Bushfires, Climate Change and Logic

It is undeniable that the Australian bushfires of 2019-20 were some of our worst. The terrible number of lives lost and properties destroyed exposes this reality. This morbid arithmetic is reinforced by the profound heartache, not felt by Australians since the 2011 Queensland floods and the 2009 Victorian bushfires. Journalists and politicians have relentlessly, and with unsettling zeal, declared the disaster to be “unprecedented”. However, although severe, by no objective measure were the 2019-20 bushfires without precedent. That is to say, it’s all happened before.

There is no perfect way to measure the severity of a bushfire event. Nevertheless, the most important parameter to any Australian is loss of life. In this regard, the 2009 Victorian bushfires are Australia’s worst with 173 fatalities. Tragic though our loss of thirty-four lives this year may be, there is no comparison to the harrowing toll of 2009 and many other more distant crises as far back as the nineteenth century.

This essay appears in April’s Quadrant.
Click here to subscribe

Although loss of life is the most meaningful comparator, it is not scientific. It is heavily influenced by circumstance, such as fire location and the variable success of fire-fighting and early warning efforts. Chance and external factors are also at play, as shown by the case of the C-130 water bomber that crashed on January 23, killing three US firefighters. Further still, one would expect more casualties in 2020 Australia with its 25 million people, compared to 1939 when Australia’s population was only 7 million.

In the absence of any perfect parameter, the most scientific comparator is that of total area burned. This objectively demonstrates the extent of bushfires and, compared with other parameters, is impacted far less by confounding factors resulting from the passage of time.

The 20 million hectares burned in 2019-20 cannot compare with the figure of almost 120 million hectares that burned in 1974-75. By January 26 this year, 19.4 million hectares had burned, which is far less than the 37 million to 46 million hectares shown by the satellite record to be the historical average. These figures are quoted by Dr Bjorn Lomborg, president of the Copenhagen Consensus Center. Lomborg acknowledges climate change, but is famous for countering alarmism with rational argument. His commentary in The Australian on February 1 exposes the reality that this bushfire event has ample precedent.

Why must we compare this year’s disaster against those in the past? The exercise is deeply unsettling to any compassionate Australian. Yet, increasingly over recent decades, we have been left with no choice but to engage in this most odious of contests. For there are those among us—politicians, journalists and citizens—who insist on keeping score. Every natural disaster that strikes and every hot or cold day is used to further a political and ideological climate-change agenda. In the death and devastation wrought by the bushfires, many people identified another opportunity to further their ideology.

There is actually one aspect of this bushfire event that is, in fact, unprecedented. It was that during the 2019-20 bushfires it became acceptable to use the misery and deaths of fellow Australians to advance a political agenda.

During past disasters it was acknowledged by mainstream politicians and journalists that discussion of climate change in the midst of crisis was not fitting. The reason is simple—climate change is a heated political issue and its debate does nothing to address the situation at hand. Amidst disaster, all efforts must focus on saving lives and resolving the crisis. Preaching climate change during a catastrophe is an exercise in scoring political points from tragedy. It is opportunistic and vile.

Formerly, only The Greens would preach climate change ideology as their fellow Australians died. Now, even New South Wales state Liberal parliamentarians are at it. Previously, only journalists of the radical Left would take advantage of disaster to pursue climate-change politics. In 2020 it is commonplace across the media. Meanwhile, legions of inner-city environmentalists, horrified by their first sight or smell of bushfire smoke, have taken to social media to preach their green faith.

The most astonishing feature of the tawdry politicking was the brazen hyperbole. Previously, climate change was only ever labelled a contributing factor to any given disaster. Suddenly in 2019-20, it is “the cause” of bushfires. This was most starkly demonstrated by Greens senator Jordan Steele-John who declared the government to be “arsonists”. These claims were parroted by politicians, journalists and the twitterati. Disturbingly, the well-known causes of bushfires were ignored or denied. It was shocking to witness Orwellian media commentary disputing the well-documented roles of accumulated fuel loads and arson in bushfires.

With the bushfire crisis over, it is now time for political discussion. Out of respect for the bushfire victims, the publication of this article was left until now, when Australians are no longer dying. Politicians, journalists and social media users should take note of this display of common decency.

The widespread acceptance of the outrageous claim that climate change “caused” the bushfires has the potential to inflict tremendous harm upon our nation. If we are blinded by this lunacy and fail to address the real causes of bushfire, efforts to avoid future catastrophe will be misdirected and more Australians will die. If we accept these fires as evidence of climate change and are fooled into believing that bushfires can be avoided by radical carbon dioxide reductions, then our economy and standard of living will be ravaged to no benefit.

Given what is at stake, the newly-mainstream view that the 2019-20 bushfires were “caused” by climate change must be repudiated. What follows is a treatise that employs only rigorous logic to counter the ideological claims that climate change “caused” the bushfires. We will employ rational argument to combat emotion, invective and hysteria. We will not attempt to refute the existence of climate change—that is an entirely different debate.

Necessary and sufficient causes

In probing the claim that climate change caused the 2019-20 bushfires, we must examine the nature of causation. A simple method is to consider the concept of “necessary” and “sufficient” causes, as used in medicine and epidemiology to evaluate disease causation.

A cause that is “neither necessary nor sufficient” can only be weakly implicated in the causation of an event, if at all. It cannot trigger the event alone and is not required to be present for the event to occur. On the other hand, a cause that is “both necessary and sufficient” is of critical importance in the causation of an event. Such a cause must always be present to trigger the event and can do so by itself. In between these two extremes are causes that must be present for an event to occur, but cannot trigger it alone (“necessary but not sufficient”) and causes that can independently cause an event, but are not essential to its causation (“sufficient but not necessary”).

Rural Australians grow up with the annual threat of bushfires and learn that a bushfire needs three things—oxygen, fuel and heat. This is known as the Fire Triangle. Air normally comprises about 20 per cent oxygen and is always present in sufficient quantity to sustain a bushfire. Fuel in the context of a bushfire is provided by vegetation.

The third component, heat, refers to the spark that ignites the fire. Although some fires are lit by lightning, about 85 per cent are due to humans, typically via machinery, power lines or arson. In fact, arson is responsible for up to 50 per cent of our bushfires. Recent news reports have attempted to rewrite history by claiming that arson has a negligible role in bushfire causation. These claims are intentional falsehoods, also known as lies.

When we apply the “necessary and sufficient” framework to the Fire Triangle, it is clear that each of the three components may be labelled as “necessary but not sufficient” causes of bushfire. That is, each component is necessary to start every bushfire, but none of them can start a bushfire alone. For example, every bushfire needs fuel (it is necessary), but abundant dry vegetation will not cause a bushfire without oxygen and a spark (it is not sufficient).

The “necessary and sufficient” model can be expanded to include “contributory causes” of an event. These are conditions that can have some role in the causation of an event and may be labelled as “neither necessary nor sufficient”.

In the case of recent bushfires, contributory causes would include a high ambient temperature, strong winds, prolonged drought and inadequate hazard reduction. However, none of these factors is required for a bushfire to occur (they are not necessary), nor are they able to trigger one alone (they are not sufficient). As well as being contributory to the causation of bushfires, these factors are key determinants of bushfire severity, whether by enhancing the available fuel load or via another mechanism.

It is now pertinent to assess how climate change fits into this framework of bushfire causation. This may be easily gauged by asking two questions. Is climate change required for all bushfires to occur (is it necessary)? And can climate change independently trigger a bushfire (is it sufficient)? The answer to both of these questions is obviously “No”.

Bushfires existed before the Industrial Revolution and can therefore occur without climate change. Further, bushfires cannot start without all components of the Fire Triangle. Thus, climate change is “neither necessary nor sufficient” in the causation of bushfires. This conclusion and the reasoning behind it are elementary, but are apparently necessary in this new age when politicians, journalists and tertiary-educated Australians claim that climate change “caused” the 2019-20 bushfires.

However, it is possible that these climate-change catastrophists were merely caught up in their own hysteria. Perhaps they were engaging in hyperbole and didn’t actually mean that climate change “caused” the bushfires, but that it was merely a contributory cause? If so, they didn’t say this.

