or frequency due to the accumulation of properties just above the 1% AEP. However, there is very little information and research on the extreme rainfall scenarios that drive floods of this rarity and beyond at catchment scales, despite Australia having a long history of devastating floods. The hazard component of financial riverine flood risk models requires numerical estimates of river flood AEPs. These estimates also need to be linked to related coastal and estuarine storm surge that can exacerbate near-coastal flooding. These are strongly influenced by sea level rise changes that incorporate regional variations to the global trends, along with adjustment factors to cater for sea level variations linked to the naturally occurring climate oscillations that could be in effect at the time of the storm surge. For river flooding, the general trend is for an increase in flood risk related to increases in daily and sub-daily rainfall intensity. The minimum rate of increase is in the order of 7% per degree Celsius of warming – acknowledging that the rate of increase for smaller catchments could be 15% or more per degree Celsius of warming if the rainfall is produced by discrete convective systems (Guerreiro et al. 2018). Only a small number of detailed dynamical studies has been conducted on the impact of future extreme rainfall events on Australian river flooding. For the larger and less populated Australian river catchments, a 7% increase per degree Celsius of warming is applied as the minimum likely future catchment-wide rainfall increase. Catchments which respond to sub-daily rainfall events could well experience significantly higher increases in catchment- wide rainfall (Wasko et al. 2016, Guerreiro et al. 2018). These catchments include many rivers flowing through the main urban centres, and most east coast river systems. For larger rivers, which respond to multi-day rainfall events, the predicted trends in 20-year flood return periods are available for a few rivers. They are assumed to also apply to rarer floods. The trends in return periods of floods for the larger river systems have some of the largest uncertainties of all the climate change-affected severe weather phenomena because the hydrological responses of the major river systems are affected by numerous factors (Sharma et al. 2018). In the lower reaches of several east coast river systems, the effects of rising sea levels and storm surge must also be considered. A recent paper by Dyer et al. (2019) outlined how this approach is applied to selected east coast river systems. The key factors relevant to the changing climatology of river flooding in Australia include the following (noting there are many more factors to consider including mitigation strategies, changing vegetation types and other demographic factors): 1. Natural variability across Australia has masked trends. However, Wasko et al. (2016) showed that total rainfall and maximum precipitation intensity are expected to increase with temperature, regardless of future changes in storm sizes. This will increase flood severity for the smaller catchments, but the impact is less clear for the larger catchments.
2. Natural climate variability can modulate forced climate change effects on river flooding.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 47 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
3. Increases in maximum extreme precipitation from hours to days will increase flood risk in the future. Extreme precipitation increases by ~7% per degree Celsius warming with sub-daily extremes and rarer events potentially having larger increases.
4. Therefore, small catchments (e.g., urban areas) will likely see stronger increases in flash flooding. Changes in flood frequency and severity are more uncertain in larger catchments.
5. Future work should focus on coupling high-resolution climate models with hydrologic models to better understand the hydrologic response to increasing extreme precipitation. Additionally, more research on changing risks in high-impact flood events (e.g., 1-in-100-year event) is needed to mitigate societal and financial risks to future flood events For the short east coast rivers that respond to sub-daily rainfall, all flood return periods are expected to become more frequent. These events are expected to vary by catchment according to their size and the meteorological situations that produce the more significant floods. Fowler et al. (2020) point out that as well as increased flood frequency, the flood volumes and peak flow rates should also increase, leading to non-linear increases in the damage produced by changes to extreme short- duration rainfall.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 48 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.3 Damaging Hail
Section Summary
In the November 2019 edition of this report, it was stated that large and giant hail risks should increase over central to south-eastern New South Wales, including the Australian Capital Territory, and the central to eastern parts of Victoria. New, independent research has strengthened these conclusions, and points to further marked increases in hail risk over the south-east of Queensland and north-east of New South Wales, which further exacerbates the multi-peril risks across this region. Insurance Council of Australia data shows hailstorms are the most frequently occurring damaging weather phenomena in Australia. However, inhomogeneous historical data have hampered the identification of trends in the occurrence of large and giant hail events across Australia. IAG analyses conducted in 2019 reveal crop losses for pea size hail and above. Motor claims commenced with hail in the 2-3cm range, with vulnerabilities varying with motor vehicle type and make. Building damage commenced with hail sizes of 4cm and above. Damage surveys indicate that Australian domestic properties can get cracked tiles and dented metal roofs (Parackal et al. 2014) when hail is approximately 5cm in diameter. The most rigorous Australian hailstorm climatology and trend analysis to date was conducted by Soderholm and Warren (2020 – personal communications). Their analysis was based on 20 years of eastern Australian radar observations. Preliminary results show evidence of a substantial increase in the frequency of damaging hailstorms across the major urban areas of eastern Australia, extending from the Sunshine Coast in Queensland down beyond Wollongong in New South Wales. These studies also confirmed the existence of strong inter-annual variability linked at least in part to the ENSO (Soderholm et al. 2017). There are emerging signs, backed-up by insurance claims records, that Melbourne may be experiencing an increase in damaging hail events. The rarity of large and particularly giant hail for cities such as Canberra, Perth and Adelaide makes the short available period of radar data inadequate to identify trends. However, historical insurance claims also identify an increasing trend of hail-related damage for these locations. Overall, insurance claims records indicate a southward extension of the primary hail risk regions of Australia.
5.3.1 Background
Insurance Council of Australia claims records show hail is the most frequently occurring cause of major damage across Australia. However, there is low confidence in long-term observed trends in hail distribution because of data inhomogeneities and inadequacies in monitoring systems (Field et al. 2012). The primary historical record of hail within Australia is the Australian Bureau of Meteorology’s severe storms archive. Reports in this database are not homogeneous in space and time with limited quantitative information on hail size and distribution. The information is heavily influenced by the location and population distribution across the country (Allen and Allen 2016). A more systematic
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 49 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
severe weather monitoring application, called WeatheX, has been implemented by CLEX partners, including Monash University, the Bureau of Meteorology and other research and industry partners including IAG. WeatheX is an attempt to, over time, build a more comprehensive database of severe weather events nationally. Details of the app can be found here: https://climateextremes.org.au/weathex/.
Observational Studies In recent years, within Australia, there has been some detailed analyses of volumetric weather radar data from the major population centres. This unique, approximately 20-year, record has already identified some key drivers of observed trends. The work of Warren et al. (2020) and Soderholm et al. (2017) leads the way, with the focus on eastern Australia. Soderholm et al. (2017) developed an 18-year hail climatology across south-east Queensland as a part of the Coastal Convective Interactions Experiment (CCIE) (Soderholm et al. 2016), limited to a 100km range around the Marburg weather radar. The study made use of the Maximum Expected Size of Hail (MESH) algorithm. They found there were three dominant synoptic situations conducive to the formation of damaging hailstorms for this region, with different spatial and temporal characteristics associated with each situation. Figure H1 summarises the key findings.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 50 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure H1 Analysis of hailstorm spatial distribution under different combinations of thunderstorm synoptic type (TST) (columns) and sea-breeze presence (rows). Averaged mean-sea-level pressure and 500 hPa temperature for each TST are shown in (a) – (c) and across all hailstorm cases in (d). Sea-breeze day subsets for each TST are shown in (e) – (g), with all hailstorms on sea-breeze days shown in (h). The spatial distributions of non-sea-breeze day subsets for each TST are presented in (i) – (k), with all hailstorms on non– sea-breeze shown in (l). The spatial distributions of hailstorms for each TST on days with and without a sea breeze are shown in (m) – (o) and the distribution of all hailstorms shown in (p). Locations Brisbane (BNE), Esk (ESK) and Boonah (BNH) are marked. Location of Marburg radar is shown with a black diamond and range rings given at 16, 40 and 80 km for (e) – (p) and study region outlined in (a) – (d). Source: Soderholm et al. (2017).
They also confirmed the existence of a strong relationship between the ENSO and hailstorm frequency for this region. El Niño phases tended to be associated with a greater number of damaging hailstorm days, while fewer hailstorms occur in La Niña phases. Warren et al. (2020) extended this study and expanded it to include the Greater Sydney region, using the MESH analysis tool. They found damaging hailstorms occurred on average 26 and 32 days per
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 51 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
year in Brisbane and Sydney, respectively, across the 8-year period of 2009-2015. These results were based upon a systematic analysis of volumetric radar data, including a thorough data homogenisation process to overcome known calibration errors of individual radar by using a stable satellite-based reference. Statistical analysis by Warren et al. (2020) determined that a MESH value of 32mm provided reasonable skill for discriminating regions with property loss. Preliminary work to extend this analysis to other capital cities is well underway, although results have not yet been published (Soderholm, 2020 personal communications).
Hail Related Damage Studies Within Australia, IAG has undertaken insurance claims-based studies to relate property (including motor, residential and commercial property) and crop damage, to their causes. Most of these analyses are not published. Hail sizes of less than 2cm can result in agricultural losses. For example, large quantities of small hail can cause crop losses, the extent of which depends on the stage in the crop growth cycle, and the type of crop. Analyses reveal crop losses commence once hail reaches pea size. Motor claims started when hail reach the 2-3cm range, with vulnerabilities varying with motor vehicle type and make. Building damage commenced in the 4cm and above range. Damage surveys indicate that for approximately 5cm hail, Australian domestic properties can sustain cracked tiles and dented metal roofs. The precise minimum size hail needed to crack solar photovoltaic panels has not been quantified. It is known that damage has occurred to these panels in several giant hailstorms events that affected each of the major capital cities. Wind and direction of approach also affect hail damage. The James Cook University Cyclone Testing Station report (Parackal et al. 2014) on the 2014 Brisbane hailstorm states “It was noted that the winds in the area had a significant influence on the direction of travel of hail and the impact on building performance. The most obvious evidence of this is that window damage far exceeded roof damage, suggesting that there was a significant horizontal component in the trajectory of the hail. Furthermore, windows on the windward side of the buildings (often to the south) were broken while windows on other sides were not. Even windows that were reasonably well-protected by awnings were broken, again indicating the effect of wind on hail trajectory. It was also noted that where there was tile damage from hail, this was often on steeper tiled roofs. Once again, this appears to illustrate that there was a significant horizontal component in the hail trajectory”. Once a roof suffers hail damage there may be ongoing water ingress damage for the weeks following the hailstorm, continuing until the damage is repaired. Occasionally damage from the storm may not be evident until there is a subsequent heavy rainfall event. This finding implies that any increases in thunderstorm-related wind gusts will also exacerbate hail damage. Changes in the approach angles of future climate thunderstorms could expose different regions and property mix to hail damage. The University of Queensland and the Bureau of Meteorology are continuing to use wind information from Doppler radar to explicitly model hail fall trajectories. This work will provide a new technique for quantifying the effects of wind on hailstorm impact angle (Brook, 2020 personal communication).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 52 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Modelling Studies Simulating hail in climate models is challenging because climate models are not able to resolve the processes (e.g. strong convective updrafts) that are necessary for hail development. Several approaches have been developed to derive hail risk estimates that either use hail observations and claims data (Changnon and Changnon 2000, Xie et al. 2008, Changnon 2009, Barthel and Neumayer 2012); remote sensing data (Witt et al. 1998, Cecil 2009); algorithms that relate large-scale environmental conditions to large hail development (Brooks 2009, Allen et al. 2015, Prein and Holland 2018); or high-resolution numerical models (Mahoney et al. 2012, Brimelow et al. 2017). Although there is considerable uncertainty, current indicators are that climate change will lead to a decrease in the frequency of small hail and an increase for large hail (see Figure H2: Mahoney et al. 2012, Brimelow et al. 2017, Dessens et al. 2015), mainly as a result of higher atmospheric melting levels in warmer climates. Prein and Heymsfield (2020) analysed reanalysis and radiosonde data to evaluate trends between 1979 and 2010. They found that there were significant trends in the height of the melting level and the WCL depth across Australia (see Figures EP4 and EP5 in Section 5.2). A change in melting level will affect the distribution of snow, damaging hail and heavy rainfall. In general, a higher melting level will lead to decreases in large hail – although not necessarily giant hail – and less snowfall. Increases in melting level across Australia for the warm season (December to May) were typically 30-60m/decade, while the increases in the deep tropical areas exceeded 140m/decade. Hail observations from the USA and Australia show that giant (> 50mm in diameter) damaging hail rarely occurs in environments with melting level heights above 4.5/3.8km because of enhanced melting. There is consensus that climate change is leading to a higher potential for extreme convective storms (Gensini and Mote 2015, Púčik et al. 2017, Rasmussen et al. 2017). Brimelow et al. (2017) used a cloud model that explicitly simulates hail embedded in climate model fields to investigate potential changes in hail risk over North America (Figure H2). They showed that the hail damage potential is increasing over southern regions in March, April and May and over northern regions (north of 50°N) and the Rocky Mountains in June, July and Figure H2 Frequency of spring hail over Colorado and August due to increasing buoyancy in the the High Plains. Green bars show the relative frequency future climate. of different hail sizes in the present climate (1979-2000) In the subtropical eastern and south- and red bars show the distribution in the future climate eastern regions of the USA, in contrast, (2041-2070) under the A2 business-as-usual scenario. Source: Brimelow et al. (2017). they showed a substantial decrease in hail
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 53 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
damage potential due to the increase in melting of hailstones. Recent advances in the explicit modelling of hailstone microphysics by Kumjian and Lombardo (2020) has confirmed the importance of weak horizontal flow and updraft geometry within the hail growth layer for large hail formation. This dynamical understanding is expected to lead to further improvements in climate model simulations of hail.
5.3.2 Changes in Australia
Observed changes It should be noted that in Australia “large hail” is defined as hail sizes from 2.0cm through to 4.9cm. “Giant hail” covers hail sizes of 5.0cm and greater. IAG’s claims experience shows that most damage to property and motor vehicles starts to occur once hail reaches 2.0cm in diameter, and the damage increases significantly as the size reaches then exceeds 5.0cm. Cyclone Testing Station damage surveys show hail of ~2.0cm damages skylights, air-conditioning units, and windows (mainly in old houses). Broken windows and skylights lead to significant water damage from the wind-driven rain. When hail reaches 4.0+cm, it starts to damage roof tiles and metal cladding. The IAG claims experience in Australia and New Zealand shows that large quantities of small hail – pea size and greater – can also produce significant damage to crops, and domestic and commercial buildings, the latter mainly through the blocking of gutters which leads to increased water ingress and, in extreme cases, roof collapse. Hail size, quantities and changes to storm tracks are therefore important considerations in any climate change risk assessments. It is necessary to consider potential changes in both large and giant hail, as well as trends in the frequency, duration and tracks of supercells that produce the greatest damage. Because Australia’s main urban areas are skewed along the coast, trends that increase hail risks in near-coastal regions exacerbate the damage potential of the hailstorms. Observed changes in large hail in Australia are strongly influenced by changes in observational practices and an increase in population density. Figure H3 shows the Bureau of Meteorology annual
Figure H3 Annual frequency of large hail (diameter larger than 2.5cm) from the Bureau of Meteorology Severe Storm Archive dataset (source Prein and Holland 2018).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 54 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
hail reports from 1979 to 2015; the general increasing trend of hail observations displayed is more likely to be caused by changes in the observation and storm spotter networks rather than an increase in underlying hail occurrence. Soderholm and Brook (personal communication, May 2020) have applied the calibration techniques from Warren et al. (2020) to the entire Australian radar archive. Using this record, they are undertaking comprehensive cell tracking for locations where there is sufficient available data to detect trends. For a couple of radars, the record length is around 20 years, albeit with some missing data periods. They have produced annual time series of hailstorm occurrence (using MESH thresholds > 30mm and 35dbz areas > 30km2). The analysis is ongoing, and allowances need to be made for periods of missing data. Preliminary results show a marked increase in hailstorm trends across a large portion of eastern Australia that requires further investigation. For most central east coast radars (Gympie, Wollongong, Marburg, Terrey Hills, Newcastle, Mt Stapylton) there has been a significant increase in the number of hailstorm cell hours and hail days. This signal is present across a diverse number of sites and radar hardware, suggesting that it is a robust trend for the region. The results for the more southern cities of Adelaide, Hobart and Perth are less clear. Trends are not yet identifiable in the volumetric radar data, which only dates back to 2010 for Perth’s Serpentine radar (the original radar dates back to 1997). However, the lower frequency and significant inter- annual variability of hailstorms mean these locations require longer periods of record to identify trends. The Melbourne records hint at an increasing trend in event days, although further work is required before this can be considered a real trend. IAG’s claims data, combined with the Bureau of Meteorology’s Severe Weather database and old media reports, highlights that major damaging hail events for Perth, Adelaide and Melbourne have been increasing in the most recent decade. When looking at the areas with the biggest impacts, there are no historical analogues prior to these major events, and these areas have been well-populated for the past century or more. For Perth, the March 2010 hailstorm set a record for giant hail within the greater Perth region. Although newspaper records are available for the past 140 years, unfortunately, the high-quality radar record only dates to this event. For Adelaide, the November 2016 large hail event set a record for hail size in the greater Adelaide area. In Melbourne, the largest hail event on record occurred in March 2010 with additional giant hail events occurring on Christmas Day 2011, December 2017 and January 2020. Finally, Canberra and the Australian Capital Territory experienced the largest and most damaging hailstorm on record in January 2020. Southern Sydney also experienced giant hail from this event, making this the first recorded event to affect three Australian capital cities. All these events produced damage that surpassed the damage from previous large hail-producing thunderstorms by very large margins. Sydney and Brisbane have experienced multiple large-to-giant hail events throughout their history, so the trends observed by Soderholm (personal communication, 2020) and Warren (2019) are particularly concerning as these cities have experienced several billion-dollar events in the past. In an outbreak without precedent in its long historical record, Sydney has recently experienced five major hailstorm events: an event with large volumes of small hail on Anzac Day (25 April) 2015; a large hail event through the northern suburbs in February 2017; a large-to-giant hail event centred on Campbelltown on 15 December 2018; a multi-cell giant hail event across large parts of Sydney on 20 December 2018; and a giant hail event across southern suburbs in January 2020. However, the April
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 55 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
1999 giant hailstorm across eastern Sydney remains the costliest Australian hail event on record and likely produced the largest reported hail size in the Greater Sydney region. In New Zealand, the South Island town of Timaru experienced New Zealand’s most damaging ever hailstorm by a large margin in November 2019. Allen and Karoly (2014) showed that ENSO has a significant impact on the distribution and frequency of severe thunderstorm environments in Australia, and this was confirmed by Soderholm et al. (2017) for south-east Queensland. This also agrees with the assessments of Prein (2019) who showed a significant increase in large hail risk in south-eastern and south-western Australia during El Niño events. However, there remains uncertainty in the details of how ENSO modulates severe convection, including hail across Australia.
Projected changes Little research has been done to investigate potential future changes in Australian hail and severe convective storm risk. However, the techniques identified in Prein et al. (2017b) and Warren et al. (2020) could be applied to climate models to use hail environments as a way to track future changes in the frequency of hail events. It should be noted that insurance records highlight the marked dominance of warm-season hail events over cool-season events for major damage-producing storms across Australia. Allen et al. (2014a, b) assessed changes in hail environments in two coarse-resolution GCM simulations. They showed that severe thunderstorm environments significantly increased in both the models for northern and eastern Australia by the end of the century under an RCP8.5 scenario. This is in response to an increase in the convective available potential energy (CAPE5) from higher moisture and warmer SSTs in the proximity of hail environments. An increase in CAPE is a likely consequence of climate change (e.g. Romps et al. 2014, Gensini and Mote 2015, Prein et al. 2017b). However, other environmental changes, such as changes in wind shear and freezing/melting level heights, are also essential for future hail risk determinations, as are the effects of topographical and meteorological triggers. A more systematic analysis of hail environments than in Allen et al. (2014a, b), with multiple GCMs at higher spatial and temporal resolution, is necessary to assess future changes in hail risk and uncertainties. Advice from the Bureau of Meteorology indicates they now have environmental diagnostics relevant for hail from 12 CMIP5 GCMs, one CMIP6 GCM and BARPA (downscaled from CMIP5 at ~12 km resolution). This includes wind shear, melting level height and SHIP (Significant Hail Parameter) with the potential to add other diagnostics. In time, trends may be identified with more confidence. To estimate changes in hail risk on local scales, high-resolution climate models are necessary to resolve land-sea contrasts and topographic gradients that are important for hail development (Soderholm et al. 2017). Leslie et al. (2008) used a 1km climate model over the Greater Sydney Basin to assess changes in hail risk between 2001 and 2050 for an RCP4.5 scenario. They showed significant increases in the key characteristics of severe hailstorms over this region.
5 CAPE is the amount of energy a parcel of air would have if lifted a certain distance vertically through the atmosphere. CAPE is effectively the positive buoyancy of an air parcel and is an indicator of atmospheric instability, which is used as an indication of severe weather potential.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 56 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.3.3 Regional Interpretations for Risk Assessment for Australia
This section focuses on the large and giant hail components of severe thunderstorms as these are the dominant drivers of severe thunderstorm risk in Australia. It must be noted that defining hail risks is far more complex than simply tracking changes in hail size. For example, the physical storm footprint and associated damage are markedly different between a short-lived pulse thunderstorm and a long-lived supercell thunderstorm, even though they may both produce the same hail size. Moreover, other important risk drivers such as severe wind squalls, intense rainfall and related water ingress and flash flooding also need to be considered. Projected changes in damaging hail over Australia, defined as hail of 2.0cm or greater, are highly uncertain due to a lack of research and deficiencies in validation data, as discussed. Climate change assessments in this report are therefore primarily based on results from other regions, particularly the USA, and large-scale drivers across Australia that are related to damaging hail occurrence. The factors concerning the changing climatology of damaging hail in Australia include: 1. The future climate trend projections were determined by interpreting a range of disparate data, including insurance claims data combined with meteorological reasoning.
2. Emerging research using volumetric radar data spanning up to 20 years by Soderholm and Brook (personal communication, 2020) suggests a significant increase in damaging hail events for a region extending from the Sunshine Coast in Queensland down beyond Wollongong in New South wales. There are tentative signs that Melbourne may be experiencing an increase in damaging hail events with insurance claims records pointing to this increase being real. The rarity of large and particularly giant hail for cities such as Perth and Adelaide makes the short available period of radar data inadequate to identify trends. Further work is required to assess the long-term radar datasets for the Perth and Adelaide regions.
3. Studies also confirmed the existence of substantial inter-annual variability linked at least in part to the ENSO (Soderholm et al. 2017, Soderholm and Brook personal communication 2020).
4. Incomplete observational hail data were supplemented by extracting trends in hail environments from the ERA-Interim reanalysis dataset (Prein and Holland 2018). The ratios between observed large and giant hail for Bureau of Meteorology forecast districts – where there are sufficient observations – were also considered. This analysis was provided by Prein (2019) based upon his analysis of the Australian Severe Thunderstorm database.
