Topic Options
#1404802 - 01/02/2017 08:44 Models versus windows
fractonimbus Offline
Weather Freak

Registered: 07/11/2001
Posts: 96
Loc: Canberra
At what point do forecasters decide they got their forecasts completely wrong? Today, for example, in Canberra. the forecast is 90% likelihood of 20-30mm of rain. The sky shows little evidence of such an event, and the radar shows a long band of rain 50-80km north of here, moving resolutely eastwards. One could be forgiven for thinking this forecast is already rather wrong. It's nice to refer to your models, but a quick look out the window sometimes does a better job. Surely, the purpose of a forecast is not to report the results of models, it is to alert people to the actual weather?

#1404829 - 01/02/2017 11:55 Re: Models versus windows [Re: fractonimbus]
Seabreeze Offline
Weatherzone Moderator

Registered: 18/09/2005
Posts: 10328
Loc: South West Rocks, NSW
The Canberra forecast ended up getting updated at 9:16am.
Just having a quick look, I see ACCESS-R and EC have rain coming through the ACT (mainly this afternoon). While GFS keeps most of the rain to the north, out of the ACT.
It's a fairly thin line of rain, another 10kms further south and some of those northern suburbs would be under the line of rain now.

But for future reference on BOM's forecasts, the rain chance part in this case means 90% chance of more than 0.2mm, and the amounts stated mean 50% chance of getting 20mm or more and a 25% chance of getting 30mm or more.
On the updated forecast the amounts have been changed to 3 to 10mm, so 50% chance of getting 3mm or more and a 25% chance of getting 10mm or more.
South West Rocks, NSW Mid North Coast:
April 2018 Rainfall: 47.8mm (Apr Avg. 169.9mm) // April 2018 Raindays: 7 (Apr Avg. 12.9 raindays)
Year-to-date Rainfall: 705.6mm (Jan-Apr Avg. 670.4mm) // Year-to-date Raindays: 46 (Jan-Apr Avg. 55.8 raindays)

#1404834 - 01/02/2017 12:58 Re: Models versus windows [Re: fractonimbus]
fractonimbus Offline
Weather Freak

Registered: 07/11/2001
Posts: 96
Loc: Canberra
Thanks, very useful information.

#1410851 - 05/03/2017 09:15 Re: Models versus windows [Re: fractonimbus]
retired weather man Offline
Weatherzone Addict

Registered: 01/07/2007
Posts: 4485
Loc: Wynnum
At the request of the MOD, I am using this thread instead of the current special weather event thread to make a further comment regarding the current PERCEIVED problem of forecast inaccuracies in SE Qld.

I don't envy Ken is reading all this currently and responding appropriately, particularly his comment about people complaining about it not raining in their backyard.

I feel the recent introduction of Met Eye and other similar apps is helping to cause this problem of expectation. I have been following Met Eye and ECMWF for our local area ( based on Brisbane Airport ) where it is specified certain amounts of rains will fall between certain hours.

When Met Eye was introduced in quite a blaze of publicity a year or so back a greater degree of accuracy was promised and so a high expectation of "now I will see what will happen in my back yard" prevailed.

So I feel this is part of the problem. As Ken would attest to - to make a model that accurate taking into account all local topography, continually updating newly built buildings that would alter local wind flows resulting in varying local rain patterns - would take a computer system so big it would not fit into a standard city building and would literally need sensors on every street corner.
Wyn Nth 2018-Jan12.2(158),Feb264.4(146),Mar217.0(126),Apr64.8(96)YTD557.4(527.3),

#1410903 - 05/03/2017 12:46 Re: Models versus windows [Re: fractonimbus]
Brett Guy Offline
Meteorological Motor Mouth

Registered: 05/10/2010
Posts: 5072
Loc: Bently Park, Cairns
Barring the cyclone prediction at the beginning of the wet our forecasts have been pretty good in FNQ this year. And obviously the cyclone forecast was not model based so doesn't really affect this topic. The last couple of days the rain hasn't been as much as predicted but overall it has been fairly close most of the time.
Then again. I never look at meteye anyway.

Edited by Brett Guy (05/03/2017 12:46)

#1411431 - 09/03/2017 08:45 Re: Models versus windows [Re: fractonimbus]
rainthisway Offline
Weather Freak

Registered: 09/11/2015
Posts: 803
Loc: Mutarnee.....70km north of Tow...
I believe models are a good indication but the problem is conditions can change rapidly sometimes so things dont pan out as expected. For the most part the forecasts are accurate.

Mutarnee, Southern Wet Tropics - 70km North of Townsville and 40km south of Ingham.

MTD (Apr 2018]: 10mm (Ave 136mm)
Mar 2018: 679.9mm (avg 329.6mm)
YTD 2018 1787.3mm (Avg 1952.1mm)


Who's Online
9 registered (bbowen, dormant, kathrynsview, RC, petethemoskeet, Steve O, AndrewM, 2 invisible), 333 Guests and 3 Spiders online.
Key: Admin, Global Mod, Mod
Today's Birthdays
AJB, cyclone steve, daniel01, Gregors, grinspoon
Forum Stats
29543 Members
32 Forums
23882 Topics
1489261 Posts

Max Online: 2925 @ 02/02/2011 22:23
Satellite Image