While compiling the Autumn Temperature Outlooks piece – I noticed what looked like an aberration in the contouring of the Australian Bureau of Meteorology (BoM) Autumn minimum temperature anomaly maps here – features that were repeated in the monthly minimum anomaly maps for March – April – May 2011.
I am talking about the area near the SW corner of the Northern Territory (NT)- a remote region with sparse data where two stations in particular seem to be at the centre of aberrant contouring over many months – and I have checked back as far as 2007 where the feature shows very obviously month after month.
The two stations concerned are; Walungurru – Number:15664 Opened: 2001 – and Giles Meteorological Office Number:13017 Opened: 1956. Of course the two stations are very different – Walungurru at a remote community – only has a few years of data, and of course averages over the 1961-1990 period would have to be estimated from neighbours and this site suggests no neighbours are close.
I have not found a photo of the instruments site or found exactly where BoM Station Walungurru Number:15664 is located other than Lat: 23.27° S, Lon: 129.38° E, Elevation: 454 m. – and when I clicked on a map icon for details I read – “The information for this station 015664 is not available now, but it may be available in the future”.
This image from GoogleEarth shows the layout at Walungurru / Kintore – the Lat-Lon plots just under the red “T”. It is a possible location seeing that public servants would attend the sewage/water works. I would expect to see a fenced enclosure.
Giles on the other hand is a purpose built – I expect state of the art – worlds best practice meteorological station.
This map of the Minimum Temperature Anomaly for Autumn 2011 has the approximate locations of Walungurru and Giles marked, the two places are ~220 km apart. We see in this map how the Walungurru data forms an anomalously warm pimple in the contouring while Giles is the opposite, forming a cool pit.
This pattern is too common to be natural when roaming back through monthly Minimum Temperature Anomaly maps here – mainly in the cooler seasons.
Here are some links to what I mean:
You get the point. So often anomalous – I think it stands out that the BoM map construction has incorporated a problem.
This could be caused if the BoM estimated Walungurru anomalies for the 1961-1990 standard reference period assuming too great a warming in the minimum temperature. That could explain why the Walungurru site so often produces a “warm pimple” in the contouring which because of the mathematics of the contouring produces a dipole effect “cool pit” in neighbouring Giles, in effect preventing the more reliable Giles data having its proper influence.
And I have only picked one small area of the map.