Federal Court of Australia

Sharma by her litigation representative Sister Marie Brigid Arthur v Minister for the Environment [2021] FCA 560

ORDERS

THE COURT ORDERS THAT:

1.    The applicants’ application for an interlocutory injunction is dismissed.

2.    The claims made by each of the applicants (other than those made on behalf of the represented persons) for a quia timet injunction, are dismissed.

3.    The parties consult and, on or before 3 June 2021, file proposed orders addressing the matters dealt with at paragraph 520 of the Court’s reasons for judgment.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

BROMBERG J:

1    The applicants claim that the first respondent, the Commonwealth Minister for the Environment (Minister) owes them and other Australian children a duty of care. They also claim an injunction to restrain an apprehended breach of that duty. In assessing whether a duty of care exists, the law of negligence focuses upon the foreseeability of harm and the relationship between the person who has caused or contributed to the harm (or will do so) and the persons who have or may be harmed.

2    That is the focus of these reasons. They commence with an introduction to the parties, their respective cases and the conduct which the applicants say is subject to a duty of care – a decision by the Minister made under the Environment Protection and Biodiversity Conservation Act 1999 (Cth) (EPBC Act) to approve the extraction of coal from a coal mine. Details about the application for approval are then given in Section 2 of these reasons. In Section 3, my reasons turn to consider the evidence about the degree of risk and the magnitude of the risk of harm feared by the applicants. The foreseeability and likelihood of that harm arising and being caused or contributed to by carbon dioxide (CO2) emissions in the Earth’s atmosphere generated by the combustion of coal from the coal mine is also considered.

3    Section 4 of these reasons addresses the legal principles applicable to establishing the existence of a duty of care and the statutory scheme in which the Minister is empowered to approve or not approve a “controlled action” such as the expansion of a coal mine. My reasons then divide to consider reasonable foreseeability of harm and those features of the relations between the Minister and Australian children which support a finding that a duty of care exists (Section 5 – The Affirmative Salient Features) and those features that do not (Section 6 – The Negative Salient Features). In Section 7, I conclude that the existence of a duty of care is established and should be recognised by the law of negligence. In Section 8, I deal with and reject the application for an injunction to restrain an asserted apprehended breach of the duty of care by the Minister. The further necessary steps to finalise this litigation are then addressed in Section 9.

1.    The Parties and their Claims

4    The applicants in this proceeding are eight Australian children: Anjali Sharma, Isolde Shanti Raj-Seppings, Ambrose Malachy Hayes, Tomas Webster Arbizu, Bella Paige Burgemeister, Laura Fleck Kirwan, Ava Princi and Luca Gwyther Saunders (the applicants). The applicants are all children residing in Australia. As a consequence of their youth, the proceeding is brought by their litigation representative Sister Marie Brigid Arthur, a Sister of the Brigidine Order of Victoria. The applicants bring the proceeding on their own behalf and as a representative proceeding under Division 9.2 of the Federal Court Rules 2011 (Cth), representing children who ordinarily reside in Australia (the Represented Children) as well as “other Represented Children”, being children residing anywhere in the world. During the course of the hearing the applicants confined their claims for relief to themselves and the Represented Children, that is, the Australian Children. I will refer to the applicants and the Represented Children collectively as the Children.

5    The Minister is an officer of the Commonwealth within the meaning of s 75(v) of the Constitution, and relevantly, the Minister responsible for administering the EPBC Act.

6    The second respondent is Vickery Coal Pty Ltd, a wholly owned subsidiary of Whitehaven Coal Pty Ltd. Whitehaven holds development consent under the Environmental Planning and Assessment Act 1979 (NSW) (EPA Act) for a coal mine in northern New South Wales, known as the Vickery Coal Project (the Approved Project). Although approved some time ago, coal production from the Approved Project is yet to commence. The Approved Project occupies a site within the Gunnedah and Narrabri local government areas, approximately 25 kilometres north of Gunnedah in New South Wales.

7    On or around 11 February 2016, Whitehaven applied to the Minister to expand and extend the Approved Project in accordance with s 68 of the EPBC Act (the Extension Project). Vickery replaced Whitehaven as the proponent of the Extension Project on 17 July 2018. If approved, the Extension Project would, amongst other things, increase total coal extraction from the mine site from 135 to 168 million tonnes (Mt). When combusted, the additional coal extracted from the Extension Project will produce about 100 Mt of CO2.

8    The Minister has before her the decision to approve or refuse the Extension Project under s 130(1) and s 133 of the EPBC Act. This proceeding concerns that decision.

9    In this proceeding the applicants claim that the Minister owes each of the Children a duty to exercise her power under s 130 and s 133 of the EPBC Act with reasonable care so as not to cause them harm. That duty of care is said to arise by reason of the existence of a legal relationship between the Minister and the Children recognised by the law of negligence.

10    The applicants apprehend that the Minister will fail to discharge the duty by exercising her discretion in favour of the approval of the Extension Project. The applicants seek declaratory and injunctive relief designed to preclude the Minister from failing to discharge the duty of care they claim she has.

11    The particular harm relevant to the alleged duty of care is mental or physical injury, including ill-health or death, as well as economic and property loss. The applicants assert that the Children are likely to suffer those injuries in the future as a consequence of their likely exposure to climatic hazards induced by increasing global surface temperatures driven by the further emission of CO2 into the Earth’s atmosphere. The feared climatic hazards include more, longer and more intense bushfires, storm surges, coastal flooding, inland flooding, cyclones and other extreme weather events.

12    The applicants allege that such harm will occur in the future and mainly towards the end of this century when global average surface temperatures are forecast to be significantly higher than they are currently. Broadly speaking, it is at that time that, unlike today’s adults, today’s children will be alive and will be the class of persons most susceptible to the harms in question. Indeed, the applicants say that today’s children will live on Earth during a period in which, if CO2 concentration continues to increase, some harm is very probable, serious harm is likely and cataclysmal harm is possible. This seems to be the basis for the proceeding being directed to providing relief to children, as distinct from all persons. On this basis, the applicants say that the Children are vulnerable to a known, foreseeable risk of serious harm, which the Minister can control, but they cannot. In addition, the applicants say that by her position in the Commonwealth Executive, the Minister has special responsibilities to Australian children.

13    The applicants say that if the Minister approves the Extension Project, carbon presently stored safely underground at the mine site of the Extension Project will be extracted, combusted and emitted as CO2 into the Earth’s atmosphere and will materially contribute to CO2 concentration.

14    The applicants accept that by this proceeding they seek that the Court recognise a novel duty of care. They say that the salient features of the relationship between the Minister and the Children support the recognition of the posited duty. Further, they say that such a duty raises a natural extension of the historical development of the law of tort in making responsible a person with the ability to cause or control harm to their “neighbour”. They say today’s adults have gained both previously unimaginable power to harm tomorrow’s adults, and the ability to control that harm. The applicants seek the aid of the Court to impose a correlative responsibility to protect them from what they say is a serious threat of irreversible future harm.

15    The Minister does not dispute that climate change presents serious threats and challenges to the environment, the Australian community and the world at large. However, the Minister denies the existence of a duty of care as alleged. The Minister denies that injury to the Children from the approval of the Extension Project is reasonably foreseeable and says that the relevant salient features point overwhelmingly against the recognition of the novel duty of care contended for by the applicants. Additionally, the Minister contends that if a duty of care exists, there is no reasonable apprehension that the duty will be breached and for that and other reasons no proper basis to grant injunctive relief. The Minister contends that the proceeding should be dismissed.

16    The applicants also sought an interlocutory injunction to restrain the Minister from exercising her power under s 130 and s 133 of the EPBC Act pending the hearing and determination of the proceeding. It was only in relation to this limited aspect of the applicants’ claim that Vickery sought to be joined as a respondent to the proceeding and participate at the hearing.

17    As a matter of case management, and with the consent of the parties, the hearing of the interlocutory injunction was adjourned to, and heard in conjunction with, the final hearing. This course was facilitated by the Minister providing an undertaking to the Court not to make a decision under s 130 and s 133 of the EPBC Act before the conclusion of the final hearing. The Minister later extended that undertaking to effectively facilitate the publication of these reasons. Ultimately, it has not been necessary for me to determine the application for an interlocutory injunction and for that reason I will dismiss that application.

2.    The Application for Approval to Extend the Coal Mine

18    The relevant background to the Approved Project and the Extension Project was not in dispute and I have drawn the following account from the Statement of Agreed Facts filed by the parties and the from parties’ respective submissions.

19    An initial proposal to develop the coal mine north of Gunnedah was made by Whitehaven in 2014 under the EPA Act. This is the Approved Project, and it was approved as a ‘State Significant Development’ within the meaning of s 89C(1) (now s 4.36(1)) of the EPA Act on 19 September 2014). That initial application did not invoke the operation of the EPBC Act. That is because on 17 May 2012, a delegate of the Minister determined that the proposed action was not a ‘controlled action’ under s 75 of the EPBC Act, if implemented in a particular manner. It therefore did not require the Minister’s approval under the EPBC Act.

20    The Approved Project sought to extract further coal buried deeper in the ground than in past mining activities on the site. It had ambitions of extracting of 135 Mt of coal over a 30-year period, at a rate of up to 4.5 Mt of run-of-mine (ROM) coal per year. In addition, associated developments were proposed which would facilitate the transportation of ROM coal on public roads to Whitehaven’s existing coal handling and preparation plant (CHPP). This facility enables coal to be processed and loaded onto trains for rail transport to the Port of Newcastle.

21    Despite these ambitions, coal production at the mine has not yet commenced.

22    As set out above, on or around 10 February 2016, Whitehaven applied to the Minister to extend the Approved Project in accordance with s 68 of the EPBC Act. The focus of this proceeding is the application for the Extension Project. The proposed actions of the Extension Project include:

(i)    an increase in the total coal extraction from the site of the Approved Project from 135 to 168 Mt;

(ii)    an increase in the peak annual extraction rate from 4.5 to 10 Mt per annum (Mtpa) of coal and an additional disturbance area of 776 hectares; and

(iii)    the development of a new CHPP and train-load-out facility at the site of the Approved Project (both of which would process coal from other nearby mines), which would involve:

(a)    stockpiling and processing a total of 13 Mtpa of ROM coal;

(b)    production of up to 11.5 Mtpa of metallurgical and thermal coal products;

(c)    transportation of up to 11.5 Mtpa of coal from the rail load facility, the rail spur line and via the public rail network to Newcastle for export to other countries;

(d)    development of a new rail spur to connect the load out facility to the main Werris Creek to Mungindi Railway line;

(e)    construction of a water supply borefield and associated infrastructure; and

(f)    changes to the final landform in certain specified ways relating to the overburden emplacement areas and pit lake void.

23    The Extension Project will cause, directly or indirectly, emissions of greenhouse gases, particularly CO2. These estimated emissions are referred to in terms of CO2 equivalent (CO2-e) emissions. Direct greenhouse gas emissions occur from sources that are owned or controlled by the relevant entity or development (referred to as Scope 1 emissions). Indirect greenhouse gas emissions arise from the generation of purchased energy products (principally electricity) by the relevant entity or development (referred to as Scope 2 emissions). Other indirect greenhouse gas emissions arise from sources that are not owned or controlled by the relevant entity or development but are nonetheless a consequence of its mining activity (referred to as Scope 3 emissions).

24    Over the course of its life, the Extension Project will, compared with the Approved Project, lead to the following levels of greenhouse gas emissions:

(i)    an overall reduction of approximately 1 Mt of CO2-e in Scope 1 emissions;

(ii)    an overall increase of approximately 0.15 Mt CO2-e in Scope 2 emissions; and

(iii)    an overall increase of approximately 100 Mt CO2-e in Scope 3 emissions.

25    Those actions will take place over a period of 26 years, with one year projected for construction. In this context, the Minister’s delegate determined that the Extension Project constituted a ‘controlled action’ under s 75(1) of the EPBC Act. The relevant controlling provisions were s 18 and s 18A, and s 24D and s 24E (relating to listed threatened species and communities and water resources respectively). As a consequence of declaring the Extension Project a ‘controlled action’, the Minister is required to assess the application under s 130(1) and s 133 of the EPBC Act. Section 130(2) of the EPBC Act prescribes that the proposed action is assessed either pursuant to a bilateral agreement or pursuant to Pt 8 of the Act. The Extension Project was assessed pursuant to a bilateral agreement between the Commonwealth and the State of NSW (Bilateral Agreement) which accredits the assessment process under the EPA Act.

26    In May 2020, the NSW Department of Planning, Industry and Environment (NSW Department) provided its assessment report (NSW Department Report) in accordance with the Bilateral Agreement. A number of environmental, social and economic factors were considered in the NSW Department Report. It found that the possible adverse environmental impacts associated with the Extension Project were outweighed by the public interest in granting its approval. On balance, the NSW Department Report concluded that the Extension Project was acceptable under certain conditions.

27    Given the status of the Extension Project as a ‘State Significant Development’ under the EPA, the extension application was also assessed by the NSW Independent Planning Commission (IPC) for development consent. The IPC is the designated development consent authority of the Extension Project site under cl 8A of the State Environmental Planning Policy (State and Regional Development) 2011 and s 4.5(a) of the EPA Act. On 12 August 2020, the IPC granted development consent for the extension project, subject to certain conditions (Development Consent) and published its Statement of Reasons for Decision (IPC Report).

28    The Development Consent and the NSW Department Report were provided to the Minister on 14 August 2020. Generally, the receipt of the assessment report provides the Minister with 30 business days, or such longer period as she specifies in writing, to decide whether to approve the application. However, on 9 December 2020, a delegate of the Minister extended this time to 30 April 2021 pursuant to s 130(1A) of the EPBC Act. Further and as previously indicated, in the context of this proceeding the undertaking given by the Minister was further extended.

3.    The Risk of Harm

29    The relief the applicants seek depends upon the Court being satisfied that the approval of the Extension Project by the Minister involves a risk of future injury to each of the Children. The risk of injury alleged by the applicants extends to many forms of what may broadly be described as climatic hazards. Each of these hazards, bushfires being one example, are alleged to be events which climate change will induce in terms of either frequency, ferocity or geographical spread. The risk of harm in question in this case is therefore harm induced by climate change and, more specifically, harm induced by increases in the Earth’s average surface temperature. The applicants alleged that such harm will occur in the future and mainly towards the end of this century when global surface temperatures are forecast to be significantly higher than they are currently.

30    In a nutshell, the applicants’ case is that the scientific evidence demonstrates the plausible possibility that the effects of climate change will bring about a future world in which the Earth’s average surface temperature (currently at about 1.1°C above pre-industrial temperature levels) will reach about 4°C above pre-industrial temperature levels by about 2100. Supported by unchallenged expert evidence, the applicants contended that a 4°C future world may come about in one of two ways: first, where the greenhouse effect upon the Earth’s increasing temperature is driven by an approximately linear relationship between increased human emissions of CO2 and increased temperatures, and second, in circumstances where continuing human emissions of CO2 will result in ‘Earth System’ changes, which diminish the Earth’s current ability to reflect heat, absorb CO2, and retain CO2 currently held in carbon sinks, triggering ‘tipping cascades’ which propel the Earth into a 4°C trajectory. That scenario was referred to in the evidence as “Hothouse Earth”. Under this scenario, humans will lose the capacity to control climate change and global surface temperatures will continue warming even if human emissions of CO2 are curtailed.

31    Further, the unchallenged evidence of the applicants is that the best available outcome that climate change mitigation measures can now achieve is a stabilised global average surface temperature of 2°C above pre-industrial levels. However, at that temperature and beyond, there is an exponentially increasing risk of the Earth being propelled into an irreversible 4°C trajectory because of ‘Earth System’ changes.

32    Given the plausible prospect of Earth’s temperature stabilising at 4°C or greater if stabilisation at 2°C is not achieved, the applicants contended that 100 Mt of CO2 emissions, attributable to the Extension Project, will be significant and material to future increased global average surface temperatures. This, in turn, will expose the Children to a greater risk of injury.

33    To enable an understanding of the different climate scenarios or the “future worlds” in which that risk of harm to the Children is to be assessed, it is necessary to consider the evidence relevant to those elements of the applicants’ case to which I have just referred.

34    Most of the evidence to which I will refer was given by Professor Steffen in his report dated 7 December 2020. Professor Steffen is an eminent specialist with over 30 years’ experience in climate and ‘Earth System’ science research and teaching. Neither his expertise nor the opinions he gave were challenged. A brief account of his experience and expertise is set out in the Schedule to these reasons.

35    The opinions which Professor Steffen gave were sourced in both his own substantial research and that of other specialists in the field. To a large extent, his evidence relied upon the research and climate change modelling published by the Intergovernmental Panel on Climate Change (IPCC). As a factsheet published by the Minister’s Department states, the IPCC is the leading international body for assessing scientific research on climate change and is acknowledged by governments around the world as the most reliable source of advice on climate change. The IPCC was established in 1988 to provide the world with a clear scientific view on the current state of knowledge on climate change and its potential environmental and socio-economic impacts. The IPCC is organised into three working groups and a taskforce that focuses on greenhouse gas emissions. The main role of each working group is to summarise the state of knowledge on climate change in reports published by the IPCC, known as IPCC Assessment Reports. To ensure that those reports are credible, transparent and objective, the reports must pass through a rigorous two-stage scientific and technical review process before being accepted by the IPCC Plenary which is constituted by representatives of member countries of the United Nations and the World Meteorological Organisation.

36    The following account of the evidence is also taken from reports prepared by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) and the Australian Bureau of Meteorology (BoM). Neither the expertise of the relevant authors of the CSIRO or BoM publications, nor the opinions contained therein, were in contest.

3.1    The Effect of Greenhouse Gases upon Earth’s Surface Temperature

37    The greenhouse effect describes the relationship between the atmospheric concentration of greenhouse gases and global average surface temperature. The Earth’s surface absorbs energy from the sun in the form of visible and ultraviolet radiation, and discharges some of this energy back into space in the form of infrared radiation (heat). CO2 is a greenhouse gas. CO2 absorbs a significant proportion of the outgoing radiation and re-radiates some of it back into the lower atmosphere (troposphere) and into the Earth’s surface, thus warming the surface and lower atmosphere.

38    It is well-established that, when burned to produce energy, fossil fuels such as coal produce greenhouse gases, particularly CO2.

39    Emissions of CO2 from industrial sources (currently about 90%) and land-use change (currently about 10%) have raised the atmospheric concentration of CO2 and the global average surface temperature by 1.1℃ compared to pre-industrial levels. From pre-industrial levels to the present, the combustion of coal by humans is estimated to have produced around 1,000 gigatonnes (Gt) of CO2 out of a total of 2,180 Gt emitted by human activity generally. That is, the combustion of coal has contributed about 46% of the total emission of CO2. Professor Steffen estimates that this has contributed about 0.5℃ of the total of 1.1℃ temperature rise from the reference date up to the present date. The commonly used reference date for climate change related parameters as defined by the IPCC is the 1850-1900 average, or, where data is available for individual years, 1876. This is referred to as “pre-industrial”.

40    Increasing emissions of CO2 from the Earth’s surface increase the concentration of CO2 in the atmosphere, which intensifies the greenhouse effect. In other words, the more outgoing infrared radiation (heat) is trapped and re-radiated by CO2, the more the Earth’s surface and lower atmosphere are warmed. Other greenhouse gases such as methane and nitrous oxide also influence global average surface temperature.

41    Professor Steffen opined that there is an approximately linear relationship between human emissions of CO2 from all sources and the increase in global average surface temperature (subject to the non-linear impact of feedbacks, which are discussed below). In the absence of the non-linear effects of feedbacks, further emissions of CO2 from human activities (combustion of fossil fuels and land use) will increase the global average surface temperature at a rate of about 1℃ for every 1,800 Gt of CO2 emitted).

42    The concentration of atmospheric CO2 is currently rising at a rate of about 2.5 ppm (parts per million) per year and this is driving increasing temperatures at the rate of 0.24℃ per five-year period or nearly 0.5℃ per decade. If this rate continues throughout this century, by 2100 the global average surface temperature will reach about 5℃ above the pre-industrial level.

43    At some point in the future, increases in global average surface temperature will likely slow and then stabilise for a multi-decadal period. The rate at which global surface temperature will stabilise depends upon a number of factors. These include, the cumulative CO2 emitted by human activities since the beginning of the Industrial Revolution and also the feedbacks within the ‘Earth System’ that strengthen or weaken the trajectories of CO2 and temperature. I turn then to explain the ‘Earth System’, feedback processes and what Professor Steffen referred to as the “tipping cascade”.

3.2    The Earth System, Carbon Sinks, Feedbacks, the Tipping Cascade and ‘Hothouse Earth’

44    Professor Steffen described the Earth as a single complex system in which the biosphere, and increasingly human activities, play a vital role in the stable functioning of the planet as a whole. He explained that the ‘Earth System’ (a conceptual construct developed to explain the processes on Earth which cycle materials and energy) is defined as “the suite of interlinked physical, chemical, biological and human processes that cycle (transport and transform) materials and energy in complex, dynamic ways within the system” (Earth System).

45    As explained by Professor Steffen, within the Earth System there are numerous natural ‘sub-systems’ which:

(i)    filter most of the damaging ultraviolet radiation from the sun, allowing life to flourish on the surface of the Earth;

(ii)    facilitate the movement of freshwater around the Earth, providing the necessary rainfall for ecosystems to flourish; and

(iii)    absorb CO2 from the atmosphere, which regulates the Earth’s energy balance. Plants perform this role as they photosynthesise.

46    The role of atmospheric CO2 in the Earth System is that it acts as the thermal regulator, a fundamental controller of the surface temperature of the planet. The ‘carbon cycle’ describes the movement of carbon between land, atmosphere and oceans. It is shown in Figure 2 in Professor Steffen’s report, replicated here:

The global carbon cycle showing the movement of carbon between land, atmosphere and oceans in billions of tons (gigatonnes - Gt) of carbon per year. Yellow numbers are natural fluxes, red are human-driven fluxes, and white are stored carbon.

47    As is depicted in Figure 2, there are natural features of the environment, including the oceans and land-based sources (eg the Amazon rainforest), which absorb more CO2 than they produce (referred to as “carbon sinks”). About 55% of human emissions of CO2 are absorbed by land and ocean carbon sinks. The remaining 45% that is left in the atmosphere is the primary driver of the increasing global average surface temperature. Land and ocean carbon sinks “fall far short” of absorbing the increased burden on the system caused by human emissions of CO2. This is depicted by Figure 3 of Professor Steffen’s report and is consistent with the joint report prepared by the CSIRO and BoM, entitled State of the Climate: 2020 (CSIRO and BoM report). Figure 3 shows human emissions of CO2 from 1850 to 2018 and the partitioning of this additional CO2 in the Earth System among the atmosphere, the land (vegetation and soils) and the ocean:

 The human emissions of CO­₂, primarily from the combustion of fossil fuels, are partitioned among the atmosphere and carbon sinks on land and in the ocean. The “imbalance” between the total emissions and total sinks reflects imprecisions in our measurements and understanding, primarily of the land and ocean sinks. Source Friedlingstein et al. (2019) and CSIRO and BoM (2020).

49    It is important to understand that within the carbon cycle there are processes known as ‘feedbacks’ which accelerate, and have the potential to further accelerate, the warming of the Earth’s average surface temperature. Examples include:

    melting ice, including the melting of Arctic sea ice and the loss of ice from the Greenland and Antarctic ice sheets. Melting Arctic sea ice will uncover darker seawater, which absorbs more sunlight and accelerates warming. Melting permafrost also releases CO2 and methane into the atmosphere;

    forest dieback, which concerns degradation through drought, heat and fire affecting large biomes such as the Amazon rainforest and boreal forests in Siberia and Canada. Increasing drought and heat will increase fire frequency, causing bushfires that will emit CO2 presently stored in the Earth’s forest systems; and

    changes in circulation patterns, such as the Atlantic Ocean circulation of the northern hemisphere jet stream. A warming ocean affects global ocean and atmospheric circulation, global and regional sea levels and uptake of anthropogenic CO2 and causes losses in oxygen and impacts on marine ecosystems.

50    According to Professor Steffen and the IPCC, feedback processes accelerate the warming of the Earth System by destroying the Earth’s ability to absorb CO2 or reflect heat. These feedback processes thus compound climate change arising from human emissions of CO2 and other greenhouse gases, producing a non-linear trajectory of increasing temperatures.

51    As the global average surface temperature rises towards 2℃ and beyond, the risk of such feedbacks being activated increases. Because many feedback processes are interconnected, triggering one feedback process may have a rippling effect on others. Professor Steffen referred to this as a tipping cascade. If this tipping cascade is activated, Professor Steffen opined that humans will lose the capacity to control the trajectory of climate change, leading to a much hotter Earth. He refers to this as the Hothouse Earth scenario.

52    Hothouse Earth is one of the future world scenarios that I will shortly explain. Before I do that, there are a few other matters to note which Professor Steffen’s report addressed.

53    Assuming that the stabilisation of CO2 is not affected by non-human factors such as ‘feedback processes’, the stabilisation of CO2 in the Earth’s atmosphere requires that human emissions of CO2 reach net zero. Professor Steffen stated that reaching net zero is a pre-requisite for global average surface temperature to stabilise. However, there will be a lag between global average surface temperature stabilising and the stabilisation of atmospheric CO2 of several decades at least and possibly up to a century. That is because of the time needed for the heat content of the major components of the Earth System – land, ocean, ice and atmosphere – to equilibrate, with a net transfer of heat from the ocean to the atmosphere.

3.3    Effects to Date of Human Emissions of CO2

54    Professor Steffen was asked to describe the effects to date of human emissions of CO2 in Australia and globally. His evidence was as follows:

The human emissions of CO2 (and other greenhouse gases, although CO2 is the most important) have already changed Earth’s climate in very many significant ways. As an overview, the planet’s atmosphere and ocean are heating at an increasing rate, polar ice is melting, extreme weather events are becoming more extreme, sea levels are rising, and ecosystems and species are being lost or degraded.

(a)    The most important impacts of climate change to date on Australia include the following (CSIRO and BoM 2020):

•    Australia’s climate has warmed on average by 1.44 ± 0.24°C since national records began in 1910, leading to an increase in the frequency of extreme heat events. Summer extreme temperatures are increasingly breaching 35°C and even 40°C in most of our capital cities and many regional centres.

•    There has been a decline of around 16 per cent in April to October rainfall in the southwest of Australia since 1970. Across the same region, May–July rainfall has seen the largest decrease, by around 20 per cent since 1970.

•    In the southeast of Australia there has been a decline of around 12 per cent in April to October rainfall since the late 1990s.

•    There has been a decrease in streamflow at the majority of streamflow gauges across southern Australia since 1975.

•    Rainfall and streamflow have increased across parts of northern Australia since the 1970s.

•    There has been an increase in extreme fire weather, and in the length of the fire season, across large parts of the country since the 1950s, especially in southern Australia.

•    There has been a decrease in the number of tropical cyclones observed in the Australian region since 1982.

•    Oceans around Australia are acidifying and have warmed by around 1°C since 1910, contributing to longer and more frequent marine heatwaves.

•    Sea levels are rising around Australia, including more frequent extremes, that are increasing the risk of inundation and damage to coastal infrastructure and communities.

(b)     The effects of climate change are clear and unequivocal around the planet - on every continent and in every ocean basin. The most important impacts of climate change to date globally include the following (IPCC 2013):

•    Warmer and/or fewer cold days and nights over most land areas.

•    Warmer and/or more frequent hot days and nights over most land areas.

•    Increases in the frequency and/or duration of heat waves in many regions.

•    Increase in the frequency, intensity and/or amount of heavy precipitation (more land areas with increases than with decreases).

•    Increases in intensity and/or duration of drought in many regions since 1970.

•    Increases in intense tropical cyclone activity in the North Atlantic since 1970.

•    Increased incidence and/or magnitude of extreme high sea levels.

Global observational evidence published since the IPCC Fifth Assessment Report in 2013 reinforce these trends. For example:

•    Measurements from satellite altimeters show a climate-change driven acceleration of mean global sea level over the past 25 years (Nerem et al. 2018). Averaged globally over the past 27 years, sea level has been rising at 3.2mm/year. But for the past five years, the rate was 4.8mm/year, and for the 5-year period before that the rate was 4.1mm year (Canadell and Jackson 2020, based on data from the European Space Agency and Copernicus Marine Service).

•    Climate change is rapidly increasing the thermal stress for coral reefs as measured at 100 coral reef locations around the world. The level of thermal stress during the 2015-2016 El Niño was unprecedented over the period 1871-2017 (Lough et al. 2018).

•    Intense tropical cyclone activity has increased from 1980 to 2016. Storms of 200 km/hr have doubled in number, and storms of 250 km/hr have tripled in number (Rahmstorf et al. 2018).

3.4    Future Effects – The Future World Scenarios

55    In his evidence, Professor Steffen outlined the approach adopted by climate scientists to project how continued CO2 emissions from human activity might affect the Earth System in the future and what the impacts of any such change (including on the level at which Earth’s surface temperatures stabilise) might be:

(a)    The most common approach involves quantitative projections by reference to Earth System models based on mathematical descriptions of the major features of the Earth System and their interactions.

(b)    The models are driven by projected human emissions of greenhouse gases and land-use change, as well as natural drivers of climate change such as solar radiation.

(c)    The outputs of the models provide insight into the risks presented by different levels of climate change, often characterised by changes in global average surface temperature.

(d)    The analysis is supplemented by evidence from past changes in the Earth System (such as the melting of the ice caps during previous warm periods) which may provide insights as to how the Earth System might change in the future.

56    One such model is the representative concentration pathway (RCP), which accounts for the full suite of greenhouse gases and land use over time. RCPs are framed in terms of “radiative forcing”, which refers to the change in energy levels in the Earth system due to particular drivers of climate change. Radiative forcings which are larger than zero indicate global warming, while radiative forcings which are smaller than zero indicate global cooling.

57    The IPCC has published four RCPs: RCP 2.6, RCP 4.5, RCP 6.0 and RCP 8.5. The numbers refer to the radiative forcing in the year 2100. Each RCP consists of a data set which includes a set of starting values and the estimated emissions up to 2100. Each data set is based on historic information and a set of plausible assumptions about future economic activity, energy sources, population growth and other socio-economic factors. The four RCPs cover a range of emission scenarios with and without climate mitigation policies. For example, RCP 8.5 is based on minimal effort to reduce emissions. RCP 2.6 requires strong mitigation efforts, with early participation from all emitters followed by active removal of atmospheric CO2. RCP 2.6 is described by the IPCC Synthesis Report (2014) as a stringent mitigation scenario. RCP 4.5 and RCP 6.0 are described as “intermediate scenarios” and RCP 8.5 as a scenario with “very high emissions of greenhouse gases”. The IPCC stated that scenarios without additional efforts to constrain greenhouse gas emissions lead to pathways ranging between RCP 6.0 and RCP 8.5.

58    Professor Steffen stated that the lowest RCP (2.6) would result in a global average surface temperature rise of below 2°C by the year 2100, while the highest RCP (8.5) would lead to a temperature rise of 4°C or more by 2100. The continuum of projected increasing global average surface temperature under each scenario from 2046 to 2100 is shown in Table 2.1 of the IPCC Synthesis Report (2014). It should be noted, however, that the reference point used here is not the pre-industrial level. Instead, changes in temperature have been calculated by reference to the 1986-2005 period:

59    In his evidence, Professor Steffen proposed three possible climate futures, which he correlated to the IPCC RCPs as I will later explain. First, it is convenient to give an outline of the main characteristics of each of Professor Steffen’s three scenarios. “Scenario 1” forecasts that global average surface temperature will stabilise in the second half of this century “at, or very close to, 2°C” above the pre-industrial level. The Minister contended that there was some ambiguity in Professor Steffen’s specification of the temperature at which global average surface temperatures would stabilise for “Scenario 1” and contended that he really meant below 2°C and around 1.8°C. For the reasons later given, I do not accept that contention. I will call Professor Steffen’s “Scenario 1” – a “2°C Future World”. It is equivalent to the RCP 4.5 scenario. Each of those scenarios are based on a linear relationship between future emissions of CO2 and increased global average surface temperature.

60    Under Professor Steffen’s “Scenario 2”, it is projected that global average surface temperature will stabilise late this century but more likely early into the 22nd century at, or very close to, 3°C, above the pre-industrial level. I will call this Scenario “3°C Future World”. According to Professor Steffen, that Scenario is approximately equivalent to the upper end of RCP 6.0 envelope of temperature scenarios. The scenario is premised on present national policy settings guiding future emissions trajectories.

61    “Scenario 3” forecasts that global average surface temperature will continue to rise throughout this century with a temperature of about 4°C above the pre-industrial level by late this century, but with the surface temperature likely continuing to rise into the 22nd century. Professor Steffen called this Scenario Hothouse Earth. For convenience and consistency, I will call it a “4°C Future World”. In terms of temperature outcomes at or around the end of this century, this scenario corresponds with the IPCC’s RCP 8.5 which forecasts a 4°C or more temperature rise by the end of this century.

62    However, the two scenarios differ in the paths they each take to reach a similar conclusion about temperature at the end of this century. RCP 8.5 is based on human emissions of CO2 being the dominant driver of temperature rise, whereas Professor Steffen’s scenario is non-linear by reference to the impact of human CO2 emissions and is premised upon feedback processes being activated and adding significant amounts of greenhouse gases to the atmosphere and playing an important role in the ultimate temperature rise.

63    Professor Steffen did not propose a possible scenario of his own which correlated with RCP 2.6. He did however give consideration to that scenario. He stated that RCP 2.6 is consistent with the Paris Agreement signed within the United Nations Framework Convention on Climate Change in 2015 (“Paris Agreement”) target of limiting temperatures to well below 2℃ with the ambition to limit temperature to 1.5°C above the pre-industrial average. Professor Steffen predicts that the target of 1.5℃ is now very likely to be “inaccessible without significant overshoot” (temperatures rising above 1.5℃) followed by a drawdown of CO₂ from the atmosphere by natural means (such as reforestation), industrial means (such as carbon capture and storage) or both.

64    In this context which includes consideration by Professor Steffen of some six years’ worth of data about emissions since the IPCC published its RCPs, Professor Steffen opined that the lowest temperature increase that can realistically be contemplated today is that the global average surface temperature will stabilise at, or very close to, 2°C above pre-industrial levels. This is Professor Steffen’s “Scenario 1” and what I have called a 2°C Future World and reflects RCP 4.5.

65    Professor Steffen’s analysis essentially contemplated that there are only two future worlds now likely to be accessible: either a 2°C Future World or a 4°C Future World. In Professor Steffen’s opinion, RCP 8.5 appears to be increasingly unlikely as renewable energies become cheaper and begin to replace fossil fuels at large scales. However, as indicated already, Professor Steffen opined that essentially the same temperature level (about 4°C by about 2100) envisaged by RCP 8.5 will be reached if the 4°C Future World scenario becomes the reality. Professor Steffen considered a 4°C Future World as plausible given sufficient levels of human emissions of CO₂. However, if certain mitigation measures are taken, Professor Steffen suggested that a 2°C Future World is also plausible. Although Professor Steffen identified a 3°C Future World as a possibility, he opined that there is a “very significant risk” that a 3°C Future World is not accessible because there is a danger that “strongly non-linear feedbacks will be activated by a 3°C warning”. In other words, Professor Steffen forecasts that a tipping cascade will likely be activated by a 3°C temperature rise. He stated that that could occur at “even lower” temperatures, noting that a 2°C temperature rise could trigger a 4°C Future World trajectory but the probability of such a scenario was “much lower” for a 2°C rise than for a 3°C rise. He alternatively expressed this by saying there was “a small (but non-zero) probability of initiating a tipping cascade at a 2°C temperature rise”. Professor Steffen’s assessment is supported by the IPCC’s projection of a “moderate” risk of feedback processes being triggered at a 2°C temperature rise. Professor Steffen opined that this risk will undoubtedly rise with a 3°C temperature increase.

66    A fundamental point made by Professor Steffen’s analysis is that if sufficient measures are not taken to reduce human emissions of CO2 so as to stabilise surface temperature at 2°C, global average surface temperatures will then enter an irreversible 4°C Future World or Hothouse Earth trajectory. Professor Steffen opined that ‘feedback processes’ will be activated by a 3°C (or even lower) temperature rise with a consequent “significant risk” that a tipping cascade will be triggered taking the global average surface temperature beyond 3°C and onto the 4°C Future World trajectory. That is depicted in Figure 4 of Professor Steffen’s report.

3.4.1    Effects of a 2℃ Future World

67    In relation to each of the three scenarios postulated by Professor Steffen, he described the projected global impacts followed by a description of the impacts in Australia. He noted that the risks and impacts described were linked to the stabilisation of the global average surface temperature for each of the three scenarios. He emphasised that stabilisation will take multiple decades at a minimum and stated that, therefore, the risks and impacts described were relevant to the current generation of children and the following generation or two.

Scenario 1: Stabilisation at a rise in global average surface temperature of about 2℃ above the pre-industrial level (IPCC 2018).

•    37% of the global population will be exposed to extreme heat at least once every five years. This will have severe impacts on human health and wellbeing, as well as on worker productivity.

•    Sea-level will rise by 0.46 m by 2100, leading to large increases in coastal flooding, saltwater intrusion in low-lying areas, and more damaging storm surges. The most vulnerable countries include small island states, Bangladesh, low-lying Southeast Asian cities and settlements, and many regions along the African coast.

•    99% of coral reefs will be dead from severe bleaching; this means that the Great Barrier Reef will cease to exist as we know it today, as well as other coral reefs around the world.

•    A decline of 3 million tonnes in marine fisheries, with the most severe impacts on developing countries that rely on marine fish for a large fraction of protein in their diets.

•    Ecosystems will shift to a new biome on 13% of Earth’s land, leading to large rates of extinctions as well as a surge in invasive species as individual organisms migrate in response to a changing climate.

•    6.6 million square kilometres of Arctic permafrost will thaw, releasing large amounts of CO2 and methane to the atmosphere, accelerating the warming trend.

•    7% reduction in maize harvests in the tropics, with the poorest countries suffering the most damaging impacts.

•    16% of plant species will lose at least half of their current range, leading to significant within-ecosystem changes as well as an increase in extinction rates.

For Australia, Scenario 1 would significantly increase the likelihood in any given year of extreme weather events (King et al. 2017): (i) 77% likelihood of severe heatwaves, power blackouts and bushfires; and 74% likelihood of severe droughts, water restrictions and reduced crop yields. More generally, CSIRO and BoM 2020, have used simulations from the latest generation of climate models to project changes to Australia’s climate over the next few decades. These projections would thus be relevant for a 1.5-2℃ world, and thus provide useful insights for Scenario 1:

•    Continued warming, with more extremely hot days and fewer extremely cool days.

•    A decrease in cool season rainfall across many regions of the south and east, likely leading to more time spent in drought.

•    A longer fire season for the south and east and an increase in the number of dangerous fire weather days.

•    More intense short-duration heavy rainfall events throughout the country.

•    Fewer tropical cyclones, but a greater proportion projected to be of high intensity, with ongoing large variations from year to year.

•    Fewer east coast lows particularly during the cooler months of the year. For events that do occur, sea level rise will increase the severity of some coastal impacts.

•    More frequent, extensive, intense and longer-lasting marine heatwaves leading to increased risk of more frequent and severe bleaching events for coral reefs, including the Great Barrier and Ningaloo reefs.

•    Continued warming and acidification of its surrounding oceans.

•    Ongoing sea level rise. Recent research on potential ice loss from the Antarctic ice sheet suggests that the upper end of projected global mean sea level rise could be higher than previously assessed (as high as 0.61 to 1.10 m global average by the end of the century for a high emissions pathway, although these changes vary by location).

•    More frequent extreme sea levels. For most of the Australian coast, extreme sea levels that had a probability of occurring once in a hundred years are projected to become an annual event by the end of this century with lower emissions, and by mid-century for higher emissions.

3.4.2    Effects of a 3℃ Future World

68    The effects forecast by Professor Steffen for a 3℃ Future World were as follows:

Scenario 2: Stabilisation at a rise in global average surface temperature of about 3℃ above the pre-industrial level. Here I focus on projected impacts on Australia of this scenario, based on a recent assessment by the Australian Academy of Sciences (Hoegh-Guldberg et al. 2020, and references therein):

•    Many of Australia’s ecological systems, such as coral reefs and forests, would be unrecognisable, accelerating the decline or Australia’s natural resources through the loss or change in the distribution of thousands of species and ecological processes. (As noted for scenario 1, the Great Barrier Reef will no longer exist at temperature rises of 2℃ or more).

•    Much larger climate change-driven changes to water resources are likely, leading to increasingly contested supplies for natural flows, irrigated agriculture and other uses.

•    At 3℃, living in many Australian cities and towns would be extremely challenging due to more frequent and severe extreme weather events, including much higher temperatures and more severe water shortages.

•    Sea levels will rise by 0.4 to 0.8 metres by 2100 and by many metres over subsequent centuries. These changes will cost hundreds of billions of dollars over coming decades as coastal inundation and storm surge increasingly impact Australia’s coastal communities, infrastructure and businesses. Between 160,000 and 250,000 properties are at risk of flooding when sea levels rise to 1 metre above pre-industrial.

•    The probability of large-scale extreme events, such as large storms, floods, droughts, hail storms, tropical cyclones, heatwaves and other climate-related phenomena will increase rapidly.

•    High fire danger weather will increase significantly, leading to more catastrophic fire seasons such as the 2019/2020 Black Summer fires.

•    Grain, fruit and vegetable crops will suffer more severe reductions in yields in a 3℃ world, and rising heat stress will negatively affect extensive and intensive livestock systems.

•    Rural communities will face increasingly harsh living conditions due to increasing debt from diminishing crop yields, insurance losses from worsening extreme weather events, and more challenging working conditions due to increasing extreme heat.

•    Australia at 3℃ will be hotter, drier and more water stressed with impacts on water security, availability, quality, economies, human health and ecosystems. Many locations in Australia in a 3℃ world would be very difficult to inhabit due to projected water shortages.

•    Multiple impacts of a 3℃ world would damage the health and wellbeing of Australians. These include escalating heat stress, more frequent and intense bushfires, reduced access to food and water, increasing risk of infectious disease, and deteriorating mental health and general wellbeing.

3.4.3    Effects of a 4℃ Future World

69    The projected effects of a 4℃ Future World were described by Professor Steffen as follows:

Scenario 3: The Hothouse Earth scenario, with stabilisation in the 22nd century at a global average surface temperature level at least 4℃, and probably higher, above the pre-industrial level. There has been much less research on the impacts of a 4-5℃ temperature rise in global average surface temperature. However, a few of the potential impacts that could arise from such a high level of warming were summarised in Steffen et al. (2018: Supplementary Information). These include:

•    Multiple impacts on agricultural regions, including depletion of soil fertility, changes in water availability and loss of coastal agricultural lands, with the risk of widespread starvation in the most vulnerable regions and/or large migrations out of those regions, increasing the risk of conflict elsewhere.

•    Destruction of coral reefs from ocean warming and acidification, and consequent loss of livelihoods for those communities and societies dependent on reefs.

•    Amazon rainforest at risk of conversion to savanna from both climate and land-use change. This would lead to large releases of CO2 to the atmosphere as well as large increases in extinction rates of species that depend on the rainforest.

•    Tropical drylands at risk of becoming too hot and dry for agriculture, and too hot for human habitation. This has very large implications for many regions in Africa in particular, but also parts of Asia and much of Australia (see below).

•    Very large risks from coastal flooding to transport, infrastructure and coastal ecosystems. Economic damages could trigger regional or global economic collapse as major coastal cities on all continents become uninhabitable.

•    Reliability of South Asian (Indian) Monsoon vulnerable to high aerosol loading and to the warming of the Indian Ocean and adjacent land. Well over 1 billion people in south Asia depend on a reliable monsoon system. Failure of the monsoon would very likely lead to large-scale starvation, migration and conflict.

•    Mountain glaciers melting at rapid rates, changing amount and timing of run-off. Freshwater resources of over 1 billion people at risk.

•    Large changes to riparian and wetlands, with loss of water of some places and increased flooding in others.

For Australia, the corresponding impacts (harms) of Scenario 3 are:

•    Much of Australia’s inland areas (savanna and semi-arid zones) will become uninhabitable for humans, except for artificial enclosed environments.

•    The southeast and southwest agricultural zones will become largely unviable, due to extreme heat and a reduction in cool season rainfall. This would lead to a large depopulation of regional Australia.

•    Australia’s large coastal cities (Brisbane, Sydney, Melbourne, Adelaide, Perth) will suffer increasing inundation and flooding from storm surges as sea level rises to metres above its pre-industrial level over the coming centuries. This will drive severe economic challenges, both because of direct damage from flooding and the large costs of adaptation.

•    The Great Barrier Reef will no longer exist.

•    Most of the eastern broadleafed (eucalypt forests) will no longer exist due to repeated, severe bushfires.

3.4.4    What Needs to Be Done to Achieve a 2℃ Future World

70    Professor Steffen’s evidence also addressed the probability of a 2℃ Future World and what would need to be done to achieve it and thus (on his analysis) avoid a 4℃ Future World.

71    Professor Steffen opined that there is a 67% probability of achieving a 2℃ Future World if cumulative CO2 emissions from 2021 onwards are restricted to about 855 Gt of CO2 (equivalent to about 20 years of emissions at 2019 rates). That would require net-zero emissions by 2050 by all major emitting countries.

72    Professor Steffen referred to research by McGlade and Ekins (2015) which, using a ‘carbon budget framework’, concluded that there was a 50% probability of the world meeting a 2℃ temperature target if a global CO2 emissions budget of 1,100 Gt of CO2 was achieved for the 2011-2050 period. Professor Steffen noted that this carbon budget was somewhat higher than the budget of 855 Gt of CO2 which he had used in his own analysis (on the basis of a 67% probability). McGlade and Ekins analysed the available global fossil fuel “reserves” and “resources”, defining “resources” as all of the fossil fuels that are known to exist and “reserves” as a subset of “resources”, being those fossil fuels that are currently “economically and technologically viable to exploit”. McGlade and Ekins showed that if all of the world’s existing fossil fuel “reserves” were burnt, about 2,860 Gt of CO2 would be emitted and that about 2,000 Gt of these emissions would come from the combustion of coal. This level of emissions is about 2.5 times greater than the allowable carbon budget for reaching a 2℃ temperature target. On that basis, McGlade and Ekins concluded that globally, 62% of the world’s existing fossil fuel reserves need to be left in the ground, unburnt, and, having performed a regional analysis, it was concluded that over 90% of Australia’s existing coal reserves cannot be burnt to be consistent with a 2℃ temperature target.

73    The definition of “reserves” used by McGlade and Ekins would appear to include the 100 Mt of coal from the Extension Project, it being “economically and technologically viable to exploit now”. On the basis of the carbon budget analysis used by McGlade and Ekins to predict a 50% probability of meeting a 2℃ Future World, Professor Steffen offered this conclusion:

The obvious conclusion from the carbon budget analysis above is that currently operating coal mines must be phased out as soon as possible (preferably no later than 2030), and that no new coal mines, or extensions to existing coal mines, can be allowed.

3.5    Deliberation and Conclusions

74    The following plausible scenarios were demonstrated by that evidence:

(i)    the Paris Agreement target of limiting global average surface temperature to well below 2°C, with the ambition to limit temperature to 1.5°C above the pre-industrial level, is now unlikely to be achieved without significant overshoot;

(ii)    the best future stabilised global average surface temperature which can be realistically contemplated today, is 2°C above the pre-industrial level; and

(iii)    if the global average surface temperature increases beyond 2°C, there is a risk, moving from very small (at about 2°C) to very substantial (at about 3°C), that Earth’s natural systems will propel global surface temperatures into an irreversible 4°C trajectory, resulting in global average surface temperature reaching about 4°C above the pre-industrial level by about 2100.

75    Furthermore, the evidence demonstrates that the risk of harm to the Children from climatic hazards brought about by increased global average surface temperatures, is on a continuum in which both the degree of risk and the magnitude of the potential harm will increase exponentially if the Earth moves beyond a global average surface temperature of 2°C, towards 3°C and then to 4°C above the pre-industrial level.

76    The applicants also seek to establish propositions which are in contest. Those propositions are directed to the extent that 100 Mt of CO2 from the Extension Project will materially contribute to the Children’s risk of being injured by one or more of the hazards induced by climate change.

77    Whether the emission of 100 Mt of CO2 from the Extension Project would increase the risk of harm to the Children is relevant to two aspects of the case. First, it bears on whether a duty of care should be recognised and, in particular, to the question of whether it is reasonably foreseeable that the emission of the 100 Mt of CO2 will increase the risk of the Children being harmed. Second, it is relevant to whether I should grant the injunction the applicants seek. For that purpose, I will need to be satisfied (to the extent later discussed) that it is likely that the emission of the 100 Mt of CO2 will cause the Children harm which, relevantly, is an inquiry as to whether it is likely that the emissions will materially contribute to that harm.

78    As French CJ said in Amaca Pty Ltd v Booth (2011) 246 CLR 36 at [41], “[t]he risk of an occurrence and the cause of an occurrence are quite different things”. Ordinarily, risk is assessed prospectively and causation is assessed retrospectively. However, because, for the purposes of the injunction, I may need to address the prospect of the Minister’s conduct causing harm to the Children, any causation assessment will necessarily be prospective rather than retrospective.

79    The submissions of the parties as to the prospective connection between the Minister’s impugned conduct (the emission of 100 Mt of CO2) and the increased risk of harm to the Children, were largely made by reference to a causation inquiry and not particularly directed to the risk-focused assessment required by the reasonable foreseeability inquiry. Despite that, the following discussion will assist in determining each of the inquiries I may need to make. My conclusions as to foreseeability inquiry and the causation inquiry (in so far as it has been necessary to come to a conclusion) are given later.

80    The applicants contended that the 100 Mt of CO2 from the Extension Project would make a material contribution to future increases in the global surface temperature and thus the degree and magnitude of the risk of harm faced by the Children. That was put in two ways although primary reliance was placed on the second. First, the applicants contended that the approval, extraction, export and combustion of carbon from the Extension Project will emit a material quantity of CO2 into the atmosphere. They contended that the more CO2 that is emitted, the higher the level of CO2 concentration will be before it reaches its zenith. The higher the level of CO2 concentration when it reaches its zenith, the worse the harm to today’s children will be.

81    The Minister responded to that contention by quantifying the increase in global temperature that 100 Mt of CO2 would cause. Assuming a purely linear relationship between increased emissions of CO2 and increased temperature, the calculation was available by reference to Professor Steffen’s evidence that further emissions will increase global average surface temperature at a rate of about 1℃ for every 1,800 Gt of CO2 emitted. The emission of 100 Mt of CO2 would therefore result in an increase of one eighteen-thousandth of a degree Celsius.

82    The Minister contended that an increase of that magnitude was de minimis, which I take to mean negligible (see Bonnington Castings Ltd v Wardlaw [1956] 1 All ER 615 at 618-619 (Lord Reid)). To make good that contention, the Minister contended by way of example that if it were to be assumed that global average surface temperature would otherwise stabilise at 2℃, it would logically follow that, with the addition of 100 Mt of CO2, the temperature would instead stabilise at 2.00005℃. It was then said that there was simply no evidence before the Court about what that magnitude of increase meant in terms of measurable risk. It was suggested that climate change modelling does not operate at a sufficient level of specificity to provide an answer.

83    The wealth of scientific knowledge demonstrated by the evidence before me suggests that science is likely capable of providing that answer. However, I am unable to say that the evidence itself demonstrates the extent, if any, that a fractional increase in average global temperature of the kind in question poses an additional risk of harm to the Children. But that conclusion does not answer the way in which the applicants put their case. They argue that it is the accumulation of CO2 which causes exposure to the risk of harm and accumulated CO2, including the contribution to that accumulation which the 100 Mt of CO2 will make, that will bring about increased temperatures and the harm that the evidence demonstrates will follow. In that way, the applicants say there will be a material contribution to injury.

84    The second way the case was put by the applicants was to adopt what an economist might call a marginal analysis. This contention was made by reference to the contribution that 100 Mt of CO2 may have on the level at which the global average surface temperature will stabilise. In that respect, the applicants first relied on Professor Steffen’s evidence that CO2 emissions from the Extension Project “would increase the level at which atmospheric CO2 concentration is eventually stabilised, and thus would increase the level at which the global average surface temperature is eventually stabilised”. The applicants then relied on the Future World scenarios identified already and the propositions set out at [74] above including that there is a risk, moving in degree from very small to very substantial as the global average surface temperature increases from 2℃ to 3℃ above the pre-industrial level, that a ‘tipping cascade’ will trigger a 4℃ Future World trajectory. The applicants contended that once global average surface temperatures reach or exceed 2℃ above the pre-industrial level, the risk of a 4℃ Future World increases exponentially and that with that heightened realm of risk in prospect, the emission of an additional 100 Mt of CO2 is material. On that basis and given that the evidence demonstrates an increase in both the degree and magnitude of risk of harm to the Children as between a 2℃ Future World and a 4℃ Future World, the applicants contended that the emission of 100 Mt of CO2 in the context of the risk profile just described, is a material contribution to the risk of exposure to harm.

85    The Minister sought to challenge that submission in a number of ways. First, the Minister characterised the applicants’ case as dependent upon demonstrating that the 100 Mt of CO2 from the Extension Project would be emitted outside the available budget of emissions necessary to meet a 2℃ target. The Minister contended that it is likely that the 100 Mt of CO2 would be emitted compliantly with the Paris Agreement and thus within a lower than 2℃ target.

86    Putting aside for the moment what I think is a mischaracterisation of the applicants’ case, there is not sufficient evidence before me on which I could conclude that there is no real prospect of the 100 Mt of CO2 being burnt outside the available fossil fuel budget necessary to meet a 2℃ target. The Minister called no evidence. The Minister essentially contended that the Court should infer that the 100 Mt of CO2 would likely be emitted in accordance with the Paris Agreement. There is no sufficient basis for that inference. The Minister relied upon little else than speculation, in circumstances where the evidence showed that at least one of the potential consumers of the coal is not a signatory to the Paris Agreement.

87    Further and in any event, there is evidence before me which tends to support the proposition that the 100 Mt of CO2 will not be emitted as part of the available carbon budget necessary to achieve a 2℃ target. Professor Steffen’s opinion was that it was “obvious” from the carbon budget analysis, that “no new coal mines, or extensions to existing coal mines, can be allowed”. There can be no doubt that in making that statement Professor Steffen had the Extension Project in mind. True it is that he did not go on to explain why, but to say it is “obvious” by reference to the carbon budget analysis he relied on implies that the reason is to be found in his prior reliance on the study made by McGlade and Ekins, who had analysed the position for Australia and had calculated that over 90% of Australia’s existing coal reserves cannot be burnt to meet a 2℃ target. That observation reveals the logic behind Professor Steffen’s conclusion and it is logic which may be relied upon irrespective of whether the conclusion he proffered was based upon his specialist expertise. If there is no capacity to include 90% of existing Australian reserves of coal in the carbon budget, it seems unlikely that a capacity for new reserves to be included exists. Even “existing” reserves, by which Professor Steffen must have meant those already being exploited, logically have only a 1 in 10 chance of being included in the budget. There is no evidence sufficient to support a contention that the 100 Mt of CO2 from the Extension Project is earmarked for some priority treatment relative to other coal sufficient to put it in the top 10% of candidates for inclusion in the budget.

88    I should say that, whilst the applicants’ contention about risk is stronger on the basis of there being a real prospect of the 100 Mt of CO2 being emitted on or after average surface temperature has reached 2℃, the contention does not depend upon that. The contention depends upon the plausible prospect that surface temperature will reach a point where a ‘tipping cascade’ will be triggered even by a fractional increase in temperature. As that fractional increase will be the product of an accumulation of CO2, it is not essential to the applicants’ contention that the 100 Mt of CO2 is emitted outside of the ‘carbon budget’. What is essential is that the emission does not occur after the ‘tipping cascade’ is triggered. No one contended for that proposition and, on the evidence I do not think it was available.

89    The Minister also suggested that the applicants’ position relied upon demonstrating that a 2℃ Future World was the most likely scenario and that the applicants had overstated Professor Steffen’s evidence on that point because, when properly analysed, Professor Steffen was really saying that a stabilised average global temperature of about 1.8℃ was the most likely scenario. There are some differences in the way that Professor Steffen has described the stabilised average global temperature for his “Scenario 1”. It is variously described as “at, or very close to, 2℃”, “around 2℃”, “approximately 2℃”, “a 2℃ target”, and on one occasion he said “approximately equivalent to, or slightly higher than the upper Paris [A]ccord target of ‘well below 2℃’”. Read in context, the better view is that when Professor Steffen was referring to the stabilised average global temperature for his “Scenario 1” he meant 2℃ or slightly lower but not “well below 2℃” and not the upper target of the Paris Agreement.

90    In any event, the applicants did not say that a 2℃ Future World is the most likely scenario. Their contention was that a 2℃ Future World is a plausible possibility in circumstances where at temperatures at or slightly lower than 2℃, there is a small (but non-zero) probability that a tipping cascade will trigger a 4℃ Future World trajectory. Professor Steffen’s unchallenged evidence establishes that trajectory as a plausible scenario, should the global average surface temperature exceed 2℃ or slightly lower. That was a necessary element of the applicants’ contention and it was established.

4.    Does the Minister owe the Children a DUTY OF CARE?

91    The applicants, who are all less than 18 years of age, contend that the Minister owes a duty of care to them and the class of persons they represent. The class description was originally identified as children born before the date the proceeding was filed who ordinarily reside in Australia or elsewhere, but during the course of the proceeding the applicants limited the relief sought to children residing in Australia. I have therefore proceeded on the basis that the relief claimed, including the scope of the duty of care claimed, is limited to Australian children including the applicants.

92    Although formulated a little differently by the applicants’ Amended Concise Statement (and with my adjustment to take into account that relief is now limited to Australian children), the content of the posited duty as described by the applicants’ submissions is the duty of the Minister to exercise her power under s 130 and s 133 of the EPBC Act with reasonable care to not cause the Children harm resulting from the extraction of coal and emission of CO2 into the Earth’s atmosphere. The type of harm that the applicants assert the duty should cover is mental or physical injury, including ill health or death, as well as damage to property and economic loss.

93    As formulated by the applicants, the duty would extend to any decision under s 130 and s 133 of the EPBC Act involving the extraction of any amount of coal. However, the evidence and submissions made were not directed to any extraction of coal but were focused specifically on the Extension Project and the Minister’s prospective decision to approve or not approve the extraction of 33 Mt of coal and the consequent emission of 100 Mt of CO2, which the applicants assert will make a reasonably foreseeable contribution to climate change and the risk of harm that the applicants fear. The applicants’ case cannot support the establishment of a duty in respect of the Minister’s approval of the extraction of any amount of coal, no matter how small. That is because reasonable foreseeability of harm is an essential pre-condition to the existence of a duty of care. It was not the applicants’ case that it is reasonably foreseeable that the extraction and combustion of any amount of coal would cause the Children injury. The description of the asserted duty was not limited to the Extension Project and was not expressly limited by a reasonable foreseeability requirement. Such a requirement must, however, be implicit in the applicants’ description of the duty of care asserted.

94    I will proceed on the basis that the duty of care asserted is not confined to the approval of the Extension Project but extends to an approval of the extraction of coal which foreseeably exposes the Children to harm. However, I can only conveniently assess whether such a duty exists by reference to the evidence and that evidence and, in particular, the evidence going to the reasonable foreseeability inquiry is specific to the Extension Project. I will therefore confine the findings I will make about the existence of a duty of care to the approval of the Extension Project. If those findings give rise to a duty of care that can be described in terms which extend beyond the Extension Project, I will consider a wider description after further submissions are made by the parties as envisaged by my concluding remarks in Section 9. For present purposes I will refer to the duty of care asserted by the applicants as “the posited duty of care” meaning a duty on the Minister to take reasonable care in the exercise of her statutory powers not to cause the Children harm arising from the extraction of coal from the Extension Project and the consequent emission of CO2 into the Earth’s atmosphere.

95    The existence of a duty of care is a necessary condition of liability in negligence: Brookfield Multiplex Ltd v Owners Corporation Strata Plan 61288 (2014) 254 CLR 185 at [19] (French CJ). The applicants do not identify any authority holding that the posited duty of care exists in directly comparable factual circumstances. They ask the Court to find what is in such circumstances referred to as a “novel” duty of care.

4.1    Ascertaining whether a Novel Duty Exists – the Applicable Legal Principles

96    Whether a novel duty of care exists is to be ascertained by reference to a multi-factorial assessment in which considerations (salient features) relevant to the appropriateness of imputing a legal duty upon the putative tortfeasor are assessed and weighed. I discussed the appropriate approach to the ascertainment of a novel duty of care in Plaintiff S99/2016 v Minister for Immigration and Border Protection (2016) 243 FCR 17 at [201]-[229]. The principles there discussed are not in contest and were relied upon by the parties. For convenience the discussion of those principles is here updated but largely repeated.

97    A salient features approach was adopted by Allsop P (with whom Simpson J agreed) as applicable to determining whether a novel duty of care exists in Caltex Refineries (Qld) Pty Ltd v Stavar (2009) 75 NSWLR 649, at [102]. Relevantly, his Honour said this (emphasis added):

This rejection of any particular formula or methodology or test the application of which will yield an answer to the question whether there exists in any given circumstance a duty of care, and if so, its scope or content, has been accompanied by the identification of an approach to be used to assist in drawing the conclusion whether in novel circumstances the law imputes a duty and, if so, in identifying its scope or content. If the circumstances fall within an accepted category of duty, little or no difficulty arises. If, however, the posited duty is a novel one, the proper approach is to undertake a close analysis of the facts bearing on the relationship between the plaintiff and the putative tortfeasor by references to the “salient features” or factors affecting the appropriateness of imputing a legal duty to take reasonable care to avoid harm or injury.

98    At [103] his Honour set out a list of seventeen salient features. They are these:

(a)    the foreseeability of harm;

(b)    the nature of the harm alleged;

(c)    the degree and nature of control able to be exercised by the defendant to avoid harm;

(d)    the degree of vulnerability of the plaintiff to harm from the defendant’s conduct, including the capacity and reasonable expectation of a plaintiff to take steps to protect itself;

(e)    the degree of reliance by the plaintiff upon the defendant;

(f)    any assumption of responsibility by the defendant;

(g)    the proximity or nearness in a physical, temporal or relational sense of the plaintiff to the defendant;

(h)    the existence or otherwise of a category of relationship between the defendant and the plaintiff or a person closely connected with the plaintiff;

(i)    the nature of the activity undertaken by the defendant;

(j)    the nature or the degree of the hazard or danger liable to be caused by the defendant’s conduct or the activity or substance controlled by the defendant;

(k)    knowledge (either actual or constructive) by the defendant that the conduct will cause harm to the plaintiff;

(l)    any potential indeterminacy of liability;

(m)    the nature and consequences of any action that can be taken to avoid the harm to the plaintiff;

(n)    the extent of imposition on the autonomy or freedom of individuals, including the right to pursue one’s own interests;

(o)    the existence of conflicting duties arising from other principles of law or statute;

(p)    consistency with the terms, scope and purpose of any statute relevant to the existence of a duty; and

(q)    the desirability of, and in some circumstances, need for conformance and coherence in the structure and fabric of the common law.

99    Stavar has been followed in intermediate appellate courts (Makawe Pty Limited v Randwick City Council [2009] NSWCA 412 at [17] (Hodgson JA) and [92]–[94] (Simpsons J); Hoffmann v Boland [2013] NSWCA 158 at [31] (Basten JA) and [127]–[130] (Sackville JA); Ku-ring-gai Council v Chan [2017] NSWCA 226 at [68] (Meagher JA; McColl JA and Sackville AJA agreeing at [1] and [115] respectively); Fuller-Wilson v State of New South Wales [2018] NSWCA 218 at [14] (Basten JA; White JA and Emmett AJA agreeing at [90] and [102] respectively)) and in this Court (Hopkins v AECOM Australia Pty Ltd (No 3) [2014] FCA 1043 at [26] (Nicholas J); Carey v Freehills [2013] FCA 954 at [313]–[317] (Kenny J)). As Kenny J stated in Carey, by reference to Makawe and Hoffmann, the salient factors listed by Allsop P were not exhaustive (at [316]). It is not necessary to make findings in relation to each factor. Rather, as Basten JA said at [31] of Hoffmann, the features provide a “valuable checklist” of the kinds of factors that can be of assistance: “[e]ach involves considerations of varying weight; some will be entirely irrelevant”, and it is necessary to “focus upon the considerations which are relevant in the circumstances of the particular case.”

100    Kenny J’s discussion in Carey traces the rejection in the High Court of the doctrine of proximity as a determinative factor and the adoption of a salient features approach to the determination of whether a novel duty of care is established. Her Honour relevantly said this at [313] (emphasis in original):

Where a duty of care is claimed to have arisen in a new circumstance or with respect to a new category of relationships, Australian law now requires a multi-factorial approach in assessing whether a duty of care has indeed arisen. As the New South Wales Court of Appeal noted in Caltex Refineries (Qld) Pty Ltd v Stavar [2009] NSWCA 258; (2009) 75 NSWLR 649 (‘Caltex v Stavar’) 675 [101], the High Court has rejected the doctrine of proximity as a determinative factor in deciding whether a duty of care existed, as well as “the two stage approach in Anns v Merton London Borough Council [1977] UKHL 4; [1978] AC 728 based on reasonabl[e] foreseeability, the expanded three stage approach in Caparo Industries Plc v Dickman [1990] 2 AC 605 [(‘Caparo v Dickman’)] and any reformulation of the latter two”. See, for example, Hill v van Erp at 210 (McHugh J), 237-239 (Gummow J), Perre v Apand Pty Ltd [1999] HCA 36; (1999) 198 CLR 180 at 193-194 [9]- [10] (Gleeson CJ), 197-198 [25]-[27] (Gaudron J), 208-213 [70]-[83], 216 [93] (McHugh J), 268 [245]-[247], 273 [255], 285 [280]-[287] (Kirby J), 303 [330]-[335] (Hayne J), 319 [389], 324 [398]-[400], 326 [406] (Callinan J); Sullivan v Moody (2001) 207 CLR 562 at 577-580 [43]-[53] (Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ); Graham Barclay Oysters Pty Ltd v Ryan [2002] HCA 54; (2002) 211 CLR 540 at 583 [99] (McHugh J), 625 [234]-[236] (Kirby J); and Stuart v Kirkland-Veenstra [2009] HCA 15; (2009) 237 CLR 215 at 260 [132] (Crennan and Kiefel JJ).

101    To those authorities may be added the support for a multi-factorial approach found in Brookfield at [24] (French CJ) where his Honour said (citing Sullivan v Moody (2001) 207 CLR 562 at [50]) that “different classes of case raise different problems, requiring a ‘judicial evaluation of the factors which tend for or against a conclusion, to be arrived at as a matter of principle’”.

102    It is recognised in the authorities that cases in which the defendant is a repository of a statutory power or discretion are in a special class of case (see, e.g., Sullivan v Moody at [50] (Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ); Hunter and New England Local Health District v McKenna (2014) 253 CLR 270 at [17]-[18] (French CJ, Hayne, Bell, Gageler and Keane JJ)). Liability in special cases is sometimes limited or negated, for reasons of policy (c.f. D’Orta-Ekenaike v Victoria Legal Aid (2005) 223 CLR 1 at [102] (McHugh J)).

103    Intermediate appellate courts have recognised that “[t]here is no authoritative guidance from the High Court for the determination of when a common law duty of care exists with respect to the exercise of statutory power” (Hunter Area Health Service v Presland (2005) 63 NSWLR 22 at [7] (Spigelman CJ); see also Sutherland Shire Council v Becker [2006] NSWCA 344 at [19] (Giles JA), [82] (Bryson JA)). The absence of a guiding principle has also been recognised by Crennan and Kiefel JJ in Stuart v Kirkland-Veenstra (2009) 237 CLR 215 at [131].

104    However, recent judgments of the High Court that have adopted a salient features methodology exemplify the approach to be taken and identify the factors which ordinarily are of the greatest significance when determining whether a novel duty of care is established in respect of the exercise of statutory power. As Spigelman CJ in Presland said at [10], the salient features approach to the exercise of a statutory power is exemplified in the joint judgment of Gummow J and Hayne J in Graham Barclay Oysters Pty Ltd v Ryan (2002) 211 CLR 540 (later endorsed by Gummow, Hayne and Heydon JJ in Stuart at [112]-[113]) where at [146]-[149] their Honours said this (footnotes omitted):

The existence or otherwise of a common law duty of care allegedly owed by a statutory authority turns on a close examination of the terms, scope and purpose of the relevant statutory regime. The question is whether that regime erects or facilitates a relationship between the authority and a class of persons that, in all the circumstances, displays sufficient characteristics answering the criteria for intervention by the tort of negligence.

Where the question posed above is answered in the affirmative, the common law imposes a duty in tort which operates alongside the rights, duties and liabilities created by statute.

…

An evaluation of whether a relationship between a statutory authority and a class of persons imports a common law duty of care is necessarily a multi-faceted inquiry. Each of the salient features of the relationship must be considered. The focus of analysis is the relevant legislation and the positions occupied by the parties on the facts as found at trial. It ordinarily will be necessary to consider the degree and nature of control exercised by the authority over the risk of harm that eventuated; the degree of vulnerability of those who depend on the proper exercise by the authority of its powers; and the consistency or otherwise of the asserted duty of care with the terms, scope and purpose of the relevant statute.

105    Two important observations flow from that passage. First, whether a duty is owed by a statutory authority requires a close examination of the terms, scope and purpose of the relevant statutory regime. That point has been repeatedly emphasised in the authorities: Graham Barclay Oysters at [78] (McHugh J) and at [213] (Kirby J); Stuart at [113] (Gummow, Hayne and Heydon JJ); Sullivan v Moody [55]-[62] (Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ). Second, whilst the ultimate question is whether a requisite relationship exists between the statutory authority and a class of persons, the criteria for assessing whether that relationship exists, and thus whether the tort of negligence will intervene, is to be found in the salient features of that relationship. I would respectfully agree with Spigelman CJ who at [11] of Presland stated that four matters (salient features) of significance came out of the above passage:

    the purpose to be served by the exercise of the power;

    the control over the relevant risk by the repository of the power;

    the vulnerability of the persons put at risk; and

    coherence.

106    It is necessary, however to bear in mind, as Gummow and Hayne JJ observed at [145] of Graham Barclay Oysters that it is “[t]he totality of the relationship between the parties … [which] is the proper basis upon which a duty of care may be recognised”. As the Minister correctly contended, whether a requisite relationship which gives rise to a duty of care is established must be assessed by reference to all the relevant salient features, although the starting point should be the statute and the nature of power conferred upon the respondent.

107    What is further emphasised by the authorities is that in determining whether a novel duty arises it is appropriate and necessary to reason analogically from decided cases (Crimmins v Stevedoring Industry Finance Committee (1999) 200 CLR 1 at [76] (McHugh J); Brookfield at [25] (French CJ)). The search is for principle. Whilst, as the High Court said in Sullivan v Moody at [49], there are “policies at work in the law which can be identified and applied to novel problems”, the law of torts “develops by reference to principles, which must be capable of general application, not discretionary decision-making in individual cases”.

108    That observation from Sullivan v Moody was elaborated upon by Nettle J in a helpful summation of the elements which influence the determination of whether a duty of care is owed. In King v Philcox (2015) 255 CLR 304 at [80], his Honour said this (references omitted):

As Deane J concluded in Jaensch, the question of whether a duty of care is owed in particular circumstances falls to be resolved by a process of legal reasoning, by induction and deduction by reference to the decided cases and, ultimately, by value judgments of matters of policy and degree. Although the concept of “proximity” that Deane J held to be the touchstone of the existence of a duty of care is no longer considered determinative, it nonetheless “gives focus to the inquiry”. It does so by directing attention towards the features of the relationships between the parties and the factual circumstances of the case, and prompting a “judicial evaluation of the factors which tend for or against a conclusion” that it is reasonable (in the sense spoken of by Gleeson CJ in Tame) for a duty of care to arise. That these considerations may be tempered or assisted by policy considerations and value judgments is not, however, an invitation to engage in “discretionary decision-making in individual cases”. Rather, it reflects the reality that, although “[r]easonableness is judged in the light of current community standards”, and the “totality of the relationship[s] between the parties” must be evaluated, it is neither possible nor desirable to state an “ultimate and permanent value” according to which the question of when a duty arises in a particular category of case may be comprehensively answered.

109    In summary:

(1)    The approach to determining whether a duty of care exists is multi-factorial (Stavar at [102]-[103]; Makawe at [17], [92]–[94]; Hoffmann at [31], [127]-[130]; Carey at [313]-[317]; Brookfield at [24]).

(2)    The seventeen factors listed by Allsop P in Stavar are a valuable checklist as to the kinds of matters that may be relevant in a multi-factorial analysis (Hoffmann at [31]; Carey at [316]). But they are not exhaustive, not all considerations will be relevant in each case, and the considerations that are relevant will be of various weights (Carey at [316]; Stavar at [104]).

(3)    The case where the respondent is a repository of statutory power or discretion is a special class of case, which raises its own problems (Sullivan v Moody at [50]; McKenna at [17]-[18]). However, the correct approach remains multi-factorial (Presland at [7], [9]-[10]; Becker at [19] and [82]; Stuart at [131]-[133].

(4)    In such cases, however, certain factors listed in Stavar assume especial relevance. Coherence with the statutory scheme and policy considerations are of critical importance (Stuart at [113]; Presland at [11]; Crimmins at [93]; Graham Barclay Oysters at [146]). So, too, may be control, reliance, vulnerability, and the assumption of responsibility (see, variously, Stuart at [133]; Graham Barclay Oysters at [81], [149], [151]; Presland at [11]; Sutherland Shire Council v Heyman (1985) 157 CLR 424 at 486 (Brennan J) and 498 (Deane J); Pyrenees Shire Council v Day (1998) 192 CLR 330 at [115] (McHugh J) and [168] (Gummow J); Crimmins at [93], [104], [108] (McHugh J)).

110    Some further, more general observations about the law of negligence should also be kept in mind. The broad principle which underlies liability in negligence is stated in the famous speech of Lord Atkin in Donoghue v Stevenson [1932] AC 562 at 580 and, in particular, its reference to the neighbourhood principle (emphasis added):

At present I content myself with pointing out that in English law there must be, and is, some general conception of relations giving rise to a duty of care, of which the particular cases found in the books are but instances. The liability for negligence, whether you style it such or treat it as in other systems as a species of “culpa,” is no doubt based upon a general public sentiment of moral wrongdoing for which the offender must pay. But acts or omissions which any moral code would censure cannot in a practical world be treated so as to give a right to every person injured by them to demand relief. In this way rules of law arise which limit the range of complainants and the extent of their remedy. The rule that you are to love your neighbour becomes in law, you must not injure your neighbour; and the lawyer's question, Who is my neighbour? receives a restricted reply. You must take reasonable care to avoid acts or omissions which you can reasonably foresee would be likely to injure your neighbour. Who, then, in law is my neighbour? The answer seems to be—persons who are so closely and directly affected by my act that I ought reasonably to have them in contemplation as being so affected when I am directing my mind to the acts or omissions which are called in question.

111    Referring to that passage in Perre v Apand Pty Ltd (1999) 198 CLR 180 at [100], McHugh J said this (emphasis in original, references omitted):

In determining whether the defendant owed a duty of care to the plaintiff, the ultimate issue is always whether the defendant in pursuing a course of conduct that caused injury to the plaintiff, or failing to pursue a course of conduct which would have prevented injury to the plaintiff, should have had the interest or interests of the plaintiff in contemplation before he or she pursued or failed to pursue that course of conduct. That issue applies whether the damage suffered is injury to person or tangible property or pure economic loss. If the defendant should have had those interests in mind, the law will impose a duty of care. If not, the law will not impose a duty.

112    The enduring importance of the neighbourhood principle espoused by Lord Atkin to the analysis of whether a duty of care exists may be observed in the conclusion expressed by French CJ, Gummow, Hayne, Crennan and Bell JJ in Sydney Water Corporation v Turano (2009) 239 CLR 51 at [53]; see also Nettle J in King at [79]; and see Balkin R and Davis JLR, Law of Torts (5th edition, LexisNexis Butterworths, 2013) pp 202-203.

113    A second observation made in Donoghue v Stevenson, but on this occasion by Lord Macmillan at 619, is also of importance, particularly in the context of a court being asked to recognise a novel duty (emphasis added):

In the daily contacts of social and business life human beings are thrown into, or place themselves in, an infinite variety of relations with their fellows; and the law can refer only to the standards of the reasonable man in order to determine whether any particular relation gives rise to a duty to take care as between those who stand in that relation to each other. The grounds of action may be as various and manifold as human errancy; and the conception of legal responsibility may develop in adaptation to altering social conditions and standards. The criterion of judgment must adjust and adapt itself to the changing circumstances of life.

114    Referring to that passage, Gleeson CJ in Tame v New South Wales (2002) 211 CLR 317 at [14] stated that the issue of reasonableness is “at the heart of the law of negligence” and that “[r]easonableness is judged in the light of current community standards”: see further Nettle J in King at [92] and at [97] where his Honour assessed what is reasonable by reference to “contemporary social conditions and community standards”. Those observations are echoed by McHugh J who at [97] of Tame stated that in the law of negligence a risk “was regarded as unreasonable and one to be prevented only if reasonable members of the community would think it sufficiently great to require preventative action”. Those observations were also referred to by Basten JA in Stavar at [160], his Honour observing at [163] that the existence of a duty depends on matters of both fact and evaluative judgment.

115    As is apparent from the observation of McHugh J at [100] in Perre (set out at [111] above), the perspective from which the existence of the duty of care is to be assessed is prospective. That is so because the “ultimate issue” is whether the alleged tortfeasor should have had the interests of the claimant in contemplation before it pursued or failed to pursue a course of conduct: see further Stavar at [177] (Basten JA).

4.2    The Law’s Adaptation to Altering Social Conditions – The Early Environmental Cases

116    The recognition of a novel duty of care represents a development in the law. As the common law develops in this manner, new legal rights are conferred on some persons and corresponding legal obligations are imposed on others. Lord Macmillan’s famous statement in Donoghue v Stevenson, extracted above at [113], recognises that the common law will respond to human errancy by imposing legal responsibility and, driven by the standards of the reasonable person, sensitive as they must be to the changing circumstances of human existence, the “conception of legal responsibility may develop in adaptation to altering social conditions and standards” (Donoghue v Stevenson at 619 (Lord Macmillan)).

117    In this case, the law is being asked to respond to altering social conditions brought about by human interference to the natural environment. The deterioration of social conditions brought about by the degradation of the habitat or the environment in which people live and on which they rely has been a constant impetus for the development of the common law. It is instructive to briefly consider the history of the development of the torts of nuisance and negligence with a focus upon the early cases which had to grapple with how the law should address the ever-increasing capacity of human beings to alter the environment to the detriment of others with whom that environment is shared. Relevantly, some of the early cases involved environmental damage done in the performance of a statutory authority or power. The following brief review of the early cases largely adopts that which was helpfully provided in the applicants’ written submission.

118    As early as the 12th century, the assize of nuisance lay for loss of profit through the defendant’s interference with incorporeal rights, such as the plaintiff’s rights of way, watercourse or pasture on the plaintiff’s land but exercised over other land (see Kiralfy AKR, Potter’s Historical Introduction to English Law and its Institutions (4th ed, Sweet & Maxwell, 1958) p 420).

119    By the 13th century, the assize of nuisance was also used for interference with the enjoyment of the plaintiff’s land by making that land unusable or uninhabitable, such as from fumes, fires, the diversion of a watercourse or raising a mill pond so that it floods the plaintiff’s land (Kiralfy (1958) p 420). For example, in Dalby v Berch (1330) Y.B. Trin. 4 Edw. III, fo. 36, pl. 26, the claimant was awarded damages for pollution that had rendered his house uninhabitable as a result of noxious gases from the defendant’s lime-kiln.

120    In Hulle v Orynge (1466) Y.B. Mich. 6 Edw. IV, fo. 7, pl. 18 (the Case of the Thorns), a majority of the King’s Bench held that if a person damages another’s property, there is a tort even if the action that caused such damages was itself lawful. In writing a concurring opinion, Pigot (a lawyer) is reported to have held that “if a man has a fish-pond in his manor and he empties the water out of the pond to take the fishes and the water floods my land, I shall have a good action, and yet the act was lawful”. Similarly, Brian (a lawyer) is reported to have held that “[w]hen any man does an act, he is bound to do it in such a manner that by his act no prejudice or damage is done to others”: see Fifoot CHS, History and Sources of the Common Law: Tort and Contract (Stevens and Sons Limited, 1949) pp 195-197.

121    During the 16th and 17th centuries, the action on the case supplanted the assize (Kiralfy (1958) p 423). Actions on the case made relief available for harm suffered by “offensive trades” (Kiralfy (1958) p 424). In William Aldred’s Case (1610) 77 ER 816, the plaintiff claimed the defendant had erected and used a pigsty too close to his house such that the stink (or the “stopping of the wholesome air”), among other things, made his own house unbearable to live in (at 821). In Boynton v Gill (1640) Rolle’s Abr. Nusans, fo. 90, pl. 7, the court held that where a trade was an annoyance it must be carried out in ‘waste places’ where no one would suffer damage: (Kiralfy (1958) p 425).