5. The IAG-sponsored Greater Sydney hail modelling work of Leslie et al. (2008) was applied to the Greater Sydney and surrounding regions out to the year 2050. It serves as a guide for the +2°C hail risk estimates for this region. Consideration is also given to the work of Brimelow et al. (2017) for North America which showed a predicted trend to larger hail sizes in continental and mid-latitude regions and decreases in subtropical environments under the RCP8.5 climate change scenario due to the increasing heights of the melting levels in these regions.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 57 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
6. IAG’s increasing hail claims experience for the combined Perth, Adelaide, Canberra and Melbourne areas is used as tentative evidence of the likely southern shift in the areas of greatest hail risk. It is noted that much longer lengths of record are needed to derive statistically significant results. It is also important to note that by the time there is sufficient length of claims data to demonstrate statistical significance, the changes will have been in effect for a decade or more.
The main trend imposed on the current Australian climatology of damaging hail is to increase the hail risk for the +2°C scenario for the south-east Queensland and coastal New South Wales regions and shift the hail risk regions southwards down the eastern coast and ranges, particularly for the +3°C scenario. The impacts would also lead to increased damage to crops across this region. These changes are primarily based on:
• The preliminary trends observed in the volumetric radar dataset by Soderholm and Brook (2020) while acknowledging the limitations in this dataset.
• The predicted southward shift of the subtropical jet stream.
• The general increases expected in atmospheric instability (CAPE) with increased heating and stronger updrafts expected in severe thunderstorms in a warmer environment.
• Changes in thunderstorm micro-physics should reduce large hail in tropical and near-tropical areas with lesser effects on damaging hail southwards.
• This effect should progressively shift the highest hail risk regions southwards in the +3°C and higher warming scenarios.
• Changes in thunderstorm triggers in a warming climate linked to the southward shift of the subtropical ridge axis and deepening of the eastern Australian inland trough.
• Observed tentative trends in large and giant hail using IAG’s claims experience across southern regions, while acknowledging the limitations in this data.
Figure H4 summarises the regional changes to large and giant hailstorm frequencies across Australia between the base climatology and the approximately +3°C climate change scenario. Based on large-scale drivers, the largest projected changes to the current climate hail climatology are likely to be concentrated over south-eastern Queensland, north-eastern, eastern and south-eastern New South Wales, including the Australian Capital Territory, and eastern Victoria. The regions that dominate the relative rates of increase in the future are those with historically fewer damaging hailstorms, i.e., Melbourne; the mountainous parts of eastern Victoria; the southern to central ranges and coastal plains of New South Wales; and the Australian Capital Territory. These trend estimates are consistent with the predicted deepening of the east coast trough; southward shift of the north-east New South Wales to southern Queensland dry line; and increased coastal moisture availability from the observed rapidly warming East Australian Current. The southward shift of the subtropical ridge axis could also affect the location, frequency and intensities of
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 58 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
the severe thunderstorm-triggering coastally-trapped southerly changes that frequent the New South Wales and south-east Queensland coasts. Increases are also applied to the warm season hail events that affect agricultural and west coastal areas of south-west Western Australia. However, the starting point is from an inadequate number of historical events, so trends cannot be identified with confidence. Here the west coast trough is expected to deepen and lead to a southward shift of the location of the coastal crossing point when the trough is at its deepest. This is typically when it triggers severe thunderstorms, due to increasing low level shear, coupled with rapidly-rising maximum temperatures and increased moisture availability from the warming Leeuwin Current off the west coast of Western Australia, and the warmer seas off the Pilbara and Gascoyne coasts. An expected higher hot season melting level in this region could partially offset other factors. However, this is only expected to reduce the occurrence of smaller size hail. The impact on damaging large and giant hail is less clear. There are tentative indications that the steering flows of severe thunderstorms are becoming more meridional during the warmer seasons. Studies available for other parts of the world indicate these risks are most likely to increase. These increases might be partially offset by reduced hail risks for large parts of tropical and central Australia, including most of Queensland where increased daytime heating and CAPE could be offset by the rising atmospheric melting levels, based on research from the USA (Brimelow et al. 2017).
Figure H4 Schematic graphs showing the relative change in large and giant hailstorm frequency between the 1950s and the +3°C climate change scenario.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 59 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.4 East Coast Lows
Section Summary
ECLs are weather systems that periodically bring heavy rain and/or high winds to eastern Australia. The most significant impacts occur near the coast due to topographical enhancement of rainfall that can last for 24 hours or more. There are several disparate definitions as to what constitutes an ECL. Therefore, when drawing conclusions from any particular study, the nature of the lows included need to be considered. Their impact range is typically from Fraser Island to eastern Victoria and Tasmania.
Observational and modelling studies point to a decline in the numbers of winter ECLs. There is less certainty on the trends of the subset of ECLs that produce the most damage - namely the warm cored or hybrid ECLs that have intense mesoscale vortices near the coast and produce intense lines of extreme rainfall producing thunderstorms. They may also produce extreme winds and damaging high energy wave trains from directions different to those experienced in the more commonly occurring types of lows.
Insurance claims data highlights a marked increase in the frequency of the most damaging ECL events during the most recent decade. Four of the six most costly ECLs observed over the past 45 years occurred in the most recent decade. What may have been the most damaging ECL during the past 220 years occurred in 1867, and there are no future climate studies available to determine the most likely trends in these rare but most extreme events.
5.4.1 Background
ECLs are weather systems that periodically bring heavy rain and/or high winds to eastern Australia, with the highest impacts near the coast. There are several classification methods used by researchers (Dowdy et al. 2019), which leads to substantial differences in their climatology. They typically affect areas from Fraser Island to eastern Victoria and Tasmania. Other nearby areas may experience similar systems. ECLs are noted for their extreme characteristics and high impacts. These characteristics have been described in several studies (e.g., Holland et al. 1987, Hopkins and Holland 1997, Mills et al. 2010, Dowdy et al. 2013a, Pepler et al. 2014, Louis et al. 2016, Dowdy et al. 2019, Cavicchia et al. 2020). These include:
• Heavy rains on the poleward side, resulting from moist tropical air being advected around the low and into the coastal region. These rains may extend for some distance down the coast, but the most damaging extreme rains are often constrained to a narrow rainband (~100km across) and are marked by a sharp transition to fine conditions to the north of the low centre.
• Damaging winds that may exceed TC force and are typically located in a narrow (usually <100km) zone poleward and westward of the centre. The Sygna ECL produced wind gusts up
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 60 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
to 170 km/h at Nobbys Signal Station, Newcastle, equivalent to that experienced in a Category 3 TC.
• Localised ocean currents and waves may cause considerable damage to coastal foreshores, including structures. The direction of approach of the highest energy waves is a key determinant on the degree of coastal erosion and inundation, as demonstrated by the severe coastal erosion event at Wamberal in July 2020 associated with a physically very large and intense central to northern Tasman Sea low.
• Storm surge is rarely much over 1.0m, as the primary damaging wind direction is along the coast – the exception may be from surges produced by a small, intense low right on the coast when wind direction varies from the more common directions.
• ECLs have a wide range of sizes and lifetimes, from small, intense storms that last 10-15 hours (e.g., the Sygna storm, Bridgeman 1986) to large, long-lived systems (e.g., the Pasha Bulker ECL of June 2007, Mills et al. 2010, and Sydney ECL of April 2015) that may affect the east coast for days.
• They are often accompanied by rapid intensification, some of which satisfies the ‘bomb’ criteria of Sanders and Guyakum (1980). There is radar evidence that the most damaging ECLs have produced mesoscale cores (see Figure EC1) similar to the eye of TCs that have severely affected coastal communities (Pasha Bulker ECL 2007, April 2015 ECL).
Figure EC1 Weather radar imagery from the Williamtown / Lemon Tree Passage radars showing the mesoscale cores of the Pasha Bulker ECL June 2007 (left) and April 2015 ECL (right). Source: Mills et al. (2010), Bureau of Meteorology.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 61 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
The conditions associated with the development of intense ECLs may include (Figure EC2):
• A characteristic cut-off low and/or split jet configuration (e.g., Holland et al. 1987, Dowdy et al. 2013a, 2019).
• The presence of a strong anticyclone or ridge on the poleward side, with the ECL cradled in an Easterly Dip being one of the more common configurations (e.g., Holland et al. 1987).
• Development of intense low centres over an area of strong SST gradients associated with eddies along the East-Australian Current (typically >4°C in 50km) (e.g., Hopkins and Holland 1997, McInnes et al. 1992, Buckley and Leslie 2000, Chambers et al. 2014, 2016).
• Pepler and Dowdy (2020) show that the vertical structure of ECLs is also important when assessing their impacts. Deeper ECLs are more likely to produce extremely heavy rainfall during their developmental phase. ECLs with only marked surface features are more likely to be associated with damaging winds and heavy rainfall.
5.4.2 Climatology
Three known observational databases have been developed, and two of these are still available (Speer et al. 2009, Pepler and Coutts-Smith 2013). However, a systematic study of these systems is hampered by the variety of definitions employed. The Bureau of Meteorology defines ECLs as intense lows off and near the coast (http://www.bom.gov.au/New South Wales/sevwx/facts/ecl.shtml). More objective definitions include the following:
• Holland et al. (1987) introduced three types of lows based on their basic characteristics and development mechanisms;
• Hopkins and Holland (1997) introduced a fourth type of low, then focused on Type 2 ECLs, which they defined based on their rainfall and wind characteristics;
• Speer et al. (2009) defined six types of Maritime lows based mainly on their Figure EC2 Schematic of the environmental flow surface characteristics (these included around one type of ECL: Solid lines indicate the systems that would not typically be surface pressure pattern; darker (lighter) shading considered an ECL); and indicates convective (stratiform) cloud; the yellow arrow indicates subsiding dry air flow; and the blue
arrows indicate rising, moist tropical air. Modified • Cavicchia et al. (2020) applied a from Holland et al. (1987). characterisation scheme based on
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 62 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
dynamical structures. They applied three classifications to climate models to allow for a more systematic analysis – cold-cored, warm-cored and warm-cored hybrid cyclones.
Others have developed automated ECL identification methods based on various surface pressure characteristics (Browning and Goodwin 2013, 2016, Dowdy et al. 2011, Murray and Simmonds 1991). Dowdy et al. (2013a, 2013b) defined ECLs empirically based on upper-level characteristics known to be associated with their development. Pepler et al. (2015) compared some of these automated systems and used three such systems in considering potential future changes. All of these latter approaches are not capable of identifying the small, intense mesoscale structures within the broader ECLs that have historically produced the greatest damage, despite their small size (Figure EC2). Dowdy et al. (2019) provides an overview of the range of classification approaches that are, or have been, in use and used this to produce a generalised definition of ECLs and Intense ECLs, while noting a range of subtypes can be useful for various specific purposes. The definition is as follows: ECLs are cyclones near south-eastern Australia that can be caused by both mid-latitude and tropical influences over a range of levels in the atmosphere; Intense ECLs have at least one major hazard associated with their occurrence, including extreme winds, waves, rain or flooding. The variety of definitions is reflected in assessments of the impacts of such storms. While there is no doubt about systems with major impacts such as the 1974 Sygna Storm (Bridgeman 1986) or the 2007 Pasha Bulker Storm (Mills et al. 2010), different studies have attributed a wide range of characteristics to ECLs based on measures of damage produced (Hopkins and Holland 1997, Speer et al. 2009, Callaghan and Power 2014). Key points to note include:
• Depending on the method used, the estimated number of systems per year can vary from 2.5 to 22. The lower numbers of ECLs generally relate to those ECLs that cause significant damage. The larger numbers refer to a wider range of low types, many of which occur far enough off the east coast of Australia to produce limited or no damage over land.
• The proportion of east coast floods or heavy rains caused by ECLs has been assessed to vary from 16% to 57%.
• Some studies include transitioning TCs as ECLs; others have them as separate systems. Given this level of ambiguity, the techniques developed by Cavicchia et al. (2020) and Pepler and Dowdy (2020) are welcome as these are consistent approaches applicable to both reanalyses and climate change modelling studies. A useful next step would be to evaluate their over-land impacts to enable better assessment of future trends in risks to near coastal communities. Despite the concerns expressed above, some useful climatological information is available.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 63 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Seasonal occurrence ECLs (Figure EC3, left) exhibit a maximum occurrence in winter (Hopkins and Holland 1997, Callaghan and Power 2014) and this is also reflected by a secondary peak in deaths from freshwater flooding (Figure EC3, right). This overall seasonal variation may be regionally- and system- dependent. Hopkins and Holland (1997) found the maximum occurred in summer in a small region along the east coast of New South Wales. The maximum from tropical transition also occurs in summer (Speer et al. 2009, Callaghan and Power 2014).
Figure EC3 Seasonal ECL occurrence variation. Left panel: Monthly occurrence of ECLs, TIs (Tropical Interaction – essentially transitioning tropical lows or cyclones) and Major Floods (the sum of the previous two). Right panel: bimodal peak in deaths associated with major freshwater flooding events. Modified from Callaghan and Power (2014).
Interannual and interdecadal variation There is good evidence for a close association between ECLs and related flooding and freshwater drownings, with the ENSO cycle (Figure EC4, Chiew et al. 1998, Power and Callaghan 2016, Browning and Goodwin 2016). Micevski et al. (2006) and Power and Callaghan (2016) also found a longer period fluctuation in phase with the IPO, which they concluded that this resulted in long-period variations in decadal occurrences of 7 to 26 floods, 2 to 15 ECLs and 3 to 114 deaths.
Figure EC4 Increased occurrence of coastal floods,
east coast lows, Tropical Interactions and related deaths during El Niño and La Niña years. Source: Power and Callaghan (2016).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 64 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.4.3 Changes in Australia
Observed changes Hopkins and Holland (1997) found a significant increase in the number of ECLs between 1958 and 1992. This increase is most likely due to natural fluctuation in frequency. This trend was also not reproduced in the modelling study by Browning and Goodwin (2016). Hopkins and Holland (1997) also found no trend in general heavy rain occurrences and a decrease in the frequency of extreme convective rain events. By comparison, Powers and Callaghan (2016) found a significant (50%) increase in the frequency of major floods since 1860 arising from ECLs (Figure EC5). Some of this increase will be due to increasing populations and improved observing systems. Curiously, a closer examination of the changes indicates that rather than a linear trend, there was a marked shift upwards around 1950 followed by nearly constant, or perhaps declining, frequency. This agrees with Franks (2002), who found a notable climatic change from low to high flood frequency around 1945. Pepler et al. (2016) found a small decline in the frequency of ECLs using the Twentieth Century Reanalysis. Ji et al. (2017) used three different tracking algorithms and found that the decline mainly occurred in winter with an increase in early spring. It is important to note that changes in overall frequencies may not correspond well to the changes in the major hazards they pose. The extremely damaging ECLs (those that have the potential to produce a billion dollars or more of damage) tend to occur on decadal to multi-decadal time scales (Sygna ECL 1974, Pasha Bulker ECL 2007, April 2015 Newcastle ECL). Hence the historical record is insufficient to identify trends in these rarest events.
Figure EC5 Annual frequency of ECLs (black) and Tropical Interactions (grey) from 1860. Modified from Power and Callaghan (2016).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 65 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Guerreiro et al. (2018) found that the rate of increase in intense rainfall with durations of an hour have been observed to be two to three times the expected rate of increase using Clausius-Clapeyron scaling (which is based only on the increased water-holding capacity of air with increased temperature). These findings were based on a study of observed extreme hourly rainfall changes between 1966-1989 and 1990-2013. The projected increase in rainfall rates at hourly timescales are in the order of 15% per degree Celsius of climate warming – more than double the 7% increase expected for wider-area rainfall, spanning a day or more. So even if the total number of ECLs declines over time, those that do occur may be expected to have increasing potential to produce greater flash flooding, fast response river and stream flooding, and wind-water ingress damage.
Projected changes Future variations associated with anthropogenic climate change have been assessed by Dowdy et al. (2013b, 2019), Pepler et al. (2016) and most recently by Cavicchia et al. (2020). The consensus of the available studies is for a significant decrease in the frequency of ECLs during the 21st century. However, these assessments show the declines are most pronounced for ECLs that occur during the cooler months of May to October. Historically the greatest observed damage to Australian property and loss of life has been associated with the less common warm-cored or hybrid ECLs. The strongest observed winds tend to occur with warm-cored ECLs, followed by hybrid ECLs. Warm- cored ECLs also have a strong tendency to produce the heaviest rainfall. Cavicchia et al. (2020) separate out trends in the cold-cored, warm-cored and hybrid-cored ECLs using a suite of 12 down-scaled climate models under the IPCC A2 scenario and compared the period 1990- Figure EC6 Box plots of relative changes of ECLs in the region 25oS to 2009 with a future climate 40oS, 150oE to 160oE for a 12-member climate model ensemble period 2060-2079. Their showing differences between the period 1990-2009 and 2060-2079 for research findings predict a an A2 scenario for the whole year (left), cool season (May-October – decline in both warm-cored centre) and warm season (November-April – right). The boxes show the and cold-cored ECLs, with inter-quartile range, the solid line the ensemble median, and whiskers the most extreme data points. Source: Cavicchia et al. (2020). the largest decline in the cooler months. Hybrid
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 66 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
cyclones showed a tentative increase in the warmer months, although also declining in the cooler months (see Figure EC6). The geographical trends in future climate ECLs were found to be inhomogeneous across the analysis domain. The main declines were further into the Tasman Sea with onshore ECLs – those forming over the rapidly-warming East Australian Current – showing evidence of some increasing trends, mostly through the warm season (Figure EC7), noting there remain substantial climate model uncertainties. It is the mesoscale onshore features of ECLs that have been observed to produce the greatest damage through heavier, short-duration rainfall and stronger wind gusts. The more synoptic- scale characteristics off the coast tend to be associated with higher storm surges, wave impacts and coastal erosion. Therefore, future climate projections need to consider both onshore and offshore structures when assessing changing impacts under a warmer climate. Although overall numbers of ECLs across the western half of the Tasman Sea appear most likely to decline in a warming climate, the more damaging aspects of ECLS may not decline. Model predictions show increasing wind speeds for warm-cored and hybrid ECLs, along with increasing rainfall, meaning the magnitude of their impacts are more likely-than-not to increase (Dowdy et al. 2019). To date, model predictions of changes to maximum intensity have been indeterminate.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 67 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure EC7 Relative changes in the ECL track densities between the historical (1990-2009) and future (2060- 2079) climate simulations for an A2 scenario, for the three classes of ECLs for the cool and warm seasons. Stars indicate where 75% or more of the model members agree on the sign of the change. Source: Cavicchia et al. (2020).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 68 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Rising sea levels are likely to exacerbate coastal and estuarine flooding and erosion associated with ECL events significantly. Further research focusing upon the more extreme events and their impacts is required to clarify more accurately what these future impacts are likely to be and their geographical variations along the coast and ranges.
5.4.4 Regional Interpretations for Risk Assessment for Australia
Risk assessment models for ECLs needs to cover all wind, rain and oceanic-related damage drivers and rely on consistent definitions for each type, and climatology databases as described earlier. The classification scheme used by Cavicchia et al. (2020) offers a more structured approach than those of earlier schemes and highlights inhomogeneities across the Tasman Sea. Future analyses will need to be undertaken at resolutions high enough to resolve the most damaging elements of ECLs, particularly the smaller subset with intense mesoscale structures close to the coast. The estimated changes to ECLs for risk assessment purposes need to be based solely on those types known to produce significant damage and/or river and flash flooding. Insurance claims from the Insurance Council of Australia and IAG’s claims history show a marked increase in damage from catastrophic ECL events – those that have produced insured losses of $250 million or more (normalised to 2017 dollars). The near-shore mesoscale structures of these ECLs were critical in causing the associated damage. These most destructive ECLs are listed in Table EC1. Historically the most destructive type of ECLs has occurred with frequencies rarer than once per decade. In the most recent decade, there have been four of these events. Until climate models can resolve the near- coastal damaging mesoscale structures of ECLs, extreme caution needs to be applied to risk modelling based on trends in all ECLs across a much broader region out to 160oE.
Table EC1 ECL events that have produced insured losses of $250 million or more (normalised to 2017 dollars) based on Insurance Council of Australia claims data, supplemented by IAG claims data scaled up in accordance with IAG’s market share.
Date Approximate Normalised (2017 dollars) Insured Loss
May 1974 ~$1,000 million
Feb 1978 $269 million
June 2007 $2,200 million
April 2015 $1,060 million
April-May 2015 $250 million
June 2016 $422 million
February 2020 ~$900 million
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 69 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
The more common, but predominantly offshore, types of ECLs are of less interest as they result in little or no damage. Trends in these non-damaging ECLs can mask the more important but harder to identify trends in the rarer and very damaging ECLs. The ECLs of greatest interest tend to be most closely aligned to those described in Holland et al. (1987) and Mills et al. (2010). Trends in “bomb” ECLs, along the lines of Sanders and Guyakum (1980), are particularly relevant. It is noted that none of the future-climate modelling studies has been completed at a high enough resolution to resolve the intense small-scale secondary cores within the larger ECLs – i.e., the structures that typically produce major damage, as observed in the Pasha Bulker ECL in 2007 and the April 2015 Newcastle ECL. The modelled changes in the intensity and duration of their associated rainfall and their projected future increases are also not in a form that is usable for quantifying trends in risks.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 70 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.5 Bushfire
Section Summary
Bushfire risk is different to other weather-related natural disasters because it relies on several biophysical determinants which include fuel quantity, fuel moisture content, weather conditions and an ignition source. Humans influence bushfire risks extensively through fuel management, active and passive suppression activities and a range of built environment strategies. Therefore, real trends in bushfire risk in a warming climate extend well beyond meteorological trends. Extensive historical building loss data indicates the most extreme loss events for properties occur when the FFDI reaches or exceeds values of 75 (Extreme FFDI Category). These property losses coincide with FFDIs well in excess of the 99th percentile for those locations. An IAG study looking at the ratios between the most extreme fire weather conditions – the 99th percentile FFDI – for the 1960s compared to the 2010-2019 decade found the largest increases across eastern and south- eastern Australia with a very large increase along the New South Wales south coast. The prolonged extreme fire weather across southern and eastern Australia in spring to summer 2019-2020 was unique in both the extent and persistence of extreme fire weather conditions. Many locations with individual FFDI values exceeding 100 (Catastrophic FFDI). There are signs that the interval between the most devastating fires in Australia is shortening, noting that this does not mean that every year will have catastrophic bushfires. A 2019 National Environmental Science Programme study using an RCP8.5 scenario and 12 downscaled regional climate model runs (NARCLiM) showed significant increases in FFDI over 25 (Very High) for virtually all of Australia for 2060-2079. Another NARCLiM study incorporating upper-level instability through an index called C-Haines found there were statistically significant trends in the mean number of days that C-Haines and FFDI were >=8 and >=25, respectively for both summer and spring. However, the largest increases were during spring. The increasing length of the fire season could reduce or shift the window of opportunity for fuel-reduction burning to winter or shorter periods in spring. Increased fire seasons will also adversely affect shared resources between Australia and the USA.
