The Metservice "Forecasting"

[Clipper 346]

It is rubbish. Just Sh*t. Always has been. Will it ever improve? Take the last weekend as an example. Wrong wrong wrong.

 

I look at metvuw and they apear to be much more accurate, aways seem spot on in terms of direction, and much much better at rain prediction.

 

We hauled out the boat this weekend to do the bottom (and some other odd jobs). The metervice was telling me heavy rain all saturday, and showers sunday, even on saunday mornign. Metvuew said no rain on sunday, and was right. such that we got 2 coats of antifouling on.

 

I appreciate that weather prediction is not an exact sicence, but other people seem better at then the metservice. A drunk monkey with a typewriter can produce a better forecats then metservice, and I'm sick and tired of it.

[Guest]

As you know, in Wellington they only need to look out the window to see their forecast is wrong.

[Murky 1]

Agreed, had a good run with Swellmap as well. No rainfall component - which is often very relevant - but gives the wind as a continuous line, which can be an improvement on (or perhaps I should say, a complement to) Metvuw's three slices per day.

 

Gave up on Metservice when I found I was missing out on doing a lot of stuff or taking conservative options on days that proved to be perfectly good for doing it.

 

A drunk monkey with a typewriter

Great quote - did you have a candidate in mind?

[BelowPAR 0]

In a wack way my job (and hobbies) is/are basically reading forecasts . I have a weekly thing that is a compilation of the metvuw charts, turned into numbers on excel. Underthis is a weekly forecast from the guru. Then take the slightly adjusted avg (I favour Metvuw and the guru by different amounts for set windranges Guru is good for light stuff, Metvuw for heavy). In a seperated chart you compare the current data with the adj average. Then its just a matter of crossing off the forecasted values, when they become the current data, and adjusting the forecast to suit. i.e. if the adj avg forecast for today is 13kts NW, and tomoro is 20kts NW, yet we see a trend toward 20kts at lunch time today, then the forecast for tomoro is crossed off, and the next day pulled forward one. Obvious stuff huh?

Well for fri/sat last week the adj avg had the wind touching NNW, then pulling back. At 6.30 after a kite at Orewa, the wind went dead north, not meant to happen untill following morning, then about 1 am it went NW for a spell, then promptly buggered off. Cross out sats forecast (its happend already), Pull in the Sat night and it says nothing, nothing, pressure, nothing.

Common sense, but surely thats what is meant to be done for us?

[khayyam 78]

yup, wind forecasts are crap. i did read an interesting article about it recently, though -- even when there's good money involved (ie for electricity generation, rather than sailing), really good wind forecast models just don't really exist yet:

 

http://www.economist.com/node/16295634

[Guest Dry Reach]

Forecasting in NZ is difficult, even with modelling.

 

as the data is added to it will get better but...

 

with the isolation of our large coastline, that is heavily exposed to the westerly dominant airstreams ,it will alway be difficult to determine how these front's etc come ashore and what they will do before the reach the east coast.

 

I think they do a good job with a highly unperdictable prediciment. and i thibnk you will find they are 90% plus accurate.

 

http://en.wikipedia.org/wiki/Weather_fo ... _forecasts

[Guest]

Wikipedia is not a reliable source of information.

I could change the attached link to say "DR is an ass".

Does this therefore make it true?

[Guest]

"Forecasting in NZ is difficult, even with modelling."

 

Quite the opposite is the case. Forecasting in NZ is a lot easier than most other places as you don't need to worry about surrounding land masses. Something coming over the ocean is much easier to understand than the same thing over an irregular landscape with it associated heating issues.

[Guest Dry Reach]

"Forecasting in NZ is difficult, even with modelling."

 

Quite the opposite is the case. Forecasting in NZ is a lot easier than most other places as you don't need to worry about surrounding land masses. Something coming over the ocean is much easier to understand than the same thing over an irregular landscape with it associated heating issues.

 

 

well the experts say the big, WARM, land mass to the east(aka Aussie) has a rather large effect on our weather (anticyclones) along with our close proximity to another large "COLD" mass called the South Pole (cold fronts)

 

so i would disagree. NZ is very difficult to get right!

[Guest Dry Reach]

Wikipedia is not a reliable source of information.

I could change the attached link to say "DR is an ass".

Does this therefore make it true?

 

 

Well Slacko if you are an expert on Ass's then i will bow to your superior knowledge!

 

and did i say it was a reliable source? or did i use the link and a reference for people to understand weather modeling.

 

far to presumptuous slacko. Now get back to your Keel pouring!

 

by the way. Why would you invest so much time in rebuilding acropp design? are you restoring to original?

[Fogg 427]

I agree that other, offshore weather services seem to predict NZ weather better than the domestic services. With the excpetion of Metvuw. I use Windguru mostly. They even have an app for the iPhone too.

