Interesting

Archive of General Weather
Forum rules
These topics are a read-only archive and may be subject to out-of-date information.

For today's weather discussion head to: New Zealand Weather & Climate
Locked
jrj
Posts: 462
Joined: Tue 18/03/2003 18:00
Location: Havelock North, New Zealand
Has thanked: 0
Been thanked: 0

Interesting

Unread post by jrj »

Metservice claims high weather prediction accuracy
(NZ Herald today)

MetService was 84 per cent accurate in predicting heavy snow in the year ended June 30, it says in its annual report. It also claims it was 87 per cent accurate in its predictions of heavy rain and 83 per cent for severe gales.

However, 23 per cent of its heavy rain warnings were false alarms, as were 30 per cent of heavy snow warnings and 25 per cent of severe gale warnings.
tich
Posts: 3467
Joined: Sat 15/03/2003 18:32
Location: Christchurch (St Albans)/Akaroa
Has thanked: 21 times
Been thanked: 91 times

Unread post by tich »

We must also look at intensity and spread of events when judging how accurate warnings turn out to be. For example look at the late September heavy snow event in Canterbury this year. Warnings were issued I think the day befor,e for heavy snow in the high country down as low as 400 metres. What eventuated was heavy snow down to about 200 metres. Alot of farming country in Canterbury lies between 200 and 400 metres.

The general public with little meteorological knowledge is likely to blame MetService for not predicting the low level heavy snow. However, those of us (hopefully including most farmers) who understand weather systems more and can see the possibilities of how certain systems can affect the various parts of the country in coming days, could've seen that the situation was one that could easily dump heavy snow to low levels in inland Canterbury.

Also MetService heavy rain warnings tend to cover the likelihood of large scale events, ie those that could affect whole districts, eg Westland or Northland. Less easy to predict are localised heavy downpours, such as those induced by convergence zones and convective downpours/thunderstorms (the latter may be easy to predict for a region, but where exactly they hit is very difficult to forecast)
User avatar
NZstorm
Posts: 11333
Joined: Mon 10/03/2003 19:38
Location: Grey Lynn, Auckland
Has thanked: 342 times
Been thanked: 361 times

Unread post by NZstorm »

Met. Service are considerably more accurate than they used to be.

I guess they can thank new technology for that. ;)
Manukau heads obs
Posts: 12305
Joined: Mon 10/03/2003 16:30
Location: West Coast Road, Manukau Peninsula, North Island
Has thanked: 0
Been thanked: 0
Contact:

Unread post by Manukau heads obs »

the latter may be easy to predict for a region, but where exactly they hit is very difficult to forecast)

well, if they use people's local knowledge more, i.e weather enthusiasts, then they can narrow it down
i.e for this greater district area, I can tell you where the convective CB's etc occur and who will get them most....and as the situation unfolds, I can tell whats likely to happen, etc ,etc...to a degreee, lol......ie local fine tune forecasting../
Image
Brian Hamilton, weather enthusiast. My weather dataEmail: [email protected]
Locked