The Bradford Hill criteria for causation

Nevertheless, let us proceed to address both claims—that climate change “caused” the 2019-20 bushfires and that it is a contributory cause of more severe bushfires. For both assertions, climate change is the proposed causative agent and we are left with two questions to answer. Does climate change “cause” bushfires? And how important is climate change as a contributory cause to more severe bushfires?

To answer these questions, we may borrow from the Bradford Hill criteria for causation. In 1965 the British epidemiologist Sir Austin Bradford Hill proposed nine criteria as a framework with which to systematically analyse the strength of a causal relationship between a proposed causative factor and a specific disease. Although Hill’s criteria have certain limitations, they are still taught to students of medicine and epidemiology. For our purposes, they act as a neat vehicle by which logic may be applied to the question of bushfire causation.

As we are removing Hill’s criteria from their home in epidemiology, it is not possible to sensibly address all nine criteria. However, it is instructive to examine the proposed relationship between climate change and bushfires through the prism of five Bradford Hill criteria—Temporality, Biological Gradient, Consistency, Plausibility, and Specificity. Further, Reversibility has been proposed as a logical tenth criterion to add to the original nine and serves as an effective tool with which to evaluate bushfire prevention.

Temporality and Biological Gradient

Temporality refers to the relationship in time between cause and effect, namely that the proposed cause must precede the nominated effect. To satisfy this criterion and to build a case for a causal relationship, climate change must be shown to have preceded bushfires. However, bushfires have occurred since time immemorial, whereas anthropogenic climate change began only after the Industrial Revolution. As the effect (bushfires) has preceded the proposed cause (climate change) a causal relationship is impossible.

To assess the role of climate change as a contributory cause of worse bushfires including those of 2019-20, it is helpful to address Temporality in conjunction with the criterion of Biological Gradient. This criterion proposes that a causal relationship is likely to be stronger if there is a dose-response relationship between cause and effect. That is, a greater degree of climate change should result in more severe bushfires. Notwithstanding recent hysteria, this has not occurred.

Although a hefty 20 million hectares were burned this year, almost 120 million were burned in 1974-75. Further, Bjorn Lomborg notes that the total area burned has declined by approximately one-third since 1997. He notes that on average, 11 per cent of the Australian land mass would burn in 1900, whilst only 5 per cent burns now. Not only has the alleged cause not preceded the effect (no temporal relationship), but worse fires have occurred in the past when there was a lesser degree of climate change (no Biological Gradient). These two truths expose how insignificant climate change must be in the causation of bushfires.

In examining Temporality, we also encounter the great flaw in logic that has become the hallmark of our global discourse on climate change—post hoc ergo propter hoc. This Latin phrase translates to “after this, therefore because of this” and refers to the logical fallacy of stating that as event X occurred before event Y, therefore, event X caused event Y. This invalid thought process can be applied to any sequence of events. For example: a child was diagnosed with autism after receiving a vaccine and, therefore, vaccination causes autism.

Climate-change alarmists regularly dabble in post hoc ergo propter hoc “reasoning”. If any given day is hotter than average, it is attributed to climate change because it occurred after climate change. Never mind that it has been hotter before. Similarly, the 2019-20 bushfires occurred after climate change and therefore they must be the result of it. Never mind that bushfires preceded the Industrial Revolution and were worse in the past. The use of post hoc ergo propter hoc is immediately apparent to anyone with an understanding of causation. It is an indictment of the cognitive abilities of climate-change alarmists or their integrity that they are either collectively oblivious to their novice error or wilfully engaging in deception.

Consistency

Consistency refers to the notion that a causal relationship is more likely if it is observed often, that is, it is reproducible. So, if a causal relationship were to exist, climate change should result in the regular and increasingly frequent occurrence of severe bushfire events. However, most fire events since 2008-09 and most in the decades prior have been quite routine in the historical context of Australian bushfires. So the claim that climate change is causing more frequent and severe bushfires is not at all reproducible.

It is a fair rebuttal to argue that the expectation for each bushfire event to be worse than the previous is an excessively high burden. However the criterion of Consistency does not demand this, but merely requires that there be some observable trend over time if climate change is to be considered a material factor in the causation of bushfires. A once-in-ten-years worse-than-average event represents no such trend.