5.5.1 Background
Bushfire risk to property is determined by a complex number of biophysical determinants, including fuel quantity, fuel moisture content, weather conditions and an ignition source. Local topographic influences are also very important, although secondary to the prevailing fire weather conditions in the event of an extreme bushfire. The influence of people on bushfire risk is many-fold, ranging from being one of the sources of ignition, through to fuel management, prescribed burning programs, and other vegetation management or agricultural practice changes. Other important factors are community education and preparedness activities, active fire suppression activities during bushfires, and advance bushfire risk management techniques using tools such as land planning, building codes and retrofitting/self-contained sprinkler installations.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 71 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
The Country Fire Authority (CFA) and Fire Indices Department of Sustainability and Environment (DSE) conducted a study The McArthur Forest Fire Danger Index (FFDI: into the sources of Victorian fire ignition McArthur 1967), commonly used to monitor fire weather types. The study included over 23,000 in Australia, is based on daily temperature, wind speed, fires that occurred between 1997 and humidity, and a drought factor (Lucas 2010). 2009. They found 35% were caused by arson, 19% were escaped fires, 5% Vertical instability, as measured using an index called were accidental ignitions, 5% were C-Haines, has also been shown to be an important lightning ignitions, 5% were started by factor affecting bushfire severity, particularly in spring machinery and the remainder (~30%) and summer, in south-eastern Australia (Di Virgilio et al. had unknown causes (Penman et al. 2019). 2014a). However, despite being one of many sources of bushfire ignition, a study by Dowdy and Mills (2012) showed that lightning accounted for 90% of the total area burnt by fires. They attributed this to the fact that lightning-induced bushfires tended to occur more evenly across the landscape in native vegetation rather than being close to roads and urban areas where more rapid-fire containment activities are possible. This makes lightning a very important contributor to bushfire risk in the Australian landscape. Further research by Clarke et al. (2019), investigating anthropogenic and lightning-caused wildfire in south-eastern Australia, found that fire weather conditions are highly correlated with the likelihood of fire ignition. This is particularly true for lightning (57%) and power transmission line ignitions (55%), while the built environment tends to be the largest driver for human-related ignitions. Climate change can either directly or indirectly influence many of these factors in complex and interacting ways. These factors, and strategies to address them, are being studied widely, with new research appearing each year. Due to the complexity of bushfire sources, the main focus of this section will be the weather and climate-related components of bushfire risk, particularly the most severe end of the spectrum as this is where most of the property loss and fatalities occur (Toivanen et al. 2019). A study by Blanchi et al. (2014) revealed that over 50% of all fatalities occurred on days where the FFDI exceeded 100 (the current threshold for declaring a day as ‘catastrophic’). For these days, over 60% of all fatalities occurred within structures, and 78% occurred within 30m of the fuel source.
5.5.2 A Global View
The complex interactions between fuel, weather conditions, and ignition sources make it challenging to simulate all the factors relevant to defining bushfire risk. In an observational study covering 1979 to 2013, Jolly et al. (2015) showed that fire weather seasons have lengthened by 18.7% globally, resulting in a doubling of the global burnable area affected by long fire weather seasons. In their 2009 review paper, Flannigan et al. stated that climate change would lead to a general increase in area burned and fire occurrence. Still, the spatial variability of this result is large. Moritz et al. (2012) used global climate models (GCMs) to project fire risk for 2010 to 2039. They found an increase in fire risk in warm climates and a small decrease in tropical and subtropical climates. However, GCM agreement for this result is low, indicating large uncertainties in changing fire hazards for small global warming increases.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 72 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.5.3 Observations and Trends for Australia
Tracking trends in bushfire risk involves identifying and tracking all the contributing factors. These include fuels, drought/dryness, fire weather, ignition sources, and a range of advanced fire mitigation factors, including the ability to conduct adequate controlled burns in peri-urban areas (Clark and Evans 2019). To date, fuel quantities, types, and moisture content changes have not been tracked in a detailed, homogeneous fashion. However, efforts are underway to quantify current conditions better, then reconstruct trends backwards in time. For example, Haverd et al. 2018 investigated the rate of fuel accumulation following a fire event. Increasing attention is being given to understand and provide better future climate projections of the responses of ecosystems, notably vegetation. For example, Medlyn et al. (2015) conducted ecosystem-scale Free-Air CO2 Enrichment (FACE) experiments. In particular, the University of Western Sydney has commenced the first Australian focused research project to identify the effects of rising carbon dioxide levels on Australian eucalypt forest; details can be found here: https://www.westernsydney.edu.au/hie/EucFACE. Results from these studies have begun to emerge (e.g. Jiang et al. 2020) with more to become available in the coming years. Fuel moisture trends to date have primarily relied on general empirical relationships. However, research has commenced to identify and quantify trends in fuel moisture levels in native forests. The results of these studies will emerge over the next few years. A preliminary study by Boer et al. (2020) revealed an upward trend in eastern Australian temperate forest area that is in a critically dry state (Figure BF1). A comment in Nature by Sanderson and Fisher (2020), indicates that the simplified views of changing fire risks in global climate models, which focuses on selected weather parameters, may be under-estimating the future bushfire risks, as many of the complex factors affecting bushfire risks are currently not adequately captured in current-generation global climate models.
Figure BF1 Forest area in critically dry fuel state, eastern Australia (1990–2019). Annual variation in the duration and cumulative area of large forest patches (>100,000 ha) in a critically dry fuel state. The horizontal black line indicates the 30-year mean value; light and dark grey bands indicate mean value ± 1 and 2 standard deviations, respectively. To identify forest areas in a critically dry state, spatially explicit predictions of fine dead fuel moisture content (DFMC) were based on gridded daily vapour pressure deficit, and a threshold of DFMC <10% was used. Source: Boer et al. (2020).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 73 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
The FFDI commonly used to monitor fire weather in Australia is based on daily temperature, wind speed, humidity, and a drought factor (Lucas 2010). Vertical instability, as measured using an index called C-Haines, has also been shown to be an important factor affecting bushfire severity, particularly in spring and summer, in south-eastern Australia (Di Virgilio et al. 2019). The seasonal 90th percentile FFDI trends show increases at almost all observation sites monitored by the Bureau of Meteorology with significant increases at most of the sites for at least one season within the period 1973-2017 (Figure BF2). The increase is particularly strong in south-eastern Australia and is primarily related to temperature increases (Dowdy 2018). The increases have been most significant in spring followed by summer for southern Australia with the largest increases in winter for the tropical regions.
Figure BF2 Map of trends in seasonal 90th percentile FFDI from 1973 to 2017. Marker size is proportional to the magnitude of the trend. Reference sizes are shown in the legend. Filled markers represent statistically significant trends. Red represents an upward trend and blue a downward trend. Source: Harris and Lucas (2019)
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 74 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Another way to assess trends in the most destructive fire weather conditions is to compare the 99th percentile FFDIs for the decade 1960-1969 with those of the most recent (2010-2019) decade (Dowdy 2018). Figure BF3 shows the ratio of the 99th percentile FFDI for the 2010-2019 decade, relative to the 1960-1969 decade. Most areas in Australia have increasing FFDIs relative to 50 years ago; the exceptions being parts of the interior of Australia, the south coast of West Australia and small parts of Tasmania. The greatest increases in FFDI have been observed across eastern Australia, central and inland Victoria, the tropics and western parts of West Australia.
Figure BF3 Ratio of the 99th percentile of FFDI for the decade 2010-2019 relative to 1960-1969, based on the Bureau of Meteorology’s gridded FFDI dataset. Source: IAG based on the BoM data described in Dowdy (2018).
Societally and economically, severe fire conditions that can lead to extreme bushfires with devastating effects are particularly important. These events typically have FFDI>40, resulting in a very high chance of house destruction (Bradstock and Gill 2001). Information suggests an increasing occurrence of extreme bushfires in recent decades (Sharples et al. 2016). Historical building loss data indicates the most extreme loss events occurs when the FFDI reaches or exceeds values of 75 (Extreme FFDI Category). These property losses coincide with FFDIs well in excess of the 99th percentile for those locations (Blanchi et al. 2010). It is therefore important that future climate modelling studies of trends for these destructive fire weather conditions focus on the fire dangers above the 99th percentiles for regions across Australia. There also need to be corresponding studies
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 75 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
on the changing demographics and forest structures in the wildland-urban interface to separate non- climatological influences. There has been a tentative reduction in the time between the most catastrophic bushfire events in Australia’s history – those that have destroyed many hundreds to thousands of properties and led to a substantial large loss of lives across south-eastern Australia. These were the devastating Victorian bushfires of January 1939 (71 deaths, 650 buildings destroyed), then the Ash Wednesday fires of February 1983 (75 deaths, 3,700 homes and buildings lost) (Miller et al. 1984). After this, Victoria experienced the deadly Black Saturday fires of February 2009 (173 deaths, 3,500 buildings destroyed) (Teague et al. 2010). Most recently many parts of Australia experienced the spring – summer fires of 2019-2020 (34 deaths, 9,352 buildings including 3,500 homes, destroyed, along with record poor air quality across major eastern and south-eastern cities and the largest area of forest burnt in historical times). There are no measurements of the FFDI for the 1939 bushfires, which had burnt the largest area of temperate forest prior to the most recent (2019-20) event. However, maximum temperatures across large parts of the state were over 45°C, and the Royal Commission investigation into these fires indicated it was extremely dry with strong winds allowing the fires to jump from mountain to mountain, indicating marked spotting and extreme bushfire behaviour. For the Ash Wednesday bushfires, the BNHCRC6 modelling for Mt Gambier produced an FFDI estimate of 92, almost twice the 99th percentile FFDI value for that location. For the Black Saturday bushfires, the FFDI estimate at Kinglake was 104, over four times the 99th percentile value. The prolonged extreme fire weather across southern and eastern Australia in spring – summer 2019- 2020 was unique in both the extent and persistence of extreme fire weather conditions, with many locations having individual FFDI values exceeding 100 (Catastrophic FFDI). Figure BF4 shows the accumulated FFDI percentages and FFDI deciles for December 2019 as one indicator of the severity and longevity of this event. The Special Climate Statement issued by the Bureau of Meteorology should be consulted for more detail (BoM 2020). Detailed investigations of the 2019-2020 fire season are ongoing. Still, it is apparent there were multiple periods of catastrophic fire weather conditions and significant losses of property across every state and territory in Australia. One measure of widespread risk from fire weather is the occurrence of high FFDI values, at the same time, across Australia (cumulative values are shown in Figure BF5). Since 1950 there has been an increase in the rate of such days for various FFDI thresholds, with steeper rises particularly evident during the major 1982-83 and 2019-20 fire seasons. Note that FFDI over 25 is considered Very High, and FFDI above 50 is Severe.
6 https://www.bnhcrc.com.au/
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 76 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure BF4 Accumulated FFDI percentages (top) and FFDI deciles (bottom) for December 2019 compared to the long-term mean for 1950-2018. Source: Australian Bureau of Meteorology/NSW Bushfire Risk management Research Hub calculated based on data as described in Dowdy (2018).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 77 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure BF5 Graphs illustrating trends in widespread fire danger across Australia from 1950 to 2019. Each graph shows a marker for each day where every Australian state and territory recorded an FFDI above the thresholds indicated above the graph. An increase in the steepness of the curve indicates an increase in the rate of these events Source: University of Wollongong, based on the Australian Bureau of Meteorology data described by Dowdy (2018). Although humans cause a significant proportion of bushfires, dry lightning strikes are the largest natural source of ignition of bushfires and account for the largest areas burnt. A recent study by Dowdy (2020) investigated thunderstorm environments in Australia for the period 1979-2016. They showed that there are observed increases in the days when there are thunderstorm environments, associated with low rainfall for the spring and summer fire seasons for southern and south-eastern Australia (Figure BF6).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 78 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure BF6 Long-term maps of thunderstorm environments by season from December 1979 to November 2016. The left column shows the average number of thunderstorm environments with the right column showing the change in the number of environments from the first half to the second half of the study. Changes are shown if statistically significant at the 90% confidence level. Source: Dowdy (2020)
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 79 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Penman et al. (2014a) noted that there are strong links between weather and human-caused ignition. This means periods of elevated fire danger conditions will lead to more ignitions, all other things being equal.
5.5.4 Projected Changes in Australia
Detailed, high-resolution studies have been undertaken for Australia. A study by the National Environmental Science Programme (2019), using a high (RCP8.5) emission scenario and an ensemble of 12 downscaled regional climate model simulations (NARCLiM), shows significant increases in the critical fire weather risk factor (defined here as the percentage change in FFDI over 25 (Very High)), for virtually all of Australia for the future climate period 2060-2079 (Figure BF7). The slower rates of change along the southern coastline of Western Australia and the eastern coastal fringe is due to the locally moderating effects of the sea breeze. This moderating effect is not likely to extend to the rarest of extreme events. So even in these regions, there is a growing risk of extreme to catastrophic bushfires over the coming decades. As concerning as these predictions are, researchers are becoming more aware of additional factors that need to be considered to identify regions of greatest bushfire risk in a warming climate (Sanderson and Fisher 2020). Even the most recent climate-fire-vegetation models (Sanderson and Fisher 2020) do not produce fires as extensive as those observed in the 2019-20 season. Therefore, challenges remain when trying to simulate all future climate fire conditions and events.
Figure BF7 Percentage change in Very High or FFDI >25 days between 1990-2009 and 2060-2079 for a high greenhouse gas emission scenario (RCP8.5) from a 12-member ensemble using the downscaled NARCLiM dataset. Coloured regions indicate at least two-thirds of the models agree on the direction of change. Source: NESP Bushfires and Climate Change 2019 based on Dowdy et al. (2019).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 80 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
There is high confidence that climate change will lead to an increase in the average FFDI and a higher frequency of days with severe fire danger or higher across southern and south-eastern Australia (Figure BF8), although with greater variability along parts of the east coast where the changes are less confident (CSIRO and Bureau of Meteorology 2015, Moritz et al. 2012, Clarke et al. 2011, Clarke et al. 2016, Clarke and Evans 2019, Dowdy et al. 2019).
Figure BF8 Projected changes in the annual cumulative FFDI between 1990-2009 and 2060-2079, based on the six downscaled NARCLiM model members that verify best against observed FFDI data. Source: Clarke and Evans (2019)
The increased frequency of days with a High FFDI will result in reduced intervals between catastrophic fire events, higher fire intensities, slower fire extinguishments, and an increase in fire spread (Hennessy et al. 2007). Hennessy et al. (2005) estimated that by 2050 the frequency of extreme fire danger would increase by 15-70% across south-east Australia. Similar increases are predicted for the Bay of Plenty, Wellington, and Nelson regions in New Zealand (Pearce et al. 2005). Recent research into the most extreme and uncontrollable bushfires – including those where there is considerable support from increased upper-level instability and sometimes enhanced by the formation of pyro-cumulonimbus – predicts that they will significantly increase in frequency in the future. Di Virgilio et al. (2019) investigated trends in both FFDI and C-Haines (see Fire Indices insert in Section Summary) between 1990-2009 and 2060-2079 using downscaled (the NARCLiM suite of model runs) GCMs for south-eastern Australia. They found there were statistically significant trends in the mean number of summer and spring days that C-Haines and FFDI were >=8 and >=25 respectively (days when both upper air and surface measures of bushfire severity are significant). However, the most substantial increases are projected to occur through spring. These findings are broadly similar to results from Dowdy et al. (2019) which investigated projected trends in FFDI and C-Haines using three different modelling approaches (including the NARCliM suite of models). They found future increases in the south as well as some regions of decrease in the east of Australia.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 81 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Under a 4°C warming, the frequency of days with FFDI>40 has the potential to increase by more than 200% in eastern Australia during the fire season (Clarke et al. 2011). Lucas et al. (2007) suggested that the number of Extreme fire danger days (FFDI>50) could increase by 100-300% by 2050, relative to 1990. However, at present, GCMs do not have sufficient resolution to robustly identify trends in the most catastrophic events. Furthermore, a lengthening of the fire season could reduce or shift the window of opportunity for fuel- reduction burning to winter or shorter periods in spring (Hennessy et al. 2007, Clarke and Evans 2019). These changes are due to increasing temperatures and drying in these regions and changes to the nature of the rainy seasons in forested areas. A slower change in fire hazard is expected in tropical and monsoonal north Australian regions (medium confidence) (CSIRO and Bureau of Meteorology 2015). However, some parts of tropical Australia have experienced major increases in FFDI relative to the 1960-1969 decade (Figure BF3). There is limited information in the literature about trends in the rarer and most catastrophic bushfires that typically drive most of the property risk and loss of life. Coupled atmosphere-fire simulations of actual extreme fire weather events are emerging. Toivanen et al. (2019) highlight the complex interactions between the relatively short-lived but most destructive periods of bushfires when most deaths and damage occur, and meteorological weather systems that produce them and cause the rapid expansion of fire areas following wind changes. They also highlight the importance of vertical instability and spotting of fires as they move through eucalyptus forests. Further research into factors contributing to increased spotting of bushfires is emerging, such as the findings of Storey et al. (2020). They showed that weather conditions at the surface and aloft, along with the vegetation and topography all have key influences. These complex interactions are lost in simplified approaches and are not accounted for in Bushfire Attack Level (BAL) approaches commonly used in land planning and building in fire-prone lands. Encouraging results applying statistical models calibrated against historical fire events are being developed to more accurately determine the bushfire risks faced by properties and assets in fire- prone regions (Price et al. 2015 (Figure BF9), Price and Bedward 2019). These techniques can be used to determine the most effective controlled burning fuel reduction programs for each region. To date, these techniques have concentrated on short-duration fire events. Further work is needed to extend this research to prolonged fire events and to extend the fire spread models to capture bushfire behaviour under the most catastrophic fire weather conditions.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 82 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure BF9 Risk map for the Sydney region showing the probability of extant fires spreading to each census block. The background is a true-colour satellite image showing the urban areas in grey, peri-urban in speckled green and grey, and forest (in green) surrounding the centre. Source: Price et al. (2015)
Future climate predictions across Australia – from the available science – are based primarily on the expected trends in the bushfire weather, as measured by the FFDI and C-Haines indices. The Bureau of Meteorology and CSIRO maps (2016, 2018) serve as a simplified visual guide to the observed trends, while the works of Clarke and Evans (2019), Dowdy et al. (2019) and Di Virgilio et al. (2019) provide a basis for the changing bushfire risk across many densely populated Australian districts. It should also be noted that the Australian Fire Danger Rating System Program is developing a more comprehensive bushfire rating system that better describes bushfire behaviour in different environments across Australia (Matthews et al. 2019). This Fire Behaviour Index may provide new insights into trends and areas at greatest risk from bushfires in a changing climate. For Tasmania, the work of Fox-Hughes et al. (2015) supplements the previously mentioned studies which focus more over the mainland of south-eastern Australia. Model projections show a broad increase in fire danger across Tasmania, but with substantial regional variation. The increase was smaller in western Tasmania (district mean cumulative fire danger increasing at 1.07 per year) compared with parts of the east (1.79 per year) for example. There were also noticeable seasonal
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 83 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
variations. Small changes occurring in autumn, while spring saw a steady increase in the area subjected to the 99th percentile fire danger (from 6% in 1961–1980 to 21% by 2081–2100). There is significant scope for next-generation models like NARCliM (versions 1.5 and 2.0) and BARPA (Bureau of Meteorology Atmospheric Regional Projections for Australia) to provide a more comprehensive and granular base for establishing fire weather risk indices and understanding the impacts of climate change on fire weather. However, they need to be extended across all of the populated regions of Australia, and trends in all parts of the bushfire fire danger spectrum need to be addressed, particularly the rare Extreme and Catastrophic components of the fire weather spectrum. Exacerbating conditions caused by more unstable upper air conditions and changes to the probability of ignition also need to be considered when refining regional predictions of changes in bushfire risk. The effectiveness of potential bushfire prevention and suppression techniques must also be factored in.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 84 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.6 Oceans: Sea Level Rise, Temperature Anomalies and Extreme Sea Levels
Section Summary
The main change since the release of the first edition of this report is to extend the observed rate of sea level rise to the end of 2019 and to confirm that there is an observed acceleration of these rises. New information has been included, highlighting timeframes when historic 1% AEP extreme sea level events become more frequent or even yearly events. Within Australia, the CoastAdapt website provides one set of sea level rise projections. An expanded set of projections is available from the US NOAA Sea Level Rise Report. This report provides a range of planning scenarios, intended to span the range from the lowest justifiable projection (a linear extrapolation of the historically observed 3mm/year of Global Mean Sea Level Rise) to an extreme scenario consistent with some of the highest physically grounded projections in the literature. Commentary provided may be particularly useful in highlighting the broad range of plausible scenarios, particularly in the most damaging and irreversible upper-end projections. The significance of these projections may be diminished by the term “average” which is biased in favour of the unachievable low-end projections. Most of the new research presented focuses on the dynamics of the Greenland and Antarctic ice sheets and ice shelves. Changes to these stand to produce the most irreversible and accelerating trends of sea level rise over centennial time frames. A summary of the key points concerning sea level rise and extreme sea levels includes:
• Globally, sea level has risen by ~220mm from 1880 to 2019, although there is considerable uncertainty in the early part of this record. • The rate of sea level rise is accelerating, from 1.4mm/year over the period 1901–1990 to 3.6mm/year for the period 2006–2015 with an above-trend rate of increase of 4.7mm/year for the period January 2014 to March 2020. • From a damage to property perspective, the changes to extreme sea levels, their magnitude and AEP are more critical than changes in the mean sea level. • IPCC projections indicate that by the end of this century global sea levels could increase by up to 0.84m (0.61-1.10m likely range) for the RCP8.5 scenario, relative to a 1986-2005 baseline. Other studies gaining credence point to plausible global sea level rises of over 2.0m within a century and multi-metre sea level rises by 2300. • Sea levels will continue to rise throughout the 21st and 22nd centuries due to ongoing ice losses from the Greenland and Antarctica ice sheets. However, the rates of increase will be determined by the rates of global emissions over the next few decades. Much of Australia is projected to continue to experience above global average sea level rise throughout the 21st century with rates up to 30% above the global average.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 85 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.6.1 Background
Arguably the most critical of all the changes produced by the rapidly changing climate are those associated with the oceans: sea level rise, ocean heatwaves and extreme sea levels. Land planning decisions around our coasts and estuaries determine where whole communities will be established, along with the supporting infrastructure. Once they are in place, it is challenging to protect them from the irreversible impacts of sea level rise that will continue for centuries, irrespective of the actions taken globally over the next few decades. The magnitude of these challenges, however, will be determined by global actions to reduce emissions over the next few decades. Retreat and relocation may become the only viable options. To safeguard our communities into the next century, land planning decisions need to consider higher sea level rise scenarios over longer timeframes rather than the mean or median projected rises for a specific point in time. Climate change impacts on the world’s oceans take many forms. They produce SST anomaly patterns that can alter the global atmospheric circulations, along with marine heatwaves that can severely impact marine ecosystems. Significant sea level anomalies can trigger compound events, where multiple damage drivers can affect countries or communities beyond those experienced from single events. These compounding events are often responsible for cascading impacts that extend well beyond the region directly affected by the events. They can also produce recurring weather events that can lead to significant levels of damage and hamper communities’ ability to prepare and respond due to the short time frame between successive events. Sea levels are changing at different rates around the world, and extremes of sea level are changing at record rates. In addition to this, dissolving carbon dioxide is leading to increasing ocean acidification that will affect all marine and coastal ecosystems, threatening food security for nations globally. Although very important socially, financially and ecologically, this aspect of climate change will not be covered further in this section as the focus is on more direct damage and disruption to property and society. Sea level rise is a combination of the thermal expansion of warming ocean water, meltwater run-off from ice sheets and glaciers, changes to terrestrial water storage, altered ocean currents and wind regimes. Changes in Earth’s gravity and rotation with the rise and subsidence of different parts of the Earth’s crust from the shifting masses will have varying impacts across the globe. When considering damage to property associated with sea level rise during the coming century, the main concern is not what is likely to happen to the mean sea level but how this will affect the changes to extreme sea levels, their magnitude and AEP. In multi-centennial time frames the greatest concern shifts to ongoing changes to the mean sea level. It is also noted that irrespective of the emissions pathway followed, sea levels will continue to rise throughout the 21st and 22nd centuries and most likely for millennia due to ongoing ice losses from the Greenland and Antarctica ice sheets (IPCC 2019). These changes can occur as non-linear step function changes happening within the space of a decade and are irreversible over human lifetime scales. Globally, the sea level has risen by ~220mm from 1880 to 2019, although there is considerable uncertainty in the early part of this record (Dangendorf et al. 2019). Based upon complex analyses of trends from tide gauge records and satellite altimetry, the rate of increase is accelerating (Nerem et al. 2018, Dangendorf et al. 2019), from 1.4mm/year over the period 1901–1990 to 2.1mm/year (1970– 2015), 3.2mm/year (1993–2015), and 3.6mm/year for 2006–2015 (IPCC 2019). Recent years show
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 86 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
an above-trend rate of increase of 4.7mm/year for the period January 2014 to March 2020. The rate of increase since 1993 is close to the upper limit of the IPCC AR5 projected changes (IPCC 2019). It is very likely that areas that are already affected by coastal erosion and inundation will be more severely affected in the future (Field et al. 2012). Increased frequency of coastal inundation for Sydney, Australia, has recently been documented by Hague et al. (2020). They show that in Sydney, frequencies of minor coastal inundation have increased from 1.6 to 7.8 days per year between 1914 and 2019 and they attribute over 80% of the observed coastal inundation events between 1970 and 2015 to anthropogenic increases in GMSL. They project that impact-producing coastal inundations will occur weekly by 2050 under high- and medium- emission/sea-level rise scenarios, and daily by 2100 under high emission scenarios. Local sea level rise can deviate significantly from the global mean due to differential heating of ocean areas; changing wind systems and ocean currents; and land movement. Satellite altimetry spanning 28 years shows considerable regional variations in the observed rate of sea level rise (Figure SL1). Climate models project that by the end of the century global sea levels could increase by 0.43m (0.29-0.59m likely range) for the RCP2.6 scenario and 0.84m (0.61-1.10m likely range) for the RCP8.5 scenario (IPCC 2019), relative to a 1986-2005 baseline. Figure SL2 summarises the range of future sea level rise projections out to 2300 with the inset showing projections out to 2100 (IPCC 2019).
