Australia’s Bureau of Meteorology (BoM) has dismissed recent allegations of widespread errors in its long-term temperature data collection, following a series of claims published by a major newspaper. The newspaper’s reporting was based on a so-called “forensic audit” that purportedly uncovered numerous inaccuracies in the Bureau’s temperature records, which have indicated that Australia’s climate has warmed by more than 1.5°C since 1910.
The audit, conducted by two retired professionals—a former scientist and a former computer engineer—asserted that the Bureau’s temperature data showed signs of “obvious patterns of human intervention,” missing or duplicated entries, and anomalies described as “data failures.” Among the cited examples was the duplication of minimum temperature readings from August 1907 with data from August 1867 at Sydney’s Observatory Hill, a claim flagged by the newspaper’s environment editor as requiring explanation.
However, BoM officials emphasized that data prior to 1910, including the disputed August 1867 record, is not included in the Bureau’s main climate dataset, known as ACORN-SAT (Australian Climate Observations Reference Network – Surface Air Temperature). The dataset focused on controlling for local changes, such as station relocations, to ensure accurate long-term climate trend analysis. The Bureau found that fewer than 200 out of seven million observations showed any duplication, with only a handful occurring after 1910, and determined the overall impact on the temperature record to be statistically insignificant.
A Bureau spokesperson suggested that the likely source of the duplications was the manual transfer of handwritten logs to punch cards during the 1950s or 1960s, a process that introduced errors unrelated to contemporary data collection or analysis. They also noted a lack of further duplicate months found across the entire network of 112 weather stations contributing to ACORN-SAT.
Another notable claim from the audit involved a sequence of 17 consecutive days recording a maximum temperature of 31°C at Palmerville, Queensland, in 2004, which was described as “data failure” due to its improbability. The Bureau countered that the observations were recorded to the nearest degree at that time, and given the naturally lower temperature variability in tropical regions, such repeated values were plausible.
The audit’s methodology was not fully disclosed, raising questions about the robustness of its findings. Furthermore, the Bureau highlighted that ACORN-SAT has undergone two independent reviews and that its methods are published in peer-reviewed scientific literature. The ‘forensic audit’ report was sent to the Bureau via independent Tasmanian MP Andrew Wilkie, who after reviewing the Bureau’s response said he maintained his trust in the agency’s climate data and expressed discomfort at how the newspaper’s coverage might have implied doubts about the Bureau’s integrity.
The Bureau of Meteorology’s temperature records have long been targeted by climate change sceptics and critics of emission reduction policies. Previous challenges have often surfaced outside scientific forums, typically in media outlets with editorial lines skeptical of climate science. Scientists and former Bureau staff have described these repeated attacks as distractions that divert resources and hinder efforts to address climate risks.
Despite the controversy, climate experts continue to emphasize that multiple independent datasets confirm Australia’s warming trend, which aligns with global patterns and highlights significant risks to ecosystems, infrastructure, and public health. The Bureau reiterated that the claims of data manipulation were unfounded and based on misunderstandings of data processes and historical record-keeping.