[w44vi 17]

forecast for wellington harbour from sunday was 30kt gusting 40kt at mid day we were out from around 11-4pm biggest gust we got was a little over 20 kts, don't know weather other parts of the harbour may have had stronger wind

[wheels 543]

The real asses are Metservice. They can't get todays weather correct. So why the hell spend millions of dollars on a super computer for long term weather prediction, when they can't get todays weather correct, which is what most NZ'ers are interested in. Metvu is good. Although I find it accurate for 3 days ahead, not so accurate for 5 days and 7 days is an indication of what weather events are out there. Watching those events and observing their trends and directions then give you an indication of what might be appraoching and then at 5 days you can say, we have X coming and then at 3 days we have a pretty good idea that it's going to be such and such and then with a really good accuracy, it is a pretty good chance it would be spot on.

As for Met service, well it is supposed to be miserable and Raining here today. We have a cloudless blue sky day with not a breath of wind.

I understand what DR is trying to say. The smaller local weather events are very hard to predict in NZ. But most of that comes from understanding the local area's. Instead of Metservice spending 12mill on a computer, they should be spending 1mill on understanding the country better and modeling that. Like for instance, what happens in Marlborough with certain weather systems, so that when a southerly is blowing, they would know that it is going to be a fine day here in Marlborough.

And before someone say's, but there new computer will help with that, well actually no it won't. Because that is not what they are going to be using it for.

And my really big question is, when other services can give far better weather info without the aid of a super computer and we as uneducated can use that info to predict our weather far more accurately than the Metservice, then why have we got them at all. My personal view is (and it is personal view and may be incorrect) is that Metservice is working outside their actual roll of Daily weather prediction and are playing in the area of NIWA in trying to understand atmosperic change etc, because it is indulging their own interests. They need to get back to simply working on our immediate weather and leave NIWA to the atmopspheric changes. Now here is a unique thought, why don't they combine the two

[Guest]

what cracks me up is the metservice is never wrong in their eyes. You ask why? Well let me explain. On the old site they (met service) had a tollerance on the wind pridictions of plus or minus 10 knots! so this means if you have 10 kts you could have anything from 0-20kts. Even funnier is if they fore cast 5 kts variable this could be anything from a 5kts vacuum to 15kts blow.

 

I cannot find any tollerances on the new site but I could be blind.

 

localy I use a combination of met vuw, BOM austaralia and personal instinct.

 

happy days all.

[Grinna 2]

Forecasting in NZ is difficult, even with modelling

 

Well, how come MetVUW can get it right(er) while MetService struggle so badly then? MetVUW use a model ... its just its a different model from the MetService one. Wind Guru use a model too. Hell, Ken Ring has a better success rate than MetService and he predicts the weather conditions up to a year in advance based on the moon.

 

The issue isn't whether you model the data to make predictions or not, because they all do. The issue is whether or not the predictions you make are worth a tin of fish or not. Predicting the weather in NZ is not easy and the further out you predict the hairier the predictions become. We all accept that. The point is, weather prediction can be done by smaller businesses a LOT better than it can be done by our "premier" weather institute .... the one that's funded by taxpayer money.

[BelowPAR 0]

At any windstrength you can expect gusts up to 50% of the wind strength. Metservice account for this, but they make the error in treating it as 50% of the windspeed, not converting to the square relationship. If someone would point this out...

 

They even explain it on their site somewhere.

[Murky 1]

Still the same challenges apply to all forecasting services and yet Metservice would be more consistently over-reading in my experience (their prediction is often approaching double the actual conditions), which can be the difference between going out or not.

[wheels 543]

At any windstrength you can expect gusts up to 50% of the wind strength. Metservice account for this, but they make the error in treating it as 50% of the windspeed, not converting to the square relationship. If someone would point this out...

 

They even explain it on their site somewhere.

In other words, they take an educated guess and then apply a very wide fudgefactor margin, while placing the fineprint in an obscure place so as their Butt remains covered.

[Guest Dry Reach]

forecast for wellington harbour from sunday was 30kt gusting 40kt at mid day we were out from around 11-4pm biggest gust we got was a little over 20 kts, don't know weather other parts of the harbour may have had stronger wind

 

 

No forecast is given for wellington harbour only!

 

the forcast is for Wellington Harbour from sinclair head to terakarie Heaqd. so yes the harbour may have been peaking at 20 but the south coast would/could peak at 35-40.

 

they dont do forecasts for specific parts of a harbour or "where you are sailing at the time.

 

But yes Metservice are less accurate than other such as wind guru or VHFChannel 23...

 

which also breaks down the "live" wind for Lyall Bay, Brothers, Karori Light, mana, and stephens Island,. From memory.

 

So sail (listen to this prior to going out) with this channel on and get updated every 5-1o minutes!

 

They also do 3 and 5 day forecasts which are very accurate or...

 

 

buy a portable VHF for about $130 (cobra)

[w44vi 17]

No forecast is given for wellington harbour only!

 

recreational marine forecast fore wellington harbour

 

http://www.metservice.com/marine/recrea ... wellington