Figure SL1 Observed satellite altimeter trends in sea level rise from 1992 to 2020. Source: NOAA Laboratory for Satellite Altimetry 2020, accessed in June 2020.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 87 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
It should be noted that there are many contributing factors relating to sea level rise (IPCC 2019), few of which are well modelled. These factors mean there is a potential for the long-term sea level rise to be significantly higher than the projections produced by climate models that use simple cryosphere modules. Factors include: the acceleration of glaciers feeding ice shelves once the ice shelves collapse (e.g. the collapse of the Larsen A and B ice shelves in 1995 and 2002 respectively); the effects of lakes and meltwater on the surface of ice shelves; the melting and destabilisation of the bases of ice sheets in contact with the warming of the Circumpolar Deep Water; and the collapse of ice cliffs into the oceans. Partially offsetting this is the expected increase in snowfall over elevated regions of Antarctica and Greenland, although this is not likely to be enough to offset losses through other mechanisms. Interactions between tropical warm ocean events linked to ENSO and IOD changes and the Southern Ocean are also poorly understood. However, it is considered an extreme El Niño event in 1940 was the trigger for the grounding line retreat of the ice sheets in the Pine Island catchment of West Antarctica, which is still ongoing.
Figure SL2 Projected sea level rise (SLR) until 2300. The inset shows an assessment of the likely range of the projections for RCP2.6 and RCP8.5 up to 2100 (medium confidence). Projections for longer time scales are highly uncertain, but a range is provided (low confidence). For context, results are shown from other estimation approaches in 2100 and 2300. The two sets of two bars labelled B19 are from an expert elicitation for the Antarctic component (Bamber et al. 2019) and reflect the likely range for a 2ºC and 5ºC temperature warming (low confidence). The bar labelled “prob.” indicates the likely range of a set of probabilistic projections. The arrow indicated by S18 shows the result of an extensive sensitivity experiment with a numerical model for the Antarctic ice sheet combined, like the results from B19 and “prob.”, with results from Church et al. (2013) for the other components of SLR. Source: IPCC (2019).
Processes controlling the timing of future ice shelf loss and the extent of ice sheet instabilities could increase Antarctica’s contribution to sea level rise to values substantially higher than the IPCC’s “likely” range on century and longer timescales. This ice sheet instability is irreversible with recovery time scales in the hundreds to thousands of years, and then only once the climate forcing triggering these instabilities are brought under control (IPCC 2019).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 88 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Emerging studies focusing on the Antarctic ice sheet, indicate that there are early signs of an acceleration of melting and a speed-up of glacial flow. Details of the history, ice mass loss, complex Antarctic ice sheet and glacier structures, and the dynamic behaviour of the Antarctic ice sheet are explained in Bell and Seroussi (2020). Estimated contributions to sea level rise from the varied ice loss mechanisms differ considerably. They are not well incorporated into most cryospheric models but amount to several metres within the next couple of centuries. Modelling studies (Figure SL2) indicate a multi-metre rise in sea level by 2300 (2.3–5.4m for RCP8.5 and 0.6–1.07m under RCP2.6), indicating the critical importance of reduced emissions to limit sea level rise (IPCC 2019). It should be noted that the ability of the world to follow an RCP2.6 scenario is becoming highly unlikely as emission and observed warming trends are already diverging above the RCP2.6 pathway. Global temperature increases beyond 2°C increase the risk of rapid increases in sea level due to melting of significant parts of ice sheets in Greenland and Antarctica. The global sea level could potentially rise by ~70m if both ice sheets completely melt (Alley et al. 2005), although this would take centuries to occur and current estimates are very uncertain. However, it is noted that transitions in ice sheet behaviour can happen rapidly once tipping points are crossed. The collapse of the Larsen A and B ice shelves (1995 and 2002, respectively) are pointers to what could occur elsewhere around Antarctica once the buttressing floating ice shelves thin and weaken. As discussed in Section 4.3, moderate changes in the mean can have dramatic effects on extremes – including the effects of storm surge and wave action. Wahl et al. (2017) investigated mid-century changes in the return period of the 100-year extreme sea level under the RCP4.5 emission scenario. For Australia, they showed that the historic 100-year extreme sea level height could occur annually for many sites along the Australian coastline (see Figure SL3). Rapid changes are predicted especially for the densely populated Australian east and south-east coasts, as well as the New Zealand North Island and around Dunedin in the South Island.
Figure SL3 Changes in the 100-year return period of extreme sea level height until mid-century under RCP4.5. Source: Wahl et al. (2017).
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 89 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
A global view of changes in extreme sea level events for RCP 2.6 and 8.5 scenarios is shown in Figure SL4. The focus is on what has been described as a historical centennial event (HCE). The approximate year in which these extreme events become annual events is illustrated by the darkness of the circle at various locations around the world.
Figure SL4 The effect of regional sea level rise on extreme sea level events at coastal locations. (a) Schematic illustration of extreme sea level events and their average recurrence in the recent past (1986–2005) and the future around 2100. (b) The year in which HCEs are expected to recur once per year on average under RCP8.5 and RCP2.6, at the 439 individual coastal locations where the observational record is sufficient. The absence of a circle indicates an inability to perform an assessment due to a lack of data but does not indicate an absence of exposure and risk. The darker the circle, the earlier this transition is expected. White circles (33% of locations under RCP2.6 and 10% under RCP8.5) indicate that HCEs are not expected to recur once per year before 2100. (c) An indication of locations at which this transition of HCEs to annual events is projected to occur more than 10 years later under RCP2.6 compared to RCP8.5. Source: IPCC (2019)
Wave heights Studies for the North and South Atlantic Oceans have shown that extreme wave heights, which contribute to extreme sea level events, coastal erosion and flooding, have been increasing at the rate of 1.0cm/year over the period 1985 to 2018 (IPCC 2019). Extreme wave heights are associated with
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 90 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
the most intense low-pressure systems. TCs (Section 5.1), ECLs (Section 5.4) and mid-latitude frontal low-pressure systems are the source of these extremes for Australian coastal waters.
Marine heatwaves Marine heatwave events are periods of extremely high ocean temperatures that exceed the local 99th percentile observed SST. These events adversely affect marine ecosystems and organisms in every ocean basin, including corals, sea grasses and kelps that help to protect coastal zones from severe oceanic impacts. For the period 1982 to 2016, the observed number of heatwave events have doubled in frequency, are more intense, and more extensive (IPCC 2019). The IPCC concluded, with very high confidence, that these heatwaves are projected to further increase in frequency, duration, spatial extent and intensity. Climate models project a 50-time increase in the frequency of marine heatwaves by 2081– 2100, relative to 1850–1900. Over the same timeframe, their intensity is expected to increase about 10-fold. These increases would continue through the 22nd century. Studies focusing on Australian marine heatwaves have been produced by Oliver et al. (2014, 2018).
5.6.2 Changes in Australia
Observed changes Most of Australia experienced above global average sea level changes during recent decades, with hotspots in the Tasman Sea and northern and western Australia (Figure SL1) where rates were up to twice the global average. To understand regional rates of sea level rise better, several factors need to be considered: the influence of ENSO on sea level rise; the influence of Glacial Isostatic Adjustment (ongoing movement of land once burdened by ice-age glaciers); local land movement produced by geological and hydrological factors; and atmospheric pressure effects.
The CoastAdapt website (https://coastadapt.com.au) provides access to observed sea level trends and variability at local government level around Australia. These trends are based on satellite observations over the adjacent oceans and seas since 1992. Tide gauge records are also available for a limited number of locations around the coast. The Bureau of Meteorology maintains the Australian Baseline Sea Level Monitoring Network and access to the data from this network can be found at the following link: http://www.bom.gov.au/oceanography/projects/abslmp/abslmp.shtml
Projected changes for Australia Much of Australia is projected to continue to experience above global average sea level rise throughout the 21st century with rates up to 30% above the global average (Figure SL2: IPCC 2019). NCCARF developed regional sea level rise projections for each local government area around Australia, for four different climate change scenarios. These projections, including a range of uncertainties, can all be found at: https://coastadapt.com.au/sea-level-rise-information-all-australian- coastal-councils. Caution is advised in using the RCP2.6 projections from this website as we, the authors of this report, believe these projections are unachievable. The RCP4.5 scenario is considered
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 91 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
by the authors the most optimistic outcome, assuming substantial actions are taken globally within the next few years. The higher RCP scenarios could be viewed as the more realistic future climate outcomes. Scenarios with projections higher than that of the RCP8.5 should be considered for the most critical developments. A sample plot of sea level rise projections for the Gold Coast Local Government Area (LGA) is shown in Figure SL5 for the RCP4.5 and 8.5 scenarios out to the year 2100, along with the observed satellite data for this area. Corresponding inundation maps for each area are available for download for the RCP scenario. However, these maps show changes in the mean sea level and not the extreme sea level.
Figure SL5 Observed satellite derived sea level (green lines) and projected sea level rises out to 2100, including uncertainty limits, for the RCP 4.5 and 8.5 scenarios for the Gold Coast LGA. Source: CoastAdapt (https://coastadapt.com.au).
The highest rate of increase in sea level is simulated in the Tasman Sea and is attributed to long-term and ongoing changes in the South Pacific Gyre (Oliver et al. 2014, 2015). Different emission scenarios differ in the amount of regional sea level rise but are consistent in the spatial patterns. Slangen et al. (2012) identified the steric contribution (sea level rising because of changes in ambient temperature and salinity) as the dominant source of regional variability. However,
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 92 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
this assessment might change with improved representation of ice sheets and glaciers in next- generation earth system models that can simulate extreme ice loss scenarios. An alternative view of sea level rise can be found in the US NOAA Sea Level Rise Report (Sweet et al. 2017). This report provides a broad range of planning scenarios, that span the range from the lowest justifiable projection (a linear extrapolation of 3 mm/yr of GMSL rise) to an extreme scenario consistent with some of the highest physically grounded projections in the literature. The intermediate-high and high emission scenarios from the NOAA report produce rates of sea level rise significantly above those contained in the CoastAdapt website. These should also be considered by those assessing risks in the most susceptible areas or for projects that are designed to remain viable well into the next century. There are few studies on what the regional changes in extreme sea levels are likely to be considering changes in the weather systems that produce them. There is very little information available on expected changes in the associated direction of approach of the highest energy waves in future climates or changes in their duration. The importance of this directional change of approach of high energy waves is illustrated by the major coastal erosion experienced along the Wamberal New South Wales coast in July 2020. The erosion was due to an ECL event that produced a major wave train from a more south-easterly direction than historically experienced. An intense weather system does not have to produce record wave heights to be extremely damaging if the direction of coastal impact differs from the directions historically experienced along that particular coastline.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 93 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
5.7 Connected Extremes: Compound and Clustered Events
Section Summary
Connected extremes are defined as the simultaneous occurrence of events, or a sequence of events, affecting a region, multiple regions, states and even countries. Connected extremes can include, for example, a major TC during a pandemic. This report only covers interconnected meteorological-hydrological-oceanographic events. Connected extremes can result in impacts far greater than the individual events alone. The science of connected extremes is in its infancy. However, some research findings are starting to emerge, including:
• Robust observational trends in hydrological extremes indicate growing evidence that anthropogenic forcing will lead to further increases in hydroclimatic variability across a wide range of timescales. • Concurrent extremes, such as prolonged drought and increased severe bushfires, are also likely to occur more frequently in the future. • Concurrent freshwater flooding and storm surge are increasingly likely under climate change. • Within Australia, there has already been a diverse range of connected events. These include the 2019-2020 extreme bushfire season; the triple TC impacts on the Pilbara coast in 1980; and, the south-east Queensland floods over the December 2010-January 2011 period. How the risks of these connected events will change in a warming world has not yet been investigated in any depth.
5.7.1 Background
The literature uses the term ‘compound event’ to describe events that are connected physically or show statistical dependency. In 2012 the Special Report of the IPCC (Seneviratne et al. 2012) introduced the following definition of compound events. “(1) Two or more extreme events occurring simultaneously or successively, (2) combinations of extreme events with underlying conditions that amplify the impact of the events, or (3) combinations of events that are not themselves extremes but lead to an extreme event or impact when combined.” Alternatively, Leonard et al. (2014) defined compound events as an extreme impact that depends on multiple statistically dependent variables or events. This section considers only interconnected meteorological-hydrological-oceanographic events. There can be connections between non-interconnected events, such as pandemics, geological events or other phenomena, but they are not included in this discussion. Coincident hazards that are naturally part of each severe weather phenomena are not considered compound or connected events for this report. For example, coincident rainfall, severe winds, storm surge, etc., that are part of a TC have already been discussed in previous sections. These events are also already considered by catastrophe risk models commonly used by the insurance industry. Instead, this section focuses on the currently available scientific understanding of how multiple severe
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 94 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
weather phenomena, either of the same or different types, are connected through time and space. Areas of future research needed to improve understanding are highlighted. Insurers must understand possible multiple impacts across a portfolio from days to weeks to years, and these portfolios may span several states or countries. Therefore, understanding the connectedness of these multiple severe weather phenomena and how they may change with climate change is important from both a claims and financial management perspective. The simultaneous occurrence of events, or a sequence of events, affecting a region, multiple regions, states and even countries can cause impacts far greater than the individual events alone. Merz et al. (2009, 2016) and Schröer and Tye (2018), for example, showed that the cumulative effect of moderate flooding could result in higher impacts than rarer single extreme events. Likewise, multi- hazard events such as combined heat and drought (Livneh and Hoerling 2016) or extreme rainfall combined with anomalous warmth (Huang et al. 2018) can produce impacts much more severe than would have resulted from individual events. Connections among extremes operate across scales. These can range from short-timescale cold-pool outflows triggering neighbouring thunderstorms to long-timescale variability of the global circulation that favours concurrent, clustered and distant events (Villarini et al. 2013). An example is the serial clustering of tropical and extratropical cyclones (Mumby et al. 2011, Vitolo et al. 2009). An Australian illustration of a clustered event is provided by the sequential impacts of Severe TCs Amy (January 1980), Dean (January/February 1980) and Enid (February 1980) on the east Pilbara coast within a five-week period. The impacts across the same general region produced damage in some towns, e.g., Goldsworthy and Port Hedland, that was exacerbated due to the earlier events. The repairs to the buildings from the impact of TC Amy were not completed before TC Dean passed near the town of Port Hedland and over the severely damaged town of Goldsworthy. Winter storms sweeping down coastal regions (Haigh et al. 2016), or self-reinforcing blocking high- pressure systems driving weather systems away from an already dry area (Seneviratne et al. 2010, Sillman and Croci‐Maspoli 2009), are other examples of high connectivity across events. Geographically distant simultaneous events can be “teleconnected”. An example of teleconnected events is the 2010 Russian heatwave and Pakistan flood, which were linked through a high-amplitude wave pattern of the mid-latitude jet stream (Lau and Kim 2010, Galarneau et al. 2012). Moreover, connections can be non-stationary over time (Wahl et al. 2015). The specific attributes of connected events that drive impacts are not well understood. But the timing between events appears to be a key factor. Fussell et al. (2017), for example, showed that a second TC landfall before recovery from an earlier TC landfall could trigger permanent population migration out of the region. Further, the particular sequencing of hydrologic extremes (e.g., a rapid transition from drought to flood) can greatly amplify human and environmental consequences (e.g., Oakley et al. 2018, Li and Ye 2015). Moreover, regions where annual mean rainfall is dominated by a few intense events – such as California – are particularly vulnerable to connected extreme events (Diffenbaugh and Giorgi 2012, Swain et al. 2018). Ultimately, the relevant scales of connected extremes will be those that intersect with scales of key societal or ecological exposure such as water, energy and food networks (Franzke 2017). Robust observational trends in hydrological extremes have recently emerged in observations (Diffenbaugh et al. 2017), and there is growing evidence that anthropogenic forcing will lead to further increases in hydroclimatic variability across a wide range of timescales (e.g., Pendergrass et al. 2017,
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 95 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Swain et al. 2018). For example, Swain et al. (2018) investigated future climate trends in sequential events similar to the Californian transition from extreme drought over the 2012-2016 period to a wet extreme in 2016-2017. They found a 25% to 100% increase in connected extreme dry-to-wet precipitation events by the end of the century, despite only modest changes in mean precipitation. Such hydrological cycle intensification would seriously challenge California’s existing water storage, conveyance and flood control infrastructure. Concurrent extremes are also likely to occur more frequently in the future. For example, concurrent drought and heat (Agha Kouchak et al. 2014, Sharma and Mujumdar 2017), may already be increasing on a regional basis (Diffenbaugh et al. 2015). Wahl et al. (2015) also showed that concurrent freshwater flooding and storm surge are increasingly likely under climate change. An indication of the types of hydrological events with a strong connection to the large-scale environment is illustrated by the sequence of floods experienced across the south-east of Queensland in the period December 2010 to January 2011. The large-scale environmental setting was that of a strong La Niña. This positive phase of the ENSO caused a prolonged period of heavy rainfall over Queensland, as well as in other parts of Australia (Figure CE1). Near-record SSTs were recorded off the Queensland coast in late 2010. December 2010 was Queensland's wettest December on record, with record-high rainfall totals reported from 107 rainfall stations. The initial bout of floods was triggered by an outbreak of extremely heavy rainfall across south-east Queensland over the period 18-20 December 2010. This produced major floods affecting multiple rivers and towns across eastern and south-eastern Queensland. Before these floods had time to subside, a combination of surface and middle-level lows led to several more days (4-12 January 2011) of extreme rainfall across south-east Queensland. This second, independent yet interconnected, event initially produced extreme flash flooding through Toowoomba, followed by major catastrophic river flooding of the Lockyer Valley. The Wivenhoe and Somerset Dams were unable to contain the full extent of the rainfall runoff from these events. Subsequent dam overflows, combined with the Lockyer Valley floodwaters, produce major flooding of the Ipswich to Brisbane region. Independently, these flood events would have been major, but occurring sequentially contributed to the floods becoming a major catastrophe for this part of Australia.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 96 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure CE1 Rainfall deciles for the three months from 1 November 2010 to 31 January 2011. Source: Australian Bureau of Meteorology. The topic of connected extremes was identified as a critical research need in the Climate Science Special Report of the Fourth National Climate Assessment (Kopp et al. 2017). There has been a surge in research focused on connected extremes over the past decade. Most of it focused on concurrent co-located multi-variate extremes such as extreme wind and rain, or extreme heat and drought, rather than on the physical processes connecting events in space and time and their predictability.
5.7.2 Changes in Australia
To date, there have been few robust investigations into compound and clustered events that have occurred in Australia. This makes it difficult to identify trends in these events that may lead to more catastrophic damage than would otherwise be expected. The following section describes some relevant Australian examples of connected and clustered events as possible future research directions.
Extensive Droughts and Bushfires Possibly the best indication of what could be considered a geographically widely distributed clustered event would be the 2019-2020 “Black Summer Bushfires” in Australia. The damaging bushfires started in September 2019 and ended in February 2020. There were multiple outbreaks of geographically separate major bushfires. Initially, the bushfires were centred over east to south-east Queensland. Later, new major bushfires occurred further southwards in the Mid North Coast of New South Wales. Subsequent events then erupted across geographically disparate locations ranging from Kangaroo Island (South Australia) to the Adelaide Hills, to Mallacoota and east Gippsland
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 97 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
(Victoria), to the New South Wales South Coast, and to the Australian Capital Territory and surrounding parts of New South Wales. These events are summarised in Table CE1. All these events occurred in a large-scale environment of prolonged record heat and drought that affected large parts of Australia throughout 2019 and into 2020. The compound impacts of these multiple fire outbreaks produced damage that surpassed that of previous catastrophic bushfire events, including the Black Saturday fires of February 2009 and the Ash Wednesday fires of February 1983.
Table CE1 List of the major damaging bushfire events in Australia from September 2019 to January 2020. Source: IAG internal reports
Bushfire start date State Areas with significant property damage
4 September 2019 Queensland Peregian Beach, Stanthorpe, Applethorpe
6 September 2019 New South Wales Tenterfield, Drake and Tabulum
8 October 2019 Queensland Rural east and south-east Queensland
8 October 2019 New South Wales Busbys Flat, Rappville, Long Gully, Wyan
12 October 2019 South Australia Riverland, Murraylands, North Districts, Lewiston, Loxton and Nain
6 November 2019 Queensland Cobraball, Bungundarra, Cooroibah, Adelaide Park, Yeppoon and Woodbury
8 November 2019 New South Wales Nymboida, Rainbow Flat, Koorainghat, Willawarrin and Old Bar
11 November 2019 South Australia Port Lincoln, Stansbury and Yorketown
21 November 2019 Victoria Rochester, Natte Yallock and Buchan
3 December 2019 New South Wales Balmoral, Grose Valley and Gospers Mountain
5 December 2019 Western Australia Two Rocks, Yanchep, Wanneroo
15 December 2019 South Australia Mount Lofty Ranges, Adelaide Hills, Kangaroo Island
19 December 2019 Victoria Marthavale, Barmouth Spur, Tambo Crossing, Mallacoota and east Gippsland
30 December 2019 Tasmania Elderslie, Mangans
31 December 2019 New South Wales Eden, Bega, Batemans Bay, Batlow, Southern Highlands
16 January 2020 Victoria Mallacoota, Bendoc, Clayton South, Glen Alvie, Sarsfield, Tamboon
24 January 2020 New South Wales Namadgi National Park, Orroral Valley, Clear Range, Wyndham, Colinton, Bredbo, Narooma, Mount Darragh
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 98 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
There is no historical analogue to this clustered series of damaging bushfire events. It poses the question of whether these multi-state, multi-month bushfire events will become more common in a warming climate, and how they will change in the future. The extent of the observed extreme fire weather exceeded that predicted by the CMIP6 models for Australia out to the end of this century (Sanderson and Fisher 2020). Only four of the CMIP6 models include a prognostic fire module (EC- Earth3-Veg, CESM2, CNRM-ESM2-1 and MPI-ESM1- 2-LR). Sanderson and Fisher found that for these models, there can be substantial regional biases that make projections or formal climate change attribution statements difficult. For New South Wales, the scale of the recent fires is unmatched in the current and future CMIP6 simulations. Similar biases are evident in other Australian territories. It is therefore critical to note that the complex dynamics of fuel accumulation, vegetation dynamics and their interactions with climate under transient CO2 concentrations, as well as impacts of land management and human ignitions, are likely to result in fire behaviour patterns not represented in historical records, or future climate simulations. Other scenarios of compound events based upon observed events from the past also require more detailed investigation. Here, we will concentrate on TCs and bushfires; ECLs; and droughts and floods.
Tropical Cyclones and Bushfires The south-west of Western Australia experienced a compound bushfire, severe wind, storm surge, and heavy rainfall event during TC Alby in April 1978 (Figure CE2). Such an event has the potential to be far more extreme in a warming climate and could be expected to have catastrophic consequences for this region. The rarity of these events makes the determination of their likely recurrence and impacts challenging to quantify. However, due to their high impact potential, they are worthy of scenario analyses. These scenarios could then serve as a basis for developing mitigation and adaptation strategies to lessen the potential impact.
Figure CE2 Track of TC Alby, April 1978. Source: Australian Bureau of Meteorology.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 99 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
East Coast Lows An illustration of how extensive the damage from multi-day, multi-peril, multi-state events can be is given by the major ECL event of June 1867. This event produced the record floods of the Hawkesbury River that peaked in the town of Windsor. An internal IAG research project investigated this event using all available newspapers and other sources. Extensive damage was reported across four states over five days (Figure CE3). Leading up to this event, widespread, good rains fell across eastern Australia. This rain resulted in full rivers and high levels of soil moisture, laying the groundwork for the severe flooding from the ECL event. It is apparent from the large-scale weather system that produced the record flooding that the event had two phases. The initial intense period resulted in the flooding of rivers in the Goulburn, Yass, Braidwood and inland parts of northern and western New South Wales (west of the ranges), and the Hunter and Clarence Rivers through Grafton. In addition to the flooding, severe thunderstorm activity was reported at Mudgee. Then the second, even more, intense phase of the weather system struck. The available data suggested that this was likely a powerful, extensive, multi-centred ECL during its most intense period. There have been no analogues to this weather system for the last 220 years as records indicate the size of these floods eclipsed those experienced and documented in the 1799-1800, 1806 and 1857 severe flood events. If these events were to occur today, the consequences would be catastrophic. Yet little research is available to quantify how these extreme and very rare events are likely to be affected by on-going climate change. Multiple ECLs can affect New South Wales in close succession, such as the ECLs that occurred in the May-June period of 1974. In more recent times another sequence of highly damaging ECLs occurred. Over the period 3-7 June 2016, an intense ECL formed that produced insured damage of $432 million. Only two weeks later, 18-20 June, a second major damage-producing ECL occurred. Research is needed to identify the large-scale environments conducive to these sequential events.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 100 5. CHANGES TO EXTREMES FOR DIFFERENT WARMING SCENARIOS
Figure CE3 Summary of the major impacts of the major ECL event spanning the period 21-24 June 1867. Source: Internal IAG research.
From Drought to Flooding Rains The year 2020 has also demonstrated how a major transition can occur from extreme heat, drought and bushfire weather (September 2019 to February 2020), to widespread floods, with the transition occurring within a week. In addition to this, in January 2020, the first connected hailstorm event to affect three capital cities – Melbourne, Canberra and Sydney – occurred. All three events produced billions of dollars of damage. Some communities affected by the 2019-20 bushfires were affected by large hail and damaging floods. There is a pressing need to understand the connections between these apparently separate yet connected extreme events.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 101
6. CONCLUSIONS
This report provides a comprehensive review of the latest climate change science, with a focus on the Australian region, and on weather and climate extremes that drive property risk. Climate change is not just about the future: there is already solid evidence that there have been measurable changes to weather and climate extremes from the warming experienced to date. Although the impacts of climate change at larger geographical scales and longer time frames are quite well understood, they can mask significant nuances at regional and local scales. That has driven the approach in preparing this report. It combines the extensively available literature with well-considered expert judgement to gain a better understanding of how climate change influences the various extreme weather events across Australia (as summarised in Table ES1). Using the available continental scale information, this report derived regional scale insights that aim to provide meaningful direction and trigger further collaboration and coordination among communities, academia, businesses and government. The key assessments are: 1. The frequency of named tropical cyclones in the Australian region has declined in recent decades, but the details of how this will project into the future is uncertain. Globally a slight reduction in TC frequency is expected. However, over the past 30 years, the proportion of the most destructive TCs has increased at the expense of weaker systems, and this change is expected to continue. The frequency of landfalling TCs throughout the western South Pacific region has increased.
There has already been a poleward shift in the regions where TCs reach their peak intensity and this is expected to continue. TC risk, therefore, is expected to increase most rapidly in the south-east Queensland / north-east New South Wales regions, followed by the coastal districts south of Shark Bay in Western Australia. Marginal decreases in risk for wind impacts may occur in some northern regions.
Planning for inland penetration of TCs should be based on substantial increases in both rainfall rate and affected areas. Winds are also likely to decay more slowly, so increased wind- driven rainfall ingress should be expected both inland and along the coast. More intense storms, combined with rising sea levels, point to increasing storm surge impacts, and these may be very substantial in some regions.
2. Intense, short-duration, rainfall is expected to increase almost everywhere in Australia, resulting in more frequent flooding in urban areas and in small river catchments. Storm rainfall totals from both ECLs and tropical systems are also expected to increase, leading to increased flood risk in the larger river catchments. More work is required to understand and confidently assess these changes.
3. Areas at risk of large and giant hail should progressively shift southwards. The largest increase in risk is likely to be in the region inland from the Hunter River, south through the central and southern New South Wales highlands, and central-to-eastern Victoria. Slower increases are assessed to affect the south-west of Western Australia. At the same time, severe hail risk is expected to decrease in northern-to-central Queensland, extending further southward with ongoing increases in warming.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 102 CONCLUSIONS
4. The multi-day impacts of ECLs on the south-eastern seaboard of Australia are expected to increase because of wind-driven rainfall ingress, and flash and fast-response riverine flooding. This effect will be compounded by rising impacts from storm surge, waves, and coastal erosion. The most damaging warm-cored and hybrid-cored ECL activity, most common in summer to autumn, is expected to increase. Other types of ECLs, notably the winter-spring systems, are expected to decrease. There is limited understanding of the subset of rare extreme ECLs that drive the majority of property damage over land.
5. Longer periods of drought and coincident extreme heatwaves will exacerbate the bushfire risk, noting these will be driven in part by future climate changes to the ENSO cycles. Bushfire risk, as measured by the trends in meteorologically-based fire danger indices, is likely to increase in almost all locations nationally, leading to more frequent and extreme events, and longer fire seasons. The rate of increase varies by location and will depend on weather system changes and site-specific factors at regional scales.
6. Sea level rise is expected to accelerate around the Australian coastline, but at differing rates. Notably, past assessments of sea level rise are lower than recent observations. Sea level rise will contribute substantially to escalating impacts from storm surge and the effects on coastal natural systems, buildings and infrastructure. The greenhouse gases that are already present will cause sea level rises to continue for the next couple of centuries, even if there are significant emission reductions globally through the coming decades.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 103
GLOSSARY
Term Definition Source
Advanced Dvorak ADT is a computer-based algorithm that determines TC NOAA Technique (ADT) intensity using geostationary Infrared satellite imagery and is based on the traditional Dvorak technique. It provides uniformity compared with manual methods.
Annual Exceedance An annual exceedance probability (AEP) is the probability of Probability (AEP) an event occurring in any given year. i.e. A 1% AEP means there is a 1% chance in any given year of the event occurring. This is often used to describe the probability of river flooding at a location.
Australian tropical Details of all tropical cyclones that are known to have BOM cyclone database occurred in the Australian region (90oE-160oE south of Indonesia and PNG) are contained in a database maintained by the Bureau of Meteorology. After a tropical cyclone has occurred, tropical cyclone meteorologists reanalyse the cyclone data and compile what is known as the 'best track' and a report.
http://www.bom.gov.au/cyclone/tropical-cyclone-knowledge- centre/databases/
Bushfire Attack The Australian Standard, AS 3959, divides bushfire prone NSW RFS Level (BAL) areas into six bushfire attack levels (BAL), based on the region, slope and vegetation, to classify the severity of a building’s potential exposure to ember attack, radiant heat and direct flame contact.
https://www.rfs.nsw.gov.au/plan-and-prepare/building-in-a-bush-fire- area/building-after-bush-fire/your-level-of-risk
Carbon capture and The process of trapping carbon dioxide at its emission CSIRO storage (CCS) source, transporting it to a storage location, and isolating it technology there.
Clausius-Clapeyron Clausius-Clapeyron equation relates saturation vapour relationship pressure to the air temperature. Vapour pressure is the pressure exerted by water vapour alone. As air warms, its capacity to hold water increases at the Clausius-Clapeyron rate (CC, approximately 7% °C−1) https://www.theweatherprediction.com/habyhints2/646/
CLEX Centre of Excellence for Climate Extremes (CLEX) is an CLEX international research consortium of five Australian
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 104 GLOSSARY
Term Definition Source
universities and a network of national and international partner organizations supported by the Australian Research Council.
https://climateextremes.org.au/
CLIVAR CLIVAR (Climate and Ocean: Variability, Predictability and CLIVAR Change) is one of the four core projects of the World Climate Research Programme (WCRP)
http://www.clivar.org/
CMIP5 models, Coupled Model Inter-comparison Project (CMIP) is to WCRP CMIP6 understand past, present and future climate changes arising from natural, unforced variability or in response to changes in radiative forcing in a multi-model context. It includes assessments of model performance during the historical period and quantifications of the causes of the spread in future projections. In addition to these long-time scale responses, experiments are performed to investigate the predictability of the climate system on various time and space scales as well as making predictions from observed climate states. An important goal of CMIP is to make the multi-model output publicly available in a standardized format. The number refers to the generation of climate models in that particular time.
https://www.wcrp-climate.org/wgcm-cmip
Decadal Prediction Decadal Prediction Large Ensemble (DPLE) is a set of NCAR Large Ensemble simulations carried out at NCAR to support research into modelling (DPLE) near-term Earth System predictions. The DPLE comprises 64 distinct ensembles, one for each of 62 initialization times (November 1 of 1954, 1955, ..., 2016, 2017). For each start date, a 40-member ensemble was generated by very slight random perturbations in the atmospheric initial conditions. The initial conditions for the atmosphere and land models were obtained from the Large Ensemble (LENS) simulations at corresponding historical times.
http://www.cesm.ucar.edu/projects/community-projects/DPLE/
Dvorak analysis An analysis procedure, named after Vern Dvorak, for AMS determining tropical cyclone / hurricane intensity from cloud patterns in satellite images.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 105 GLOSSARY
Term Definition Source
East Coast Low East Coast Lows are intense low-pressure systems which BOM (ECL) occur on average several times each year off the eastern coast of Australia, in particular, southern Queensland, New
South Wales and eastern Victoria
El Niño-Southern El Niño refers to the warming of the oceans in the equatorial BOM Oscillation (ENSO) eastern and central Pacific and Southern Oscillation is the changes in the climate associated with this warming (hence 'Southern Oscillation Index' to measure these changes). 'ENSO' is used to describe the whole suite of changes associated with an 'El Niño' event - to rainfall, oceans, atmospheric pressure etc.
Ensemble modelling Ensemble modelling is a process where many computer models (which may be from difference weather centres or variants of the one model) are used to predict the range of possible outcomes by examining the aggregate output from the many models.
Fast response A fast response catchment is a river catchment that responds catchments rapidly (usually within 6 hours) to heavy rain. They tend to produce floods of short duration with rapidly rising water levels (e.g. high peak discharge).
Forest Fire Danger The McArthur Forest Fire Danger Index (FFDI) was Wikipedia Index (FFDI) developed in the 1960s by CSIRO scientist A. G. McArthur to provide a consistent index of the weather component of danger of fire in Australian forests. The index combines a measure of dryness, based on rainfall and evaporation, with meteorological variables for wind speed, temperature and humidity.
GFDL Geophysical Fluid Dynamics Laboratory (GFDL) at the National Oceanic and Atmospheric Administration (NOAA)
https://www.noaa.gov/
https://www.gfdl.noaa.gov/about/
IBTrACS IBTrACS (International Best Track Archive for Climate UCAR Stewardship) provides global tropical cyclone best track data in a centralized location to aid our understanding of the
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 106 GLOSSARY
Term Definition Source
distribution, frequency, and intensity of tropical cyclones worldwide.
https://climatedataguide.ucar.edu/climate-data/ibtracs-tropical-cyclone- best-track-data
Indian Ocean Dipole The IOD is the difference in sea surface temperature BOM (IOD) between two areas (or poles, hence a dipole) – a western pole in the southern Arabian Sea (western Indian Ocean) and an eastern pole in the eastern Indian Ocean south of Indonesia.
Insurance Council of The Insurance Council of Australia is the representative body ICA Australia (ICA) of the general insurance industry in Australia.
https://www.insurancecouncil.com.au/
Interdecadal Pacific The IPO is a large-scale, long period oscillation that NIWA Oscillation (IPO) influences climate variability over the Pacific Basin. The IPO operates at a multi-decadal scale, with phases lasting around 20 to 30 years.
Intergovernmental www.ipcc.ch IPCC Panel on Climate The IPCC is the United Nations body for assessing the Change (IPCC) science related to climate change and prepares assessment reports about the state of scientific, technical and socio- economic knowledge on climate change and its impacts.
La Niña The extensive cooling of the central and eastern Pacific BOM Ocean. In Australia (particularly eastern Australia), La Niña events are associated with an increased probability of wetter conditions.
Latitude of Lifetime The Latitude of LMI. Maximum Intensity (LLMI)
Lifetime Maximum The Lifetime Maximum Intensity (LMI) is the maximum Intensity (LMI) intensity, defined by the greatest difference in pressure between the central pressure of a TC and the environmental pressure, of a TC during its lifetime.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 107 GLOSSARY
Term Definition Source
NARCliM The New South Wales and ACT Regional Climate Modelling NARCliM (NARCliM) Project is a research partnership between the New South Wales and ACT governments and the Climate Change Research Centre at the University of New South Wales. It has produced an ensemble of regional climate projections for SE Australia that can be used by the New South Wales and ACT community to plan for the range of likely future changes in climate.
https://climatechange.environment.New South Wales.gov.au/Climate- projections-for-New South Wales/About-NARCliM
NCCARF The National Climate Change Adaptation Research Facility (NCCARF) works to support decision makers throughout Australia as they prepare for and manage the risks of climate change and sea-level rise. NCCARF has a national focus across Australia to build resilience to climate change in government, NGOs and the private sector. https://www.nccarf.edu.au/
Pre-industrial levels 'Pre-industrial levels' refer to the period before the start of the IPCC industrial revolution, notionally 1850-1900.
RCP2.6 scenario The RCP2.6 scenario is very stringent pathway considered NIWA by the IPCC, that requires removal of some of the CO2 present in the atmosphere.
RCP8.5 scenario The RCP8.5 scenario is essentially ‘business as usual’ with NIWA very high greenhouse gas concentrations out to 2100 and beyond. The Special Report on Emissions Scenarios (SRES) A2 pathway used in the AR4 IPCC report is similar to the current RCP8.5 scenario.
Representative A RCP is a greenhouse gas concentration (not emissions) NIWA Concentration trajectory adopted by the IPCC (based on levels pre 1750). Pathway (RCP) Lower numbers relate to lower GHG concentrations.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 108 GLOSSARY
Term Definition Source
Saffir Simpson A 1 to 5 classification scheme for US hurricane intensity AMS Intensities based on the maximum one-minute average surface wind speed and the type and extent of damage done by the storm. The intensities for 1 to 5 are: 1) 64-82 kt, 2) 83-95 kt, 3) 96- 112 kt, 4) 113-136 kt, and 5) 137 kt and higher. These categories are used by weather forecasters in North America to characterize the intensity of hurricanes for the public. The Saffir Simpson classification is NOT the system used by the BOM in Australia.
Scatterometer A scatterometer is a microwave radar sensor which NASA measures the reflection (or scattering effect) produced while scanning the surface of the Earth from an aircraft or a satellite. This can be used to measure global sea-surface wind speed and direction.
https://winds.jpl.nasa.gov/aboutscatterometry/
Southern Oscillation The Southern Oscillation Index (SOI) is calculated from the BOM Index (SOI) monthly or seasonal fluctuations in the mean sea level air pressure difference between Tahiti and Darwin.
South Pacific The SPCZ is a band of low-level convergence, cloudiness Wikipedia Convergence Zone and precipitation extending from the east of Papua New (SPCZ) Guinea south-eastwards towards French Polynesia and as far as the Cook Islands. The SPCZ is a portion of the Intertropical Convergence Zone (ITCZ).
Tropical cyclones Tropical cyclones are intense low-pressure systems which BOM form over warm ocean waters at low latitudes that must have 10-minute average wind speeds around more than half the circulation of 63 km/h (34 knots) or more. Tropical cyclones can cause extensive damage as a result of the strong wind, flooding (caused by either heavy rainfall or ocean storm surges). If they attain maximum mean winds above 117 km/h (63 knots) they are called severe tropical cyclones. In the north-west Pacific severe tropical cyclones are known as typhoons and in the north-east Pacific and Atlantic/Caribbean they are called hurricanes. The severity of a tropical cyclone is described in terms of categories ranging from 1 (weakest) to 5 (strongest) related to the maximum mean wind speed as shown in this table.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 109 GLOSSARY
Term Definition Source
Note: corresponding wind gusts are also provided as a guide. Stronger gusts may be observed over hilltops, in gullies and around structures.
Category Strongest Typical Effects Gust (km/h)
1 < 125 Damaging winds. Negligible house damage. Damage to some crops, trees and caravans. Craft may drag moorings.
2 125 - 164 Destructive winds. Minor house damage. Significant damage to signs, trees and caravans. Heavy damage to some crops. Risk of power failure.
3 165 - 224 Very destructive winds. Some roof and structural damage. Some caravans destroyed. Power failures likely.
4 225 - 279 Significant roofing loss and structural damage. Many caravans destroyed and blown away. Dangerous airborne debris. Widespread power failures.
5 > 279 Extremely dangerous with widespread destruction.