122    Until the recognition in modern times of negligence as a tort in itself, many actions on the case that today would be described as negligence were historically described as nuisance (Hargrave v Goldman (1963) 110 CLR 40 at 61-62 (Windeyer J)). The modern tort of negligence can be traced back to Mitchil v Alestree (1676) 1 Vent 295. In that case, a man who brought an unruly horse into Lincoln’s Inn Fields (to break the horse in) was held liable for an injury to a passer-by. The claim was based not on particular knowledge of the animal’s unruliness, but on the broad ground that the whole operation was clearly likely to lead to someone being hurt (Kiralfy (1958) p 387).

123    In the 12th to 17th centuries, the power of humans to cause injury was generally limited by physical proximity, except where nature provided an intermediate causal agent, such as water, fire, air or wild animals. By 1768, at about the time the Industrial Revolution was commencing, the author of Buller’s An Institute of the Law Relative to Trials at Nisi Prius wrote that “[e]very man ought to take reasonable Care that he does not injure his Neighbour; therefore, where-ever a Man receives any Hurt through the Default of another, though the same were not wilful, yet if it be occasioned by Negligence or Folly, the Law gives him an Action to recover Damages for the Injury so sustained”: see Cornish WR, Banks S, Mitchell C, Mitchell P, Probert R Law and Society in England 1750-1950 (2nd ed, Hart Publishing, 2019) p 461.

124    During the 19th century, the various actions on the case were developed into the separate torts of nuisance and negligence.

125    Private nuisance cases increased in response to the impact of the Industrial Revolution on the local environment. In 1808, a visitor of Manchester said “the steam engine is pestiferous, the Dyehouses noisome and offensive, and the water of the river as black as ink or the Stygian Lake” (McLaren JPS, “Nuisance Law and the Industrial Revolution – Some Lessons from Social History” (1983) 3(2) Oxford Journal of Legal Studies p 164-165). In 1835, when a French statesman known as de Tocqueville visited Manchester, he observed, “[a] sort of black smoke covers the city. The sun seen through it is a disc without rays” (McLaren (1983) p 165).

126    In Attorney-General v Council of the Borough of Birmingham (1858) 70 ER 220, the Court granted an interim injunction against public sewage works ordered by an Act of Parliament pursuant to which Birmingham Council was dumping sewage into the River Tame which ran through the plaintiff’s property. An undertaking was also given “to prevent the pollution of the river Tame, so as to render it injurious to the inhabitants of the houses adjoining its course, and also to prevent its being so polluted as to become offensive and unfit for use” (at 228).

127    In Hole v Barlow (1858) 4 CBNS 334 it was held that, as long as the defendant’s use of the land was itself proper and convenient, then such use could not be a nuisance even if it interfered with the plaintiff’s use and enjoyment of their land. That view was not followed in Bamford v Turnley (1862) 122 ER 27. In that case, the defendant’s burning of bricks in a kiln emitted noxious fumes to the surrounding area, making his neighbours and their servants ill. In delivering separate reasons from the majority, Bramwell B also rejected the argument that the defendant’s actions were lawful because they were for the public benefit (at 33-34).

128    In St Helens Smelting Co v Tipping (1865) 11 ER 1483, the House of Lords dismissed an appeal from the verdict of a jury awarding damages to the plaintiff on the basis that certain noxious vapours emitted by the defendant’s smelting plant damaged the plaintiff’s trees, hedges and plants. The copper smelting plant had drastic effects on the environment. In delivering judgment, Lord Westbury LC (with whom Lord Cranworth and Lord Wensleydale agreed) affirmed the jury’s verdict and rejected the appellant’s argument that it was entitled to carry on copper smelting with “impunity” (as the whole neighbourhood was a manufacturing neighbourhood), despite the fact this may have resulted in the “utter destruction, or the very considerable diminution” of the value of the plaintiff’s property (at 1487).

129    The courts also recognised potential issues of causation at a time when pollution was widespread. The English Court of Appeal held in Crossley and Sons Ltd v Lightowler [1867] LR 2 Ch App 478 that the fact that a stream was fouled by others was not a defence to a suit to restrain the fouling by the defendant. In this respect, Lord Chelmsford LC stated (at 483): “[t]he Defendants cannot justify their interference with the Plaintiffs’ right to have the water of [the stream] in the state in which it would be without their additional pollutions”.

130    Concern about the risks inherent in industrial activities led to the imposition of strict liability for hazardous activities in some cases. In Fletcher v Rylands (1865-1866) LR 1 Ex 265 at 280 (upheld on appeal in Rylands v Fletcher (1868) LR 3 HL 330), Blackburn J gave as a particular instance of a person who should have an action for damages, “[t]he person … whose habitation is made unhealthy by the fumes and noisome vapours of his neighbour’s alkali works”.

131    The 19th century also saw the development of negligence as an independent tort. Sir Percy Winfield attributed its rise to “industrial machinery. Early railway trains, in particular, were notable neither for speed nor for safety. They killed any object from a Minister of State to a wandering cow, and this naturally reacted on the law” (Winfeld PH, “The History of Negligence in the Law of Torts” (1926) 42(2) The Law Quarterly Review p 195). The power to cause harm was enhanced by the Industrial Revolution, so that it no longer relied on natural intermediaries such as fire, water or wild animals, but extended to industrial poisons and pollutants (albeit still often borne by air or water) and machines.

132    By 1856, the courts had provided a general definition of negligence: “the omission to do something which a reasonable man, guided by those considerations which ordinarily regulate the conduct of human affairs, would do, or doing something which a prudent and reasonable man would not do” (Blyth v Birmingham Waterworks (1856) 156 ER 1047 at 1049 (Alderson B)).

133    By 1883, the development of a requirement of a ‘duty of care’ in negligence was well underway. In Heaven v Pender (1883) QBD 503 at 509, Brett MR stated the circumstances in which a relationship was created such that “a duty arises to use ordinary care and skill to avoid such danger”. However, it should be noted that although Cotton and Bowen LJJ agreed in the result in that case, they did not agree with the broader principle proposed by Brett MR (at 516-517).

134    In the 19th century, the growth of industrialised activity by persons or bodies acting under statutory powers saw actions against them in which it was consistently held that the conferral of authority or discretionary power under statute was no defence to a private law action in tort. In Weld v The Gas-Light Company (1816) 171 ER 442, Lord Ellenborough stated “I am clearly of opinion, that where any Company, such as the Gas-Light Company, is entrusted with the execution of a power from which mischief may result to the community, they are bound to execute it as innocently as they can …” (at 442).

135    In Geddis v Proprietors of the Bann Reservoir [1878] 3 App Cas 430, the House of Lords held the reservoir proprietors liable to neighbouring landowners when releasing water onto the land which destroyed the landowner’s crops. Lord Blackburn held (at 455- 456):

[I]t is now thoroughly well established that no action will lie for doing that which the legislature has authorized, if it be done without negligence, although it does occasion damage to anyone; but an action does lie for doing that which the legislature has authorized, if it be done negligently. And I think that if by a reasonable exercise of the powers, either given by statute to the promoters, or which they have at common law, the damage could be prevented it is, within this rule, “negligence” not to make such reasonable exercise of their powers.

136    Courts of equity would grant an injunction to restrain tortious conduct, although done in performance of a statutory authority or discretionary power, provided the conduct was not required by Parliament. In Attorney-General v Colney Hatch Lunatic Asylum (1868) LR 4 Ch App 146, the Court of Appeal in Chancery decreed that an injunction should be granted (but suspended for three months) to restrain visiting magistrates of an asylum, who were exercising a statutory power, from allowing the sewage from that asylum into a stream. Lord Hatherley LC rejected an argument that Parliament was responsible because it had conferred the power to build the asylum (at 159). To similar effect, Selwyn LJ observed at 165 that “an Act of Parliament merely authorizing the erection of such an asylum cannot justify an interference with the rights of neighbours to the extent contended for”.

137    That historical review of the cases is of some assistance. However, there are limitations which should be recognised. Although of contextual assistance, many of the cases concern the law of nuisance and not the law of negligence with which I am here concerned. Secondly, later developments in the law must be brought to account. Nevertheless, the cases reviewed demonstrate the willingness of the common law to respond to changing social conditions including those brought about by the increasing power of human beings to cause harm to others. That is the context in which the applicants contend that because today’s adults have gained previously unimaginable power to harm tomorrow’s adults, the common law should now impose correlative responsibility.

4.3    The Methodology of Development of the Common Law

138    The applicants do not shy from the proposition that their case calls for a development in the law. They accept that legal principles which control the capacity of a court to develop the law are applicable. They contend, however, that the common law is not a set of static rules. By reference to Windeyer J in Skelton v Collins (1966) 115 CLR 94 at 135, the applicants assert (and I accept) that the common law “is a body of principles capable of application to new situations, and in some degree of change by development”. The applicants’ submissions recognised that in the development of the law the doctrine of precedent should provide necessary stability (referring to Lord Goff’s observations in Kleinwort Benson Ltd v Lincoln City Council (1999) 2 AC 349 at 378), but emphasised that precedent will not always trump the need for desirable change in the law and that in developing the common law judges must “necessarily look to the present and to the future as well as to the past” (referring to observations made in Perre at [92] by McHugh J as adopted in Brodie v Singleton Shire Council (2001) 206 CLR 512 by Gaudron, McHugh and Gummow JJ at [108]).

139    Both parties accept that in the development of the law of negligence by the recognition of a novel duty of care the method of development is analogical. It is, as mentioned above, appropriate and necessary to reason analogically from decided cases.

140    The applicants submitted that their case proceeds by analogical reference to two categories of negligence. First, the applicants referred to the Speirs line of authority (Caledonian Collieries Ltd v Speirs (1957) 97 CLR 202), which holds that statutory powers must be exercised with reasonable care and that the common law may impose liability for harm caused by their negligent exercise. Second, the applicants referred to what they called the Rylands v Fletcher line of authority, which holds a person liable for harm caused by dangerous things which escape from that person’s land.

141    The Minister does not challenge the foundational proposition said by the applicants to flow from each of the lines of authority upon which the applicants relied. The Minister contended, however, that there is no relevant analogy between this case and the two lines of authority relied upon by the applicant. The Minister drew upon various distinctions between this case and the lines of cases relied upon by the applicants. Conversely, the applicants emphasised the similarities and sought to diminish the significance of any differences.

142    Neither approach is to be criticised. Each involved analogical reasoning. It is not necessary to reason from any particular line of authorities, although the Speirs line is a helpful source of guidance. Having regard to the salient features relevant to this case, it is informative to reason analogically from those cases which are most closely analogous to this case. That is the essence of the approach taken by the parties and the approach that I will adopt.

4.4    The Salient Features to Be Considered

143    The exercise in which I am engaged is that of undertaking a close analysis of the facts which bear upon the relationship between the Minister and the Children. That analysis is to be conducted by reference to those salient features which indicate whether a legal duty should be imposed upon the Minister to take reasonable care to avoid harm or injury to the Children: Stavar at [102] (Allsop P). That expression of the Court’s task as well as the description of the duty as a ‘duty of care’ are themselves revealing of the nature of the relationship required to impute to one person a duty of care in respect of another. The law often imposes legal obligations upon persons charged with the responsibility to care, look out for or, at the least, do no harm to the interests of others. What is it in the facts of this case that tends to affirm a conclusion that the Minister bears a responsibility to look out for and take care to avoid her conduct inflicting harm upon the Children? Why, to adopt the question flowing from the neighbourhood principle and which McHugh J in Perre at [100] posed as the “ultimate issue”, should the Minister have the interests of the Children in contemplation when she exercises her power to approve the extraction of coal and its consequential emission of carbon into the atmosphere? If the Minister should have the interests of the Children in mind “the law will impose a duty of care” and “[i]f not, the law will not impose a duty of care” (Perre, at [100] (McHugh J)). The simplicity expressed by that query somewhat belies the underlying complexity required for an answer by a close analysis of the facts against the salient features. But in the search for that answer the subject of the question (should A have had the interests of B in contemplation) should not be lost.

144    In order to focus the later discussion on those salient features which require detailed consideration, it is convenient that I now identify the salient features relied upon by each of the parties in support of the argument each made.

145    The applicants emphasised the degree and nature of control able to be exercised by the Minister to avoid harm (‘control’), the vulnerability of the Children (‘vulnerability’), the reasonable foreseeability and nature of the harm (‘reasonable foreseeability’) as well as a recognised category of relationship between the Minister and the Children (‘recognised relationship’) as of especial importance. They contended that each of those salient features supported the recognition of the posited duty of care. There are a number of salient features which need to be considered. I ultimately conclude that each of those salient features tend to support a duty of care being recognised. Those salient features are affirmative of a duty of care being recognised and for that reason I shall refer to them by that designation. They are dealt with in the following section.

146    The Minister contended that the posited duty was extraordinary, submitting that there is no precedent for a duty analogous to the duty contended for by the applicants. Of the salient features addressed, the Minister argued that incoherence and inconsistency with the EPBC Act and public law principles (‘coherence’) was determinative. The Minister contended that ‘reasonable foreseeability’, ‘control’, the salient features of ‘proximity’, ‘reliance and responsibility’ as well as ‘indeterminacy’ all supported the rejection of the duty for which the applicants contend. Insofar as I have concluded that any of those salient features tend towards the rejection of the posited duty I have addressed those in the section headed “The Negative Salient Features”.

147    There is one salient feature I should mention now. I regard ‘coherence’ as having especial importance to the outcome of this proceeding. My discussions of that salient feature appears much later in these reasons. The statutory scheme of the EPBC Act is of critical relevance to ‘coherence’ as a salient feature. However, the statutory scheme has broader relevance and an appreciation of it is necessary for my discussion of other of the salient features including ‘control’. For that reason, an outline of the statutory scheme will follow.

148    There is one matter that I determine in my discussion of ‘coherence’ that has a consequent impact upon the scope of my discussion about each of the salient features. I have concluded that ‘coherence’ is determinatively against a duty of care being recognised which would require the Minister to take reasonable care to avoid harm to the Children beyond harm by way of personal injury. In other words, ‘coherence’ precludes the recognition of a duty of care extending to property damage or pure economic loss to which the Children may be exposed. My discussion of each of the salient features is premised on that conclusion.

4.5    The Statutory Scheme

149    The power which is the subject of the posited duty is a statutory power and for that reason the relevant statutory context provided by the EPBC Act looms large in my consideration of ‘coherence’ as a salient feature. The statutory power here in question is the power of the Minister to determine whether or not what the EPBC Act refers to as a “controlled action” should or should not be approved. That power is given by s 130 and s 133 of the EPBC Act. The statutory circumstances in which an action becomes a controlled action and the statutory context in which the Minister’s statutory discretion to approve or not approve such an action need to be outlined. I will do that commencing with the objects of the EPBC Act.

150    The objects of the EPBC Act include providing for the protection of the environment, especially those aspects of the environment that are matters of “national environmental significance”: s 3(1)(a). Section 3(1)(b) states that a further object is the promotion of “ecologically sustainable development” through the conservation and “ecologically sustainable use” of natural resources. Each of those terms used in s 3(1)(b) is defined. Section 528 provides the meaning of “ecologically sustained use” as the “use of the natural resources within their capacity to sustain natural processes while maintaining the life-support systems of nature and ensuring that the benefit of the use to the present generation does not diminish the potential to meet the needs and aspirations of future generations”. The principles of “ecologically sustainable development” are given meaning by s 3A which provides:

Principles of ecologically sustainable development

The following principles are principles of ecologically sustainable development:

(a)    decision-making processes should effectively integrate both long-term and short-term economic, environmental, social and equitable considerations;

(b)    if there are threats of serious or irreversible environmental damage, lack of full scientific certainty should not be used as a reason for postponing measures to prevent environmental degradation;

(c)    the principle of inter-generational equity—that the present generation should ensure that the health, diversity and productivity of the environment is maintained or enhanced for the benefit of future generations;

(d)    the conservation of biological diversity and ecological integrity should be a fundamental consideration in decision-making;

 (e)    improved valuation, pricing and incentive mechanisms should be promoted.

151    Other objects listed in s 3(1) include to promote the conservation of biodiversity (para (c)); to provide for the protection and conservation of heritage (para (ca)); and to assist in the co-operative implementation of Australia’s international environmental responsibilities (para (e)).

152    With those objectives, the EPBC Act establishes a protective scheme for those aspects of the environment specified by Pt 3 which include but are not limited to matters of “national environmental significance”. That is done, in substance, by providing the Minister with the capacity to determine whether the taking of action which has or is likely to have a “significant impact” on a protected aspect of the environment should be permitted to proceed: s 11. Broadly speaking, that is achieved by prohibiting a person from taking an action with such an impact unless the Minister has given approval for the action under Pt 9 of the EPBC Act or decided that approval is not required. A regime for determining whether or not approval is required is established by Pt 7. That Part requires that certain proposed actions be referred to the Minister for the Minister to decide whether any of the provisions in Pt 3 would prohibit the taking of the action.

153    If a person proposes to take an action that the person thinks may be or is a “controlled action”, the person must refer the proposal to the Minister for a decision as to whether or not the action is a controlled action: s 68(1). An action is a “controlled action” if the taking of the action by the person without approval under Pt 9 would be prohibited by a provision of Pt 3 of the EPBC Act: s 67. The provision of Pt 3 that would prohibit the action (if not approved) is the “controlling provision”: s 67. A person must not take a controlled action unless an approval is in operation under Pt 9 for the purposes of the relevant provision of Pt 3: s 67A. A person who has engaged, engages or proposes to engage in conduct consisting of an act that constitutes an offence or other contravention of the EPBC Act or the regulations may be restrained by injunction on the application of the Minister or an “interested person”: s 475. An “interested person” includes an Australian citizen whose interests have been, are or would be affected by the conduct or who has engaged in a series or activities for protection or conservation of, or research into, the environment at any time in the 2 years immediately before the conduct: s 475(6).

154    The aspects of the environment which are sought to be protected from an action with a “significant impact” upon them specified in Div 1 of Pt 3 are referred to as “matters of national significance” (see ss 74(2), 77(1)(a)(iii), 78B(5), 132(d)(ii)). Those matters are World Heritage properties (Subdiv A); National Heritage places (Subdiv AA); wetlands of international importance (Subdiv B); listed threatened species and communities (Subdiv C); listed migratory species (Subdiv D); a Commonwealth Marine Area (Subdiv F); and the Great Barrier Reef Marine Park (Subdiv FA). Beyond those “matters of national significance”, other aspects of the environment are also specified under Div 1 of Pt 3. Protection is there afforded to the environment generally from any nuclear action taken by a constitutional corporation, the Commonwealth or an agency of the Commonwealth (Subdiv E) and also to a water resource affected by a coal seam gas development or a large coal mining development of a constitutional corporation, the Commonwealth or an agency of the Commonwealth (Subdiv FB). Additionally, matters of national environmental significance prescribed by regulations made under the EPBC Act are also protected (Subdiv G). Division 2 of Pt 3 sets out further matters which require approval. They are acts of the Commonwealth or a Commonwealth agency taken within or outside Australia which have or are likely to have a “significant impact” on the environment and acts in respect of Commonwealth land and Commonwealth heritage places located overseas.

155    The phrase “significant impact” is not defined but s 527E relevantly provides the following meaning for the word “impact”:

(1)    For the purposes of this Act, an event or circumstance is an impact of an action taken by a person if:

   (a)    the event or circumstance is a direct consequence of the action; or

 (b)    for an event or circumstance that is an indirect consequence of the action—subject to subsection (2), the action is a substantial cause of that event or circumstance.

156    As is apparent from the matters listed above, not all aspects of the environment are the subject of the scheme for approval established by the EPBC Act. The subject matters of those aspects of the environment which are covered suggests that constitutional limitations upon the legislative power of the Commonwealth Parliament shaped that coverage. It is particularly apparent, including by reference to the terms of ss 137, 138, 139 and 140, that the EPBC Act was enacted largely relying upon s 51(xxix) of the Constitution so as to give effect to Australia’s obligations under a number of environmental and world heritage treaties or conventions.

157    It is pertinent to note that whilst protection is afforded to various listed species and their habitats, neither the health, wellbeing nor survival of human beings, nor their habitats (by which I mean homes or private real property) are protected aspects of the environment directly specified by Pt 3. Actions with significant impacts on those subject matters are not prohibited subject to either the Minister’s approval or the Minister’s decision that approval is not required.

158    That people and in particular future generations of people, should be able to enjoy the “health, diversity and productivity” of the environment is, however, a matter emphasised by the “principle of inter-generational equity” expressed in s 3A(c). Further, “protection of the environment” is an object of the EPBC Act (s 3(1)(a)) and the definition of the term “environment” given by s 528 refers specifically to “people and communities”. The term is there defined as follows:

environment includes:

 (a)    ecosystems and their constituent parts, including people and communities; and

 (b)     natural and physical resources; and

 (c)     the qualities and characteristics of locations, places and areas; and

 (d)     heritage values of places; and

(e)     the social, economic and cultural aspects of a thing mentioned in paragraph (a), (b), (c) or (d).

