I respond to this letter in the post:
BOM finally explains! Cooling changed to warming trends because stations “might” have moved!
For the record, the entire response is below.
Media enquiry: Comparison of temperature charts
From: Graham Lloyd, The Australian
Date: 22 August 2014
Homogeneity adjustments are standard analyses carried out by agencies around the world.
As outlined in the Bureau’s previous correspondence, each temperature recording is the result of a range of choices — such as choice of instrument, choice of calibration, choice of instrument enclosure and choice of enclosure siting.
All of these elements are subject to change over a period of 100 years, and such non-climate-related changes need to be accounted for in the data. Put another way, one needs to determine whether changes recorded in unhomogenised temperature records are real or artificial — there can be no implicit assumption that such records are unaffected by changes in recording practice over time. This implicit assumption is often uncritically presented on internet blogs.
The international literature strongly concludes that data homogenisation is necessary for the creation of temporally consistent data sets. There is a wide range of non-climate related influences on historical data that must be taken into account, and which are described in the links provided above.
The FAQ linked to above provides graphical examples of when homogeneity adjustments are required in temperature time series, and may assist you in interpreting the following information.
In the specific cases mentioned:
Bourke: the major adjustments (none of them more than 0.5 degrees Celsius) relate to site moves in 1994 (the instrument was moved from the town to the airport), 1999 (moved within the airport grounds) and 1938 (moved within the town), as well as 1950s inhomogeneities that were detected by neighbour comparisons which, based on station photos before and after, may be related to changes in vegetation (and therefore exposure of the instrument) around the site.
Amberley: the major adjustment is to minimum temperatures in 1980. There is very little available documentation for Amberley before the 1990s (possibly, as an RAAF base, earlier documentation may be contained in classified material) and this adjustment was identified through neighbour comparisons. The level of confidence in this adjustment is very high because of the size of the inhomogeneity and the large number of other stations in the region (high network density), which can be used as a reference. The most likely cause is a site move within the RAAF base.
Rutherglen: the major adjustments in minimum temperature data are in 1966 and 1974. Both were detected through comparisons with neighbours. The nature of the change is consistent with the site moving from a location near the main experimental farm buildings (which are on a small hill) to its current location on low-lying flat ground (minimum temperatures are normally higher on slopes than on flat ground or in valley bottoms).
While the three examples quoted here, all involve stations where the homogenised data have a stronger positive trend than the unhomogenised data. There are examples where the reverse is true. For example, at Mackay (where the site moved from the post office to its current location, on a small hill near the waterfront), the trend in minimum temperatures is +0.40 C/decade in the raw data but only +0.18 C/decade in the adjusted data, as the new site is climatologically much warmer for minimum temperatures than previous sites were.
More generally, the Bureau of Meteorology maintains both homogenised and unhomogenised data sets in parallel for the purposes of comparison (amongst other fit-for-purpose needs).
An extensive sensitivity study was carried out to determine the influence of homogeneity adjustments on national temperature trends. These comparisons have determined that from 1950 to present — the period during which most of the warming has occurred in Australia — homogeneity adjustments have little impact on national trends and changes in temperature extremes.
This comparison work is publicly available, and can be found on the ACORN-SAT website
The Bureau of Meteorology’s practices in developing and maintaining temperature data sets have been ranked as amongst the best in the world by an independent panel of international experts in 2012.