ABOUT one week after Central Queensland University put out a media release announcing the publication of our latest paper on medium-term rainfall forecasting, John Abbot and I received a phone call from the ABC, from a rural reporter.
Part of the interview with Professor Abbot can be heard online here, http://www.abc.net.au/news/2014-02-21/nrn-forecast/5275122
ABC journalists Cassie Hough and Kathleen Calderwood have also written a short article based on the interview in which they effectively capture the key points, http://www.abc.net.au/news/2014-02-21/nrn-forecast/5275122
And following is a succinct summary of the interview by one of the media monitoring services:
“New research from the University of Central Qld shows medium-term rainfall forecasts may be more accurately predicted by a form of artificial intelligence, which aim to predict the weather through pattern detection, and predicts not what the weather will do day-to-day, but in a month or more. John Abbot, researcher, says he was motivated to do the research after the huge flood at the start of 2011.
Abbot talks about how his research works, and the problems with modern-day weather forecasting systems, and says farmers and mine-operators need accurate data as opposed to vagaries. Abbot says he would like the Bureau of Meteorology take more interest in what he is doing.”
The paper can be downloaded from Science Direct, with full open access available until March 26, here,
http://elsarticle.com/1ej97n3
[Open access until March 26, 2014]
Highlights include:
1. Artificial neural networks are developed to forecast monthly rainfall for localities in Queensland.
2. Comparative utility of input variables, including local maximum and minimum temperatures, is tested in the artificial neural network.
3. The Inter-decadal Pacific Oscillation, IPO, is found to have utility for rainfall forecasting.
4. Artificial neural networks forecasts are superior to Australian Bureau of Meteorology official seasonal and monthly forecasts.
Stop griping bazza and provide a link.
Just as an example of real-life complexity and the dangers of literal thinking:
Since 1882 my region's wettest autumn by far occurred in the latter stages of a neutral year, leading into an El Nino (which was wet enough). In fact, our rainfall records won't neatly fit any theoretical pattern based on ENSO. The two biggie El Ninos - 1982-3 and 1997-8 - gave rainfall around or above the mean...and that rainfall was considerably higher than in the super-double La Nina of 1916-17! There's value in knowing about SOI etc year to year, but it's limited value. Pretending that the great dry periods of that pre-1950 era - independent of El Nino etc - never happened will take some stupendous spinning. (I know, I know...there are some stupendous spinners around here.)
My guess is that there were more things stifling Eastern Australia's rainfall in the first half of the twentieth century than we know much about. People have started to give names to these other influences - let's hope they treat them with a sane mix of curiosity and skepticism.
Which system was BoM using in that 2012-13 report?
BTW, the seasonal forecast for winter/spring 2013 from BoM was 80% chance of wetter than average.
Not sure how you want to work that in to your farmer argument?
What extraordinary irony - no gambling here says Cohenite at a blog preoccupied with downplaying the risk to the planet from AGW.
What ignorance from sp shooting off on seasonal climate funding without bothering to research the millions that farmers have invested over 2 decades , and saved many times more from better informed decisions such as whether or not to invest or defer, for example by a grain farmer in a new header with a forecast of 65% chance of a lower than average spring rainfall.
What blind loyalty from Glen Michel not taking the obvious cue to ask Jen what % accuracy she is offering not knowing that she will respond that is too simple a metric and someone might understand it.
Bazza: "BOM Annual Report 2012-13 Seasonal Forecast Success rate 65%"
I have a 50% chance of getting it right by flipping a coin - how million dollars did BOM spend to get it to 65%?
BoM spring outlook2013 - bullish at 80% locally SSTs were never going to support their projections.Covered their arses quickly when real time caught up with them.50:50 is the best that the bureau puts up.Heads or tails?
Bazza that's no good - Debs needs 100%