The only remaining rebuttal to the absence of reproducibility is to speculate that such consistency will emerge in the future when the global temperature has increased further. However, intrinsic to this argument is the acceptance that the role of climate change in bushfire causation has thus far been too minor for an observable pattern to emerge.

Underlying Consistency is the principle that no single experiment is ever sufficient to prove causation. No scientific study is methodologically perfect and there is always some residual possibility that chance may be responsible for the observed outcome. To mitigate this risk and to build a case for causation, studies are repeated and their outcomes assessed for reproducibility.

Their disregard for this fundamental scientific principle allows climate-change alarmists to proclaim that the 2019-20 bushfires are stand-alone evidence of climate change. To satisfy the criterion of Consistency there must be a series of similar events that demonstrate reproducible outcomes. Given that the last severe fire event was eleven years ago, climate-change alarmists are again relying on an isolated event to prove their hypothesis.

Plausibility

Plausibility suggests that a causal relationship is more likely if there is an identifiable biological mechanism by which the proposed cause exerts its effect. The mechanisms for the components of the Fire Triangle require no explanation. Similarly, the means by which drought, poor land management and arson contribute to bushfires are well-understood, despite Orwellian attempts by politicians and the media to revise our hard-won understanding of bushfire science.

So, is there a biological mechanism by which climate change contributes to bushfires or exacerbates them? Most of the politicians and commentators who discredited themselves by preaching their ideology amidst tragedy never explained how climate change actually caused the 2019-20 bushfires. Those few who did apply some modicum of thought proposed various mechanisms, namely “a longer bushfire season”, “longer and more severe drought” and “hotter temperatures”.

It is impossible to utilise the length of the bushfire season to measure the severity of bushfire events. “Length of bushfire season” is not an officially recorded statistic and there is no means by which a bushfire season’s start and end may be absolutely and consistently defined, particularly given that bushfires often occur in winter. Rather, “bushfire season” is a qualitative neologism for the period of the year in which bushfires and the weather that fosters them are most severe. It is fruitless and grossly unscientific to attempt to discern any change in a parameter whose limits are undefined and for which no historical record exists.

The suggestion that climate change causes bushfires through drought is also likely a falsehood. Professor Andrew Pitman, the Director of the Centre of Excellence for Climate Extremes, neatly exposed this myth in June 2019:

… as far as the climate scientists know there is no link between climate change and drought … there is no reason a priori why climate change should make the landscape more arid … If you look at the Bureau of Meteorology data over the whole of the last one hundred years there’s no trend in data. There’s no drying trend. There’s been a trend in the last twenty years, but there’s been no trend in the last hundred years, and that’s an expression of how variable Australian rainfall climate is.

In fairness to Professor Pitman, it must be noted that he later partially recanted his statement. Although, one must consider if his half-hearted reversal was akin to an infidel recanting when faced with the headsman’s axe, given the recent academic assassination of Dr Peter Ridd for daring to dissent from popular climate-change catastrophism.

The absence of evidence does not disprove a hypothesis, as the evidence may just not yet have been found. Thus, the data underpinning Professor Pitman’s statement does not rule out climate change as a cause of worse drought, but it does highlight the current absence of any demonstrable relationship. It is thus supremely dishonest to declare drought as a mechanism by which climate change exacerbates bushfires.

However, if we are for a moment to assume that climate change does contribute to bushfires via worse drought, the weakness of this relationship is still readily exposed. Initially, it is evident that drought is “neither necessary nor sufficient” in the causation of bushfires and is, itself, only a contributory cause. Further, we know that worse droughts, such as the Federation Drought, occurred before climate change. Thus the Bradford Hill criteria of Temporality, Biological Gradient and Consistency don’t apply to the hypothetical relationship between climate change and drought. As such, climate change is proposed to cause or exacerbate bushfires through its status as a minor contributing factor to drought, which is itself neither necessary nor sufficient in the causation of bushfires. Hardly convincing.