159    Paragraph (a) addresses “people” directly and, by reason of para (e), the social, economic and cultural aspects of the subject matters otherwise dealt with by the definition must also be understood as dealing with the interests of people. From that definition and the statute read as a whole, the conclusion may be drawn that the object of the EPBC Act is not the protection of environment per se but the protection of the interests of human beings in the environment including, in particular, those aspects of the environment which are specified in Pt 3.

160    As the “controlled actions” of relevance to the application to approve the Extension Project are addressed by ss 18, 18A, 24D and 24E of the EPBC Act, additional reference should be given to those provisions including so as to illustrate, beyond the outline already given, the scheme of the EPBC Act for regulating action taken in respect of a protected subject matter. In this respect I adopt, with some variation, most of what was helpfully outlined in the Minister’s written submission.

161    Sections 18 and 18A protect certain categories of listed threatened species and ecological communities by prohibiting an action that will have, or is likely to have, a significant impact on those species or ecological communities. Contravention of s 18 results in the imposition of a civil penalty and s 18A creates offences. Subsections 19(1) and 19(2) exclude the operation of s 18 and s 18A in relation to an action if there is a relevant approval to take the action by the person in operation under Pt 9 of the EPBC Act.

162    Sections 24D and 24E protect water resources from action that involves a coal seam gas development or a large coal mining development by prohibiting an action if it has, will have or is likely to have a significant impact on a water resource. Contravention of s 24D results in the imposition of a penalty and s 24E creates offences. Subsections 24D(1) to (3) do not apply if an approval to take the action is in operation under Pt 9: s 24D(4)(a). Similarly, the action described in ss 24E(1) to (3) does not constitute an offence if an approval of the action under Pt 9 of the EPBC Act is in operation: s 24E(4)(a).

163    Division 1 of Pt 9 of the EPBC Act governs the process by which the Minister may approve a controlled action under Pt 3. Relevantly, the Minister may approve for the purpose of a controlling provision the taking of controlled action after receiving the “assessment documentation” relating to the controlled action: s 133(1) of the EPBC Act. Part 8 provides for the assessment of impacts of controlled actions in order to inform decisions made as to whether the taking of the action should be approved. Various methods of assessment including “environmental impact statements” (Div 6 of Pt 8) are provided for. The Minister must identify the method to be adopted in any particular case: s 87. However, s 83(1) provides that an assessment of the kind required by Pt 8 need not be conducted where, inter alia, a bilateral agreement between the Commonwealth and a State is operative. In such a case an assessment conducted by the State may be substituted for the assessment that would otherwise be required by Pt 8: s 47.

164    For the application to approve the Extension Project, the “assessment documentation” referred to by s 133(1) included the NSW Department Report referred to above at [26]. That was submitted to the IPC and given to the Minister pursuant to cl 6.2 of the Bilateral Agreement, which was made between the Commonwealth and New South Wales pursuant to s 47(1) of the EPBC Act: s 133(8) and s 130(2).

165    As required by s 47(4) of the EPBC Act, the bilateral agreement (cl 6.2(a)) provides that:

NSW will ensure there is sufficient Information in the Assessment Report on the impacts of a controlled action covered by this Agreement on each relevant Matter of [national environmental significance] so that the Commonwealth decision-maker may consider those impacts when determining whether to approve the action and, if so, on what conditions. The extent of the assessment will be proportionate to the level of likely environmental risk.

166    Section 130(1) imposes a duty on the Minister to decide whether or not to approve, for the purposes of each controlling provision for a controlled action, the taking of the action. Relevantly, the Minister must do so within 30 business days of receiving the assessment report (s 130(1B)(a)) or such longer time as the Minister specifies in writing (s 130(1A)).

167    Before the Minister decides whether or not to approve the taking of an action, and what conditions (if any) to attach to the approval, she must inform any other Minister whom she believes has “administrative responsibilities relating to the action” of the decision that she proposes to make and invite the other Minister(s) to comment on the proposed decision within 10 business days: s 131(1). Without limiting the comments that another Minister may give, s 131(2) provides that in response to an invitation another Minister may make comments that “relate to economic and social matters relating to the action” and those comments may be considered by the Minister “consistently with the principles of ecologically sustainable development”.

168    Section 131AA requires the Minister to give notice of her proposed decision, including any conditions that she proposes to attach to the approval, to the proponent of the action and invite the person to provide comments within 10 business days. In making the final decision as to whether or not to approve the action, the Minister must take into account “any relevant comments” provided in response to an invitation: s 131AA(6).

169    If an action involves a coal seam gas development or a large coal mining development, and the Minister believes that the taking of the action is likely to have a significant impact on water resources and may have an adverse impact on a matter protected by a provision of Pt 3, she must obtain the advice of the Independent Expert Scientific Committee on Coal Seam Gas and Large Coal Mining Development (established by s 505C) before deciding whether or not to approve the action: s 131AB.

170    The Minister may, but is not required to, publish her proposed decision and any conditions that the Minister proposes to attach to the approval on the Internet, and invite comments in writing on the proposed decision: s 131A. Section 132 provides the Minister the capacity to request that further information be provided where “on reasonable grounds” she believes she does not have enough information to make an informed decision.

171    Section 133 deals with the grant of approval and relevantly provides:

Grant of approval

Approval

(1)    After receiving the assessment documentation relating to a controlled action, or the report of a commission that has conducted an inquiry relating to a controlled action, the Minister may approve for the purposes of a controlling provision the taking of the action by a person.

(1A)    If the referral of the proposal to take the action included alternative proposals relating to any of the matters referred to in subsection 72(3), the Minister may approve, for the purposes of subsection (1), one or more of the alternative proposals in relation to the taking of the action.

Content of approval

(2)    An approval must:

  (a)    be in writing; and

(b)    specify the action (including any alternative proposals approved under subsection (1A)) that may be taken; and

  (c)    name the person to whom the approval is granted; and

  (d)    specify each provision of Part 3 for which the approval has effect; and

  (e)    specify the period for which the approval has effect; and

  (f)    set out the conditions attached to the approval.

 Note:    The period for which the approval has effect may be extended. See Division 5.

172    Section 136 of the EPBC Act deals with the matters that the Minister either must or may consider in approving or imposing conditions upon the approval of a controlled action. It provides:

General considerations

Mandatory considerations

(1)    In deciding whether or not to approve the taking of an action, and what conditions to attach to an approval, the Minister must consider the following, so far as they are not inconsistent with any other requirement of this Subdivision:

(a)    matters relevant to any matter protected by a provision of Part 3 that the Minister has decided is a controlling provision for the action;

  (b)    economic and social matters.

Factors to be taken into account

(2)    In considering those matters, the Minister must take into account:

   (a)    the principles of ecologically sustainable development; and

   (b)    the assessment report (if any) relating to the action; and

(ba)    if Division 3A of Part 8 (assessment on referral information) applies to the action—the finalised recommendation report relating to the action given to the Minister under subsection 93(5); and

(bc)    if Division 4 of Part 8 (assessment on preliminary documentation) applies to the action:

(i)    the documents given to the Minister under subsection 95B(1), or the statement given to the Minister under subsection 95B(3), as the case requires, relating to the action; and

(ii)    the recommendation report relating to the action given to the Minister under section 95C; and

  (c)    if Division 5 (public environment reports) of Part 8 applies to the action:

(i)    the finalised public environment report relating to the action given to the Minister under section 99; and

(ii)    the recommendation report relating to the action given to the Minister under section 100; and

  (ca)    if Division 6 (environmental impact statements) of Part 8 applies to the action:

(i)    the finalised environmental impact statement relating to the action given to the Minister under section 104; and

(ii)    the recommendation report relating to the action given to the Minister under section 105; and

(d)    if an inquiry was conducted under Division 7 of Part 8 in relation to the action—the report of the commissioners; and

(e)    any other information the Minister has on the relevant impacts of the action (including information in a report on the impacts of actions taken under a policy, plan or program under which the action is to be taken that was given to the Minister under an agreement under Part 10 (about strategic assessments)); and

(f)    any relevant comments given to the Minister in accordance with an invitation under section 131 or 131A; and

(fa)    any relevant advice obtained by the Minister from the Independent Expert Scientific Committee on Coal Seam Gas and Large Coal Mining Development in accordance with section 131AB; and

(g)    if a notice relating to the action was given to the Minister under subsection 132A(3)—the information in the notice.

Note:    The Minister must also take into account any relevant comments given to the Minister in response to an invitation under paragraph 131AA(1)(b). See subsection 131AA(6).

Person’s environmental history

(4)    In deciding whether or not to approve the taking of an action by a person, and what conditions to attach to an approval, the Minister may consider whether the person is a suitable person to be granted an approval, having regard to:

  (a)    the person’s history in relation to environmental matters; and

(b)    if the person is a body corporate--the history of its executive officers in relation to environmental matters; and

(c)    if the person is a body corporate that is a subsidiary of another body or company (the parent body) — the history in relation to environmental matters of the parent body and its executive officers.

Minister not to consider other matters

(5)    In deciding whether or not to approve the taking of an action, and what conditions to attach to an approval, the Minister must not consider any matters that the Minister is not required or permitted by this Division to consider.

173    In deciding whether or not to approve the taking of an action, and what conditions to attach to an approval, the Minister must consider the matters set out in s 136(1) “so far as they are not inconsistent with any other requirement” of Subdiv B of Div 1 of Pt 9. The provisions of that subdivision address decisions about World Heritage properties (s 137), National Heritage places (s 137A), Ramsar wetlands (s 138), threatened species and endangered communities (s 139), migratory species (s 140) and certain nuclear installations (s 140A). For instance and in relation to whether or not to approve the taking of an action specified in s 18 or s 18A, the Minister must not act inconsistently with Australia’s obligations under a number of international Conventions specified in s 139, including the Convention on Biological Diversity done at Rio de Janeiro on 5 June 1992. Save as to those negative stipulations, the EPBC Act does not provide any criteria as to which the Minister must be satisfied in order to grant approval.

174    The matters that “must” be considered, which are identified in s 136(1) are:

(a)    matters relevant to any matter protected by a provision of Part 3 that the Minister has decided is a controlling provision for the action;

(b)    economic and social matters.

175    In considering those matters, the Minister “must take into account” the matters set out in s 136(2). Such matters include the principles of ecologically sustainable development set out in s 3A and mentioned above. Furthermore, the Minister is permitted to also consider “whether the person is a suitable person to be granted an approval”, having regard to the person’s history in relation to environmental matters: s 136(4). The Minister must not consider any matters that she is not required or permitted by Div 1 of Pt 9 to consider: s 136(5).

176    I will return to address the proper construction of s 136, but before doing so, two matters should be mentioned.

177    First, the Minister may attach a condition to an approval of an action if she is satisfied that the condition is “necessary or convenient” for protecting a matter protected by a provision of Pt 3 for which the approval has effect (s 134(1)(a)) or protecting specifically from the action any such matter (s 134(2)(a)). The Minister may also attach a condition if satisfied that the condition is “necessary or convenient” for repairing or mitigating:

    damage to a matter protected by a provision of Pt 3 for which the approval has effect whether or not the damage has been, will be or is likely to be caused by the action (s 134(1)(b)); or

    damage that may or will be, or has been, caused by the action to such a matter (s 134(2)(b)).

178    Subsection 134(3) sets out a non-exhaustive list of conditions that may be attached to an approval. On its text, s 134 only empowers the Minister to attach a condition that is directed to protecting a matter protected by a provision of Pt 3.

179    Second, s 487 of the EPBC Act confers an extended right of standing to seek judicial review pursuant to the Administrative Decisions (Judicial Review) Act 1977 (Cth) of, inter alia, decisions made under the EPBC Act. Section 487(2) deems an individual to be a person aggrieved by the decision for the purpose of the judicial review proceeding if they are an Australian citizen or resident and, at any time in the two years immediately before the decision, have engaged in a series of activities in Australia for protection or conservation of, or research into, the environment.

180    The proper construction of s 136 of the EPBC Act has been the subject of Full Court authority with which no party took issue. In Tarkine National Coalition Inc v Minister for the Environment (2015) 233 FCR 254, Jessup J (with whom Kenny and Middleton JJ agreed), described Subdiv B of Div 1 of Pt 9, in which s 136 is found, as establishing “a closed system of the matters the Minister [is] to consider in making [her] decision and the things that should be taken into account” (at [28]). However, no party contended that the potential for harm to Australia’s children was not a matter that the Minister may permissibly take into account in deciding whether or not to approve an action. The applicants contended that such a matter fell within the expression “economic and social matters” in s 136(1)(b). The Minister did not contend to the contrary. That expression is unqualified and there is no basis for thinking that it was intended to be confined to those economic and social matters which are a beneficial rather than an adverse consequence of the “controlled action” subject to the Minister’s approval. The Act’s concern with adverse economic and social matters can be seen from ss 270(3)(c), 287(3)(c) and 464(3) as well as the objects in s 3 and, in particular, the object in s 3(1)(b) in light of the elaboration provided by s 3A(a).

181    A particular “economic” or “social” matter is not, on the authority of Tarkine, a mandatory consideration that the Minister is required to take into account or to consider. At [25]-[28], Jessup J relevantly said this:

Returning to s 136, I would make four observations about the structure and content of this section. First, subss (1) and (2) made a distinction between the matters that the Minister “must consider” (subs (1)) and the things that the Minister “must take into account” in considering those matters (subs (2)). The purpose of subs (1), as it seems to me, was to mark out the broad categories of consideration to which the Minister was required to turn his mind, and specifically to require consideration not only of the matters protected by Pt 3 of the EPBC Act but also of matters that, otherwise, appear to be of no concern under that Act, namely, “economic and social matters”. Neither para (a) nor para (b) of s 136(1) dealt, at the level of detail, with particular matters that required consideration. For example, what, if any, particular “social matter” might have required consideration in a proposal that came before the Minister was, it seems, a matter for the Minister.

Secondly, the expression “matters relevant” in s 136(1) was not defined in the EPBC Act. By contrast, the expression “relevant impacts”, used in s 136(2)(e), was defined and gave content, at the level of detail, to the Minister's obligation to take things into account. I shall return to this definition below.

Thirdly, while the range of things that the Minister was to take into account under subs (2) was extensive, with the exception of those referred to in paras (a) and (e), each was a concrete document or some similar existing artefact. In effect, what the Minister had to take into account were the contents of those documents or artefacts. This approach to regulation is to be contrasted with a situation in which the things to be taken into account were identified by description, or generically, such as, for example, where a decision-maker was required to take account of the condition of the habitat of a particular species. Subject to the exceptions mentioned, the scheme of s 136 was one in which it was assumed that specific subjects of this and similar kinds were already dealt with in the documents or artefacts referred to. The role of the Minister was to take into account the things that were before him in this way, rather than being either obliged or entitled to undertake additional research or investigations.

Fourthly, the terms of s 136(5) should be noted. While they require no further explanation, they confirm the impression that Subdiv B established a closed system of the matters that the Minister was to consider in making his decision, and the things that should be taken into account.

182    At [44], Jessup J characterised the purpose of s 136(1) as “categorical”, that is, as intended to set out the categories within which the Minister may choose to take a matter into account. As his Honour relevantly said of s 136(1) at [45]:

I do not regard this provision as the source of any obligation to take particular matters into account, in point of detail. So long as the Minister, in making his or her approval decision, proceeded by reference to the categories in s 136(1), the decision could not be assailed on the ground that some particular matter, falling within either para (a) or para (b), had not been considered. The particular matters that had to be taken into account were the concern of subs (2).

183    The reasoning of North J in Blue Wedges Inc v Minister for Environment Heritage and the Arts (2008) 167 FCR 463 at [115] is to the same effect.

5.     The Affirmative Salient Features

5.1    Reasonable Foreseeability of Harm

184    I turn then to consider reasonable foreseeability of harm as a salient feature. In doing so I will make extensive reference to the evidence about the risk of personal injury alleged by the applicants. That evidence is also relevant to other issues I need to consider but is conveniently referred to in this section. Extensive reference to the evidence is not commonly made by courts when considering reasonable foreseeability for the purpose of determining whether a duty of care exists. That is because aspects of that inquiry overlap with inquiries about breach of duty and causation of harm and it is usually more convenient for a close analysis of the evidence to be conducted in relation to the question of breach. As this is a peculiar case in which the establishment of a duty of care is being considered prior to any alleged breach or actual harm, I have made adjustments to accommodate the peculiarity. I confine my assessment to the foreseeability of the risk of personal injury and not property damage or pure economic loss for reasons alluded to above at [148] and further explained in Section 6, which deals primarily with ‘coherence’.

185    A duty of care is owed to an individual and must be considered in relation to the facts of that individual’s case: Agar v Hyde (2000) 201 CLR 552 at [66] (Gaudron, McHugh, Gummow and Hayne JJ). As their Honours went on to say at [66]-[67]:

That does not mean that the conduct of a person cannot give rise to a duty of care to many persons. Nor does it mean that a person cannot owe a duty to someone whom he or she does not know or cannot identify.

Nevertheless, the basic rule of the law of negligence is that it is “incumbent on a claimant to establish breach of an independent duty to himself as a particular individual”.

186    In assessing the reasonable foreseeability of harm for the purpose of considering whether a duty of care is owed, the Court is obliged to undertake a “generalised inquiry”: Wyong Shire Council v Shirt (1980) 146 CLR 40 at 47 (Mason J); Vairy v Wyong Shire Council (2005) 223 CLR 422 at [72] (Gummow J). In other words, the inquiry takes place at a “higher level of abstraction” than that which is required when considering a breach of a duty of care: Vairy at [71] (Gummow J, citing Glass JA in Shirt v Wyong Shire Council [1978] 1 NSWLR 631 at 639). In the context of the particular task presented by this case, the question that needs to be answered in relation to each of the Children is whether a reasonable person in the position of the Minister would foresee that the approval of the Extension Project would expose the Children to a risk of personal injury (cf. Crimmins at [223] (Kirby J)).

187    That assessment can only be made prospectively by reference to the risk of the future harm alleged. Reasonable foreseeability of that harm will be established where, at the time of the Minister’s approval, there exists a real risk of the harm occurring. A real risk is a risk which is not far-fetched or fanciful: Shirt at 48 (Mason J); McKenna at [30] (French CJ, Hayne, Bell, Gageler and Keane JJ). The test of foreseeability has been described as “undemanding”: Shirt at 44 (Mason J, citing Glass JA in Shirt v Wyong Shire Council at 641).

188    To establish a duty of care it is not necessary for a plaintiff to demonstrate that the defendant should reasonably foresee (or have foreseen) that the plaintiff or some particular person or persons may be at real risk of harm. It is, as Dixon CJ, Kitto, Taylor, Menzies and Windeyer JJ said in Chapman v Hearse (1961) 106 CLR 112 at 121, “sufficient if it appears that injury to a class of persons of which [the particular person] was one might reasonably have been foreseen” by the defendant (see also Crimmins at [223] (Kirby J)).

189    Chapman v Hearse was a case in which a rescuer who came to the aid of a person injured in a car accident was then himself injured by the driver of a car passing the scene of the first accident. The “class of persons” the Court had in mind was a class defined by the geographical and temporal proximity of a member of the class to the scene of the first accident and their “moral and social duty to render aid” (at 120). It was those common characteristics of class membership which made it reasonably foreseeable that a member of that class (a rescuer) would be exposed to the harm inflicted.

190    That case, however, was concerned with the foreseeability of a single member of a class being harmed. I am concerned with the foreseeability of some 5 million people being harmed. For the applicants to succeed in establishing that the Minister should reasonably foresee harm to each of those persons, I need to be satisfied that each member of that class is exposed to a real risk of harm from the Minister’s conduct. That forensic exercise is not without its difficulties, given the general nature of the evidence. None of the evidence about the risk of future harm is directed to any particular child or any particular sub-group of the Children.

191    However, as will become apparent when I turn to consider the foreseeability of the particular harms relied upon by the applicants, the age of a person may have a relevant nexus to the risk of exposure to those harms because susceptibility thereto tends to be age-related. Alternatively or perhaps additionally, age may have a temporal connection to the particular harm because the exposure is not immediate but delayed and only persons of a younger age are likely to be exposed to the harm or to its full intensity. A common shared circumstance of relevance to each of the Children is that they are all under 18 years of age. In relation to some of the harms in question, there is a geographical or locational element which is relevant to the risk of harm. A common shared circumstance is that all the Children are geographically located in Australia. Common circumstances such as those have enabled a conclusion that some of the events induced by climate change relied upon by the applicants expose each of the Children to a real risk of harm. The evidence of other such events permits a conclusion that some of the Children may be exposed to a real risk of harm, but it does not permit an identification, even by way of a sub-class, of who those children are.

