Costing the impacts of current climate extremes for key vulnerable sectors in Victoria
Adriana Keating, John Handmer & Joshua Whittaker
Centre for Risk and Community Safety
ISBN: 978 0 7340 4867 7
Preface This report is a product of the research project “Framing multi-level and multi-actor adaptation responses in the Victorian context.” Previous working papers in this component analysed the variety of estimates of both the aggregate costs of disasters and the cost of specific climate-related events in Victoria (Keating and Handmer, 2011b) and methodologies for assessing the costs of climate change (Keating and Handmer 2011a). This report extends this work by taking a sectoral approach. Future work on the project will include exploring how the costs estimated here may change under climate change scenarios.
This report was prepared by Adriana Keating, John Handmer and Joshua Whittaker, Centre for Risk and Community Safety, School of Mathematics and Geospatial Science, at RMIT University.
In September 2010, VCCCAR held a workshop with key Victorian government representatives and academics to elicit expert opinion and guidance on the selection of an appropriate methodology for costing current and future climate change impacts in Victoria. Experts at this workshop discussed many of the issues presented in this report and their insights are much appreciated.
Climate change is expected to increase the frequency and severity of extreme events in Victoria. Demand exists, in both government and the private sector, for estimates of the cost of these climate change impacts so that potential abatement and adaptation options can be evaluated in economic terms. In order to estimate how costs related to extreme events will increase under climate change, the current costs of extreme events must first be estimated.
More frequent and severe heatwaves, droughts, bushfires, storms, winds and floods are projected for Victoria. Several studies have looked at Victorian sectoral vulnerabilities to extreme events now and under climate change. While definitions of vulnerability and study focuses differ, there is general agreement that agriculture and health are vulnerable sectors.
There is scant data on the economic impacts of climate anomalies at the sectoral level in Victoria. Pronouncements made in the media about the cost of disasters to certain sectors seem to be made with little or no empirical backing. Economic impact assessment of disasters is an involved process that is rarely undertaken in any consistent manner. Stephenson (2010) undertook a thorough assessment of the cost of bushfires to Victoria, and the raw data from this study was used to produce the estimates presented here on the current cost of bushfires to agriculture and the timber industry. Estimates of these costs under climate change can be found in the subsequent working paper for this series, Future potential losses from extremes under climate change: the case of Victoria, Australia.
Using some available data this report makes conservative estimates of the current costs of bushfires to the Victorian agricultural and timber industries, the cost of heatwave mortality to Victoria and the costs of climate anomalies to the Victorian public sector. The report estimates that:
Bushfire costs the Victorian agricultural industry approximately $42 million per annum. If we include business disruption, the total cost to the Victorian economy is approximately $92 million per annum.
Bushfire costs the Victorian timber industry approximately $74 million per annum. If we include business disruption the total annual cost to the Victorian economy is approximately $185 million per annum.
Heatwave mortality in Melbourne results in approximately 330 deaths costs Victoria approximately $1.26 billion annually.
Climate disasters cost the Victorian public sector approximately $424 million per annum. Note this accounts for direct expenditure in terms of output and asset investments only.
These estimates are generally considered to be underestimates. A comprehensive assessment of sectoral level economic impacts from climate anomalies in Victoria would provide greater backing to climate change adaptation decision-making. Assessments of this type would be drawn from partial equilibrium analysis and would preferably utilise standard economic impact assessment methodologies as outlined in earlier work on this project (Keating and Handmer 2011a).
Many gaps remain in our knowledge regarding options for adaptation to the increased frequency and severity of extreme events under climate change. The literature has canvassed a few options, these relate to both physical and social/institutional changes. The field of climate change adaptation economics is in its infancy and is currently grappling with the complex interactions and uncertainties that confound estimates about the probable costs, benefits and distribution of climate change impacts and adaptation options.
1. Introduction 1
2. Current and future extremes in Victoria 2
3. Victoria’s key vulnerable sectors 6
4. Current cost of climate anomalies in key vulnerable sectors 11
5. Conclusion 18
6. References 20
List of Tables
Table 1: Vulnerability to climate change impacts in nine major Victorian sectors
Table 2: Costs of bushfires to Victorian agricultural industry
Table 3: Costs of bushfires to Victorian timber industry
Table 4: Estimated costs of climate related disasters to the Victorian State Budget
List of Acronyms
ABARE: Australian Bureau of Agricultural and Resource Economics
ABS: Australian Bureau of Statistics
AEMI: Australian Emergency Management Institute
AUD: Australian Dollars
CFA: Country Fire Authority
CSIRO: Commonwealth Scientific and Industrial Research Organisaiton
IPCC: Intergovernmental Panel on Climate Change
VCCCAR: Victorian Centre for Climate Change Adaptation Research
Climate change is expected to increase the frequency and intensity of extreme climatic events worldwide (IPCC 2007). Events such as bushfires, cyclones, drought, floods, heatwaves and storms are predicted to become more frequent and severe throughout Australia, with impacts varying by region, sector and social group (Henessy et al. 2007; Garnaut 2008). In Victoria, climate change is expected to lead to higher average annual temperatures, more days above 35°C, reduced rainfall and more frequent droughts, more extreme weather events such as storms, high winds and floods, an increase in the number of extreme fire danger days, more frequent bushfires, rising sea levels and storm surges (Department of Premier and Cabinet 2009).
Demand exists, in both government and the private sector, for estimates of the cost of these climate change impacts so that potential abatement and adaptation options can be evaluated in economic terms. Increasing frequency and severity of extreme weather events are expected to be one of the first significant impacts of climate change felt by the people of Victoria. In order to estimate how costs related to extreme events will increase under climate change the current costs of extreme events must first be estimated.
This report sits within the wider Victorian Centre for Climate Change Adaptation Research (VCCCAR) project entitled “Framing multi-level and multi-actor adaptation responses in the Victorian context.” This project includes a work package investigating a “preliminary economic analysis of climate change impacts.” Previous working papers in this series have explored the confounding variety of estimates of both the aggregate costs of disasters and the cost of specific events in Victoria (Keating and Handmer, 2011b), and methodologies for assessing the costs of climate change in Victoria (Keating and Handmer 2011a). This report extends this work by taking a sectoral approach. Future work on the project will include exploring how the costs estimated here may change under various climate change scenarios.
Unfortunately, establishing cost estimates to inform climate change related decision-making is a complex and uncertain exercise. The field is currently in its infancy and the estimates that do exist are plagued by issues around data availability, methodological weaknesses and modelling uncertainty in both climate and impact models (Parry et al. 2009). Estimating the costs of current climate anomalies to Victorian sectors is hampered by a lack of available estimates on the economic impacts of disasters (Keating and Handmer, 2011b). Note that while cost and/or impact estimates are sometimes available for industry specific impacts as a result of a specific climate anomaly (for example the cost of the 2009 heatwave to Victorian infrastructure), these individual numbers are insufficient for the calculation of annualised estimates of the cost of climate anomalies for specific sectors, which is the goal of this report.
This report starts with an outline of the current extremes faced by Victoria, and how these are expected to increase in frequency and severity under climate change. Extreme heat and bushfires are key hazards of concern for Victoria in the coming century. Vulnerability to increased climate anomalies under climate change has been analysed by a few studies, and the results of these are also explored. While definitions of vulnerability and differing study aims and approaches mean no strict agreement is identified, agriculture is consistently identified as a key vulnerable sector (see section 3 below).
Using available data this report makes conservative estimates of the current costs of bushfires to the Victorian agricultural and timber industries, the cost of heatwave mortality to Victoria and the costs of climate anomalies to the Victorian public sector. These estimates are generally considered to be underestimates.
2. Current and future extremes in Victoria
This section of the report provides an overview of the major climatic extremes facing Victoria now and in the future. As mentioned above, climate change is expected to increase the frequency and intensity of extreme climatic events in Victoria with the likelihood of more frequent and severe heatwaves, droughts, bushfires, storms, winds and floods.
