FOLLOWING are my initial comments in response to the release of the report by the Technical Advisory Forum on the Australian Climate Observations Reference Network (ACORN-SAT):
I NOTE that The Forum, chaired by Dr Ron Sandland formerly of the CSIRO, concurs with the Bureau that:
“There is a need to adjust the historical temperature record to account for site changes, changes in measurement practices and identifiable errors in measurement… To this end, the Forum supports the need for the Bureau’s homogenisation process to incorporate both metadata-based adjustments and adjustments based on the statistical detection of atypical observations. In the opinion of the Forum members, unsolicited submissions received from the public did not offer a justification for contesting the overall need for homogenisation or the scientific integrity of the Bureau’s climate records.”
As a member of the public who made an unsolicited submission, I would like to clarify that at no time did I suggest there was no need for adjustments, rather I have queried why there are adjustments made when, in fact, there are no documented site changes, no changes in measurement practices, and no identifiable errors. Yet adjustments are still made.
The Forum appears to have overlooked many examples of this provided in the public submissions, and published by The Australian newspaper late in 2014. For example, the Forum has completely ignored the notorious example of Rutherglen, where a slight cooling trend was converted into a warming trend, despite an absence of any metadata providing justification.
The Forum has also made no comment on the actual choice of stations for inclusion in ACORN-SAT, nor how the selection of stations has changed in recent years. For example, in his submission to the panel, retired chartered accountant Merrick Thomson showed how the choice of ACORN-SAT stations changed from 2012 to 2013 and, how this could generate a large increase in global warming.
The Forum has suggested that the Bureau consider pre-1910 data in its analysis of climatic trends.
“Recommendation 5: Further, the possible availability of pre-1910 data at south-eastern sites may allow for a comparative analysis to be performed for south-eastern Australia to assess whether the inclusion of pre-1910 data is worthwhile in attempting to understand current temperature patterns.”
This is currently listed as a low priority by The Forum, but its inclusion is nevertheless welcome, and was a key recommendation in my submission. I also recommended that all temperature series start at the same date. For example, I provided the example, in my submission, of the Bureau adding in the very hot town of Wilcannia only from 1957, when there is data available from the late 1800s.
I also welcome the recommendation that the Bureau:
“Address two key aspects of ACORN-SAT, namely: a) improving the clarity and accessibility of information provision—in particular, explaining the uncertainty that is inherent to both raw and homogenised datasets, and b) refining some of the Bureau’s data handling and statistical methods through appropriate statistical standardisation procedures, sensitivity analyses, and alternative data fitting approaches.”
I note that The Forum state in their report that: “It is not currently possible to determine whether the improvements recommended by the Forum will result in an increased or decreased warming trend as reflected in the ACORN-SAT dataset.”
I would suggest that if the committee’s recommendations were properly implemented, and the Bureau abandoned some of its more creative accounting practices (e.g. adding in particularly hot locations for later years in the time series), then it would become apparent that there has been an overall trend of cooling over much of central and eastern Australia from 1880 to 1960, more dramatic warming than previously documented from 1960 through to about 2002, while more recently temperatures have plateaued, with some evidence of a cooling trend establishing in north eastern Australia since 2002.
I note The Forum intends to operate for another two years, and urge them to be honest to their title of “The Forum” and actually meet with some of those who have so far provided unsolicited public submissions. Indeed, I urge Dr Sandland to immediately set up an open and transparent Forum process whereby these submissions can be presented allowing any accusations of scientific misconduct by the Bureau to be both defended and contested before the Australian public, and media.
The committee makes five recommendations, but puts emphasis on the importance of the first two components of the first recommendation.
I applaud the first component of the first recommendation of the committee that in full states:
“Expediting the Bureau’s current work on developing uncertainty measures in closer consultation with the statistical community. The Forum recommends the Bureau seek to better understand the sources of uncertainty and to include estimates of statistical variation such as standard errors in reporting estimated and predicted outcomes, including: quantifying the uncertainty for both raw and adjusted data; prioritising the provision of explicit standard errors or confidence intervals, which should further inform the Bureau’s understanding and reporting of trends in all temperature series maintained by the Bureau; examining the robustness of analyses to spatial variation; and articulating the effect of correcting for systematic errors on the standard error of resulting estimates.”
Of course, that such basic statistical information is not currently available is impossible to reconcile with the overall conclusion in the report that, “the analyses conducted by the Bureau reflect good practice in addressing the problem of how to adjust the raw temperature series for systematic errors.” Then again, the executive summary of The Forum’s report appears to have been written by someone straight out of the BBC television series ‘Yes Minister’.
The second component of the first recommendation is also applauded, which reads in full:
“Developing a clearer articulation of the purpose for the ACORN-SAT exercise to enhance public understanding of the program, and communicating processes for developing and using ACORN-SAT in a way that is appropriately clear, broad and supported by graphics and data summaries. In particular, the central focus on the Australian annual mean temperature anomaly as the primary end point of the ACORN-SAT exercise should be reconsidered and a broader narrative around including regional effects should be developed.”
Indeed, it has become apparent over the years that the entire focus of the work of the small ACORN-SAT unit is not the provision of higher quality individual temperature series, but the remodeling of the raw data, and the compilation of a select few station, to suggest that it is getting hotter and hotter across the Australian landmass with such announcements made with great fanfare by the Bureau’s David Jones at the beginning of each year.
Recommendation No. 2, has several components including comment that:
“Releasing the Python computer code for ACORN-SAT as a downloadable link along with all supporting documentation and listing of the technical requirements for the software. The Bureau should also monitor and gather download statistics to gauge demand for this software.”
Of course, without access to this software it has been impossible to reproduce any of the adjustments made by the Bureau. Yet if the method is scientific, it should be reproducible. For many years, the Bureau has erroneously claimed its methods are transparent. It should be noted, however, that even with the provision of this software, it will be impossible to justify ACORN-SAT because it is unclear why the Bureau chooses some stations above others for its comparisons. For example, despite endless requests for clarification, the Bureau has never explained why it uses the distant location of Hillston to make comparison, and then changes, to the raw temperature data for Rutherglen in north eastern Victoria.
Recommendation 2 also includes comment that: “Publishing a brief, plain-language (as far as possible) description of the criteria for adjustment and the basis for adjustment itself.” Of course this should have been available since the very first adjustment was made in the development of ACORN-SAT. That such a document still does not exist is evidence that ACORN-SAT is poorly documented. So, how could The Forum endorse the Bureau’s claims that it represents world’s best practice?
Dr Jennifer Marohasy
A Coruna, Spain
18th June, 2015
Jennifer Marohasy BSc PhD has worked in industry and government. She is currently researching a novel technique for long-range weather forecasting funded by the B. Macfie Family Foundation.