In my quest to verify that an increase in droughts is likely, as claimed in the Drought Exceptional Circumstances Report, I have spent a few weeks corresponding with Kevin Hennessy the lead author. Despite more than one request, he has not provided technical information, reports, or any evidence to support the claims made in the report.
Providing evidence in support of claims is fundamental to science, which is why I am so amazed that CSIRO has not responded abundantly with proof of their confident assertions that droughts are going to be more likely. As the key claims of the report can’t be validated, the only responsible course is to retract the report until such time as the claims can be demonstrated to be true, or amend the report. I therefore sent the following letter to the heads of two institutions responsible for the DECR. (Actually I have received a reply, and will report back in a few days.)
Dr G. Ayers, Chief, CSIRO Marine and Atmospheric Research
Dr G. Foley, Director of the Bureau of Meteorology
Re: Drought Exceptional Circumstances Report (DECR) 
While I acknowledge that sections of the report have validity, the reliability of key claims cannot be determined, so the report is misleading and should be retracted. The problematic claims are most prominently presented in the Summary, page 1, para. 5 (my emphasis in bold).
Observed trends in exceptionally low rainfall years are highly dependent on the period of analysis due to large variability between decades. If rainfall were the sole trigger for EC declarations, then the mean projections for 2010-2040 indicate that more declarations would be likely, and over larger areas, in the SW, SWWA and Vic&Tas regions, with little detectable change in the other regions. Under the high scenario, EC declarations would likely be triggered about twice as often and over twice the area in all regions. In SWWA the frequency and areas covered would likely be even greater.
The key claim that â€œmean projections for 2010-2040 indicate that more declarations would be likelyâ€ indicates a high level of confidence in increasing frequency and severity of droughts, an interpretation widely reported in the media. However, this level of confidence has not been demonstrated, as I show below. Therefore no level of confidence, likely or otherwise, can satisfactorily be attributed to the claim. In fact, the evidence suggests that while the models may have skill at historic temperature and rainfall, the models have little to no skill at modeling historic trends in exceptionally low rainfall events (droughts). The report therefore misleads by falsely attributing skill to claims of increasing frequency and severity of low rainfall, and should be retracted.
The evidence for your consideration is as follows:
1. The report itself contains no tests of skill at modeling historic area of exceptionally low rainfall events.
2. An interview available on the CSIROpod website (http://www.csiro.au/multimedia/DroughtReport.html) states the report is performing predictions, and that there are no clear trends in exceptionally low rainfall years (despite recent warming).
3. In a good faith attempt to determine if the models have skill at exceptionally low rainfall events, I corresponded with Mr Hennessy and subsequently obtained data from the report to test the 13 climate models. While some models had skill at some tests in some areas, overall I was unable to find a level of skill that would support the key claim above. In particular, I concur with Mr Hennessy that the long-term trends in exceptionally low rainfall are not clear. There appears to be an inverse relationship between model predictions of drought and historic records (Fig 1).
4. While I was unable to find evidence of skill, the models might pass other tests. Lead author Kevin Hennessy has not responded to my requests for evidence of skill.
5. For additional interest I attach an article critical of the report by Ian Castles, former Australian Statistician.
It could be argued that the DECR only reports model results. In this case, the key claim would, at least, be reported without a confidence attribution e.g. â€œMean projections of models for 2010-2040 included more EC declarations.â€
Predicating the key claims with â€œis more likelyâ€ communicates a certain level of confidence in increasing severity and/or frequency of low rainfall events. Confidence in model projections might be achieved by demonstrations of skill at validation tests. Validation of models is a widespread and accepted scientific practice, and would potentially justify the attachment of â€œis more likelyâ€ to the claims. However, no validation of the models for drought was reported, or supplied after requests from the lead author, or appears likely to be found.
As the authors did not establish the validity of key claims, the appropriate response is to withdraw the DECR.
David R.B. Stockwell PhD.
 1. K. Hennessy, R. Fawcett, D. Kirono, F. Mpelasoka, D. Jones, J. Batholsa, P. Whetton, M. Stafford Smith, M. Howden, C. Mitchell, and N. Plummer. An assessment of the impact of climate change on the nature and frequency of exceptional climatic events. Technical report, CSIRO and the Australian Bureau of Meteorology for the Australian Bureau of Rural Sciences, 33pp., 2008.
 See attachment interview.doc
 See attachment article.pdf
 See attachment emails.doc
 See attachment CastlesOLO.doc