The final mechanism used to lend Plausibility to a causal relationship between climate change and bushfires is the proposal that climate change increases the ambient temperature to produce worse fire conditions. This has merit. Climate change is reasonably claimed to make hot days hotter and to increase the number of hot days. It is self-evident that this would exacerbate bushfire conditions.

While this mechanism is conceptually plausible, a quantitative examination reveals otherwise. Authorities indicate that Australia has experienced a maximum of 1.5 degrees of warming to date—the difference between the current post-climate-change 26.7 degrees January average maximum temperature measured at Sydney Airport and 25.2 degrees, which must have been the January average maximum before 1.5 degrees of climate change. Although this is a simplistic summary, it does serve to highlight that climate change is unlikely to make a material impact on bushfire conditions—the 1.5 degrees difference in January average maximum.

A more authoritative argument is provided by Bjorn Lomborg, who noted in the Australian that climate change will increase Australia’s average annual area burned from 5.3 per cent to 6.0 per cent by 2100. So, while Dr Lomborg agrees that climate change will increase the magnitude of bushfires, it will only increase total area burned by 0.7 per cent of Australia’s land mass over eighty years, or less than 0.009 per cent annually.

In medical science, this is known as the difference between “statistical significance” and “clinical significance”. A drug trial may show that Drug X increases the average Australian female lifespan by one week (statistically significant), however this increase is materially meaningless (clinically insignificant), given that the average Australian female lives for 82.5 years or 4290 weeks. A difference that is detected on paper may not translate into any meaningful change.

Similarly, an annual increase in area burned of less than 0.009 per cent of Australia’s land mass would be materially insignificant. Although Dr Lomborg labels this increase as “not trivial”, he notes that 120 years ago an average of 11 per cent of Australia’s land mass burned annually, compared with only 5.3 per cent today. If climate change increases this figure to 6 per cent in 2100, it will still only be about half of what it had been 200 years previously, when a much poorer and technologically-primitive Australia thrived.

Even if one is not convinced by this refutation of climate change’s Plausibility as a cause of bushfires, it is important to note that the existence of Plausibility does not prove causation, just as its absence doesn’t disprove it. In this case, the proposed causal relationship also fails the tests of Temporality, Biological Gradient, Reversibility and Consistency and is therefore likely to be extremely weak even if one believes it to have a plausible mechanism.

Specificity

Bradford Hill’s criterion of Specificity states that a causal relationship is likely to be stronger if the proposed cause is one of only a few for the nominated effect. That is, the association between climate change and bushfires is more compelling if climate change is one of only a short list of causes. This is obviously not the case. It can also be inferred that, if there are numerous causes, then the overall importance of any one cause is diminished. Thus, even if climate change is considered to be causative, the existence of many other contributing factors degrades its individual significance.

Another way of considering Specificity is to ask: Does the designated effect have any other possible explanation? If other explanations exist, then the proposed causal relationship is not at all specific and thereby weakened. The proposed relationship may also be entirely false. So, are there other explanations for bushfires, including the 2019-20 bushfires, besides climate change? Again, it is clear that there are other well-established causes that render climate change decidedly impotent as a causative factor.

However, if we accept for a moment that climate change is a cause of bushfires, it is important to determine where it resides in the causal hierarchy. We have already noted that the components of the Fire Triangle are paramount. These may be labelled First-Order Causes, as they lead directly to bushfire. In the case of heat, the specific phenomena that ignite a fire may also be considered First-Order Causes. Farm machinery, motor vehicles, power lines, lightning and arson are key First-Order Causes in addition to oxygen and fuel (dry vegetation).

Second-Order Causes are those that enhance First-Order Causes, including drought and even criminal justice and mental health systems that fail to control arson. Other Second-Order Causes that directly exacerbate bushfires include hot temperatures, low humidity and strong winds, which are quintessential features of the Australian summer. Second-Order Causes are important, but far less so than First-Order Causes.

