SCIENCE AT THE SHINE DOME canberra 7 - 9 may 2008

Symposium: Dangerous Climate Change: Is it inevitable?

Friday, 9 May 2008

Professor Neville Nicholls
School of Geography and Environmental Science, Monash University

Neville Nicholls spent 35 years at the Bureau of Meteorology developing systems to forecast climate variations and their impacts, and researching our changing climate. In 2006 he moved to Monash University where he is an Australian Research Council Professorial Fellow. He was a lead author for the Intergovernmental Panel on Climate Change’s Fourth Assessment of Climate Change published last year. In 2005 Neville was awarded the Fitzroy Prize of the UK Royal Meteorological Society, for ‘distinguished work in applied meteorology’. He is vice-president of the Australian Meteorological and Oceanographic Society.

Water/drying and climate change

(Click on image for a larger version)

Almost exactly 250 years ago, on 24 June 1758, Samuel Johnson wrote the following in The Idler, 'It is commonly observed, that when two Englishmen meet, their first talk is of the weather. They are in haste to tell each other, what each must already know, that it is hot or cold, bright or cloudy, windy or calm.'

When Australians first meet don't talk about the weather, they talk about the 'drought', and that is something that has been with us at least as long as European settlement. In March 1791, for instance, Governor Phillip wrote back to the Colonial Secretary in London and said, 'From June until the present time so little rain has fallen that most of the runs of water … have been dried up for several months … I do not think it probable that so dry a season often occurs.' Well, Phillip was wrong.

It took us a while to burn the image of drought into the Australian culture and mind. But by the end of the 19th century this had really been well established. Henry Lawson, in 1888, was sent out to the backblocks of New South Wales to look at the areas affected by severe drought, and that left him scarred for life. When he came back he published a new poem called Beaten Back, describing the reaction of the people in the outback to the drought. It starts, 'Beaten back in sad dejection, After years of weary toil, On that burning hot selection, Where the drought has gorged his spoil.'

Later, Lawson was involved in a debate with Banjo Paterson that was conducted entirely in verse, in the pages of The Bulletin. Lawson was pointing out that the Australian climate is tough, and drought is important. Banjo Paterson was the denier. They had this fascinating debate about whether Australia had a fairly benign climate or whether drought was a really critical aspect of our climate and environment. I think Lawson won then, and I think everyone in this room now would also think that drought is important.

Droughts continue, and they always have had a lot of impact. So there was the Federation drought of 1901–02. In that year there were 140 articles published in The Sydney Morning Herald about drought; that is many more than you get nowadays. Nellie Melba was making a triumphal tour of Australia at the time, travelling round by rail, and she was horrified. She said, 'While travelling through Australia by rail, I have seen heartrending proofs of the misery caused by the drought. I have seen with my own eyes the brown, burnt paddocks extending for hundreds of miles, with no vestige of grass left upon them. I have seen starving sheep leaning against the fences too weak to move … I have seen the skeletons of cattle and sheep dotting the paddocks, and the signs of desolation and starvation everywhere. It is simply appalling.'

If you want appalling, just go and look at the pictures of Nolan and Drysdale, who were also sent out to outback Victoria and New South Wales during the droughts of the 1940s. They are also appalling.

So we have had this deep cultural impact of drought on Australia. But there are always deniers. In 1921 Mr Gullett, the Commonwealth Superintendent of Immigration, said, 'Many thousands of Australians go abroad every year on business or pleasure. The Commonwealth Immigration Office appeals to every one of them to embark with the resolve that he will on all possible occasions speak well of Australia. Let none of them speak evil. Such words as 'drought' … should be thrown overboard as the vessels put out to sea.'

But scientists were pointing out the reality of drought, and that it was a natural part of the Australian environment. The young Francis Ratcliffe, the great ecologist who was imported into Australia in the 1930s, said in 1938, 'One of the most extraordinary and at the same time most discouraging aspects of the whole matter is the reluctance, amounting almost to stubborn refusal, on the part of the Australian people to recognise the inevitability of drought. The tacit assumption that drought is an exceptional visitor to the inland country has shaped and infected public thought and official policy alike.' He pointed out, 'We just have to be used to a crushing drought about every seven or ten years.'

So it is built into our culture, it is built into our understanding of Australia, and right from early on Australian scientists were trying to understand these droughts as well, to understand what caused them.

Charles Todd, who built the overland telegraph in 1888, in the drought that scarred Henry Lawson, pointed out that our droughts tend to coincide with those of India. That is the first sign that anyone had published, as far as I am aware, of a 'teleconnection' between drought that we now know as the El Niño–Southern Oscillation, or ENSO. That was in 1888. ENSO is now well enshrined in our popular culture also, and my final quote (before I actually get to the talk) is from a poem by Les Murray, Variations on a Measure of Burns – much more recent. He wrote, 'Because drought, who is in on every forced sale, Though he may have seen the farmers granted bail, This summer has the continent in his entail, Even smashed he has seen you, That old man up a back road fumbling his mail, Gets letters from El Niño.'

(Click on image for a larger version)

So we know a lot about drought. We know it does impact Australia. I am going to talk about three things: very briefly, what causes Australian droughts; what is predicted for Australian drought; and is this dangerous climate change?

You might think, after that introduction, why would I bother talking about whether droughts are dangerous, or changes in droughts are dangerous? But all through that 200-year history I have just given you, agriculture provided 70 to 90 per cent of Australia's exports. In my childhood, in the 1950s, it was 70 per cent. Now it is 20 per cent. So is it really important anyway?

But first: what causes Australian droughts?

(Click on image for a larger version)

If life was simple and climate models could do everything we wanted them to do, we could look for what you might call the ultimate cause of droughts – natural variations or anthropogenic climate change – to explain droughts and predict them. And we might use models that could do this very perfectly to predict what might happen in the future, in the next 100 years or so.

But we can't. And I will get to why we can't, later.

So what we need is to put together a structure of research and findings that link all of the causes of drought. Really, there are only two proximate causes of drought. There is less rain, or more evaporation. But each of those can be caused by a whole variety of intermediate causes.

The first step for less rain is that you get either fewer rain days or less rain per rain day. They can be caused by changes in storm tracks or moisture advection or changes in convergence. That in turn can be caused by changes in the great modes of circulation, like ENSO and a few others.

In turn, ENSO can possibly be affected by either long-term natural variability or anthropogenic climate change, but so can those also affect, more directly, these other parts of the question: shifting storm tracks and fewer rain days.

On the other hand, more evaporation really can only be caused by a few things, like stronger wind, less humidity and more solar radiation. They in turn can be changed by a variety of intermediate causes, and in turn those can be affected by ENSO and the other modes, and in turn again by anthropogenic climate change.

So this is a whole picture of how various steps in these causal chains can lead to what we understand about the causes of drought. And we need to know these if we are going to predict how drought will change. That is a lot more complex than saying, following the experiments of John Tyndall in the mid-19th century, that if you increase the carbon dioxide in the atmosphere you get warming. That pretty simple, basic 19th century physics hasn't been changed for 150 years. Something as complex as this, though, requires a lot of work. And we haven't got this complete linkage worked out; we haven't got all those arrows worked out; we don't know the relative importance of them.

I want to just show two diagrams which talk about partial answers to partial parts of this diagram.

(Click on image for a larger version)

The first diagram is from Wenju Cai and Tim Cowan, at CSIRO Marine and Atmospheric Research. This looks at trying to predict the annual Murray-Darling Basin inflow, which is on the vertical axis, relating it to rainfall and temperature in different seasons. The top one is summer, then autumn, winter and spring.

On the left is shown the relationship between rainfall anomalies and inflow in the Murray-Darling Basin, and particularly in winter and spring there is a strong relationship – not surprisingly. You get heavy rains in winter and spring; you get a lot of inflow in the Murray-Darling catchment.

On the right are shown attempts to look at the effect of temperature, separate from rainfall. Temperature and rainfall are related, so you have to look at the residual relationship between temperature and runoff, inflow, after you have removed the effects of rainfall. In winter and spring there is a negative relationship. The warmer years lead to lower inflow, even once you have removed the effect of rainfall.

So you have got at least these two effects, these proximate causes of changes in inflow: one related to rainfall, one related to temperature. So that is filling in a little bit of the diagram I showed you before.

(Click on image for a larger version)

Another little bit of the diagram is something that is reflected in Les Murray's poem and that we have known about for quite a long time, that changes in El Niño affect rainfall in Australia.

The green line here is the annual rainfall averaged over that part of Australia south of 30°S – coastal southern Australia – over the last 56 years or so. That is the green line goes up and down.

The blue line is an indicator of ENSO, a sea surface temperature anomaly in the central equatorial Pacific, and there is a scale on the right-hand side reversed. So El Niños are the lowest points. You can see that that bounces around a lot too.

The relationships between those two from year to year are remarkably good. Rainfall over southern Australia is pretty closely related from year to year with El Niño. It is not perfect, but it is good for something in climate!

The thick lines are just linear trends of the two. Rainfall has declined about 10 per cent over the last 50 years, and that actually reflects, in some sense, a change in El Niño. We have moved towards something which you might call more El Niño-like conditions, and that can account for the 10 per cent rainfall decline.