2.1 Extreme heat / heatwaves
Extreme heat is a major cause of hazard-related fatalities in Australia. Heatwaves have killed approximately 70% as many people as all other hazards combined (Blong, 2004). Severe heatwaves in south-east Australia in 1895, 1908 and 1939 led to particularly large losses of life (QUT 2010). In late January and early February 2009, south-east Australia was affected by an exceptional heatwave that saw records set for both high day and night time temperatures and for the duration of an extreme heat event (Bureau of Meteorology 2009). Melbourne experienced three consecutive days with maximum temperatures over 43°C from 28-30 January and unusually high night-time temperatures. A record-high maximum temperature of 46.4°C was recorded in Melbourne on February 7, when bushfires ravaged the state (Bureau of Meteorology 2009). Some 374 excess deaths were recorded during the heatwave, with mortality rates highest among those aged 75 years or older. This was the seventh deadliest disaster in the world in 2009 (Munich Re 2010). The heatwave also had significant impacts on critical infrastructure, particularly power and transport systems and infrastructure (QUT 2010).
There is no standard definition of heatwave in Australia. The Bureau of Meteorology defines a heatwave as ‘a period of abnormally hot weather lasting several days’ (Bureau of Meteorology 2011a). The Victorian Department of Health states that “a heatwave is generally defined as a period of abnormally and uncomfortably hot weather that could impact on human health, community infrastructure and services” (Department of Health, 2011, pg. 2). The American Red Cross (2011) defines a heatwave as ‘a prolonged period of excessive heat, often combined with excessive humidity’. Notwithstanding the differences in definition, climate change is expected to increase the risk of heatwaves (Hennessy et al, 2007, Wang & McAllister 2011). The IPCC’s Fourth Assessment report notes that heatwaves ‘are virtually certain to increase in intensity and frequency (high confidence)’, with increasing risks to human populations and infrastructure (Hennessy et al. 2007, p. 509).
Melbourne’s current annual average of nine days over 35°C is expected to increase to 12 by 2030, 21 by 2070 and 27 by 2100 under a no-mitigation scenario (Garnaut 2008). Regardless of national or international climate change mitigation commitments, adaptation to increased heatwave stress is required. Following the 2009 heatwave Victoria developed a Heatwave Plan for Victoria (Department of Health, 2011). The development of this plan is an example of adaptation in response to climatic extremes; early warning systems, health system and hospital emergency department preparations, promotion of behavioural change and infrastructure improvements (better designed homes and cities) are all examples of heatwave adaptation options (Wang & McAllister, 2011).
Inter-decadal variation in rainfall is a characteristic that has driven the evolution of many Australian ecosystems. Dry times are generally interpreted as ‘drought’ because they interrupt agricultural systems (Smith 2004). The Bureau of Meteorology states that there is no agreed upon definition of ‘drought’ because people use water in many different ways. The Bureau of Meteorology monitors and reports on rainfall deficiencies, however it is the responsibility of the Victorian State Government to declare a Victorian drought in consideration of factors other than rainfall such as agricultural impact, ground water levels and social expectations (Bureau of Meteorology, 2011e).
The ‘Federation Drought’ of 1895-1902 devastated large parts of the country, causing livestock numbers to plummet and the lowest wheat crop yields on record (Bureau of Meteorology 2011b). Severe droughts were experienced across Australia in 1937-45 and 1965-68, with 1967 remaining the driest year on record in Melbourne (Bureau of Meteorology 2011b). The drought of 1982-83 was particularly severe in Victoria, causing dust storms and contributing to the devastating Ash Wednesday bushfires that killed 71 people and destroyed more than 2000 houses. Total losses attributed to the drought exceeded $3 billion (Bureau of Meteorology 2011b).
Since the mid-1990s, the majority of Victoria has experienced severe drought conditions, characterised by the lowest streamflow in approximately 80 years of record (Kiem and Verdon-Kidd 2010). The drought of 2002/03 was estimated to have cost approximately 1 percentage point in GDP growth, despite the fact that the farm sector accounts for only 3.5% of GDP (Horridge et al, 2003). The social and economic impacts of drought have been particularly severe for farming families and rural communities. Although not solely attributable to drought, the number of farming families declined by 9% from 112,800 in 2001 to 102,600 in 2006 (ABS 2006). Employment in the agricultural sector fell by 19% over this period, with the greatest annual fall (14%) coinciding with severe drought conditions in 2002-03 (ABS 2006). Between 2005 and 2007, average cash income for Victorian farms declined by $16,000 to $39,240 per year, while the number of farms with negative cash income increased from 20% to 35% (ABARE 2008).