192    Before embarking upon the analysis required, there are further observations made in Chapman v Hearse which are of relevance to the present case. As the Court said at 120, the test for the existence of a duty of care does not depend upon “the precise sequence of events” which lead to the injury being reasonably foreseeable. Nor is it necessary that the precise damage that may be caused be reasonably foreseeable. That is because “…it would be quite artificial to make responsibility depend upon, or to deny liability by reference to, the capacity of a reasonable [person] to foresee damage of a precise and particular character or upon [that person’s] capacity to foresee the precise events leading to the damage complained of” (at 121). Further, their Honours characterised reasonable foreseeability as “not, in itself, a test of ‘causation’” (at 122).

193    The applicants do not contend that any of the alleged harms would be occasioned as a direct result of the Minister’s conduct. What is effectively alleged is a series of steps or processes in a chain of events between the Minister’s conduct and each category of harm alleged. That is so because the risk of the harm alleged depends on future emissions of CO2 increasing global average surface temperature which, in turn, increase the frequency, ferocity or geographical range of one or more hazards (such as bushfires) in circumstances where susceptibility to harm is not entirely a function of exposure to climate-induced hazards.

194    The Minister’s contentions in respect of ‘reasonable foreseeability’ were founded upon a causal analysis, including as to the materiality of the impugned conduct to the alleged risks of harm. The Minister contended that the foreseeability of harm from the conduct of the Minister which the applicants impugn was causally negated by the complex interaction of factors that will evolve over the coming decades. She contended that each step in the causal chain of events relied upon by the applicants to connect the Minister’s conduct to the harm alleged was a contingency and that the possibility of that contingency not occurring denied the foreseeability of the harm. As noted already, ‘reasonable foreseeability’ is not a test of causation. As the discussion at the end of this section demonstrates by reference to authority, foreseeability of risk and likelihood of risk are different concepts. An event may be foreseeable even though its occurrence is improbable, including because one or more of the necessary steps in the chain of events which connect the defendant’s conduct with the alleged harm is improbable. Furthermore, “[f]oreseeability does not mean foresight of the particular course of events causing the harm. Nor does it suppose foresight of the particular harm which occurred, but only of some harm of a like kind”: Mount Isa Mines Ltd v Pusey (1970) 125 CLR 383 at 402 (Windeyer J). As Gummow J said in Rosenberg v Percival (2001) 205 CLR 434 at [64], “[t]he precise and particular character of the injury or the precise sequence of events leading to the injury need not be foreseeable”. I therefore reject the Minister’s approach to ‘reasonable foreseeability’, which essentially amounted to a contention that the mere possibility of a break in the causal chain will suffice to deny the reasonable foreseeability of harm. I accept, however, that it is necessary to consider whether the steps in the chain of events asserted by the applicants, individually or collectively, by reason of the complexity of their interactions or otherwise, deny reasonable foreseeability because they deny the existence of a real risk that the Minister’s conduct will expose the Children to the particular harm in question.

195    The chain of events commences with the Minister’s conduct itself – the lifting of a statutory prohibition enabling Vickery to extract an additional 33 Mt of coal from its coal mine. For the purposes of this exercise, approval by the Minister must be assumed. The next steps in the chain of events to be considered are the extraction of the coal, its sale, its combustion and the consequential emissions of CO2 into the Earth’s atmosphere. None of those contingencies serves to deny the real risk of harm. There is evidence of a market for the coal and its likely sale to and combustion in Japan, South Korea and Taiwan.

196    The next step in the chain is that the emission of CO2 into the Earth’s atmosphere from the combustion of that coal will increase global average surface temperature. But that is not a contingency. The proposition is not disputed and is the subject of unchallenged expert evidence. What is disputed is the significance of the temperature increase. The Minister disputes that her conduct will make a material contribution to the alleged harms occurring even if it is the case that climate change bears responsibility for those harms. That is a matter to which I will return.

197    Next, the various types of harm contended for by the applicants depend upon there being a nexus between an increase in global average surface temperature and the increased frequency or gravity of extreme climatic events such as heatwaves or bushfires which create a particular risk of harm. Some such risks are alleged to be an indirect or flow-on consequence of climate change induced phenomena. It is convenient to set out now what the Minister accepts that is of relevance to the risk here under discussion.

198    It is not in dispute that the emission of CO2 into the atmosphere bears the lion’s share of responsibility for the warming of the Earth’s surface since the Industrial Revolution. The Minister accepts and thus must be taken to have actual knowledge or understanding of the fact that, due to increased greenhouse gas emissions (primarily CO2), the global average surface temperature is increasing and that humans are primarily responsible. The Minister accepts that increases in temperature affect the environment, economy and society. Climate change exacerbates inherent risks in the Australian climate and introduces new risks. She accepts that heatwaves, droughts, bushfires, floods and tropical cyclones are all part of the Australian climate experience and that economic infrastructure in Australia’s cities and ports is vulnerable to sea level rises and storm surges. Australia’s agriculture, mining and other industries are all vulnerable to increasing frequency of severe heatwaves and intensity of drought, floods and storms. Further, the Minister accepts that terrestrial and marine ecosystems are facing serious threats from climate change, including extreme weather events, bushfires, ocean acidification and marine heatwaves. There is an acceptance that the effects of increased temperatures are likely to be compounded by climate change induced events such as severe storms, heatwaves, more extreme droughts and floods, and sea level rise. It is not disputed that these events will have impacts on the Australian economy, Australia’s natural and managed terrestrial and marine ecosystems, and on the health and wellbeing of individuals, communities and society.

199    Looking to the future, the Minister accepts that under all future emission scenarios, it is very likely that: (a) average temperatures will continue to increase and Australia will experience more heat extremes and fewer frosty days; (b) extreme rainfall events will become more intense; (c) southern and eastern Australia will experience more extreme fire-related weather; (d) the time in drought will increase over southern Australia; (e) sea levels will continue to rise throughout the 21st century, with increased frequency of storm surge events; and (f) oceans around Australia will warm and become more acidic. The Minister also accepts that the projected effects of climate change vary depending upon the extent of global emissions of greenhouse gases in coming years.

200    The nexus between increasing global average surface temperature and the harms alleged by the applicants is broadly encompassed by the facts which the Minister accepts. In any event, that nexus is dealt with by the applicants’ evidence. In relation to the risk of each particular kind of personal injury to which the applicants contend climate change exposes the Children, whether that nexus is established to the standards required in the application of the test of reasonable foreseeability is later addressed.

201    It is necessary then to identify each of the alleged risks of harm to the Children relied upon by the applicants. The particulars of the alleged harm were set out at [16] of the applicants’ Amended Concise Statement as follows:

Particulars of harm include mental or physical injury, including ill-health or death, or economic loss, from:

(a)    more, longer and more intense: (i) bushfires, storm surges, coastal flooding, inland flooding, cyclones and other extreme weather events; (ii) periods of extreme heat; (iii) periods of drought;

 (b)    sea-level rise;

 (c)    increasing loss of non-human species and ecosystems, on land and in oceans;

(d)    systemic breakdowns and overwhelming of infrastructure networks and critical services, including electricity, water supply, internet, health care, and emergency services;

 (e)    food insecurity and breakdown of food systems;

(f)    adverse impacts on: (i) national and global economies; (ii) financial markets; (iii) industries, businesses and professions; (iv) the number and quality of employment opportunities; (v) standard of living; and (vi) living costs;

 (g)    increasing smoke, heat, and disease;

 (h)     loss of clean water, clean air and nutriment (essentials);

(i)    social and political unrest, violence and scarcity as essentials are depleted, and humans try to move in search of essentials, habitable land, or both; and

(j)     mental harm caused by solastalgia, and the experience and anticipation of the above.

202    Those particulars travel beyond the risk of personal injury and include the risk of economic harm which I need not assess for reasons later given. However, the way in which the description of harm has been structured in the Amended Concise Statement makes it a little difficult to entirely separate the economic harm alleged from personal injury. The applicants’ written submissions together with the evidence relied upon provided greater specification. Under a heading “Physical and mental harm” and by reference to a categorisation made in Professor Capon’s Report, the applicants grouped the alleged physical and mental harms into a number of categories being “direct impacts”, “indirect impacts” and “flow-on impacts” on human health. “Direct impacts” were described as including specific risks of personal harm (injury or death) inflicted as a result of the occurrence of particular phenomena such as bushfires or heatwaves. “Indirect impacts” were said to comprise changes to physical systems, biological systems and ecosystem structure and function. Finally, “flow-on impacts” were described as being brought about by social, economic and demographic disruption. These categories of impacts will be elaborated on below.

203    I regard the applicants’ categories of “indirect harms” and “flow-on impacts” as intended to encompass all of the physical and mental harms listed in the particulars given at [16] of the Amended Concise Statement, other than those directly resulting from the particular climatic events listed at (a) of that paragraph. If more extensive physical or mental harms encompassed by those categories were intended to be alleged in [16] of the Amended Concise Statement, they were not sufficiently specified, there was no evidence about them (that the Court was taken to) and I could not be satisfied of their reasonable foreseeability.

204    I do not accept that the applicants have demonstrated that each of the Children is exposed to a real risk of climate change induced personal injury in relation to any of the harms in the categories of “indirect impacts” or “flow-on impacts”. Some of the Children may be exposed, but the evidence does not permit their identification either directly or as members of a clearly identifiable sub-class. Before coming to a more detailed assessment of those alleged harms, I turn to consider the injury-inducing events which I am satisfied expose each of the Children to a real risk of harm from extreme weather events brought about by climate change.

5.1.1    Heatwaves

205    The likely exposure of the Children to the risk of personal injury or death by reason of heatwaves induced by climate change was largely dealt with by expert evidence given by Dr Mallon and Dr Meyricke.

206    Dr Mallon’s expertise is set out in the Schedule, but particular aspects of his experience should be emphasised. Since 1997 Dr Mallon has worked in the field of climate change physical impact analysis, providing risk analysis to governments, the private sector and non-government organisations. He described his area of specialisation as being “the quantification of human and financial impacts from climate change, and cost-benefit-analysis of adaptation”. He is a director of two companies who provide climate change physical impact analysis. One of those, XDI Pty Ltd, is considered to be one of the world’s top four providers of physical risk analysis. The other, Climate Risk Pty Ltd, assesses how extreme weather and climate might cause harm to built assets and to communities.

207    Dr Mallon explained that models developed by Climate Risk are inherently probabilistic in nature. The main mechanism by which climate change impacts are evaluated and quantified under those models is the changing probability of events capable of breaching the design threshold of a given asset or the coping capacity of component elements. As Dr Mallon further explained, failure and strain thresholds can also be applied by those models to cohorts of people, which means that the number of people affected by extreme events can be quantified going forward. Dr Mallon stated that modelling can be utilised to analyse impacts such as flooding and forest fires. He referred specifically to the modelling developed by Climate Risk to assess heat stress and stated that the model uses metrics specifically designed to capture the circumstances which cause heat stress – metrics which are presently used as warnings by the CSIRO and the BoM – and applies them to detailed climate change modelling data. Dr Mallon observed that it thus becomes possible to forward compute the annual projected numbers of people likely to suffer discomfort or heat stress, call a doctor, or attend hospital.

208    Dr Mallon was asked to assess possible future impacts, including those of the kind identified at [16] of the Amended Concise Statement, resulting from various phenomena including climate change related extreme events (such as heatwaves) specified at [15] of the Amended Concise Statement. His Report does so but selectively. First, Dr Mallon helpfully chose to confine his consideration of future impacts to Australia’s children. He assessed the future impacts upon that cohort in relation to three broad categories – wealth, prosperity and health – and then presented his opinion in terms of the following three epochs:

    impacts in the near future (approximately 2020-2030);

    impacts in the middle of the Children’s working lives (approximately 2040-2060); and

    impacts at the end of the Children’s lives (approximately 2070-2100).

209    The first two categories (impacts on wealth and prosperity) are not relevant to this discussion. The third category is. Dr Mallon opined that there are many climate change impacts that would affect the health of the Children. Those impacts range from injury and extreme events such as cyclones to the long term impacts of smoke inhalation during bushfires. However, Dr Mallon confined his specific assessment on health impacts to the ill-effects of heat stress. He seems to have done so for a number of reasons. First, he considered that all Australians (and thus the entirety of the relevant cohort) “will be affected by increases in temperatures and especially extreme temperatures”. Furthermore, he stated that he had confined the subjects of his assessments to those in relation to which he had access to sufficiently detailed modelling upon which to form an opinion. He also referred to time constraints on the preparation of his Report as a limiting factor.

210    In assessing heat stress impacts upon the Children, Dr Mallon further confined his assessment to the third epoch, being the years 2070 to 2100. The upper temporal limit appears to have been chosen because that is about as far as current climate change models extend and because, statistically speaking, the Children will have passed the average Australian life expectancy between 2078 and 2098. The lower temporal limitation appears to be the product of two assumptions made by Dr Mallon. The first is that, in his view, RCP 8.5 (a scenario which in terms of surface temperature is comparable to what I have earlier referred to as a 4°C Future World) is the most applicable future scenario for predicting the extent and timing of future atmospheric temperature increases. That assumption is relevant to the extent of the frequency and severity of heatwaves likely to be experienced by the Children. The second assumption deals with when the burden of heat stress will likely have its greatest impact upon the Children. As Dr Mallon stated, the burden of heat stress does not fall equally on the population and hits the elderly particularly hard. Dr Mallon gave the example of the 2003 French heatwave which resulted in 14,729 excess deaths, 11,731 of which involved the death of a person over the age of 75. As Dr Mallon stated, the Children will all be over 80 years old by 2100.

211    Dr Mallon’s opinion about the likely impact of heat stress upon the Children is relatively succinct and it is helpful to set it out in full:

9.3    There is an emerging body of research to suggest that the health impacts of heat are not solely dependent on temperature, but on ‘thermal shock’ or the inability to acclimatise to heatwaves and also the inability to cool down or get respite from severe temperature (Goldie et al, 2017). The Climate Risk science team has adopted the CSIRO developed metric Excess Heat Factor (EHF) to interpret the climate change modelling data in terms of ill health. These are used for quantification of heat-stress related doctor, ambulance and hospital presentation.

9.4     The figure below shows the trends in Excess Heat Factor in Melbourne over time for one of the climate models considered. EHF includes the degree to which people are able to acclimatise to increasing temperatures ahead of a heat wave and uses an average daily temperature which captures both the daily extreme and the night time minimum.

9.5    Using EHF to predict the days where a heat wave is likely to cause ill health combined with elevated presentation rate data during actual heat waves (Nitschke, Tucker, and Bi 2007, Department of Health & Human Services 2014, Jegasothy et al. 2017) enables forward looking projections of presentations to doctor, ambulance or hospital due to heat-stress under RCP8.5, see the table below.

9.6     Based on the sample of 1% of all addresses and assuming average occupancy levels, my team has estimated the thermal shock of heat waves. In Southern states like Victoria heat stress presentations to doctors, paramedics and hospitals will more than double. But for Australia as a whole the incidents will increase by 850% or an eight fold increase.

9.7    In practical terms that means that 8 million doctor visits will be attributable to climate change driven warming, equivalent to an average 38% of the population attending the doctor due to a heat stress event. There are also expected to be 50,000 of additional hospitalisations due to heat stress.

212    In the next paragraph of his report, Dr Mallon referred to observations I have already dealt with about the burden of heat stress falling unequally. Having given the example of the 2003 French heatwave, Dr Mallon then said this at [9.8]-[9.9]:

So we can assume that the doctor visits and hospitalisations will be heavily represented by today’s children. If such ratios play out for the cohort in question, then every year, 1% of the group will be hospitalised with heat-stress exacerbated illness - an estimated tenth of that figure today. Put another way, it would imply that in the last 20 years of these children’s lives, on average one [in] five will suffer at least one heat-stress episode serious enough to require acute care in a hospital.

Some of those heat-stress events will be fatal, but as my team has not undertaken these statistical calculations or projections I am unable to offer an opinion.

213    Later in his report and in setting out a summary of his conclusions, Dr Mallon stated:

10.12    I have specifically considered the impacts on the cohort of children when they pass 75 years of age, when statistically speaking they are at a significantly heightened risk from heat-stress related health impacts. Climate change will cause an 8.5 fold increase in the probability of an average person having a heat-stress related presentation to a doctor or hospital. On the balance of probabilities it’s likely that 1 in 5 of the cohort will be hospitalised due to heat stress during the senior years. Some of these people will die due to exacerbated underlying health conditions.

214    Dr Meyricke’s expertise as an actuary is set out in the Schedule. Her specific academic and research experience in relation to climate change mitigation and also as to the risk of mortality should be noted. Her report was based on her published research on the subject titled “Climate Change, Mortality and Retirement Incomes” and co-written with Rafal Chomik: Meyricke and Chomik (2019).

215    Dr Meyricke’s report considered possible future impacts of heat and heatwaves on mortality, including analysis of the effects of such impacts on the Children. She defined a heatwave to be three or more days of unusually high maximum and minimum temperatures in any area. She stated that heatwaves significantly increase mortality across the globe. In Australia, heatwaves have caused more deaths since 1890 than bushfires, cyclones, earthquakes, floods and severe storms combined. Dr Meyricke opined that in addition to deaths, heatwaves also drive an increase in heat-related illness (or morbidity) as exemplified in the following table.

216    As Dr Meyricke stated, Table 1 illustrates the numbers of excess deaths (being the number of deaths over what would normally be expected for the same period), emergency department presentations, after-hours doctor consultations and ambulance dispatches during two separate week-long heatwaves in Melbourne. To put the incidents exemplified by Table 1 into context, Dr Meyricke stated that the 374 excess deaths in Victoria during the heatwave which commenced on 26 January 2009 represented a 62% increase in total all-cause mortality during the period of the heatwave.

217    Dr Meyricke stated that the health and mortality effects of heatwaves are more pronounced in older people, explaining that the increased vulnerability of the elderly to heatwaves relates to a combination of altered homeostatic mechanisms and the higher prevalence of chronic diseases among the elderly. To demonstrate the point, Dr Meyricke stated that, while individuals aged 75 years or older made up 6.5% of the Victorian population, in the 2009 heatwave in Victoria exemplified in Table 1:

•    61% of the 7,035 ambulance dispatches were for those 75 years or older;

•    65% of the 1,955 after-hours doctor consultations were for those 75 years or older;

•    46% of the 714 Emergency Department heat-related presentations were for those 75 years or older; and

•    66% (or 248) of the 374 excess deaths occurred in those 75 years or older.

218    Dr Meyricke acknowledged that as the risk of heat-related mortality increases with ageing, individuals currently under 18 years of age would be most at risk from heatwaves in their late adulthood (that is, around when they reach age 65, in 47 to 65 years’ time). Meyricke and Chomik (2019) estimated that excess mortality in persons over 65 years old is approximately four to six times higher than excess all-ages mortality caused by similar heatwaves. Excess mortality expected in the period from 2020 to 2040 is 1% for all ages (in Sydney), but 4% in persons over 65 years old (in Brisbane). Over the period between 2060 and 2080, excess mortality from heat is projected to be 2% for all ages, but 12% in persons over 65 years old.

219    Dr Meyricke opined that with ongoing improvement in public awareness and risk mitigation, the extent of excess mortality from climate change could be limited, for instance by the Children taking measures in mitigation such as increasing the time spent indoors over the course of their lifetimes, compared with past generations. She stated, however, that whilst human capacity to adapt to varied climates is considerable, there are absolute limits to the amount of heat exposure an individual can tolerate. Even with highly effective adaptation (for example, all time spent indoors in air-conditioned environments) there are residual risks, such as air conditioning system failure or power failure. Dr Meyricke opined, therefore, that “even with effective adaptation and risk mitigation there will still be excess mortality in future, amongst individuals currently under 18 years of age, from heatwaves”. She stated that an increase in mortality risk globally, even after allowing for adaption, is expected by the World Health Organisation (WHO), which has stated (WHO 2018)):

Overall, climate change is projected to have substantial adverse impacts on future mortality, even considering only a subset of the expected health effects, under optimistic scenarios of future socioeconomic development and with adaptation.

220    Dr Meyricke’s opinion is consistent with the prediction made by the IPCC Synthesis Report (2014) with ‘high confidence’ that climate change will lead to an increased risk of heat-related mortality compared to a baseline without climate change.