So changes in El Niño are really important to Australian rainfall and drought. If we want to predict how Australian droughts will change, we do need to know how El Niño will change.

(Click on image for a larger version)

So, to the future: this is from the IPCC [Intergovernmental Panel on Climate Change] Fourth Assessment. Sometimes what the IPCC has said about droughts gets misinterpreted. In the red box here I have just highlighted the summary of all summaries of what was said about droughts.

At the left, the phenomenon we are talking about is that the area affected by droughts increases. The IPCC said, 'But since the 1970s we think it is likely that that has happened in many regions.' Now, 'likely' has very specific definition in IPCC language. It means, 'We are more two-thirds confident.' So there is a more than 66 per cent chance that, in many regions, since the 1970s drought has increased – not all that strong.

The next one is even less strong. The next column looks at what is the likelihood that there is a human contribution to that observed trend, and we say, 'More likely than not'. 'More likely' is a coin toss. That is just slightly better than 50 per cent that there has been a human contribution to observed trends. So there is not a great deal of confidence in saying this. This is not like ocean acidification or sea-level rise or destruction of the Arctic sea ice, or the fact that we are getting more heatwaves and that is leading to increased mortality in cities. Those things are simple, compared with what we can say about droughts.

Even with regard to the future, in the final column at the right, again the word 'likely' is used. Compare that with the phenomena at the top of the table, about more frequent hot days and nights, where we are virtually certain that this is going to happen. We only think it is 'likely' that the area affected by droughts increases.

(Click on image for a larger version)

What is the basis for that assessment? This figure was published early this year, well after IPCC, but it says things which are quite similar to what was used in the IPCC report.

This is projected runoff by the middle of this century, under one of the IPCC scenarios. Where it is orange-y or yellowy, it means there has been a projected decrease in runoff, and the blue-ish, grey-ish ones are where there is a projected increase in runoff.

The places with colour are only those states where there is consistency between models. I want you to look at Australia. Australia is nearly all white. That means that over most of Australia there is no consistency between the model projections of drought. The one exception here is Tasmania and Victoria. Because of the way this group did their work, because they just looked at states, they did all of Western Australia together. If they had just done the south-west of Western Australia, that would have shown up as orange as well. So in the southern part the models do project declining rainfall, but for the rest of Australia there is no consistency between them.

So again it is very different from many other impacts of climate change. I will get back to it in a second, but I just want to address the question: are changes in drought 'dangerous climate change'? – despite the fact that it is a declining impact on our economy.

(Click on image for a larger version)

The green line here is New South Wales annual rainfall, starting in about 1950, and the orange line is the wheat yield per hectare in New South Wales. You can see that the wheat yield has increased quite dramatically over that 50 years, with improved cultivars, improved management techniques. You can also see that the interannual variations are closely related to rainfall. So a drought still causes a big drop in Australian agricultural production, and anyone who has been following the global food crisis at the moment will recognise that people attribute part of that global crisis this year to the drought in Australia reducing wheat production. So, in terms of food production, yes, drought is still dangerous.

(Click on image for a larger version)

What about forest fires? Forests burn when they are dry. This chart is for Tasmania. The vertical axis is the area burnt by forest fires – wildfires – in Tasmania, versus the summer rainfall. On the right-hand side are the years with almost no fires. They are the wet years. If you get dry years, you can get massive areas of Tasmania burning (and Victoria, by the way). So yes, droughts cause fires.

(Click on image for a larger version)

Droughts also cause suicide. This chart shows, on the vertical axis, the change in the suicide rate in New South Wales over the last 40 or so years, plotted against the change in rainfall from year to year. It is not a strong relationship but it is significant. You get a drought and you do get about an 8 to 10 per cent increase in deaths from suicide in New South Wales.

(Click on image for a larger version)

Finally, what happens to water supplies in cities? This is data for Perth. The top diagram is the last 100 or so years of rainfall in south-west Western Australia. It has declined about 15 per cent. The bottom, right-hand graph is the total annual inflow into Perth's dams – so that 15 per cent decline in rainfall has been modified into a 40 to 50 per cent decline in inflow into Perth's dams. It is just a feature of hydrological systems that a small rainfall change can be amplified in terms of inflow.

So to city supplies also, droughts are dangerous.

(Click on image for a larger version)

El Niño causes Australian droughts. El Niño and droughts have been with us for as long as anyone can remember. They are inevitable. One of the keys to predicting future drought is clearly El Niño. No matter what climate change does in general, if El Niño changes then that will change the structure of droughts.

Droughts are still dangerous, but multi-decadal prediction of them is difficult. So what do we do?

As I have said twice, it is different from many other impacts or threats of climate change. The threat of changes to drought I think is really dangerous simply because we don't exactly know what will happen. In much of Australia we don't know if the droughts will get worse, we don't know how they will change, but we do know that it is reasonable to expect such a change.

I have studied climate science for 37 years. I cannot believe that warming the world by 2° to 3° will not lead to major, major shifts in storm tracks and changes in droughts in many parts of the world, including parts of Australia; we just can't predict what they are. So what do we do?

I think we do four things – and these are somewhat different from the way you might approach the problem of coral bleaching or ocean acidification.

First, you monitor the climate, because unless we can actually see how the regional climate of Australia is changing, we don't know what to do about it.

We try to improve our scientific understanding of the causes of those variations in drought and other climate aspects, and we improve the models to predict them. And at the end of that, you can adapt.

But those last three – understand, predict and adapt – are dependent on the first one, monitoring. The Bureau of Meteorology does that monitoring. It monitors climate and weather, and has done since its establishment about 100 years ago.

Over my career the Bureau has been reduced in size by about 45 per cent by successive governments, by 'efficiency' dividends. That has seriously compromised the Bureau's ability to monitor how Australia's regional climates are varying, and to understand them. I hope to see a government that will take climate change seriously enough that it will establish a climate monitoring system able to help us do the understanding, predicting and adapting we need to do, because otherwise we will really find out if droughts and climate change are dangerous.

Discussion

Question (Philip Kuchel): It seems almost inappropriate of me to say this – it is like a physician-scientist talking to a dying patient and saying, 'You've got a wonderfully interesting disease.' But, viewed in terms of the modelling that you talked about, it is salutary to think that if you took the whole human biomass alive today, it would fit into a cube less than a kilometre on the side, and if smeared over the surface of the Earth it would be less than a micron thick. So that's less than a hundredth of the thickness of a human hair. Isn't it wonderful, or amazing, that this could be perturbing such a massive system? (Just a comment!)

Neville Nicholls: The system isn't all that massive. The atmosphere is a tiny, skinny skin on a very large orange. We don't have to do too much to affect it.

Question: In one diagram you mentioned the impact of vegetation type on drought. We Europeans have been here 200 years, with a massive change of vegetation and introduced cropping systems and grazing systems. What is the effect of that? Also, what about the interaction of biodiversity which was there hitherto and rainfall?

Neville Nicholls: It is an area of ongoing scientific research, to try to work out how much land use change in Australia, particularly, has either exaggerated climate change or the impact of climate change on things like drought, or whether it has offset any of that. I think the jury is still out. I think the consensus would be at the moment – and there are a few people here who might correct me – that we have demonstrated that there are local effects but we are still waiting for clear description that larger-scale regional effects are produced by these changes in vegetation, and changes in biodiversity.

Question (Brian Kay): Do you remember the work you did on Murray Valley encephalitis prediction over 20 years ago, when we recognised that the Murray Valley water flows, dams and everything had changed? In relation to that very useful predictive work that you did all those years ago, I take it that what we are saying now in relation to Murray Valley encephalitis exacerbation is that that disease is likely to get less. Would you like to comment on how you see that particular model, 20-odd years on?

Neville Nicholls: I certainly would. Thank you for remembering it!

Encephalitis, like most other viruses, is very strongly driven by rainfall variations. In one way I was unfortunate enough to do that paper just as drying took off – background drying, or variations in drying. I think one of the problems we face (and this is an esoteric meteorologist problem) is that we know we can make climate variation forecasts through the ENSO mechanism, and in fact using indices of ENSO built into the Australian encephalitis predictions, but all bets are off, really, with those relationships now, because they were developed under the assumption of stationarity, using data up to about the 1970s, and the background in most of those climate variables has changed. So we have more complications with predicting climate variability and its impacts because of the underlying climate change.

But I think in general you are right that, if it dries, even if it warms, then the drying will at least reduce any exacerbation you might get of some in these illnesses due to the warming.

Question (Bryan Furnass): I would like to ask a question about particulate pollution. Satellite photographs show an enormous brown cloud over central and east Asia, which is said to be due partly to the massive clearing and burning of forests and so on, and which has had a significant effect in reducing monsoon rains over south Asia and also food production. Do you think that kind of mechanism is going to affect rainfall over Australia?

Neville Nicholls: This is an area of vigorous scientific debate, and I have to say that it must have an effect – it must. It is such a big regional problem that it must have an effect. We don't know at the moment how substantial that effect is, though. But it must do something. It is a whopper of a brown cloud.