Climate change is expected to lead to drier conditions throughout the state. Droughts have become hotter since about 1973 due to higher temperatures during periods of rainfall deficiency (Nicholls 2004). A high emissions path would lead to warming of 1.8°C to 3.8°C by 2070, with a rainfall change of -25% to +3%. Warming is likely to be greater in northern regions, while greater drying is expected in southern regions (Department of Premier and Cabinet 2009).
Botterill (2004) tracks the development of drought policy in Australia and describes a change from viewing drought as a natural disaster to a normal feature of Australia’s climate. This policy shift has led to an increasing focus on farm management and resilience, but Botterill (2004) argues that this shift is constrained by a policy landscape influenced by emotive factors and (understandable) sympathy for farm hardship.
Climate, vegetation and dense settlement make bushfires a particularly destructive hazard in Victoria. There is a close connection between major devastating bushfires and severe droughts. Early events in the state’s history include the 1851 “Black Thursday” fires which burnt about 5 million hectares (one quarter of the state) and resulted in widespread destruction of faming communities, although loss of life was low; the 1939 ‘Black Friday’ bushfires, which burned around five million hectares, claimed 12 lives and destroyed around one million sheep and cattle, and the 1898 ‘Red Tuesday’ fires, which claimed 12 lives and destroyed more than 2000 buildings. Bushfires in 1926 killed 60 people and caused widespread damage to farms, homes and forests, while the 1939 ‘Black Friday’ fires burned around 2 million hectares, claimed 71 lives and destroyed 650 houses. More recent events include ‘Ash Wednesday’ (1983) and ‘Black Saturday’ (2009). The 1983 fires, which as with the other “named fires’ coincided with a severe drought, saw 47 Victorians lose their lives and more than 2000 houses destroyed. Another 28 people were killed in South Australia. In 2009, 173 people lost their lives and 2133 houses were destroyed in fires that burned on Melbourne’s outskirts and in other highly populated areas. The fires burned under the most severe fire weather conditions ever recorded, with a record high maximum temperature of 46.4°C in Melbourne, record low relative humidity and strong winds throughout the state (Bureau of Meteorology 2009; Karoly 2009).
Climate change is expected to increase the frequency and severity of extreme fire danger in south-east Australia. The IPCC’s Fourth Assessment Report states that ‘an increase in fire danger in Australia is likely to be associated with a reduced interval between fires, increased fire intensity, a decrease in fire extinguishments and faster fire spread’ (Hennessy et al. 2007, p. 515). The number of extreme fire danger days in south-east Australia is likely to increase by 15-65% by 2020 relative to 1990 and by 100-300% by 2050 for a high rate of global warming (Lucas et al. 2007; CSIRO 2009). Exposure to bushfire hazard is also set to increase. Victoria’s population is expected to grow by 2.27 million over the 20-year period to 2026, while the number of households is projected to rise by 54.6% from 2006 to 2036 (Department of Planning and Community Development 2009).
Options for adapting to increased frequency and severity of bushfire events in Victoria have not been thoroughly explored to date. The Royal Commission in response to the bushfires of February 7 2009 (VBRC 2009) stresses a need for improved land use planning in the outer metropolitan regions of Melbourne, as well as improved emergency service provision.
2.4 Floods, storms and high winds
Floods, storms and high winds have caused considerable damage in Victoria. Flood types can be loosely grouped into riverine, flash-floods and storm water drainage. Relatively minor flooding as a result of storm-water drain surcharge is not included in flood loss estimates. Long-term average flood damage costs for Victoria are estimated at $350 million per annum, with major regional flooding occurring every 10 to 20 years (Comrie 2011). Riverine flooding has tended to occur in the central, north-east and Gippsland regions, although significant events have occurred in the north and south-west. Major flooding has also occurred along the Yarra, Barwon and Maribyrnong rivers, with flash flooding occurring in urban areas.
One of Victoria’s worst flood disasters occurred in December 1934 when rain in excess of 350mm fell in the catchment for the Yarra River and 140mm fell in Melbourne within 48 hours. The Yarra and other metropolitan rivers and creeks broke their banks, inundating suburban areas and forcing evacuations. The Yarra Valley, South Gippsland and the La Trobe River District were severely flooded, with damages to buildings, livestock and crops exacerbated by gale force winds. 36 people lost their lives in the floods and in Melbourne 400 houses and factories were inundated (Bureau of Meteorology 2011b).
