Is there a “New Zealand Effect” in Australian data? Quite possibly. The Torok and Nicholls network of stations which can be downloaded from the BoM ftp site, not yet adjusted to produce the ‘official’ BoM version, shows little warming.

The blue line is the ‘official’ BoM mean temperature, the black line is the mean for all stations combined using my patented normalization method, described in a previous post.

The minimum temperature (red and orange lines are the minimums) for the Torok and Nicholls network of all Australian stations is higher at the beginning of the century than at present. Minimum temperatures are actually declining.

The minimums differ from the maximum temperatures which are virtually flat from mid century to present, then falling at about the same rate as the official temperatures beyond that.

The red and the orange represent temperatures pre and post the T&N adjustments which don’t seem to be very significant when all stations (around 133) are averaged.

When the minimum and maximum is combined (black) and compared with the official mean (blue), there is still a reduction in the warming trend, down to 0.6C/century compared with 1.4C/century, and clearly seen by comparing the 10 year moving averages.

So the question is: Is there a “New Zealand effect in Australian data? Quite possibly. The Torok and Nicholls network of stations which can be downloaded on the BoM ftp site, has not yet been adjusted to produce the ‘official’ BoM version, but shows no real warming.

The adjustments performed by T&M for station moves and inhomogeneity (the qc and analysis directories) seem to produce relatively minor changes in the means, compared with the adjustments that go into the final versions of the BoM data. Even more so, the GHCN network for Darwin shown at Smoking Gun for Darwin Zero. This results shows that the adjustments producing warming are not isolated, but appear to be a feature of the overall dataset.

Questions could be asked about the methods of normalizing the data and the uncertainty of the data due to the small number of stations at the beginning of the century. The normalization method has been shown to preserve the trend of the official BoM version in previous posts.