WeatheX WeatheX is a mobile phone-based application to allow the CLEX public to report the severity, location and timing of hail, wind damage, flooding and tornadoes. https://climateextremes.org.au/weathex/
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 110
BIBLIOGRAPHY
AghaKouchak, A., Cheng, L., Mazdiyasni, O., and Farahmand, A. (2014). Global warming and changes in risk of concurrent climate extremes: Insights from the 2014 California drought. Geophysical Research Letters, 41(24), 8847-8852. Allen, J.T. and Allen, E.R., 2016. A review of severe thunderstorms in Australia. Atmos Res 178-179: 347- 366. Allen, J.T. and Karoly, D.J., 2014. A climatology of Australian severe thunderstorm environments 1979– 2011: inter‐annual variability and ENSO influence. International Journal of Climatology, 34(1), pp.81- 97. Allen, J.T., Karoly, D.J. and Walsh, K.J., 2014a. Future Australian severe thunderstorm environments. Part I: A novel evaluation and climatology of convective parameters from two climate models for the late twentieth century. Journal of Climate, 27(10), pp.3827-3847. Allen, J.T., Karoly, D.J. and Walsh, K.J., 2014b. Future Australian severe thunderstorm environments. Part II: The influence of a strongly warming climate on convective environments. Journal of Climate, 27(10), pp.3848-3868. Allen, J.T., Tippett, M.K. and Sobel, A.H., 2015. An empirical model relating US monthly hail occurrence to large‐scale meteorological environment. Journal of Advances in Modeling Earth Systems, 7(1), pp.226-243. Alley, R.B., Clark, P.U., Huybrechts, P. and Joughin, I., 2005. Ice-sheet and sea-level changes. Science, 310(5747), pp.456-460. Arakawa, H., 1952: Mame Taifu or midget typhoon (small storms of typhoon intensity). Geophys. Mag., 24, 463–474. Bacmeister, J.T., Reed, K.A., Hannay, C., et al. 2016. Projected changes in tropical cyclone activity under future warming scenarios using a high-resolution climate model. Climatic Change, 146: 547-560. Ban, N., Schmidli, J. and Schär, C., 2015. Heavy precipitation in a changing climate: Does short‐term summer precipitation increase faster? Geophysical Research Letters, 42(4), pp.1165-1172. Bao. J., Sherwood, S.C., Alexander, A.V., Evans, J.P., 2017. Future increases in extreme precipitation exceed observed scaling rates. Nature Climate Change, doi:10.1038/nclimate3201. Barthel, F. and Neumayer, E., 2012. A trend analysis of normalized insured damage from natural disasters. Climatic Change, 113(2), pp.215-237. Bell, R.E., Seroussi, H. 2020. History, mass loss, structure, and dynamic behaviour of the Antarctic Ice Sheet. Science 367 (6484), 1321-1325. DOI: 10.1126/science.aaz5489. Bell, R., J. Strachan, P.L. Vidale, K. Hodges, and M. Roberts, 2013. Response of Tropical Cyclones to Idealized Climate Change Experiments in a Global High-Resolution Coupled General Circulation Model. Journal of Climate, 26, 7966–7980. Bell, S. S., Chand, S. S., Tory, K. J., Dowdy, A. J., Turville, C., & Ye, H. (2019). Projections of southern hemisphere tropical cyclone track density using CMIP5 models. Climate Dynamics, 52(9-10), 6065– 6079. http://doi.org/10.1007/s00382-018-4497-4. Berg, P., Moseley, C. and Haerter, J.O., 2013. Strong increase in convective precipitation in response to higher temperatures. Nature Geoscience, 6(3), p.181. Bhatia, K. T., Vecchi, G. A., Knutson, T. R., Murakami, H., Kossin, J., Dixon, K. W., & Whitlock, C. E. (2019). Recent increases in tropical cyclone intensification rates. Nature Communications, 1–9. http://doi.org/10.1038/s41467-019-08471-z.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 111 BIBLIOGRAPHY
Bhatia, K., Vecchi, G., Murakami, H., Underwood, S., & Kossin, J. (2018). Projected Response of Tropical Cyclone Intensity and Intensification in a Global Climate Model. Journal of Climate, 31(20), 8281– 8303. http://doi.org/10.1175/JCLI-D-17-0898.1. Blanchi, R., Leonard, J., Haynes, K., Opi, K., James, M., de Oliveira, F.D. 2014. Environmental circumstances surrounding bushfire fatalities in Australia 1901-2011. Environmental Science and Policy 37, Pp 192-203. http://dx.doi.org/10.1016/j.envsci.2013.09.013. Blanchi, R., Lucas, C., Leonard, J., Finkele, K. 2010. Meteorological conditions and wildfire-related house loss in Australia. International Journal of Wildland Fire 19, Pp 914-926. Blocken, B. and Carmeliet, J., 2004. A review of wind-driven rain research in building science. Journal of wind engineering and industrial aerodynamics, 92(13), pp.1079-1130. Boer, M.M., de Dios, V.R., Bradstock, R.A. 2020. Unprecedented burn area of Australian mega forest fires. Correspondence, Nature Climate Change 2020. https://doi.org/10.1038/s41558-020-0716-1. Boughton G., Falck D., Henderson D., Smith D., Parackal K., Kloetzke T., Mason M., Krupar R., Humphreys M., Navaratnam S., Bodhinayake G., Ingham S., Ginger J., (2017). Tropical Cyclone Debbie: Damage to buildings in the Whitsunday Region, Cyclone Testing Station, James Cook University, Report TR63. Bradstock, R.A. and Gill, A.M., 2001. Living with fire and biodiversity at the urban edge: in search of a sustainable solution to the human protection problem in southern Australia. Journal of Mediterranean Ecology 2: 179-195. Bridgeman H.A., 1986. The ‘Sygna’ Storm at Newcastle 12 years later, Meteorology Australia 3(5), 10-16. Brimelow, J.C., Burrows, W.R. and Hanesiak, J.M., 2017. The changing hail threat over North America in response to anthropogenic climate change. Nature Climate Change, 7(7), p.516. Brook, Jordan. University of Queensland 2020. Personal communications. Brooks, H.E., 2009. Proximity soundings for severe convection for Europe and the United States from reanalysis data. Atmospheric Research, 93(1-3), pp.546-553. Browning, S.A. and Goodwin, I.D. 2013. Large-scale influences on the evolution of winter subtropical maritime cyclones affecting Australia’s east coast. Mon. Weather Rev, 141, 2416–2431, doi:10.1175/MWR-D-12-00312.1. Browning, S.A. and Goodwin I.D., 2016. Large scale drivers of Australian East Coast Cyclones since 1851. Journal of Southern Hemisphere Earth Systems Science 66, 146–171. Bruyère, C.L., 2020: Tropical Cyclones in the DPLE Simulations. Personal Communications. Bruyère, C.L., Done, J.M., Jaye, A.B., Holland, G.J., Buckley, B., Henderson, D., Leplastrier, M., Chan, P., 2019a. Physically-Based Landfalling Tropical Cyclone Scenarios in Support of Risk Assessment. Wea. Clim. Extremes, https://doi.org/10.1016/j.wace.2019.100229. Bruyère C.L., Holland G.J., Towler E., 2012. Investigating the use of a genesis potential index for tropical cyclones in the North Atlantic Basin. Journal of Climate, 25:8611–8626. Bruyère, C.L., Jaye, A., Buckley, B.W., Prein, A.F., Done, J.M., Holland, G.J., Chan, P., Leplastrier, M., and Henderson, D., 2019b. Tropical Cyclone Predictions for Eastern Australia. AMOS Conference, Darwin 2019. Buckley, B.W. and Leslie, L.M. 2000. The Australian Boxing Day Storm of 1998 – Synoptic Description and Numerical Simulations. Weather & Forecasting Vol 15. 543-559.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 112 BIBLIOGRAPHY
Bureau of Meteorology, 2018. Joint Industry Project for Objective Tropical Cyclone Reanalysis: Final Report V1.3. http://www.bom.gov.au/cyclone/history/database/OTCR- JIP_FinalReport_V1.3_public.pdf. Bureau of Meteorology, 2020. Special Climate Statement 73 – extreme heat and fire weather in December 2019 and January 2020. http://www.bom.gov.au/current/statements/scs73.pdf. Bureau of Meteorology and CSIRO, 2016. State of the Climate 2016. Bureau of Meteorology and CSIRO. http://www.bom.gov.au/state-of-the-climate/State-of-the-Climate-2016.pdf. Bureau of Meteorology and CSIRO, 2018. State of the Climate 2018. Bureau of Meteorology and CSIRO. http://www.bom.gov.au/state-of-the-climate/State-of-the-Climate-2018.pdf. Cai, W., Santoso, A., Wang, G., Yeh, S.W., An, S.I., Cobb, K.M., Collins, M., Guilyardi, E., Jin, F.F., Kug, J.S. and Lengaigne, M., 2015. ENSO and greenhouse warming. Nature Climate Change, 5(9), p.849. Callaghan, J. and Power, S.B., 2011. Variability and decline in the number of severe tropical cyclones making land‐fall over eastern Australia since the late nineteenth century. Climate Dynamics, 37:647– 662. Callaghan, J. and Power, S.B., 2014. Major coastal flooding in southeastern Australia 1860–2012, associated deaths and weather systems. Australian Meteorological and Oceanographic Journal, 64, pp.183-213. Carmago, J.J., Giulivi, C. F., Sobel, A. H., Wing, A.A., Kim, D., Moon, Y., Strong, J. D. O., Del Genio, A. D., Kelley, M., Murakami, H., Reed, K. A., Scoccimarro, E., Vecchi, G., Wehner, M. F., Zarzyck, C. Zhao, M. 2020. Characteristics of Model Tropical Cyclone Climatology and the Large-Scale Environment. Journal of Climate. DOI: 10.1175/JCLI-D-19-0500.1. Camargo, S.J., 2013. Global and regional aspects of tropical cyclone activity in the CMIP5 models. Journal of Climate, 26:9880–9902. Cattiaux, J., Bousquet, S. M., 2020. Projected changes in the Southern Indian Ocean cyclone activity assessed from high-resolution experiments and CMIP5 Models. Journal of Climate. DOI 10.1175/JCLI-D-19-0591.1. Cavicchia, L., Pepler, A., Dowdy, A., Evans, J., Di Luca, A., Walsh, K. 2020. Future changes in the occurrence of hybrid cyclones: The added value of cyclone classification for the east Australian low- pressure systems. Geophysical Research Letters, 47, e2019GL085751. https://doi.org/10.1029/2019GL085751. Cecil, D.J., 2009. Passive microwave brightness temperatures as proxies for hailstorms. Journal of Applied Meteorology and Climatology, 48(6), pp.1281-1286. Chambers, C.R.S., Brassington, G.B., Sheng, J., Simmonds, I., and Walsh, K., 2016. Sea level pressure response to the specification of eddy-resolving sea surface temperature in simulations of Australian east coast lows. Chambers, C.R., Brassington, G.B., Simmonds, I. and Walsh, K., 2014. Precipitation changes due to the introduction of eddy-resolved sea surface temperatures into simulations of the “Pasha Bulker” Australian east coast low of June 2007. Meteorology and Atmospheric Physics, 125(1-2), pp.1-15. Chand, S. S., Dowdy, A. J., Ramsay, H. A., Walsh, K. J. E., Tory, K. J., Power, S. B., et al. (2019). Review of tropical cyclones in the Australian region: Climatology, variability, predictability, and trends. Wiley Interdisciplinary Reviews: Climate Change, 47(5), 183–17. http://doi.org/10.1002/wcc.602. Changnon, S.A., 2009. Increasing major hail losses in the US. Climatic Change, 96(1-2), pp.161-166. Changnon, S.A. and Changnon, D., 2000. Long-term fluctuations in hail incidences in the United States. Journal of Climate, 13(3), pp.658-664.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 113 BIBLIOGRAPHY
Chen, C., Cane, M.A., Wittenberg, A.T. and Chen, D., 2017. ENSO in the CMIP5 simulations: life cycles, diversity, and responses to climate change. Journal of Climate, 30(2), pp.775-801. Chen, Y.R., Yu, B. and Jenkins, G., 2013. Secular variation in rainfall intensity and temperature in eastern Australia. Journal of Hydrometeorology, 14(4), pp.1356-1363. Chiew, F.H., Piechota, T.C., Dracup, J.A. and McMahon, T.A., 1998. El Niño/Southern Oscillation and Australian rainfall, streamflow and drought: Links and potential for forecasting. Journal of Hydrology, 204(1-4), pp.138-149. Christensen J.H., Krishna Kumar, K., Aldrian, E., An, S.‐I., Cavalcanti, I.F.A., de Castro, M., Dong, W., Goswami, P., Hall, A., Kanyanga, J.K., et al. Climate phenomena and their relevance for future regional climate change. In: Stocker T.F., Qin D., Plattner G‐K., Tignor M., Allen S.K., Boschung J., Nauels A., Xia, Y., Bex,V., Midgley, P.M., eds. Climate Change 2013. The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC AR5). Cambridge, UK and New York, NY: Cambridge University Press. Church, J.A. and White, N.J., 2011. Sea-level rise from the late 19th to the early 21st century. Surveys in geophysics, 32(4-5), pp.585-602. Cirulis, B., Clarke, H., Boer, M., Penman, T., Price, O., Bradstock, R., 2019. Quantification of inter-regional differences in risk mitigation from prescribed burning across multiple management values. Int. J. Wildland Fire. Clarke, H. & Evans, J.P. Exploring the future change space for fire weather in southeast Australia. Theor Appl Climatol (2019), 136: 513. https://doi.org/10.1007/s00704-018-2507-4. Clarke, H., Gibson, R., Cirulis, B., Bradstock, R.A., Penman, T.D. 2019. Developing and testing models of the drivers of anthropogenic and lightning-caused wildfire ignitions in south eastern Australia. J. Environmental Management, Vol 235. Pp 34-41. https://doi.org/10.1016/j.jenvman.2019.01.055. Clarke, H.G., Smith, P.L. and Pitman, A.J., 2011. Regional signatures of future fire weather over eastern Australia from global climate models. International Journal of Wildland Fire, 20(4), pp.550-562. CSIRO and Bureau of Meteorology, 2015. Climate Change in Australia Information for Australia’s Natural Resource Management Regions: Technical Report, CSIRO and Bureau of Meteorology, Australia, https://www.climatechangeinaustralia.gov.au/en/publications-library/technical-report/. Cyclone Testing Station, 2018. North Queensland Study into Water Damage from Cyclones, https://www.jcu.edu.au/__data/assets/pdf_file/0009/748485/CTS_wind_driven_rain_20181018.pdf. Dangendorf, S., Hay, C., Calafat, F.M. et al. Persistent acceleration in global sea-level rise since the 1960s. Nat. Clim. Chang. 9, 705–710 (2019). https://doi.org/10.1038/s41558-019-0531-8. Deser, C., Phillips, A.S., Bourdette, V., and Teng, H., 2012. Uncertainty in climate change projections: The role of internal variability. Climate Dynamics, 38, 527–546. Dessens, J., Berthet, C. and Sanchez, J.L., 2015. Change in hailstone size distributions with an increase in the melting level height. Atmospheric Research, 158, pp.245-253. Diffenbaugh, N. S., and Giorgi, F. (2012). Climate change hotspots in the CMIP5 global climate model ensemble. Climatic Change, 114, 813–822, doi:10.1007/s10584-012-0570-x. Diffenbaugh, N.S., Singh, D., Mankin, J.S., Horton, D.E., Swain, D.L., Touma, D., Charland, A., Liu, Y., Haugen, M., Tsiang, M. and Rajaratnam, B. (2017). Quantifying the influence of global warming on unprecedented extreme climate events. Proceedings of the National Academy of Sciences, 114(19), pp.4881-4886. Diffenbaugh, N.S., Swain, D.L. and Touma, D. (2015). Anthropogenic warming has increased drought risk in California. Proceedings of the National Academy of Sciences, p.201422385.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 114 BIBLIOGRAPHY
Di Virgilio, G., Evans, J.P., Blake, S.A.P., Armstrong, M., Dowdy, A.J., Sharples, J., McRae, R. 2019. Climate Change increases the potential for extreme wildfires. American Geophysical Union. DOI: 10.1029/2019GL083699. Donat, M.G., A.J. Pitman, and S. I. Seneviratne, 2017, Regional warming of hot extremes accelerated by surface energy fluxes, Geophysical Research Letters, 44, 7011–7019, doi:10.1002/2017GL073733. Dowdy, A.J., 2018. Climatological variability of fire weather in Australia. Journal of Applied Meteorology and Climatology, 57(2), 221-234. Dowdy, A.J. 2020. Climatology of thunderstorms, convective rainfall and dry lightning environments in Australia. Climate Dynamics https://doi.org/10.1007/s00382-020-05167-9. Dowdy, A.J., Mills, G.A. 2012. Atmospheric and Fuel Moisture Characteristics Associated with Lightning- Attributed Fires. J. App. Meteor & Climatology. Vol 51. Pp 2025-2037. Dowdy, A.J., Mills, G.A., Timbal, B., 2011. Large-scale indicators of Australian East Coast Lows and associated extreme weather events. Centre for Australia Weather and Climate Research, Technical Report 37 (available from http://www.cawcr.gov.au/publications/). Dowdy, A.J., Mills, G.A. and Timbal, B., 2013a. Large‐scale diagnostics of extratropical cyclogenesis in eastern Australia. International Journal of Climatology, 33(10), pp.2318-2327. Dowdy, A.J., Mills, G.A., Timbal, B. and Wang, Y., 2013b. Changes in the risk of extratropical cyclones in eastern Australia. Journal of Climate, 26(4), pp.1403-1417. Dowdy et al. 2019. Review of Australian east coast low pressure systems and associated extremes. Climate Dynamics, https://doi.org/10.1007/s00382-019-04836-8. Dyer, A., Buckley, B.W., Conway, P., 2019. Regional Sensitivity of Flood Risk to Climate Drivers. Floodplain Management Conference Canberra, ACT 2019. EEA, 2018: European Environment Agency. https://www.eea.europa.eu/data-and-maps/indicators/global- and-european-temperature-8/assessment, accessed June 2018. Emanuel, K., 1991. The theory of hurricanes. Annu. Rev. Fluid Mech., 23, 179–196. Emanuel, K., 2017. Will global warming make hurricane forecasting more difficult? Bull. Amer. Meteor. Soc., 98, 495-501. Emanuel, K., Ravela, S., Vivant, E. and Risi, C. 2006. A Statistical deterministic approach to hurricane risk assessment. Bull. Am. Meteorol. Soc. 87, 299–314. Estrada, F., Botzen, W.J.W., and Tol, R.S.J., 2015. Economic losses from US hurricanes consistent with an influence from climate change Nat. Geosci. 8 880–4. Evans, J.P., Ji, F., Lee, C., Smith, P., Argüeso, D. and Fita, L., 2014. Design of a regional climate modelling projection ensemble experiment–NARCliM. Geoscientific Model Development, 7(2), pp.621- 629. Field, C.B. ed., 2012. Managing the risks of extreme events and disasters to advance climate change adaptation: special report of the intergovernmental panel on climate change. Cambridge University Press. Fischer, E.M., Sedláček, J., Hawkins, E. and Knutti, R., 2014. Models agree on forced response pattern of precipitation and temperature extremes. Geophysical Research Letters, 41(23), pp.8554-8562. Flannigan, M.D., Krawchuk, M.A., de Groot, W.J., Wotton, B.M. and Gowman, L.M., 2009. Implications of changing climate for global wildland fire. International journal of wildland fire, 18(5), pp.483-507.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 115 BIBLIOGRAPHY
Fowler, H.J., Lenderink, G., Prein, A.F., et al. (2020) Intensification of short-duration rainfall extremes with global warming and implications for flood hazard, Nature Reviews Earth & Environment. Fox-Hughes, P., Harris, R.M.B., Lee, G., Jabour, J., Grose, M.R., Remenyi, T.A. and Bindoff, N.L., 2015. Climate Futures for Tasmania future fire danger: the summary and the technical report, Antarctic Climate & Ecosystems Cooperative Research Centre, Hobart, Tasmania. Franks, S.W., 2002. Identification of a change in climate state using regional flood data. Hydrology and Earth System Sciences, 6(1), pp.11-16. Franzke, C. L. E. (2017). Impacts of a changing climate on economic damages and insurance. Econ. Disasters Clim. Chang. 1, 95–110. Freund, M., Henley, B.J., Karoly, D.J., Brown, J. 2020. Warming patterns affect El Niño diversity in CMIP5 and CMIP6 models. Journal of Climate. DOI: 10.1175/JCLI-D-19-0890.1. Fussell, E., Curran, S.R., Dunbar, M.D., Babb, M.A., Thompson, L., and Meijer-Irons, J. (2017). Weather- related hazards and population change: A study of hurricanes in the United States, 1980-2012. Annals of the American Academy of Political and Social Science 669, 146-167. Gallant, A.J., Karoly, D.J. and Gleason, K.L., 2014. Consistent trends in a modified climate extremes index in the United States, Europe, and Australia. Journal of Climate, 27(4), pp.1379-1394. Gallant, A.J., Reeder, M.J., Risbey, J.S. and Hennessy, K.J., 2013. The characteristics of seasonal‐scale droughts in Australia, 1911–2009. International Journal of Climatology, 33(7), pp.1658-1672. Galarneau Jr, T.J., Hamill, T.M., Dole, R.M. and Perlwitz, J. (2012). A multiscale analysis of the extreme weather events over western Russia and northern Pakistan during July 2010. Monthly Weather Review, 140(5), pp1639-1664. Gensini, V.A. and Mote, T.L., 2015. Downscaled estimates of late 21st century severe weather from CCSM3. Climatic Change, 129(1-2), pp.307-321. Gentry, M.S., Lackmann, G.M., 2010. Sensitivity of simulated tropical cyclone structure and intensity to horizontal resolution. Monthly Weather Review 138(3):688–704. Giorgi, F., Im, E.S., Coppola, E., Diffenbaugh, N.S., Gao, X.J., Mariotti, L. and Shi, Y., 2011. Higher hydroclimatic intensity with global warming. Journal of Climate, 24(20), pp.5309-5324. Grose, M.R., Narsey, S., Delage, F. P., Dowdy, A. J., Bador, M., Boschat, G., et al. (2020). Insights from CMIP6 for Australia's future climate. Earth's Future, 8, https://doi.org/10.1029/2019EF001469. Guerreiro, S.B., H.J. Fowler, R. Barbero, S. Westra, G. Lenderink, S. Blenkinsop, E. Lewis and X-F. Li, 2018, Detection of continental-scale intensification of hourly rainfall extremes, Nature Climate Change, 8, 803-807, doi: 10.1038/s41558-018-0245. Hague, B.S., McGregor, S., Murphy, B.F., Reef. R., Jones, D.A. 2020. Sea-Level Rise Driving Increasingly Predictable Coastal Inundation in Sydney, Australia. doi: 10.1029/2020EF001607. Hague, B.S., Murphy, B.F., Jones, D.A., Taylor, A.J. 2020. Developing impact based thresholds for coastal inundation from tide gauge observations. Journal of Southern Hemisphere Earth Systems Science. 2019, 69, 252–272. https://doi.org/10.1071/ES19024. Haigh, I.D., Wadey, M.P., Wahl, T., Ozsoy, O., Nicholls, R.J., Brown, J.M., Horsburgh, K. and Gouldby, B. (2016). Spatial and temporal analysis of extreme sea level and storm surge events around the coastline of the UK. Scientific data, 3, p.160107. Harris, S., Lucas, C. 2019. Understanding the variability of Australian fire weather between 1973 and 2017. PLoS ONE 14(9): e0222328. https://doi.org/10.1371/journal.pone.0222328.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 116 BIBLIOGRAPHY
Hartmann, D.L., Klein Tank, A.M.G., Rusticucci, M., Alexander, L.V., Brönnimann, S., Charabi, Y., Dentener, F.J., Dlugokencky, E.J., Easterling, D.R., Kaplan, A., et al. Observations: atmosphere and surface. In: Stocker T.F., Qin D., Plattner G‐K., Tignor M., Allen S.K., Boschung J., Nauels A., Xia Y., Bex V., Midgley P.M., eds. Climate Change 2013. The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK and New York, NY: Cambridge University Press. Haverd, V., Smith, B., Canadell, J.G., Cuntz, M., Mikaloff-Fletcher, S., Farquhar, G., Woodgate, W., Briggs, P.R., Trudinger, C.M. 2019. Higher than expected CO2 fertilization inferred from leaf to global observations. Glob Change Biol. 2020;26:2390–2402 DOI: 10.1111/gcb.14950. Haverd, V., Smith, B., Nieradzik, L., Briggs, P. R., Woodgate, W., Trudinger, C. M., Cuntz, M. (2018). A new version of the CABLE land surface model (Subversion revision r4601) incorporating land use and land cover change, woody vegetation demography, and a novel optimisation-based approach to plant coordination of photosynthesis. Geoscientific Model Development, 11, 2995–3026. Hayhoe, K., D.J. Wuebbles, D.R. Easterling, D.W. Fahey, S. Doherty, J. Kossin, W. Sweet, R. Vose, and M. Wehner, 2018: Our Changing Climate. In Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, Volume II [Reidmiller, D.R., C.W. Avery, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C. Stewart (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, pp. 72–144. doi: 10.7930/NCA4.2018.CH2.
