Tag Archives: Risk

Comments on the Latest IPCC Report and Communicating Risk

Comments by Barrie Pittock, former IPCC Lead Author, climate scientist and author:


The critical first volume of the fifth report of the Intergovernmental Panel on Climate Change is now published and the subject of wide discussion. It is the result of years of effort by many of the world’s leading climate scientists, and was subjected to approval by representatives of governments with widely differing views on what to do about climate change. These included Australia, the United States, Russia, India, China, Saudi Arabia and dozens of others.

Notable is that the new findings make it even clearer that the world is warming at an unprecedented rate and that this is almost certainly due to human pollution of the atmosphere. The report also concludes that the warming climate is increasing the severity and frequency of many extreme weather events, is changing rainfall patterns, causing rising sea levels, acidifying the oceans, and creating increasing risk for human well-being, the environment and the economy. Modern human society has developed to cope with the risks posed by normal climatic variations of the last century or so, including past variability and extremes. Adaptations include water storages, levee banks, zoning and design rules to cope with floods, coastal storm surges and other environmental hazards.

These and many other aspects of modern human society are now threatened by changes that could have serious human and economic consequences. Risks from normal climate variability include that to coastal development of ports, power stations, cities and holiday resorts, and modern agriculture, dependent on the normal temperature range and seasonal rainfall distributions. Increased risks will impose enormous extra costs on society if people and investments are to continue to be protected.

The climate system is very complex, with many cross-connections, including amplifying effects such as warming that reduces snow cover which in turn allows more solar heating of the surface and thus more warming and even less snow cover. A similar amplification is involved with loss of Arctic sea ice, and probably with ocean acidification as that threatens life in the oceans which at present absorbs a lot of carbon dioxide from the atmosphere.

So clearly climate change imposes extra risks. This has already been documented and acted upon by insurance companies, particularly the big international re-insurance companies such as Munich Reinsurance and Swiss Re, which insure other insurance companies.

A central task of the IPCC is to identify and quantify (as far as possible) the risks from climate change. Risk, as recognised in insurance and in engineering design, is in general the product of the probability of adverse impacts times their magnitude. For example, we generally insure a house against burning down even though the probability of this happening to our particular house is likely very small, because if it burnt down it would be very serious financially. The probability of it burning down may be less than 1 in 100 in any given year, yet we pay for insurance, or take other precautions like installing fire alarms and sprinkler systems, just in case.

Similarly, in the case of engineering design, for a road culvert we generally design it to take care of a flood that might occur only once in ten years. But for a bridge the cost of failure would be much larger so we design it to withstand a flood that might occur only once in 100 or more years, while for a major dam we would be extra cautious and design it to withstand a flood that might occur only once in 1000 or more years.

So even hazards that have quite low probability must be designed for or prevented if they are of great magnitude. So it should be with climate change hazards.

Take coastal zoning and design standards as an example. If we consider the risk to a new development over its anticipated life-time, for example of a house or tourist development, where the lifetime might be up to 100 years, we should design it to cope with a sea-level rise or storm surge that just might occur over that time.

This leads to a problem with the way the latest IPCC report fails to communicate the real problem of risk. The IPCC characterisation of uncertainty is stated as being from “virtually certain” (99-100%), through “about as likely as not” (33-66%), to “unlikely” (0-33%), to “exceptionally unlikely” (0-1%). I suggest that that fails to communicate the serious nature of the risk of extreme changes since an outcome labelled as “unlikely” is likely to be ignored by lay people when really a 1 in 3 chance of a disaster is really a serious risk. Uncertainty requires a more explicit risk assessment approach, so the less certain possibilities should be described as “quite possible” (0-33%) or “possible” (0-10%) and perhaps “just possible” (0-1%), or some similar description that allows for the risk of a calamity like a 5 meter sea-level rise.

Would you board an aeroplane if its chance of crashing was stated as 1 in 3, or even 1 in 10? Of course not!

A case in point is the uncertainty regarding possible rapid disintegration of the massive ice sheets covering Greenland and Antarctica. Decades back glaciologists thought this would only come about by surface warming and the slow conduction of heat to the bottom of the ice sheet, taking hundreds to thousands of years. Then the idea appeared that surface melt-water might flow through crevasses in the ice, lubricating the bottom of the outlet glaciers, which would thus flow faster due to less resistance. More recently it was thought that this might not be such a great issue, since melt-water may scour out only narrow channels, rather than lubricating the whole bottom layer.

Now that understanding has changed once more: the melt-water penetrating deep into the ice sheet will warm the ice, causing it to be softer, like butter, and thus flow faster. This could have a much wider effect and lead to great increases in flow of outlet glaciers. Large rises in sea level are thus possible, even though they are as yet poorly quantified. Indeed, there is a real risk of multi-meter sea-level rises within a century or so.

This raises two problems for IPCC: firstly that of increasing uncertainty about the rapidity of sea-level rise, and secondly how the risk implied can be communicated. Given large uncertainties about possible disastrous impacts, IPCC needs to communicate what this means in terms of risk. The risk is obviously great for large impacts even if they have only a small but appreciable probability of occurring. So climate-driven changes that have only a small but non-negligible chance of occurring do need to be taken seriously. So far IPCC has not clearly communicated this.

Time scales for damaging climate changes are also vitally important, as is their possible damaging effect on investment, either from the destruction of infrastructure or the need for costly adaptations such as massive sea walls or retreat from the coast. Insurance companies tend to understand this risk, but many other companies seem less clear.

I am reminded for instance of ski resort owners who in the 1990s were affronted when my research group identified future loss of snow cover in future decades. The resort companies accused us of wanting to shut down the ski industry, and said that if they invested in a new ski run they would make a profit in a few years, and if they invested in a new ski resort they could profit within a decade. So they were not concerned about global warming that might take several decades to damage their investments. Later, however, new resort owners actually asked us for improved estimates of loss of snow cover so that they could plan investments in artificial snow making machines.

Today in Australia new investments are being made or planned in coal mines, railways to carry the coal and new ports for exporting the coal. These new mines will take a decade or more to come on-line, by which time many of the coal consuming countries such as China may well have decided to do away with highly polluting coal-fired power stations for two reasons. First is local air pollution which poses increasing health risks, and secondly their countries will be increasingly threatened by rising sea levels and changing climates. China and India are already taking measures to deal with these problems

These potential coal consumers will thus be moving increasingly towards renewable or nuclear energy and energy efficiency. If they do that the investors in new coal mines and export facilities will have stranded assets. Thus potential investors must look ahead, not to make profits in a year or two but to anticipate longer term threats to their investments. They would do much better in the long run by investing in renewable energy technology.

One interesting irony in Australia has been the recent serious flooding of many open cut coal and indeed uranium mines in this country due to increasing heavy rain events. Such an increase was predicted by climate scientists in the early 1990s, but investors took no notice. Now they are having to build bigger tailings dams (in the case of uranium mines) and install pumps, levee banks and better drainage facilities.

It is thus time that businesses and investors took enhanced climate change seriously and planned ahead by investing more appropriately in a changing environment. And it is up to climate change scientists and environmental advocates to get businesses on side. This is serious. Scientists, environmentalists and investors need to work together. That is the second major communication problem. First we must better communicate the risk of extreme effects and secondly we must help investors to understand what this means for them.