One hundred years of drought and flooding rains

This week the Minister for Climate Change and Water, Senator Penny Wong, told viewers of the ABC Lateline program: “There is a great deal of scientific advice about the impact of climate change on rainfall, particularly in southern Australia”, and gave as her first example that “We know the IPCC said by 2050 that Australia should expect around about a 25 per cent reduction in rainfall in the southern part of ... Australia.” http://www.abc.net.au/lateline/content/2008/s2353504.htm.

With respect, the IPCC did not say this. According to the Panel’s Working Group I report (Climate Change 2007: The Physical Science Basis), which was approved by 130 governments in Paris on February 2, 2007, the average projected “precipitation response” for Australia and New Zealand south of the 30th parallel (about the latitude of Grafton) during the 21st century from the A1B emissions scenario, using a set of 21 global models, was a median decline of 4 per cent, with the middle half of the distribution ranging from minus 13 per cent to plus 4 per cent (see Chapter 11, Table 11.1, p. 856.

Projected ranges of change in annual average rainfall for 2020, 2050 and 2080 relative to 1990 are presented for several groups of regions of Australia in the “Australia and New Zealand” chapter of the IPCC Working Group II report (Climate Change 2007: Impacts, Adaptation and Vulnerability). For 2050, the range of change for that part of Australia “within 400km of western and southern coasts” is given as minus 40 per cent to nil (see Chapter 11, Table 11.4, p. 515).

However, the projections in this table do not appear to be credible. The ranges are stated to be “based on results from forty SRES emission scenarios and fifteen climate models for various locations in each region”, yet for four of the five groups of regions shown the upper and lower limits are all either plus or minus 5 per cent or 10 per cent (2020), plus or minus 13 per cent or 27 per cent (2050) and plus or minus 27 per cent or 54 per cent (2080). It defies belief that the range of rainfall change in 2080 (relative to 1990) from all of these scenarios and models could be from minus 27 per cent to plus 54 per cent for “Northern NSW, Tasmania and central Northern Territory” - and yet be from minus 80 per cent (i.e. one-fifth of the 1990 level) to nil “within 400 km of western and southern coasts”.

Senator Wong claimed “there is a very, very sound body of evidence that indicates that climate change is and will have an impact on rainfall in the Murray-Darling Basin and in southern Australia”. Yet Table 11.4 in the IPCC’s Working Group II report invites us to believe that the upper limit of projected rainfall change from 1990 in the western and southern coastal regions from all of these scenarios and models in 2020, 2050 and 2080 is nil, and that the lower limit of rainfall for the same three indicative years is, respectively, 85 per cent, 60 per cent and 20 per cent of the 1990 level.

The Table is sourced to “Suppiah et al., 2007”, a paper which is listed in the References to the Chapter as being authored by seven CSIRO scientists (including two IPCC lead authors) and as having been “accepted” by the Australian Meteorological Magazine, a publication of the Bureau of Meteorology. However, the manuscript of the paper of the same title that was subsequently published in this journal (vol. 56, no. 3, Sep. 2007) carries the name of an additional author and was stated to have been received in March 2007 and revised in August 2007 (four months after the IPCC Working Group II report was approved by governments). This raises the question of what paper, if any, was provided to IPCC expert reviewers who may have wanted to check the implausible data in Table 11.4 for themselves - and what opportunity, if any, they may have had to seek to replicate the model simulations used by the IPCC.

We now know the probable answer to that question, thanks to the publication by IPCC Working Group II of the drafts of their Report together with the reviewer comments thereon and the responses from the writing teams. For example, the First Order Draft of the “Australia and New Zealand” chapter, which was circulated for expert and government review in September 2005, included the following statement:

Reduced soil moisture is very likely, producing up to 20 per cent more droughts (soil moisture deficit in lowest 10 per cent from 1974-2003) over most of Australia by 2030 and up to 80 per cent more drought in south-western Australia (Mpelasoka et al., 2005a).

The full citation revealed that Mpelasoka et al., 2005 was a “CSIRO Technical Report, submitted”, and was the product of four CSIRO scientists (again including two CSIRO lead authors). One of the expert reviewers of this chapter - Dr Ken Ritman of the Bureau of Rural Sciences - commented that he considered the statement in the IPCC draft to be “highly uncertain given rainfall projections and lack of response in evaporation”. And, more significantly, Dr Ritman also commented “This is taken from a report not available to reviewer” (Comment 11-553).

I have made an extended comment on the subsequent history of Mpelasoka et al., 2005 and related matters on the Climate Audit blog (#44 on the “Some Quick Thoughts on CSIRO Drought Info” thread). The details do not matter: what this history shows is that, if the “Australia and New Zealand” chapter is representative, the IPCC reports are far from being the model of rigour and inclusiveness that the Panel and its more vocal supporters claim them to be.

Another expert reviewer of the “Australia and New Zealand” chapter was Graham Farquhar, Distinguished Professor in the School of Biological Sciences at the Australian National University. Before discussing some of Professor Farquhar’s comments and the responses, I need to point out that he has 250 research publications to his credit. He was a lead author for the IPCC’s Second and Third Assessment Reports and a coordinating lead author of the Panel’s Special Report on Land Use & Land Use Change and Forestry. And he was elected to the Fellowship of the Australian Academy of Science (FAA) in 1988 and of the Royal Society of London (FRS) in 1995.

Graham Farquhar’s comment on the sentence from the First Order Draft which is quoted above was as follows: