Questions over Bureau of Meteorology data

The ARC Centre of Excellence for Climate System Science does not collect primary data. We use data collected and quality controlled by agencies around the world. We examine these data to ensure processes used to quality control the data are rigorous. Specifically, we use data that have been quality controlled using methods that are reported in the peer-reviewed literature that have been subject to open and rigorous scrutiny. Best practice utilizes these methods to build high quality data sets but (and critically) also makes the original data sets available.

We were therefore disappointed by a series of articles in The Australian by Graham Lloyd reporting on some analyses by Dr Jennifer Marohasy.

Criticisms of any area of science is appropriate when backed by peer reviewed publication. We note:

  1. the criticisms by The Australian and Dr Marohasy of the Bureau of Meteorology data and associated methods used to quality control the data have not been subject to peer review. They are therefore merely the opinion of the writers and do not have scientific credibility;
  1. the methods utilized by the Bureau of Meteorology have been formally reviewed (see Trewin, 2013, published in the International Journal of Climatology, doi 10.1002/joc.3530). In addition, a rigorous independent peer-review of the methods was conducted in 2012. A panel of independent international experts found the data and methods were amongst the best in the world.

Simply, the criticisms published in The Australian are not scientifically legitimate.

A thorough critique of these issues is provided at: the Guardian and at: the Conversation

As a user of observational data, the ARC Centre of Excellence for Climate System Science is always concerned about the quality of data. Whether the data come from surface observations, ocean buoys, flux towers or satellites the processes and procedures used to quality control those data must be open to scrutiny via the peer-reviewed literature. We endeavor to limit our use of data to high quality and quality-controlled data and products. The Bureau of Meteorology observational data are an example of the highest quality data and the sorts of criticisms published of these data recently are without scientific validity.  

Andy Pitman
Director of ARCCSS

UNSW logo ANU logo Monash logo UMelb logo UTAS logo