Whither weather?

For most of this year - certainly since March 2011 - it has appeared to me that the weather forecasts for Canberra issued by the Bureau of Meteorology (BoM) have been a bit of a dogs breakfast.  By that I mean they have been all over the place.  I place great weight on the phrase "appeared to me" since I have not tried to record events in any structured sort of way.

By way of example, here is a summary of my memory of the weather forecast issued prior to 31 July for the period from Sunday 31 July to about Friday 5 August.
  • There were to be no frosts in this period (ie all minimum temperatures would be above 0);
  • There would be showers on Sunday and then showers again from Wednesday through  Friday.
The forecast issued in the afternoon of 31 July (a more or less dry day after a -2 minimum) has
  • A mininmum of -3 for Monday and 0 for Tuesday.;
  • Isolated showers on Friday.
This seems to be quite a significant level of change.  To try to get a handle on whether this variation is significant I intend to record each day for the 7 days forecast:
  • the overnight minimum forecast for each day; and
  • an indicator of likely rainfall.
After 7 days there may be a pattern emerging.  I shall update this post with comments where something appears to be worth commenting on.

Of course, the BoM forecasts, while the most authoritative are only one of the forecasts issued.  I also look at the forecasts issued by The Weather Channel (TWC).  For the first 7 days of August there is not great correlation in the minimum temperatures (see comments on this graph):
 Another way of looking at this is to do a line graph of the two series:
The following chart shows the variability of forecast rainfall for 3 dates.  Notes about interpreting the graph follow the image.
Looking at the forecast for 8 August (red columns) the BoM started forecasting on 1 August, with no precipitation expected.  For the next two days the forecast for 8 August was 'chance of showers' which rose to 'rain' in the forecast of 4 August dropping back to 'chance of showers' from the forecast on the 5th.

In essence this shows the problem: no, or very little rain forecast then downpour then drop back to very little.

WRT to rainfall the agreement between the two forecasts is better in that they are identical except for 7 August where TWC indicate chance of showers.

Adding to the confusion is a disconnect between what is on the BoM website and what is announced on the ABC TV News.  For the first period of observations (1 August) the website said -1.8 was the minimum while the presenter on TV said -3.  I will stick with what is on the website, since if it is on the internet it must be true!

Comments

Denis Wilson said…
Sorry, Martin, but I don't see any points of comparison.
I was looking for two separate lots of TWC and BOM dots.
Am I missing something?
Denis
Flabmeister said…
Denis

The dots in the graph relate the value of BoM to the value in TWC. Thus the lowest dot is for -3 on the BoM scale (vertical axis). For the same day TWC were forecasting +4 (so the dot is a fair way out on the horizontal axis).

I will add a chart with two lines (assuming I have kept the TWC data).

Martin

Popular posts from this blog

A tour of the West (part 1)

Insects from pine trees

Maslins beach rules