Held, I.M. and Zhao, M., 2011. The Response of Tropical Cyclone Statistics to an Increase in CO2 with Fixed Sea Surface Temperatures. J. Climate, 24, 5353–5364. Henderson, D. J., Smith, D. J., Boughton, G.B., Ginger, J. D., (2018) Damage and loss to Australian engineered buildings during recent cyclones, International Workshop on Wind-Related Disasters and Mitigation, Tohoku University, Sendai, Japan. Hennessy, K., Fitzharris, B., Bates, B.C., Harvey, N., Howden, S.M., Hughes, L., Salinger, J. and Warrick, R., 2007. Australia and New Zealand in ML Parry, OF Canziana, JP Palitikof, PJ van der Linder, and CE Hanson, editors. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Hennessy, K., Lucas, C., Nicholls, N., Bathols, J., Suppiah, R. and Ricketts, J., 2005. Climate change impacts on fire-weather in south-east Australia. Climate Impacts Group, CSIRO Atmospheric Research and the Australian Government Bureau of Meteorology, Aspendale. Hill, K.A. and Lackmann, G.M., 2011. The Impact of Future Climate Change on TC Intensity and Structure: A Downscaling Approach. J. Climate, 24, 4644–4661. Hirabayashi, Y., Mahendran, R., Koirala, S., Konoshima, L., Yamazaki, D., Watanabe, S., Kim, H. and Kanae, S., 2013. Global flood risk under climate change. Nature Climate Change, 3(9), p.816. Holland, G.J., 1997. The Maximum Potential Intensity of Tropical Cyclones. J. Atmos. Sci., 54, 2519–2541. Holland, G.J., and Bruyère, C.L., 2014. Recent intense hurricane response to global climate change. Climate Dynamics, 42:617–627. Holland, G.J., Lynch, A.H. and Leslie, L.M., 1987. Australian east-coast cyclones. Part I: Synoptic overview and case study. Monthly Weather Review, 115(12), pp.3024-3036. Holland, G.J., and Webster, P.J., 2007. Heightened tropical cyclone activity in the North Atlantic: natural variability or climate trend? Philos Trans R Soc Lond A, 365:2695–2716. Holmes, J.D., 2020. Land-falling tropical cyclones on the Queensland coast and implications of climate change for wind loads. Australian Journal of Structural Engineering. Doi: 10.1080/13287982.2020.1717842.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 117 BIBLIOGRAPHY
Hopkins, L.C. and Holland, G.J., 1997. Australian heavy-rain days and associated east coast cyclones: 1958–92. Journal of Climate, 10(4), pp.621-635. Höppe, P., Pielke Jr, R.A., (eds). 2006. Workshop on climate change and disaster losses: Understanding and attributing trends and projections, final workshop report. Hohenkammer, Germany, May 2006. Huang, X., Hall, A.D. and Berg, N. (2018). Anthropogenic Warming Impacts on Today's Sierra Nevada Snowpack and Flood Risk. Geophysical Research Letters, 45(12), pp.6215-6222. ICCI, 2016: Thresholds and Closing Windows: Risks of irreversible climate change. International Cryosphere Climate Initiative, Pawlet, VT, USA. IPCC, 2013: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp. IPCC, 2018: Global Warming of 1.5°C an IPCC special report on the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty. http://www.ipcc.ch/report/sr15/. IPCC, 2019: Technical Summary [H.-O. Pörtner, D.C. Roberts, V. Masson-Delmotte, P. Zhai, E. Poloczanska, K. Mintenbeck, M. Tignor, A. Alegría, M. Nicolai, A. Okem, J. Petzold, B. Rama, N.M. Weyer (eds.)]. In: IPCC Special Report on the Ocean and Cryosphere in a Changing Climate [H.- O. Pörtner, D.C. Roberts, V. Masson-Delmotte, P. Zhai, M. Tignor, E. Poloczanska, K. Mintenbeck, A. Alegría, M. Nicolai, A. Okem, J. Petzold, B. Rama, N.M. Weyer (eds.). Ishak, E.H., Rahman, A., Westra, S., Sharma, A. and Kuczera, G., 2013. Evaluating the non-stationarity of Australian annual maximum flood. Journal of Hydrology, 494, pp.134-145. Jakob, D., Karoly, D.J. and Seed, A., 2011. Non-stationarity in daily and sub-daily intense rainfall–Part 2: Regional assessment for sites in south-east Australia. Ji F, Pepler A, Browning S, Evans JP, Di Luca A (2017) Trends and low frequency variability of East Coast Lows in the twentieth century. J South Hemisph Earth Syst Sci. https://doi.org/10.22499/3.6801.001. Jiang, M., Medlyn, B.E., Drake, J.E. et al. The fate of carbon in a mature forest under carbon dioxide enrichment. Nature 580, 227–231 (2020). https://doi.org/10.1038/s41586-020-2128-9. Johnson, N.C., Xie, S.P., 2010. Changes in the sea surface temperature threshold for tropical convection. Nat Geosci, 3:842–845. Jolly, W.M., Cochrane, M.A., Freeborn, P.H., Holden, Z.A., Brown, T.J., Williamson, G.J. and Bowman, D.M., 2015. Climate-induced variations in global wildfire danger from 1979 to 2013. Nature communications, 6, p.7537. Kendon, E.J., Roberts, N.M., Fowler, H.J., Roberts, M.J., Chan, S.C. and Senior, C.A., 2014. Heavier summer downpours with climate change revealed by weather forecast resolution model. Nature Climate Change, 4(7), p.570. King, A.D., and Henley, B.J., 2018. https://andrewdking.weebly.com/global-warming-gif-and-methods.html. King, A.D., Karoly, D.J. and Henley, B.J., 2017. Australian climate extremes at 1.5 C and 2 C of global warming. Nature Climate Change, 7(6), p.412. King, A.D., Klingaman, N.P., Alexander, L.V., Donat, M.G., Jourdain, N.C. and Maher, P., 2014. Extreme rainfall variability in Australia: Patterns, drivers, and predictability. Journal of Climate, 27(15), pp.6035- 6050.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 118 BIBLIOGRAPHY
King, A.D., Lane, T.P., Henley, B.J. et al. Global and regional impacts differ between transient and equilibrium warmer worlds. Nat. Clim. Chang. 10, 42–47 (2020). https://doi.org/10.1038/s41558-019- 0658-7. King, A.D., Lewis, S.C., Perkins, S.E., Alexander, L.V., Donat, M.G., Karoly, D.J. and Black, M.T., 2013. Limited evidence of anthropogenic influence on the 2011-12 extreme rainfall over Southeast Australia. Bulletin of the American Meteorological Society, 94(9), p.S55. Knutson, T.R., Camargo, S.J., Chan, J.C.L., Emanuel, K., Ho, C-H., Kossin, J., Mohapatra, M., Satoh, M., Sugi, M., Walsh, K., Wu, L. 2019. Tropical Cyclones and Climate Change Assessment: Part I. Detection and Attribution. Bulletin of the American Meteorological Society, Doi 10.1175/BAMS-D-18- 0189.1. Knutson, T.R., Camargo, S.J., Chan, J.C.L., Emanuel, K., Ho, C-H., Kossin, J., Mohapatra, M., Satoh, M., Sugi, M., Walsh, K., Wu, L. 2020. Tropical Cyclones and Climate Change Assessment: Part II. Projected Response to Anthropogenic Warming. Bulletin of the American Meteorological Society, Doi 10.1175/BAMS-D-18-0194.1. Knutson, T.R., and Co-authors, 2010. Tropical cyclones and climate change. Nat. Geosci.,3, 157–163, doi:10.1038/NGEO779. Kopp, R.E., C. Andrews, A. Broccoli, A. Garner, D. Kreeger, R. Leichenko, N. Lin, C. Little, J.A. Miller, J.K. Miller, K.G. Miller, R. Moss, P. Orton, A. Parris, D. Robinson, W. Sweet, J. Walker, C.P. Weaver, K. White, M. Campo, M. Kaplan, J. Herb, and L. Auermuller. New Jersey’s Rising Seas and Changing Coastal Storms: Report of the 2019 Science and Technical Advisory Panel. Rutgers, The State University of New Jersey. Prepared for the New Jersey Department of Environmental Protection. Trenton, New Jersey. Kopp, R.E., Hayhoe, K., Easterling, D.R., Hall, T., Horton, R., Kunkel, K.E., and LeGrande, A.N. (2017). Potential surprises – compound extremes and tipping elements. In: Climate Science Special Report: Fourth National Climate Assessment, Volume I [Wuebbles, D.J., Fahey, D.W., Hibbard, K.A., Dokken, D.J., Stewart, B.C., and Maycock, T.K. (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, pp. 411-429, doi: 10.7930/J0GB227J. Kossin, J.P., Emanuel, K., Vecchi, G.A., 2014. The poleward migration of the location of tropical cyclone maximum lifetime intensity. Nature, 509, 349-352. Kossin, J.P., Knapp, K.R., Olander, T.L., Velden, C.S. 2020. Global increase in major tropical cyclone exceedance probability over the past four decades. PNAS, June 2020. www.pnas.org/cgi/doi/10.1073/pnas.1920849117. Kuleshov, Y.R., Fawcett, R., Qi, L., Trewin, B., Jones, D., McBride, J., Ramsay, H., 2010. Trends in tropical cyclones in the South Indian Ocean and the South Pacific Ocean. J Geophys Res Atmos, 115:D01101. Kumjian, M. R., and K. Lombardo, 2020: A Hail Growth Trajectory Model for Exploring the Environmental Controls on Hail Size: Model Physics and Idealized Tests. J. Atmos. Sci., 77, 2765–2791, https://doi.org/10.1175/JAS-D-20-0016.1. Kunreuther, H. and Michel-Kerjan, E., 2009. At war with the weather: Managing large-scale risks in a new era of catastrophes. New York, NY, MIT Press. Lau, W. K. M. and Kim, K.-M. (2012) The 2010 Pakistan flood and Russian heat wave: teleconnection of hydrometeorological extremes. J. Hydrometeorol. 13, 392–403. Lavender, S.L. and Abbs, D.J., 2013. Trends in Australian rainfall: contribution of tropical cyclones and closed lows. Climate Dynamics 40: 317-326.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 119 BIBLIOGRAPHY
Lavender, S.L., Hoeke, R.K., and Abbs, D.J., 2018. The influence of sea surface temperature on the intensity and associated storm surge of tropical cyclone Yasi: a sensitivity study Nat. Hazards Earth Syst. Sci., 18, 795-805. Lavender, S.L., and Walsh, K.J.E., 2011. Dynamically downscaled simulations of Australian region tropical cyclones in current and future climates, Geophys. Res. Lett., 38, L10705. Lee, C., S. J. Camargo, A. H. Sobel, and M. K. Tippett, 2020: Statistical–Dynamical Downscaling Projections of Tropical Cyclone Activity in a Warming Climate: Two Diverging Genesis Scenarios. J. Climate, 33, 4815–4834, https://doi.org/10.1175/JCLI-D-19-0452.1 Lee, C.-Y., Tippett, M.K., Sobel, A.H., and Camargo, S.J., 2016. Rapid intensification and the bimodal distribution of tropical cyclone intensity. Nat. Commun., 7, 10625. Leonard, M., Westra, S., Phatak, A., Lambert, M., van den Hurk, B., McInnes, K., Risbey, J., Schuster, S., Jakob, D. and Stafford‐Smith, M. (2014). A compound event framework for understanding extreme impacts. Wiley Interdisciplinary Reviews: Climate Change, 5(1), pp.113-128. Leslie, L.M., Leplastrier, M. and Buckley, B.W., 2008. Estimating future trends in severe hailstorms over the Sydney Basin: A climate modelling study. Atmospheric Research, 87(1), pp.37-51. Li, C., Zwiers, F., Zhang, X., Chen, G., Lu, J., Li, G., Norris, J., Tan, Y., Sun, Y. and Liu, M., 2019. Larger increases in more extreme local precipitation events as climate warms. Geophysical Research Letters, 46(12), pp.6885-6891. Li, X., and Ye, X. (2015). Spatiotemporal characteristics of dry-wet abrupt transition based on precipitation in Poyang Lake basin, China. Water, 7(5), 1943-1958. Lin, N.K. and Emanuel. K., 2016. Grey swan tropical cyclones. Nat. Clim. Change 6, 106–111. Livneh, B. and Hoerling, M.P. (2016). The physics of drought in the U.S. Central great plains. J. Clim. 29, 6783–6804. Lopez‐Cantu, T., Prein, A.F. and Samaras, C., 2020. Uncertainties in Future US Extreme Precipitation From Downscaled Climate Projections. Geophysical Research Letters, 47(9), p.e2019GL086797. Louis, S., Couriel, E., Lewis, G., Glatz, M., Kulmar, M., Golding, J., and Hanslow, J., 2016. New South Wales East Coast Low Event – 3 to 7 June 2016, Weather, Wave and Water Level Matters. http://new.mhl.New South Wales.gov.au/data/realtime/wave/docs//2016New South WalesCoastalConferenceLouisCourieletal_Final.pdf (Accessed March 2018). Lucas, C., 2010. On developing a historical fire weather data-set for Australia. Australian Meteorological and Oceanographic Journal, 60(1), p.14. Lucas, C., Hennessy, K., Mills, G. and Bathols, J. 2007. Bushfire weather in southeast Australia: Recent trends and projected climate change impacts. Consultancy report for The Climate Institute of Australia. Bushfire CRC and CSIRO Marine and Atmospheric Research, 79 pp. Available from http://www.bushfirecrc.com/research/downloads/ climate-institute-report-september-2007.pdf. Mahoney, K., Alexander, M.A., Thompson, G., Barsugli, J.J. and Scott, J.D., 2012. Changes in hail and flood risk in high-resolution simulations over Colorado's mountains. Nature Climate Change, 2(2), p.125. Malan, N., Reason, C.J.C., Loveday, B.R., 2013. Variability in tropical cyclone heat potential over the Southwest Indian Ocean. J Geophys Res Oceans, 118:6734–6746. Matthews, S., Grootemaat, S., Kenny, B., Hollis, J., Fox-Hughes, P., Sauvage, S., Runcle, J., Holmes, A., Cummins, N. 2019. The Australian Fire Danger Rating System. Bushfire Natural Hazards CRC. McArthur, A.G., 1967. Fire behaviour in eucalypt forests. Forestry and Timber Bureau: Leaflet; no. 107
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 120 BIBLIOGRAPHY
McInnes, K.L., Leslie, L.M. and McBride, J.L., 1992. Numerical simulation of cut‐off lows on the Australian east coast: Sensitivity to sea‐surface temperature. International journal of climatology, 12(8), pp.783- 795. McMahon, G.M., Kiem, A.S., 2018. Large floods in South East Queensland, Australia: Is it valid to assume they occur randomly? Australasian Journal of Water Resources, DOI: 10.1080/13241583.2018.1446677. Medlyn, B., Zaehle, S., De Kauwe, M. et al. Using ecosystem experiments to improve vegetation models. Nature Clim Change 5, 528–534 (2015). https://doi.org/10.1038/nclimate2621. Merz, B., Elmer, F. and Thieken, A.H. (2009). Significance of" high probability/low damage" versus" low probability/high damage" flood events. Natural Hazards and Earth System Sciences, 9(3), pp.1033- 1046. Merz, B., Nguyen, V.D. and Vorogushyn, S. (2016). Temporal clustering of floods in Germany: do flood- rich and flood-poor periods exist? Journal of Hydrology, 541, pp.824-838. Micevski, T., Franks, S.W. and Kuczera, G., 2006. Multidecadal variability in coastal eastern Australian flood data. Journal of Hydrology, 327(1-2), pp.219-225. Miller, S.I., Carter, W., Stephens, R.G. 1984. Report of the Bushfire Review Committee on Bushfire Disaster Preparedness and Response in Victoria, Australia, following the Ash Wednesday Fires of 16 February, 1983. TEN.048.001.0002. Mills, G.A., Webb, R., Davidson, N.E., Kepert, J., Seed, A., and Abbs, D., 2010. The Pasha Bulker east coast low of 8 June 2007. Centre for Australia. Weather and Climate Research, Technical Report 23 (available from http://www.cawcr.gov.au/publications/). Moritz, M.A., Parisien, M.A., Batllori, E., Krawchuk, M.A., Van Dorn, J., Ganz, D.J. and Hayhoe, K., 2012. Climate change and disruptions to global fire activity. Ecosphere, 3(6), pp.1-22. Mumby, P.J., Vitolo, R., and Stephenson, D.B. (2011). Temporal clustering of tropical cyclones and its ecosystem impacts. Proc. Natl. Acad. Sci., 108:17626–17630. Murakami, H., Mizuta, R., Shindo, E., 2012. Future changes in tropical cyclone activity projected by multi‐ physics and multi‐SST ensemble experiments using the 60km‐mesh MRI‐AGCM. Climate Dynamics, 39:2569–2584. Murray, R. and Simmonds, I. 1991. A numerical scheme for tracking cyclone centres from digital data Part I: development and operation of the scheme. Australian Meteorological Magazine, 39, 155–166. Nerem, R.S., Beckley, B.D., Fasullo, J.T., Hamlington, B.D., Masters, D. and Mitchum, G.T., 2018. Climate-change–driven accelerated sea-level rise detected in the altimeter era. Proceedings of the National Academy of Sciences, p.201717312. National Environmental Science Programme 2019. Bushfires and climate change in Australia brochure. www.nespclimate.com.au. Nicholls, R.J. and Cazenave, A., 2010. Sea-level rise and its impact on coastal zones. science, 328(5985), pp.1517-1520. NOAA National Centers for Environmental Information, State of the Climate: Global Analysis for Annual 2016, published online January 2017, retrieved on September 11, 2017 from https://www.ncdc.noaa.gov/sotc/global/201613. NOAA National Centers for Environmental Information, State of the Climate: National Climate Report for January 2019, published online February 2019, retrieved on August 20, 2019 from https://www.ncdc.noaa.gov/sotc/national/201901.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 121 BIBLIOGRAPHY
Nott, J., Green, C., Townsend, I. and Callaghan, J., 2014. The world record storm surge and the most intense southern hemisphere tropical cyclone: New evidence and modelling. Bull. Am. Meteorol. Soc. 95, 757–765. Oakley, N.S., F. Cannon, E. Boldt, J. Dumas, F.M. Ralph, 2018: Origins and variability of extreme precipitation in the Santa Ynez River Basin of Southern California. J. Of Hydrology: Regional Studies, 19, 164-176. DOI: https://doi.org/10.1016/j.ejrh.2018.09.001. Oliver, E.C.J. and N.J. Holbrook, 2014: Extending our understanding of South Pacific gyre “spin‐up”: Modeling the East Australian Current in a future climate. J. Geophys. Res.-Oceans, 119(5), 2788– 2805, doi:10.1002/2013JC009591. Oliver, E.C.J., T.J. O’Kane and N.J. Holbrook, 2015: Projected changes to Tasman Sea eddies in a future climate. J. Geophys. Res.-Oceans, 120(11), 7150–7165, doi:10.1002/2015JC010993. Oliver, E.C.J., S.E. Perkins-Kirkpatrick, N. J. Holbrook and N. L. Bindoff, 2018b: Anthropogenic and natural influences on record 2016 marineheat waves. In “Explaining Extreme Events of 2016 from a Climate Perspective”. Bull. Am. Meterol. Soc., 99(1), S44–S48, Doi:10.1175/BAMSExplainingExtremeEvents2016.1. Oliver, E.C.J., S.J. Wotherspoon, M.A. Chamberlain and N.J. Holbrook, 2014: Projected Tasman Sea Extremes in Sea Surface Temperature through the Twenty-First Century. J. Clim., 27(5), 1980–1998, doi:10.1175/jcli-d-13-00259.1. Parackal, K., Mason, M., Henderson, D., Stark, G., Ginger, J., Somerville, L., Harper, B., Smith, D., and Humphreys, M., 2014, TR 60 - Investigation of Damage: Brisbane, 27 November 2014, Cyclone Testing Station, James Cook University, Australia. https://www.jcu.edu.au/cyclone-testing-station. Parker, C.L., Bruyère, C.L., Mooney, P.A. and A.H. Lynch, 2018. The response of land-falling tropical cyclone characteristics to projected climate change in northeast Australia. Climate Dynamics. doi:10.1007/s00382-018-4091-9. Pearce, H.G., Mullan, A.B., Salinger, M.J., Opperman, T.W., Woods, D. and Moore, J.R., 2005. Impact of climate change on long-term fire danger. National Institute of Water & Atmospheric Research Limited. Pendergrass, A.G., 2018. What precipitation is extreme?. Science, 360(6393), pp.1072-1073. Pendergrass, A.G., Knutti, R., Lehner, F., Deser, C., and Sanderson, B.M. (2017). Precipitation variability increases in a warmer climate, Scientific Reports, 7(1), 17966, doi:10.1038/s41598-017-17966-y. Penman, T.D., Bedward, M., Bradstock, R.A. 2014(a). National Fire Danger Rating System Probabilistic Framework Project. Bushfire and National Hazards CDC, Australia. Penman, T.D., Bradstock, R.A., Price, O.F., 2014(b). Reducing wildfire risk to urban developments: Simulation of cost-effective fuel treatment solutions in south eastern Australia. Environ. Model. Software 52, 166-175. Pepler, A., Coutts‐Smith, A., 2013. A new, objective, database of East Coast Lows. Australian Meteorological and Oceanographic Journal 63 (2013) 461–472. Pepler, A., Coutts‐Smith, A., and Timbal, B., 2014. The role of East Coast Lows on rainfall patterns and inter‐annual variability across the East Coast of Australia. International Journal of Climatology, 34(4), pp.1011-1021. Pepler, A.S., Di Luca, A., Ji, F., Alexander, L.V., Evans, J.P. and Sherwood, S.C., 2015. Impact of identification method on the inferred characteristics and variability of Australian East Coast Lows. Monthly Weather Review, 143(3), pp.864-877.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 122 BIBLIOGRAPHY
Pepler, A.S., Di Luca, A., Ji, F., Alexander, L.V., Evans, J.P. and Sherwood, S.C., 2016. Projected changes in east Australian midlatitude cyclones during the 21st century. Geophysical Research Letters, 43(1), pp.334-340. Pepler, A., Dowdy, A. 2020. A three-Dimensional perspective on Extratropical Cyclone Impacts. 2020. Journal of Climate V 33. 5635- 5649 DOI: 10.1175/JCLI-D-19-0445.1. Pielke Jr, R.A., Gratz, J., Landsea, C.W., Collins D., Saunders M.A., Musulin R., 2008. Normalized hurricane damage in the united states: 1900–2005. Natural Hazards Review. doi: 10.1061/ASCE1527-698820089. Power, S.B. and Callaghan, J., 2016. Variability in severe coastal flooding, associated storms, and death tolls in southeastern Australia since the mid–nineteenth century. Journal of Applied Meteorology and Climatology, 55(5), pp.1139-1149. Power, S.B. and Callaghan, J., 2020. Variability and decline in the number of severe tropical cyclones making landfall over northeastern Australia since the late nineteenth century: An update. AMOS Fremantle Feb 2020 and personal comms. Power, S., Casey, T., Folland, C., Colman, A. and Mehta, V., 1999. Inter-decadal modulation of the impact of ENSO on Australia. Climate Dynamics, 15(5), pp.319-324. Prein, A.F., 2018, 2019. Variability and trends in global hail environments. Personal communications. Prein, A.F. and Heymsfield A.J., 2020. Increased melting level height impacts surface precipitation phase and intensity, Nature Climate Change. Prein, A.F., and Holland G.J., 2018. A Global Algorithm to Estimate the Risk of Damaging Hail. Journal of Advances in Modeling Earth Systems, Submitted. Prein, A.F., Langhans, W., Fosser, G., Ferrone, A., Ban, N., Goergen, K., Keller, M., Tölle, M., Gutjahr, O., Feser, F. and Brisson, E., 2015. A review on regional convection‐permitting climate modelling: Demonstrations, prospects, and challenges. Reviews of geophysics, 53(2), pp.323-361. Prein, A.F., Liu, C., Ikeda, K., Trier, S.B., Rasmussen, R.M., Holland, G.J. and Clark, M.P., 2017b. Increased rainfall volume from future convective storms in the US. Nature Climate Change, 7(12), p.880. Prein, A.F., Rasmussen, R.M., Ikeda, K., Liu, C., Clark, M.P. and Holland, G.J., 2017a. The future intensification of hourly precipitation extremes. Nature Climate Change, 7(1), p.48. Price, OF, Bedward, M (2020 in press) Using a statistical model of past wildfire spread to quantify and map the likelihood of fire reaching assets and prioritise fuel treatments. International Journal of Wildland Fire. Price, O.F., Borah, R., Bradstock, R., Penman, T. 2015. An empirical wildfire risk analysis: the probability of a fire spreading to the urban interface in Sydney, Australia. International Journal of Wildland Fire 24, 597-606. Púčik, T., Groenemeijer, P., Rädler, A.T., Tijssen, L., Nikulin, G., Prein, A.F., van Meijgaard, E., Fealy, R., Jacob, D. and Teichmann, C., 2017. Future Changes in European Severe Convection Environments in a Regional Climate Model Ensemble. Journal of Climate, 30(17), pp.6771-6794. Quinting, J.F., M.J. Reeder, J.L. Catto, 2019: The intensity and motion of hybrid cyclones in the Australian region in a composite potential vorticity framework. Qua. J. Royal Met. Soc., DOI:10.1002/qj.3430. Raftery, A.E., Zimmer, A., Frierson, D.M.W., Startz, R., and Liu, P., 2017. Less than 2 °C warming by 2100 unlikely. Nature Climate Change 7, pp.637–641. doi:10.1038/nclimate3352.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 123 BIBLIOGRAPHY
Ramsay, H.A., 2013. The effects of imposed stratospheric cooling on the maximum intensity of tropical cyclones in axisymmetric radiative–convective equilibrium. Journal of Climate, 26:9977–9985. Ramsay, H.A., Chand, S.S., and Camargo, S.J. 2018. A Statistical Assessment of Southern Hemisphere Tropical Cyclone Tracks in Climate Models. Journal of Climate 31, pp 10081-10103 Doi: 10/1175/JCLI-D-18-0377.1. Rasmussen, K.L., Prein, A.F., Rasmussen, R.M. et al., 2017. Changes in the convective population and thermodynamic environments in convection-permitting regional climate simulations over the United States. Clim Dyn. https://doi.org/10.1007/s00382-017-4000-7. Romps, D.M., Seeley, J.T., Vollaro, D. and Molinari, J., 2014. Projected increase in lightning strikes in the United States due to global warming. Science, 346(6211), pp.851-854. Sanders, F. and Guyakum, J.R., 1980. Synoptic-Dynamic Climatology of the “Bomb”. Mon. Wea. Rev., 108, 1589–606. Sanderson, B.M., Fisher, R.A. A fiery wake-up call for climate science. Nat. Clim. Chang. 10, 175–177 (2020). https://doi.org/10.1038/s41558-020-0707-2. Sanderson, B.M., Xu, Y., Tebaldi, C., Wehner, M., O'Neill, B., Jahn, A., Pendergrass, A.G., Lehner, F., Strand, W.G., Lin, L. and Knutti, R., 2017. Community climate simulations to assess avoided impacts in 1.5 and 2° C futures. Earth System Dynamics, 8(3), p.827. Schaefer, J.M., Finkel, R.C., Balco, G., Alley, R.B., Caffee, M.W., Briner, J.P., Young, N.E., Gow, A.J. and Schwartz, R., 2016. Greenland was nearly ice-free for extended periods during the Pleistocene. Nature, 540(7632), p.252. Schröer, K., and Tye, M.R. (2018). Quantifying the nuisance damage contribution of convective precipitation using weather types. Journal of Flood Risk Management. doi:10.1111/jfr3.12491. Seneviratne, S.I., Corti, T., Davin, E.L., Hirschi, M., Jaeger, E.B., Lehner, I., Orlowsky, B. and Teuling, A.J. (2010). Investigating soil moisture–climate interactions in a changing climate: A review. Earth-Science Reviews, 99(3-4), pp.125-161. Seneviratne, S.I., Donat, M.G., Pitman, A.J., Knutti, R. and Wilby, R.L., 2016. Allowable CO 2 emissions based on regional and impact-related climate targets. Nature, 529(7587), p.477. Seneviratne, S.I., Nicholls, N., Easterling, D., Goodess, C.M., Kanae, S., Kossin, J., Luo, Y., Marengo, J., McInnes, K., Rahimi, M., Reichstein, M., Sorteberg, A., Vera, C., and Zhang, X. (2012). Changes in climate extremes and their impacts on the natural physical environment. In: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation [Field, C.B., Barros, V., Stocker, T.F., Qin, D., Dokken, D.J., Ebi, K.L., Mastrandrea, M.D., Mach, K.J., Plattner, G.-K., Allen, S.K., Tignor, M., and Midgley, P.M. (eds.)]. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC). Cambridge University Press, Cambridge, UK, and New York, NY, USA, pp. 109-230. Shaevitz, D.A., Camargo, S.J., Sobel, A.H., Jonas, J.A., Kim, D., Kumar, A., LaRow, T.E., Lim, Y.‐K., Murakami, H., Reed, K., et al., 2014. Characteristics of tropical cyclones in high‐resolution models in the present climate. J Adv Model Earth Syst, 6:1154–1172. Sharma, S., Mujumdar, P. (2017). Increasing frequency and spatial extent of concurrent meteorological droughts and heatwaves in India. Sci Rep 7, 15582. Sharma, A., Wasko, C. and Lettenmaier, D.P., 2018. If precipitation extremes are increasing, why aren't floods?. Water Resources Research, 54(11), pp.8545-8551. Sharmila, S. and Walsh, K.J., 2017. Impact of Large-Scale Dynamic versus Thermodynamic Climate Conditions on Contrasting Tropical Cyclone Genesis Frequency. J. Climate, 30, 8865–8883.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 124 BIBLIOGRAPHY
Sharples, J.J., Cary, G.J., Fox-Hughes, P., Mooney, S., Evans, J.P., Fletcher, M.S., Fromm, M., Grierson, P.F., McRae, R. and Baker, P., 2016. Natural hazards in Australia: extreme bushfire. Climatic Change, 139(1), pp.85-99. Shepherd, T.G., 2014. Atmospheric circulation as a source of uncertainty in climate change projections. Nature Geoscience, 7(10), p.703. Sillmann, J. and Croci‐Maspoli, M. (2009). Present and future atmospheric blocking and its impact on European mean and extreme climate. Geophysical Research Letters, 36(10). Slangen, A.B.A, Church, J.A., Agosta, C., Fettweis, X., Marzeion, B., and Richter, K., 2016. Anthropogenic forcing dominates global mean sea-level rise since 1970. Nature Clim. Change, 6, 701–705, doi:10.1038/nclimate2991. Slangen, A.B.A., Katsman, C.A., Van de Wal, R.S.W., Vermeersen, L.L.A. and Riva, R.E.M., 2012. Towards regional projections of twenty-first century sea-level change based on IPCC SRES scenarios. Climate dynamics, 38(5-6), pp.1191-1209. Smith, A., Katz, R., 2013. U.S. Billion-dollar Weather and Climate Disasters: Data Sources, Trends, Accuracy and Biases. Natural Hazards, doi:10.1007/s11069-013-0566-5. Soderholm, J., McGowan, H., Richter, H., Walsh, K., Weckwerth,T. and Coleman, M. (2016) The coastal convective interactions experiment (CCIE): Understanding the role of sea breezes for hailstorm hotspots in eastern Australia. Bulletin of the American Meteorological Society, 97, 1687–1698. Soderholm, J.S., McGowan, H., Richter, H., Walsh, K., Weckwerth, T.M., Coleman, M. 2017. An 18 year climatology of hailstorm trends and related drivers across southeast Queensland, Australia. Q.J.R.Meteorol.Soc, 143: 1123-1135, January 2017. DOI:10.1002/qj.2995. Solomon, S., et al., (eds), 2007. Climate change 2007: The physical science basis. In: Contribution of Working group I to the fourth assessment report of the IPCC, Vol. 4. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Speer, M.S., Wiles, P. and Pepler, A., 2009. Low pressure systems off the New South Wales coast and associated hazardous weather: establishment of a database. Australian Meteorological and Oceanographic Journal, 58(1), p.29. Steffan, w., Hughes, L., and Pearce. A. 2015. The Heat is On. Climate Change, extreme heat and bushfires in Western Australia. Climate Council of Australia. Stewart, M.G., Rosowsky, D.V., Huang, Z., 2003. Hurricane risks and economic viability of strengthened construction. Natural hazards review, 4(1):12–19. Storey, M.A., Price, O.F., Sharples, J.J., Bradstock, R.A. 2020 Drivers of long-distance spotting during wildfires in south-eastern Australia. International Journal of Wildland Fire 29(6) 459-472 https://doi.org/10.1071/WF19124. Strachan J., Vidale, P.L., Hodges, K., Roberts, M., Demory, M.‐E., 2013. Investigating global tropical cyclone activity with a hierarchy of AGCMs: the role of model resolution. Journal of Climate, 26:133– 152. Strazzo, S.E., Elsner, J.B., and LaRow, T.E., 2015. Quantifying the sensitivity of maximum, limiting, and potential tropical cyclone intensity to SST: Observations versus the FSU/COAPS global climate model. J. Adv. Model. Earth Syst., 7, 586–599. Sugi, M., Murakami, H., Yoshimura, J., 2012. On the mechanism of tropical cyclone frequency changes due to global warming. J Meteorol Soc Jpn, 90A:397–408. Sugi, M., Y. Yamada, K. Yoshida, R. Mizuta, M. Nakano, C. Kodama,and M. Satoh, 2020: Future changes in the global frequency of tropical cyclone seeds. SOLA, 16, 70−74, doi:10.2151/sola.2020-012.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 125 BIBLIOGRAPHY
Sun, Y., and Coauthors, 2017. Impact of Ocean Warming on Tropical Cyclone Size and Its Destructiveness. Nature Scientific Reports, 7(8514). Swain, D.L., Langenbrunner, B., Neelin, J.D., and Hall, A (2018). Increasing precipitation volatility in twenty-first-century California. Nature Clim Change, 8(5), 427. Sweet, W.V., Kopp., R.E., Weaver, C.P., Obeysekera, J., Horton, R.M., Thieler, E.R., and Zervas. C. 2017. Global and Regional Sea Level Rise Scenarios for the United States. NOAA Technical Report NOS CO-OPS 083. Te Linde, A.H., Bubeck, P., Dekkers, J.E.C., De Moel, H. and Aerts, J.C.J.H., 2011. Future flood risk estimates along the river Rhine. Natural Hazards and Earth System Sciences, 11(2), p.459. Teague, B., McLeod, R., Pascoe, S. 2010. Final Report of the 2009 Victorian Bushfires Royal Commission. Government Printer of Victoria, PP No 332-Session 2006-2010. Tippett, M. K., Camargo, S. J., & Sobel, A. H. (2011). A Poisson Regression Index for Tropical Cyclone Genesis and the Role of Large-Scale Vorticity in Genesis. Journal of Climate, 24(9), 2335–2357. http://doi.org/10.1175/2010JCLI3811.1. Toivanen, J., Engel, C. B., Reeder, M. J., Lane, T. P., Davies, L., Webster, S., et al. (2019). Coupled atmosphere-fire simulations of the Black Saturday Kilmore East wildfires with the Unified Model. Journal of Advances in Modeling Earth Systems, 11. https://doi.org/10.1029/2017MS001245. Tory, K.J., Chand, S.S., Dare, R.A., McBride, J.L., 2013. The development and assessment of a model‐, grid‐, and basin‐independent tropical cyclone detection scheme. Journal of Climate, 26:5493–550. Trenberth, K.E., Dai, A., Rasmussen, R.M. and Parsons, D.B., 2003. The changing character of precipitation. Bulletin of the American Meteorological Society, 84(9), pp.1205-1217. Trewin, B., and Smalley, R., 2013. Changes in extreme temperatures in Australia, 1910 to 2011. In 19th AMOS National Conference, Melbourne (pp. 11-13). Ukkola A.M., Pitman A.J., De Kauwe M.G., Abramowitz G., Herger N., Evans J., and Decker M., 2018, Evaluating CMIP5 model agreement for multiple drought metrics, J. Hydrometeorology, 19, 969-988, 10.1175/JHM-D-17-0099.1. van den Broeke, M., Box, J., Fettweis, X., Hanna, E., Noël, B., Tedesco, M., van As, D., van de Berg, W.J. and van Kampenhout, L., 2017. Greenland Ice Sheet Surface Mass Loss: Recent Developments in Observation and Modeling. Current Climate Change Reports, 3(4), pp.345-356. Vecchi, G.A., Delworth, T.L., Murakami, H. et al. 2019. Tropical cyclone sensitivities to CO2 doubling: roles of atmospheric resolution, synoptic variability and background climate changes. Clim Dyn 53, 5999– 6033. https://doi.org/10.1007/s00382-019-04913-y. Velden, C., Olander, T., Herndon, D., Kossin, J. 2017 Reprocessing the Most Intense Historical Tropical Cyclones in the Satellite Era using the Advanced Dvorak Technique. Monthly Weather Review, V145, 971-983. DOI:10.1175/MWR-D-16-0312.1. Vermeer, M., and Rahmstorf, S., 2009. Global sea level linked to global temperature. Proceedings of the National Academy of Sciences, 106(51), pp.21527-21532. Villarini, G., Lavers, D.A., Scoccimarro, E., Zhao, M., Wehner, M.F., Vecchi, G., Knutson, T., 2014. Sensitivity of tropical cyclone rainfall to idealized global scale forcings. Journal of Climate 27:4622– 4641. Villarini, G., Smith, J.A., Vitolo, R., and Stephenson, D.B. (2013). On the temporal clustering of US floods and its relationship to climate teleconnection patterns. Int J Climatol, 33:629–640.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 126 BIBLIOGRAPHY
Vitolo, R., Stephenson, D.B., Cook, I.M., and Mitchell-Wallace, K. (2009). Serial clustering of intense European storms. Meteorologische Zeitschrift, 18(4), 411-424. Vose, R.S., Applequist, S., Bourassa, M.A., Pryor, S.C., Barthelmie, R.J., Blanton, B., Bromirski, P.D., Brooks, H.E., DeGaetano, A.T., Dole, R.M. and Easterling, D.R., 2014. Monitoring and understanding changes in extremes: Extratropical storms, winds, and waves. Bulletin of the American Meteorological Society, 95(3), pp.377-386. Wahl, T., Haigh, I.D., Nicholls, R.J., Arns, A., Dangendorf, S., Hinkel, J. and Slangen, A.B.A., 2017. Understanding extreme sea levels for broad-scale coastal impact and adaptation analysis. Nature communications, 8, p.16075. Wahl, T., Jain, S., Bender, J., Meyers, S. D., and Luther, M.E. (2015). Increasing risk of compound flooding from storm surge and rainfall for major US cities. Nature Climate Change, 5(12), 1093. Walsh, K.J.E., 2015. Fine resolution simulations of the effect of climate change on tropical cyclones in the South Pacific. Climate Dynamics, 45:2619–2631. Walsh, K. J. E., M. Fiorino, C. W. Landsea, and K. L. McInnes, 2007: Objectively Determined Resolution- Dependent Threshold Criteria for the Detection of Tropical Cyclones in Climate Models and Reanalyses. J. Climate, 20, 2307–2314, https://doi.org/10.1175/JCLI4074.1. Walsh, K. J. E., McBride, J.L., Klotzbach, P.J., Balachandran, S., Camargo, S.J., Holland, G.J., Knutson, T.R., Kossin, J.P., Lee, T.-C., Sobel, A., and Sugi, M., 2016a. Tropical cyclones and climate change. WIREs Climate Change, 7, 65-89, doi:10.1002/wcc371. Walsh, K., White, C.J., McInnes, K., Holmes, J., Schuster, S., Richter, H., Evans, J.P., Di Luca, A. and Warren, R.A., 2016b. Natural hazards in Australia: storms, wind and hail. Climatic Change, 139(1), pp.55-67. Warren, R.A., Ramsay, H.A., Siems, S.T., Manton, M.J., Peter, J.R, Protat, A., Pillalamarri, A. 2020. Radar-based climatology of damaging hailstorms in Brisbane and Sydney, Australia. Q.J.R.Meteorol.Soc, 146: 505-530, January 2020. DOI:10.1002/qj.3693. Wasko, C., Sharma, A., Westra S., 2016. Reduced spatial extent of extreme storms at higher temperatures. Geophys. Res. Lett., 43 (2016), pp. 4026–4032 https://doi.org/10.1002/2016GL068509. Wehner, M., Prabhat, Reed, K.A., Stone, D., Collins, W.D., and Bacmeister, J., 2015. Resolution Dependence of Future Tropical Cyclone Projections of CAM5.1 in the U.S. CLIVAR Hurricane Working Group Idealized Configurations. J. Climate, 28, 3905–3925. Weinkle, J., Maue, R., Pielke Jr, R., 2012. Historical global tropical cyclone landfalls. Journal of Climate 25(13):4729–4735. Westra, S., Fowler, H.J., Evans, J.P., Alexander, L.V., Berg, P., Johnson, F., Kendon, E.J., Lenderink, G. and Roberts, N.M., 2014. Future changes to the intensity and frequency of short‐duration extreme rainfall. Reviews of Geophysics, 52(3), pp.522-555. Westra, S. and Sisson, S.A., 2011. Detection of non-stationarity in precipitation extremes using a max- stable process model. Journal of Hydrology, 406(1-2), pp.119-128. White, N.J., Haigh, I.D., Church, J.A., Koen, T., Watson, C.S., Pritchard, T.R., Watson, P.J., Burgette, R.J., McInnes, K.L., You, Z.J. and Zhang, X., 2014. Australian sea levels—Trends, regional variability and influencing factors. Earth-Science Reviews, 136, pp.155-174. Wiese, D.N., Yuan, D-N., Boening, C., Landerer, F.W., Watkins, M.M. 2017 Antarctica mass variability time series version 1 from JPL GRACE Mascon CRI Filtered, version 1 (Physical Oceanography Distributed Active Archive Center, 2017); https://doi.org/10.5067/TEMSC-ANTS1.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 127 BIBLIOGRAPHY
Witt, A., Eilts, M.D., Stumpf, G.J., Johnson, J.T., Mitchell, E.D.W. and Thomas, K.W., 1998. An enhanced hail detection algorithm for the WSR-88D. Weather and Forecasting, 13(2), pp.286-303. Woodruff, J.D., Irish, J.L., Camargo, S.J., 2013. Coastal flooding by tropical cyclones and sea‐level rise. Nature, 504:44–52. Xie, B., Zhang, Q. and Wang, Y., 2008. Trends in hail in China during 1960–2005. Geophysical Research Letters, 35(13). Yeager, S.G., et al., 2018: Predicting near-term changes in the Earth System: A large ensemble of initialized decadal prediction simulations using the Community Earth System Model, Bull. Amer. Meteor. Soc., in press, doi: 10.1175/BAMS-D-17-0098.1. Yin, J., Gentine, P., Zhou, S., Sullivan, S.C., Wang, R., Zhang, Y., Guo. 2018 Large increase in global storm run-off extremes driven by climate and anthropogenic changes. Nature Communications. DOI: 10.1038/s41467-018-06765-2. Zhao, M., Held, I.M., Vecchi, G., Scoccimarro, E., Wang, H., Wehner, M., Lim, Y.‐K., LaRow, T., Camargo, S.J., Walsh, K., et al., 2013. Robust direct effect of increasing atmospheric CO2 concentration on global tropical cyclone frequency—a multi‐model inter‐comparison. CLIVAR Variations, 11:17–23.
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 128
CONTACTS
If you have any questions or feedback, please email [email protected].
Image credits: Kiri Armstrong & Nick Annand (front cover), Louise Kerkham (back cover - top), Peter Chan (back cover – bottom right), IAG (back cover – bottom left)
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 129 CONTACTS
SEVERE WEATHER IN A CHANGING CLIMATE – 2ND EDITION 130