Between September 2010 and February 2011, heavy rainfall associated with a strong La Niña event caused widespread flooding across Victoria. Floods in September 2010 mostly affected rural areas and townships in the state’s west and north-east. Around 35 rivers were flooded, causing damage to farms and houses and forcing thousands to evacuate. Damage caused by heavy rain and strong winds cut power to more than 40,000 homes. Many of the areas affected were flooded again in January and February 2011, in addition to parts of metropolitan Melbourne. Victoria experienced its wettest January on record in 2011, recording a state-wide average of 118.58mm compared to the long-term average of 39.72mm (Bureau of Meteorology 2011c). Significant rainfall and flooding events also occurred during February, including an ‘extreme’ event on February 4 that caused flash flooding across metropolitan Melbourne (Bureau of Meteorology 2011d). Around one-third of Victoria experienced some form of flooding or storm damage between September and February. An estimated 4000 houses were damaged by the floods, in addition to damage and disruption to farms, critical infrastructure and essential services. The floods are estimated to have cost the agricultural sector in excess of $269 million in losses and damages, with lost revenue for the tourism sector at around $176 million (Comrie 2011). Premier Baillieu was widely reported to estimate the cost to infrastructure as being well in excess of $60 billion (Australian Broadcasting Corporation 2011). The estimated gross cost of the floods is in excess of $1.3 billion (Comrie 2011).
Climate change is expected to increase the risk of damaging floods, storms and winds. The IPCC predicts an increase in the frequency of high-intensity rainfall and strong wind events in south-east Australia, with likely increases in associated damages (Hennessy et al. 2007).
Response and adaptation options to increased frequency and severity of flooding in Victoria fall into two broad categories – physical responses and institutional changes (including warning systems). SMEC (2010) suggest that infrastructure could be upgraded to prevent flood damage, and Wang & McAllister (2001) suggest updating building codes and incorporating flood considerations into land use planning. Comrie’s (2011) review of the handling of the 2010-11 Victorian flood warnings and response recommended substantial overhaul of State and local government emergency management in order to improve planning and response coordination, including an expanded flood warning system.
3. Victoria’s key vulnerable sectors
3.1 Identifying vulnerable sectors
Climate change is likely to affect key sectors in different ways and to varying degrees. Just as some regions may be more or less affected, some sectors may experience greater exposure to climatic extremes and may have limited capacities to adapt. The identification and costing of likely impacts will enable planning and preparation to minimise impacts on key sectors and facilitate adaptation.
The IPCC has identified a number of sectors that may be vulnerable to climate change in Australia (Hennessy et al. 2007). The IPCC report considers the potential impacts of climate change on: freshwater resources; natural ecosystems; agriculture; forestry; coasts; fisheries; settlements, industry and societies; indigenous people; tourism and recreation; energy; and human health. Hennessy et al. (2007) identify a number of sectors as being particularly vulnerable:
Cropping is vulnerable due to potential increases in pests, diseases and weeds and reduced yields, although significant regional differences exist. While moderate yield increases are likely in north-east Australia, cropping is likely to become non-viable at the dry margins of southern Australia if rainfall is reduced substantially or if variability increases significantly. Irrigation-dependent horticulture is also likely to be affected by reduced water availability, while warmer conditions may negatively affect temperate fruit and nuts and lead to the spread of the Queensland fruit fly Bactrocera tryoni into southern Australia. Viticulture is also likely to be negatively affected by earlier ripening and reductions in grape quality by 2030. Damage from bushfires to the viticulture industry has seen significant losses in recent years (Stephenson 2010). Pastoral and rangeland farming is likely to be affected by reduced pasture growth,1 increasing land degradation problems such as erosion and salinity, livestock heat stress, and the increasing southward movement of the cattle tick (Boophilus microplus) (Hennessy et al. 2007).
Tourism and recreation:
While some tourist destinations may benefit from drier and warmer conditions, climate change is likely to have significant impacts on snow-based tourism in south-eastern Australia. For the full range of SRES scenarios, 2020 is likely to see between 5 and 40 fewer days of snow cover per year, a rise in the snowline of between 30 and 165 metres and a reduction in the total snow-covered area of 10 to 40 percent. By 2050 it is predicted that there will be between 15 and 100 fewer days of snow cover per year, a rise in the snowline of between 60 and 570 metres, a reduction in maximum snow-depth of between 10 and 99 percent, and a decline in total snow-cover of 20 to 85 percent. The tourism sector as a whole is likely to face greater risks from increases in hazards such as flooding, storm surges, heatwaves, cyclones, fires and drought (Hennessy et al. 2007).