Third-Order Causes are another step removed from the nominated effect and less important still in its causation. El Niño and the Indian Ocean Dipole are Third-Order Causes of bushfires, as they exacerbate Second-Order Causes. Evidently, climate change is a Third-Order Cause, as its proposed mechanisms merely exacerbate Second-Order Causes, such as high temperatures. Thus it can be of only minimal importance compared with First- and Second-Order Causes. Moreover, the presence of El Niño and an Indian Ocean Dipole during the 2019-20 bushfires further diminishes the individual role of climate change.

Reversibility

It is neat to conclude with Reversibility, which has been proposed as a tenth Bradford Hill criterion. Reversibility provides that the elimination of a proposed cause should also erase the nominated effect. Given that bushfires existed before anthropogenic climate change and that worse fires occurred in the past, reversing climate change will do little to prevent bushfires or reduce their severity.

However, Reversibility has additional signif­icance as it relates directly to bushfire prevention. Indeed, bushfire prevention is achieved through the total or partial reversal of causative factors. Such is the paramount importance of the First-Order Causes that comprise the Fire Triangle that their reversal would directly prevent bushfires. This exposes the reality that effective prevention strategies should target fuel loads and fire ignition. Given that 85 per cent of fires are ignited by humans, there is immense scope for the reversal of these First-Order Causes.

On the other hand, targeting climate change will have negligible impact, given its status as a Third-Order Cause at best. Moreover, rapid climate change mitigation would be prohibitively expensive and will have no measurable effect in the foreseeable future, if at all. Certainly, addressing climate change as the bushfires raged and Australians died would have done nothing to resolve the disaster, which is the most damning indictment of those who publicly obsessed over this ideological issue during the crisis.

Another way to realise the ineffectiveness of addressing climate change for bushfire prevention is to consider the concept of “modifiable” and “non-modifiable” risk factors. This dichotomy is used by doctors in disease prevention. Modifiable risk factors can be ameliorated with some type of intervention, whereas non-modifiable risk factors cannot. Obesity, high cholesterol and old age are all risk factors for heart disease. Obesity and cholesterol can be improved with medication or lifestyle changes and are, therefore, modifiable risk factors. Old age is a non-modifiable risk factor as it cannot be reversed.

With regard to bushfires: fire ignition by humans and accumulated dry fuel loads can be mitigated and are, thus, modifiable risk factors. Conversely, humans cannot alleviate hot temperatures and low humidity which are, therefore, non-modifiable. It is common sense that only modifiable risk factors should be targeted in bushfire prevention, just as they are in disease prevention.

Even if climate change does cause or exacerbate bushfires, Australia contributes only 1.3 per cent of global emissions and any action we take would, therefore, be wholly futile. Climate change is a non-modifiable risk factor. Just as a doctor would be utterly insane to attempt reversing a patient’s age for heart disease prevention, so too would Australia be insane were it to attempt to reverse climate change to prevent bushfires.

The logical conclusion

The use of logic to address the causation of the 2019-20 bushfires has scarcely been attempted. The national discourse has been dominated by hysteria, Orwellian attempts to rewrite history and arguments predicated on post hoc ergo propter hoc “reasoning”. In 2019-20 it became acceptable for the first time to use national tragedy to advance a political agenda.

The claim that climate change “caused” the 2019-20 bushfires threatens to undermine bushfire prevention by misidentifying the causes of the crisis. If this fallacy triggers reckless efforts to mitigate climate change, our nation will suffer. For these reasons, as well as to counter the abhorrent use of death and devastation to advance an ideology, it has been necessary to rigorously discredit the claim that climate change causes bushfires. This logical process revealed nothing profound but merely reiterated what was once common sense that has somehow been unlearned by Australians.

Ultimately, climate change is, at best, a minor Third-Order Cause of bushfires that is neither necessary nor sufficient, has no Consistency or Specificity, has weak Plausibility and fails the tests of Temporality, Biological Gradient and Reversibility. Efforts to reverse climate change would have a negligible effect on bushfire prevention, as it is at most a weak Third-Order Cause and a non-modifiable risk factor. All interventions should target First-Order Causes and modifiable risk factors only. This conclusion is so obvious that it exposes the unprecedented climate change hysteria of the 2019-20 bushfire crisis to be nothing more than an opportunistic and callous attempt to advance an ideological agenda.

Jarrod Brady is a doctor and medical scientist.