221    Professor Steffen’s evidence was not directed at the detrimental effects of heatwaves as specifically as the evidence that I have already addressed. However, Professor Steffen did opine that even under his “Scenario 1” (ie a 2°C Future World), there would be a significant increase in the likelihood of extreme weather events in Australia and in particular a 77% likelihood of severe heatwaves in a given year. In relation to “Scenario 2” (a 3°C Future World), Professor Steffen referred to heatwaves increasing rapidly and included escalating heat stress as one of the impacts upon the health and wellbeing of Australians. He did not specifically refer to heat stress under his “Scenario 3” (a 4°C Future World), but it logically follows that his view about the prevalence of heatwaves and heat stress would be at least as grave as for a future world with a lower stabilised temperature.

222    A report by the Victorian Department of Environment, Land, Water and Planning entitled Victoria’s Climate Science Report: 2019 (DELWP report) predicts that in 2090 (with average temperatures in Victoria increasing between 2.8 to 4.3°C) parts of Victoria could experience days of up to 55°C in summer and 33°C in winter.

223    Referring to a study of extreme heat events in Australia between 1844 and 2010, Professor Capon stated that “[h]eatwaves are the most deadly natural hazard in Australia”. He opined that climate change is increasing the frequency and intensity of these extreme events and that risk to health can be expected to increase.

224    All of that evidence is available to the Minister and is to be regarded as being within her knowledge. As earlier mentioned, the Minister accepts that under all future emissions scenarios it is very likely that global average surface temperature will continue to increase and that Australia will experience “more heat extremes”. She accepts that the effects of increased temperatures are likely to be compounded by events induced by climate change, such as heatwaves.

225    Particularly in the absence of any challenge to the more specific evidence called by the applicants about heatwaves and their likely effect on the Children, all of that evidence should be regarded as reliable. The evidence addresses the age characteristic of the Children and makes out a sufficient link between that characteristic and the exposure to the harm in question. That is done by reference to the likely time in the lives of the Children when heatwaves are likely to be most frequent and most extreme and the time when the Children will be most susceptible to being harmed by heat stress. It establishes not only that many of the Children are exposed to a real risk of harm but that each of the Children is so exposed. I am satisfied that a reasonable person in the Minister’s position would foresee that each of the Children is exposed to a real risk of death or personal injury from heatwaves induced by climate change. It remains to consider whether it is reasonably foreseeable that the Minister’s impugned conduct will contribute to the risk of those harms. I will address that issue shortly.

5.1.2    Bushfires

226    Turning then to the risk of personal injury or death due to bushfires caused by climate change, the applicants sought to extrapolate from the harms caused by the 2019-2020 Australian bushfire season (“Black Summer fires”), as well as other major bushfires recently experienced in Australia. Reports by the CSIRO and the BoM demonstrate an increase in extreme fire weather and in the length of fire seasons across large parts of Australia since the 1950s. This is, in part, due to changes in temperature, rainfall and relative humidity, which affect fuel moisture content (dryness). Climate change also affects the amount of fuel that may be ignited in a bushfire, including via increased CO2 emissions which can alter the rate and amount of plant growth. As these trends continue, Australia is projected to experience an increase in the number of dangerous fire weather days and longer fire seasons, particularly in southern and eastern Australia. This mirrors the DELWP report, which anticipates that, by the 2050s, with continuing high emissions of CO2 (that is, RCP8.5) Bendigo, Ballarat and Shepparton will experience more than a 60% increase in the projected number of high fire danger days compared to 1986-2005.

227    Evidence was given by former State and Territory fire chiefs and senior emergency services personnel (Mr Waller, Mr Warrington, Mr Dunn and Mr Mullins) of their personal and professional experience of major bushfires in south-eastern Australia, including the Eastern Alpine bushfires in 2003, the Canberra bushfires in 2003, the Grampian bushfires in 2005-6, the Great Divide bushfires in 2006-7, the Victorian Black Saturday fires in 2009 and the Black Summer fires in 2019-2020.

228    That evidence speaks to increasing bushfire risk from the perspective of ignition, intensity and spread, and the impact on people, property and the environment. In particular, those witnesses gave evidence as to:

(i)    the increased frequency of major bushfires in south-eastern Australia;

(ii)    the impact of increased bushfire intensity (including the incidence of pyro-convective events or “fire storms”) in driving bushfire spread and reducing the defensibility and survivability of bushfires;

(iii)    the lengthening fire seasons, with a consequent impact on firefighting resources and resource-sharing;

(iv)    increasing climate-driven constraints on the implementation of planned burning and back-burning as fire mitigation tools, and the reduced impact of planned burning on bushfire spread;

(v)    the increasing incidence of bushfires in areas that were not previously prone or amenable to bushfires (such as Queensland rainforests); and

(vi)    the impact of climate change in reducing the capability of the natural environment to slow or stop fire spread (for example, through the progressive drying up of creeks and gullies that would previously have stopped or impeded fire spread).

229    Dr Mallon considered that forest fire events in recent years have shown the increases in severity and duration of forest fire events which have led to longer fire seasons and loss of life and property. He opined that, looking forward, the analysis suggested increased probability of fire conditions in many areas and, more worryingly still, increased penetration in areas not normally associated with forest fires. His report noted that as at 2019, 4% of Australian properties were exposed to forest fire. Addressing an example of a fire last summer in Nymboida where 150 homes were destroyed by fire, Dr Mallon stated that his models suggested that the probability of fire weather has increased by 17% since 1990 and that the increase could be much worse if it were to include the combined impact of increasing drought probability, which has increased by 8%.

230    Professor Steffen opined in relation to his “Scenario 1” (a 2°C Future World) that there was a 77% likelihood of severe bushfires in any given year. He opined that under that scenario there would be a longer fire season for the south and east of Australia and an increase in the number of dangerous fire weather days. Under his “Scenario 2” (a 3°C Future World), Professor Steffen opined that high fire danger weather will increase significantly, leading to more catastrophic fire seasons such as the Black Summer fires. Further, he stated that more frequent and intense bushfires would damage the health and wellbeing of Australians. For his “Scenario 3” (a 4°C Future World) Professor Steffen predicted that most of Australia’s eastern broadleafed (eucalypt) forest “will no longer exist due to repeated, severe bushfires”.

231    To demonstrate the extent of future harm to health from extreme bushfires, the applicants relied on the following evidence sourced from various studies about the health effects of the Black Summer fires and provided by Professor Capon (references removed):

Tragically, 33 people lost their lives during that bushfire season, including 9 firefighters. Epidemiologists have since estimated that the smoke from those bushfires was associated with 429 premature deaths, 3230 hospitalisations for cardiovascular and respiratory problems, and 1523 emergency department presentations for asthma. Other health impacts of fires include the long-term health sequelae of burns, impacts on eye health, substance use, and domestic and family violence. The mental health toll from the 2019-20 bushfires, including from loss of property and livelihoods, is yet to be fully calculated.

232    A scientific journal article by Alexander Filkov et al. titled Impact of Australia’s Catastrophic 2019/20 Bushfire Season on Communities and Environment. Retrospective Analysis and Current Trends provided a critical review of the health impacts from the smoke generated by the Black Summer fires. From November to January, thick smoke covered coastal cities of New South Wales, such as Sydney, which were not themselves affected by major bushfires. This culminated in 81 days of poor or hazardous air quality in Sydney during 2019, which is greater than the prior 10 years combined. The 24-hour average of PM2.5 concentrations (that is, concentrations of particulate matter 2.5 micrometres or less in diameter) in the air over most areas of Sydney during December 2019 were four times higher than the WHO Air Quality guideline value. At one point during the fires, Canberra had the world’s worst air quality. The authors observed that the increase in PM2.5 concentrations experienced were likely to induce an increase of approximately 5.6% in daily all-cause mortality, 4.5% in cardiovascular mortality and 6.1% in respiratory mortality.

233    What is made clear from this evidence is that the spread of bushfire smoke from a catastrophic fire is vast and more than capable of affecting every Australian. As noted by the authors, the smoke from the Black Summer fires spread to the whole South Island of New Zealand on 1 January 2020. By the following day, the smoke had already reached the North Island of New Zealand, affecting glaciers in the country and casting a brown tint over the snow. By 7 January 2020, the smoke had travelled approximately 11,000 kilometres across the South Pacific Ocean to Chile, Argentina, Brazil and Uruguay.

234    On the evidence before me, it is undoubtedly likely that some of the Children, and perhaps many hundreds or thousands of them, will be killed or injured by future climate change induced bushfires on the Australian continent. However, taking into account the capacity of fire smoke to spread, as demonstrated by evidence of the Black Summer fires, it is reasonably foreseeable that all Australians will be exposed to the risk of ill-health from an atmosphere polluted by smoke from one or more bushfires. That is particularly so in circumstances, which are also reasonably foreseeable, where bushfires induced by climate change will wipe off the face of the Earth most of Australia’s eastern eucalypt forests in a 4°C Future World.

235    Of the people living in Australia who are currently alive, it is the Children who are most likely to remain alive long enough to fully experience the wholesale destruction by fire of much of Australia’s forests in the latter part of this century. I am satisfied that each of the Children is exposed to a real risk of harm from bushfires. I am therefore satisfied that a reasonable person in the position of the Minister would foresee that, by reason of the effect of increased CO2 in the Earth’s atmosphere upon the increasing extent and ferocity of bushfires in Australia, each of the Children is exposed to a real risk of either death or personal injury from bushfires. The further issue of the foreseeability of the contribution to the risk of harm of the Minister’s conduct is addressed below.

5.1.3    Other ‘Direct Impacts’

236    Other climatic events relied upon by the applicants were inland and coastal flooding and cyclones. The evidence about those events was scant and, insofar as it existed, was pitched at a high level of generality which, in terms of identifying risks to health, was not sufficiently directed to the Children to enable a conclusion that each child is exposed to a real risk of injury or that some approximate number of them are at risk. The evidence mainly relied upon by the applicants was that given by Professor Capon. But that evidence barely touched upon the field in question. Insofar as it was given, it was given in the context of a global report of the IPCC which specifically acknowledged local geographical variations and in any event the only concrete example provided was sea level rises which (without specification) were said to “[threaten] population health”. Professor Steffen briefly mentioned cyclones suggesting the likelihood of fewer but more extreme tropical cyclones for Australia. There is some evidence in the report of Dr Mallon about inland and coastal flooding and cyclones but only in the context of his consideration of future damage to property. However, in the “undemanding” assessment here being undertaken it may be assumed that a reasonable person would have a general appreciation that, because of their hazardous nature, severe flooding and extreme cyclones pose a risk of harm to humans and, if those events are more frequent or more severe, that risk is increased. On that basis it may be said that some harm to some of the Children is reasonably foreseeable.

5.1.4    ‘Indirect’ and ‘Flow-on’ Impacts

237    Evidence about these two broad categories of alleged harm was given by Professor Capon. Whilst that evidence was unchallenged and I have no reason to doubt the expertise of Professor Capon or the veracity of the opinions he gave, the evidence had its limitations. The evidence given about these categories of harm was only given at a high level of generalisation with scant particular examples. The evidence does not sustain a finding that each of the Children, or some clearly identifiable sub-class thereof, is exposed to a real risk of personal injury from the events in question.

238    Professor Capon’s evidence as to harm in the “indirect impacts” category was essentially this (references removed):

27.    The second category of health impacts from climate change are indirect (also called secondary or system-mediated) health impacts. These can be further sub-categorised into (1) changes to physical systems, (2) changes to biological systems, and (3) changes to ecosystem structure and function.

28.    An example of the first sub-category of these system-mediated health impacts—changes to physical systems—is urban air pollution. In Australia, urban air pollution comes from a variety of sources including transport, energy production and manufacturing industry. However, the level of air pollution to which Australian people are exposed is also affected by weather conditions. For example, hot weather conditions can lead to higher levels of ozone formation at ground level and, therefore, increased risks to health (e.g. exacerbating childhood asthma).

29.    An example of the second sub-category of system-medicated health impacts—changes to biological systems—is the changing abundance and distribution of mosquitoes and other vectors for infectious diseases. In Australia, important endemic vector-borne viruses include Ross River and Barmah Forest. Although dengue is not currently endemic in Australia, outbreaks are associated with imported cases.

30.    An example of the third sub-category of system-mediated health impacts—changes to ecosystem structure and function—is the impact of climate change on the habitat of wild animals which can lead to new opportunities for spill-over of pathogens to domestic animals and humans.

239    As to the harms in the “flow-on impacts” category, Professor Capon said this (references removed):

31.    The final category of health impacts from climate change are flow-on health impacts. These have also been called tertiary health impacts, and are perhaps the most profound impacts. They are mediated via social, economic and demographic disruption.

32.    One important example of these flow-on impacts in Australia are the effects of prolonged drought which lead to reduced levels of soil moisture, declines in agricultural productivity, and declines in rural incomes. This affects the wellbeing of rural communities and the mental health of farmers. Psychiatrists are concerned about rising levels of depression from prolonged drought in Australia.

33.    Another important example of these flow-on health impacts is displacement from inundation in low-lying island communities with attendant risks to community wellbeing and mental health.

34.    Remote settlements are vulnerable to health impacts of climate change due to isolated location, quality of infrastructure, economic resources, limited transport and existing health vulnerabilities. Remote Indigenous communities are particularly vulnerable.

240    Having set out those categories, Professor Capon’s report then turned to the request made of him to opine as to whether any of the alleged harms particularised at [16] of the Amended Concise Statement were likely to be inflicted on the world’s children, including the Children. His broad response to that question reflected the breadth of the question. In essence, the evidence that followed amounted to no more than that some of the world’s children in some parts of the world, including Australia, may become exposed by climate change to the risk of harm, including by events induced by climate change and of the kind exemplified by the evidence extracted above, and more exposed as temperatures rise higher and higher.

241    Professor Capon gave evidence that vector-borne viruses including Ross River, Barmah Forest and Dengue Fever exist in Australia. His evidence said nothing further specifically about the Ross River or Barmah Forest viruses. His evidence suggested that a vector-borne virus like Dengue Fever will make more people ill because climate change is increasing the geographic range of mosquitos who carry such a disease. He did so by reference to a wide-ranging report on the global health consequences of climate change which itself noted that there will be “regional differences” in the extent to which individuals are put at risk. The report also noted that the effect of climatic change on infectious diseases “will further expand the geographic range of these diseases, with increases and decreases projected depending on the disease (for example malaria, dengue, West Nile virus, and Lyme disease), the region, and the degree of temperature change”. That evidence is clearly insufficient to demonstrate a real risk of increased exposure to vector-borne diseases – even Dengue Fever – for each of the Children. Although Professor Capon suggested that the geographical spread of disease will increase, neither the timing, extent or location of any spread was given and the relational exposure of that spread to the Children was not addressed. Further, the report on which Professor Capon relied noted that geographical range could also decrease, depending on the disease. The evidence was at a high level of generality and insufficiently directed to the Children to enable the conclusion of reasonable foreseeability of the particular harm contended by the applicants. I do not consider that it may be presumed that, by reason of “common sense and ordinary human experience” (King at [82] (Nettle J)), a reasonable person would have some foresight about this particular risk sufficient to cure the informational deficit presented by the evidence.

242    I come to the same view about the emergence of new human pathogens as a foreseeable risk of climate change.

243    Mental health risks were also referred to by Professor Capon. Speaking to a global report, Professor Capon suggested that diminished mental health would arise from increased human conflict brought about by displacement of human populations either due to inundation of low-lying islands or in coastal zones or, alternatively, due to declines in agricultural productivity and rural incomes because of droughts. However, that evidence was based on a global report and is so divorced from the Australian experience that I would not conclude that Professor Capon was suggesting it had application to Australia. I do not consider that the reasonable foreseeability of that harm to any of the Children is demonstrated.

244    Professor Capon also opined that mental illness, including rising levels of depression, would be caused by reduced farm productivity and income losses brought about by drought. Meyricke and Chomik (2019) referred to two studies published in 2018 and stated that climate change is likely to drive longer, harsher and more frequent droughts in parts of Australia and that the negative impacts of drought on mental health of those living in remote and regional communities is “widely evidenced”.

245    At common law mental harm may only be compensated where the harm constitutes a “recognised psychiatric illness”: Tame at [7] and [41] (Gleeson CJ) at [61]-[64] (Gaudron J) and at [201] (Gummow and Kirby JJ). The question then is whether a reasonable person in the position of the Minister should foresee that a person in the position of the Children might suffer a recognisable psychiatric injury as a result of a climate change induced drought. In relation to those of the Children whose lives and livelihoods may be severely affected by drought, reasonable foreseeability is demonstrated. However, that would only be so in relation to some of the Children and not each of them.

246    Lastly, as opined by Professor Capon, the level of air pollution is exacerbated by weather conditions, with warmer weather conditions leading to higher levels of ozone formation at ground level and, therefore, increased risks to health. These health impacts include damage to the heart, lungs and other vital organs and an exacerbation of childhood asthma. That evidence was confirmed by Meyricke and Chomik (2019), who stated that climate change is likely to exacerbate the health and mortality impacts of air pollution. In illustrating this further, the authors rely on a 2014 study which stipulates that in Sydney the influence of climate change on ozone concentrations alone is expected to cause an additional 55-65 deaths per year in 2051-2060. That evidence is geographically confined to Sydney but it may be inferred that exposure to the risk of ill-health by reason of air pollution caused by climate change would similarly be experienced in each of the major metropolitan cities of Australia. Whilst I do not come to the view that the risk is a foreseeable risk in relation to each of the Children, I would conclude that a real risk is reasonably foreseeable in relation to some and probably large numbers of the Children.

5.1.5    Conclusion on Reasonable Foreseeability of Harm

247    I have concluded that a reasonable person in the Minister’s position would foresee that, by reason of the effect of increased CO2 in the Earth’s atmosphere and the consequential increase in global average surface temperature, each of the Children is exposed, through the occurrence of heatwaves or bushfires, to the risk of death or personal injury. However, as earlier noted, the proper inquiry is narrower. What needs to be established is that the injury to the Children is a foreseeable consequence of the Minister’s approval of the Extension Project. Accordingly, I need to be satisfied that a reasonable person in the Minister’s position would foresee that a risk of injury to the Children would flow from the contribution to increased atmospheric CO2 and consequent increased global average surface temperature brought about by the combustion of the coal which the Minister’s approval would facilitate.

248    That the combustion of 33 Mt of coal from the Extension Project will contribute to an increase in atmospheric CO2 is both obvious and foreseeable. That the increase in atmospheric CO2 from the combustion of the coal sourced from the Extension Project would increase global average surface temperature may not be immediately obvious to a lay person, or even a reasonable lay person. However, the Minister must be taken to be aware of the unchallenged evidence given by Professor Steffen that CO2 emissions caused by the Extension Project would increase global average surface temperature and thus would increase the level at which that temperature is eventually stabilised.

249    The discussion at [74]-[90] above demonstrates why I consider that the emission of 100 Mt of CO2 from the Extension Project increases the risk of the Children being exposed to harm and particularly so in the realm of the risk profile which plausibly arises should the ‘tipping cascade’ be triggered and engage a 4℃ Future World trajectory.

250    Furthermore, it is relevant to note on the issue of reasonable foreseeability of harm that both the IPC Report and the NSW Department Report considered that the “Scope 3” emissions from the Extension Project (that is, the 100 Mt of CO2) would be “a significant contributor to anthropological climate change”.

251    It should be reiterated that the risk of injury flowing from the Extension Project need only be a real risk. A foreseeable risk may be a remote risk: Shirt at 46 (Mason J). As the High Court said in Chapman v Hearse at 122, reasonable foreseeability is not, in itself, a test of causation. The foreseeability of the risk of injury and the likelihood of the risk are “two different things”: Shirt at 47 (Mason J). When courts “speak of a risk of injury as being ‘foreseeable’ [they] are not making any statement as to the probability or improbability of its occurrence, save that [they] are implicitly asserting that the risk is not one that is far-fetched or fanciful”: Shirt at 47 (Mason J).

252    A risk flowing from the conduct of a defendant may be small, even “infinitesimal”, but nevertheless not be “fantastic or far-fetched” and does constitute a real risk: Shirt at 46 (Mason J). In that respect, Mason J in Shirt cited Lord Reid’s discussion about reasonable foreseeability in Overseas Tankship (UK) Ltd v The Miller Steamship Co [1967] AC 617 (The Wagon Mound (No. 2)) at 642-643. What is apparent from that discussion, and in particular the analysis made of Bolton v Stone [1951] AC 850, is that so long as the risk is real, the probability of its occurrence (even if infinitesimal) does not negate its foreseeability. Similarly, in Rosenberg v Percival, Gummow J said at [64] that “[a] risk is real and foreseeable if it is not far-fetched or fanciful, even if it is extremely unlikely to occur”.