The health sector is likely to be affected by an increase in heat-related deaths and illness, as well as food and water-borne diseases. More frequent and severe droughts may exacerbate mental health risks given the well-documented relationship between drought and mental health issues in rural Australia. Incidences of asthma may also rise if projected increases in bushfire risk lead to more frequent bushfires (Hennessy et al. 2007).
Jones and Webb (2008) assessed the vulnerability of nine major Victorian sectors to climate change. They adopted a qualitative, triple bottom line approach to assessing vulnerability, considering the potential economic, social and environmental impacts of climate change on the following sectors: primary production; minerals and resources; manufacturing; energy; building, construction and infrastructure; tourism and services; water; natural resources and biodiversity; and health. The study considers vulnerability to climate change impacts for 2030 and 2070. The authors argue that by 2030 most vulnerability will be encountered through increases in the frequency and extent of extreme events such as drought, fire, flooding and coastal storm surge, while in 2070 vulnerability will arise from the limits of adaptation being exceeded in a range of systems. They found that although the potential economic impacts of climate change may not be high on a state-wide basis, a high degree of economic impact is possible at the regional scale for the water, manufacturing and primary production sectors. The sectors most vulnerable to social impacts include water (high level of vulnerability), primary production (moderate to high), energy (moderate), natural resources and biodiversity (moderate), and health (moderate). Vulnerability to environmental impacts was found to be greatest in the water (high), natural resources and biodiversity (high) and primary production (moderate) sectors. Table 1 summarises key vulnerabilities identified for each sector.
Table 1: Vulnerability to climate change impacts in nine major Victorian sectors
Vulnerable to decreases in rainfall and increases in extreme events. Particularly vulnerable activities include irrigated agriculture (drought), forestry (fire) and dairy (to drought and heat). Adaptive capacity may be constrained by socio-economic pressures such as rising land prices, changing land use, water security and associated costs. Jones and Webb note that adaptive capacity is likely to be greater in intensively managed systems than those that are extensively managed, such as forestry and fishing, which are more likely to be subject to climatic constraints.
Minerals and resources
Low level of vulnerability to fire, flood and storm events on land and at sea. Adaptive capacity is considered high due to the high level of engineering and safety built into systems.
Vulnerable to higher costs or reductions in the supply of raw materials due to drought and fire, water shortages and power loss. Adaptive capacity is considered moderate to high on a state-wide basis; however, some regional manufacturing is highly vulnerable to chronic and/or severe water shortages.
Low level of vulnerability to drought and fire, as well as increased demand due to periods of extreme heat. The sector is considered highly adaptive.
Building, construction and infrastructure
Low level of vulnerability to increases in extreme weather events, sea level rise and associated impacts. Adaptive capacity is considered moderate to high due to the potential for retrofitting and changes to construction and planning.
Tourism and services
Low level of vulnerability overall, with some service industries such as insurance, scientific services and outdoor sport and recreation considered moderately vulnerable. Regional vulnerabilities include snow-based and coastal tourism, rural finance, transport and a range of outdoor and water dependent tourism activities.
Vulnerability is high for economic, social and environmental impacts, with water shortages affecting a range of sectors including primary production and natural resources and biodiversity. Adaptive capacity is considered moderate, influenced by the sector’s capacity for demand management, options for supply, options for multiple use and re-use, appropriate pricing to fund improvements and the willingness of users to seek innovative social and technical solutions.
Natural resources and biodiversity
High level of vulnerability, particularly in marine, coastal, estuarine, wetland and alpine environments. Adaptive capacity is considered moderate due to degradation of some ecosystems.
High level of community vulnerability to heat and cold stress, vector-borne diseases, respiratory problems, and stress and mental illness from prolonged or severe climatic events. Adaptive capacity is considered moderate due to the capacities of ambulance, hospital and community services, although it is noted that the poor and socially isolated may be highly vulnerable.