Accuracy of seasonal forecasting

[Michael]

http://www.nzherald.co.nz/section/1/sto ... d=10443866

[Manukau heads obs]

why am I not surprised to see Augie Auer name there
flipping a coint is going to tell you even less

[RWood]

The guy has a personal vendetta. Hardly objective.

[jrj]

He's still right!

[NZstorm]

Niwa's overall accuracy was 48 per cent.

Thats not a great number, international standard or not!

[RWood]

So he's going to do better? You people aren't living on this planet. I repeat - the basis for his sniping is personal, if you don't believe me I can PM about it.

[jrj]

Is he going to do better?
No.
Am I living on another planet?
No.
Do I think that NIWA's long-range forecasting is a waste of tax-payers' money?
You betcha.

[RWood]

You may - or may not - be aware that NIWA can't really give up that long-range forecasting even if it wants to, with current expectations. If you don't want the money to be spent, then lobby for a change.

TonyT might be able to establish that he has a better track record - over to him to elaborate if he wishes.

[RWood]

If you want more comment on the limitations of seasonal forecasting for NZ, refer to the paper "Are NCEP Seasonal Forecasts Useful in New Zealand?" (Francis & Renwick), Weather and Climate, 23 (2004).

Among other conclusions is one (not surprisingly) that forecast skill looks better when ENSO values are near the extremes.

[NZ Thunderstorm Soc]

Some of my forecasting is accurate:
People ask me how long is it going to rain for?
I say "Until it stops"

JohnGaul
NZTS

[RWood]

Now there's a chap with his feet on the ground.

[Manukau heads obs]

i must admit that i was surprised when niwa had the east of the NI down for normal rain for the autumn.....when we did not have any la nina type weather pattern emerging

[TonyT]

TonyT might be able to establish that he has a better track record - over to him to elaborate if he wishes.

Our own evaluations suggest we are around 65% accurate, and also find NIWA to be around 50%. But I stress that these are informal assessments and wouldnt stand up to peer review without some more solid methodology being applied. At the end of the day, the Blue Skies predictions are out there for all to see (thanks to Fencepost) so judge for yourselves. The feedback we get from Fonterra farmers is positive and encouraging, so we must be doing something right

[Michael]

Imagine if surgeons were 35% wrong

Our own evaluations suggest we are around 65% accurate, and also find NIWA to be around 50%. But I stress that these are informal assessments and wouldnt stand up to peer review without some more solid methodology being applied. At the end of the day, the Blue Skies predictions are out there for all to see (thanks to Fencepost) so judge for yourselves. The feedback we get from Fonterra farmers is positive and encouraging, so we must be doing something right

[RWood]

Dream on!

[TonyT]

Imagine if surgeons were 35% wrong

Silly argument. Surgeons work in an environment where repeatable results are found 99.999% of the time - the element of randomness in the human body is many many magnigtudes less than the element of randomnes in the atmosphere ("oh crikey, I opened up his leg and found a lung in there! Oh well, never mind, its just part of the random nature of my job").

[RWood]

I must say one can get terribly off-piste with armchair critics - most of them could benefit by doing some background research before shooting off.

And even with surgeons' advantages, there are some pretty bad surgeons around!