Understanding Climate Risk

Science, policy and decision-making

Climate shifts paper published

with 4 comments

The climate shifts paper on non-linear anthropogenic climate change in SE Australia with earlier descriptions here and herehas been published in the Journal of Geophysical Research. We had plans to make it open access  but must not have paid the invoice yet, so it’s behind a paywall here (pdf). If you can’t get access, get in touch with me via comment below or email (firstname.lastname at vu.edu.au) and I’ll send a reprint.

Following is an edit of the draft press release that should go out in the next week or so. The press release implies that this pretty much goes on everywhere – not just SE Australia. That’s what I believe from further work currently underway – the ocean runs the climate system and the atmosphere doesn’t warm quite the way people assume it does. The energy from greenhouse warming goes into the ocean first, then is re-emitted periodically into the atmosphere. The atmosphere doesn’t warm in situ (or not yet to any significant degree). It may later under increased radiative forcing, but at the moment the ocean is pretty much running the show. This pattern of non-linear change fits in with palaeoclimatic evidence from the region that I’ll get round to describing some time.

Future climate changes are shown on graphs as smooth curves. Gradually rising greenhouse gases are interpreted as causing gradual warming of the atmosphere overprinted by the ‘noise’ of climate variability. This leads to the expectation that people can adapt gradually to a smoothly changing climate.

Victoria University’s Professor Roger Jones, challenges this view with a paper analysing recent warming in south-eastern Australia published in the Journal of Geophysical Research – Atmospheres. He shows that historical relationships between rainfall, maximum and minimum temperature changed in a step-wise manner in the period 1968–73 as the region warmed abruptly. A further step change occurred in 1997 that coincided with the drier conditions of the millennial drought. Analysis showed that the warming was greater than would be expected by the drier conditions (drier soils evaporate less, leading to higher temperatures during droughts).

These step changes coincide with increases in temperatures averaged across the southern hemisphere. Climate model output for SE Australia showed similar step changes but with different timing. All models showed a stationary period of historical climate followed by an abrupt start to regional warming. Under increasing emissions similar shifts continue through to 2100. Under climate change, regional warming therefore looks more like a staircase than a smooth curve.

Regional climate in SE Australia was stationary to 1967, experiencing a step change in 1968 of 0.7°C in minimum temperature and 0.5°C in 1973 for maximum temperature. A step change in 1997 in maximum temperature of 0.8°C coincided with a statistically significant step change in global mean air temperature of 0.3°C. Changes since 1997 have affected the number of days of extreme heat, fire danger and a range of other variables.

“These findings don’t challenge the basic theory of greenhouse-induced warming,” Professor Jones said. “The long-term response simulated by the models is correct. However, some shorter term variations that have been interpreted as natural variability are in fact part of climate change. Energy is being stored elsewhere in the climate system, most probably the oceans, and released into the atmosphere in periodic bursts. Opponents of the theory go on about no warming since 1998 while conveniently overlooking the significant increase in temperature in 1997–98 that contributes to the long-term trend.”

Non-linear climate change has important implications for adaptation. Sudden increases in risk may mean that gradual adjustments fail to cope with rapidly changing extremes. If such shifts are interpreted as temporary conditions due to climate variability the frequency of extremes may continue to be under-estimated.

“Since 1997, SE Australia has experienced the same frequency of days above 35°C estimated for 2030 assuming gradual change.” Professor Jones said. “I expect to see similar shifts in future. If we understand how they are linked to energy transfer between the ocean and atmosphere, we can get early warnings of such shifts when they happen.”

Professor Jones is working with colleagues to understand the physical and economic impacts of such changes on natural and human systems.


4 Responses

Subscribe to comments with RSS.

  1. Thankyou Roger. I note that SW Australia or Perth specifically has just endured its 8th heat wave (3 days above 35 C) matching a previous summer record from 1973. This seems to coincide with your 1967 – 73 period for SE Australia. Any linkage between the two? Too far apart geographically?

    peter tuck

    March 12, 2012 at 9:01 am

    • No Peter, it’s not. The temperature effects seem to have influenced all of southern Australia. 1968 is the year that rainfall turns dry in SW WA, and a number of other phenomena (formation of lows and perhaps storm generation) follow suit. The 1968-73 thing might just be the vagaries of interannual variability in the eastern states. I have had shot at WA climate for step changes and Tmin shifts up in 1968 (preliminary results). Tmax doesn’t show a lot until 1994 (which coincides with sea surface temperature warming). There is an increase in TminARW (the ‘anthropogenic’ component of Tmin) in 1960 that may be related to the extension of the Leeuwin current – this also affects some coastal stations in southern Victoria so it’s probably a regional effect (We see Leeuwin current effects in past climate reconstructions of ocean temps from WA to western Tassie).

      I’ve just put up a post where I calculate the likelihood of the Texas heatwave under climate change or climate variability. Would be interested in testing this also for southern Australia. https://2risk.wordpress.com/2012/03/12/nothing-grows-in-texas/

      Out of interest, the underlying temperatures in the eastern states, which have been relatively cool in the last couple of wet years, are still quite warm when you allow for the rainfall affect.

      Roger Jones

      March 12, 2012 at 9:19 am

  2. Hi Roger,

    Any chance of a peek at your paper? (gareth at the web site domain) Thinking of doing something on jerks, but I need to read and understand first…



    March 26, 2012 at 7:50 pm

  3. […] on variability in this case, but independent of it. This is completely consistent with the statistical analyses I have been using to separate climate variability in temperature from the warming […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: