Another Cyclone

[Fish 0]

Metservice rain forecast models are showing a dart board like feature coming out of New Caledonia and heading towards NZ around the end of the weekend.

Two predict wind models show the same feature, but go either side of NZ.

The other two predict wind models don't show it (those are the in-house models, that don't handle cyclones anyway)

 

The media haven't started bleating on about it yet - the editors probably think there is 'weather disaster fatigue' and wont give it as much coverage as the last two.

 

Anyone actually sussed it properly?

I haven't checked any of the Pacific charts for proper details. Obviously it is too far out for any reliable forecasting of a feature like that.

[Battgirl 23]

They picked Gita hitting the South Island 10 days out. They are very good

[ScottiE 174]

media has picked it up.  All models still show it going well east.

[zk-cessnaguy 1]

this is what we use at work...

 

http://www.usno.navy.mil/NOOC/nmfc-ph/RSS/jtwc/warnings/sh1218.gif

[John B 106]

Its not often you see such wildly divergent forecasts between GFS and ECMWF. Both speed and direction. EC says direct hit on Northland on monday. GFS says well out east and a lot earlier to our latitude too.

 

 A classic in the making .. who is right, the Euros or the North Americans? Roll up roll up place yer bets  please.

 

 I tend to subscribe to GFS on serious stuff .

[Deep Purple 511]

Windy.com has it hitting dead centre on Monday

 

[John B 106]

^  Legend.

I was just trying to do that.  Thats the Euro forecast, have a look at the  GFS for the same  time,  SD.

 Er .. Puff.

[ScottiE 174]

ha - never looked at those.   

Could be a good oportunity to run a wager!

[Deep Purple 511]

Thar

[Fish 0]

I really don't think you can say one forecast or the other is any more accurate at this timeframe out.

By the sheer nature of tropical revolving storms, you can't accurately predict their path with any certainty more than a few (2 or 3) days out. That is the whole thing with them re-curving - it is not predictable.

 

When two or more separate forecasts starting agreeing on a track, that is when the forecasts start becoming more reliable.

 

That said, the in-house Predict Wind forecasts aren't good for cyclones. My logic is that those forecasts were developed and fine tuned for 'mid range' wind, i.e. good sailing breezes. The European and US forecasts are more focused around 'public good' such as predicting server weather events. Already we are seeing those forecasts with more plausible tracks than the in-house Predict Wind ones.

[John B 106]

No , but on monday we'll see . 

 Thats interesting you say that about PWE and PWG .They didn't pick up  the late season cyclones last year very early... Ella / Donna ... one or both of those.

 

 Good ole GFS did.

[DrWatson 374]

So, if the trees on the council land next to my place fall on my shed during this storm... Who pays?

[wheels 543]

 

I really don't think you can say one forecast or the other is any more accurate at this timeframe out.

By the sheer nature of tropical revolving storms, you can't accurately predict their path with any certainty more than a few (2 or 3) days out. That is the whole thing with them re-curving - it is not predictable.

 

When two or more separate forecasts starting agreeing on a track, that is when the forecasts start becoming more reliable.

Exactly right Fish. There are many variables that enter the equation. The computer modelling is only as accurate as the info punched into them. The new Super computers do a great job, but there is still some info that can never be entered because the info can't be "seen" to be able to be entered. As the event comes closer in time, the more accurate and comprehensive the info able to be entered becomes and thus the greater the accuracy.

 

There is actually very little difference between in accuracy between the various prediction programs we can subscribe to. Where the variables actually are, is in what they are actually telling us. So for instance, some will tell us the Sustained wind strength. Others may use peak or average strength. Some use average rain intensity, others the max possible. So the models look different in how thy are presented to us, but are actually ver close to being the same. And mainly this is because they all use the exact same base data.

 

This possible Cyclone I can't see affecting us much. Maybe heavy rain for the Far North. But the current weather os pushing a massive amount of cooler Southerly air up into the Pacific. Cold air is drier and so there is not as much moisture in the Air. A cyclone needs warmth from the Sea and lots of moisture laden air to survive. As it drops toward us, the thing is going to run out of steam I reckon. But that is my amateur view of the situation.

[Fish 0]

Current Metservice situation for Auckland.

This is the first time I've ever seen this, and it is the most accurate forecast they could come up with!

 

It states explicitly that there is uncertainty as to the track of the cyclone. It does say a cyclone is coming and when it is roughly expected. but instead of givign afulse sense of security around its likely track, they are just saying there is 'uncertainty'.

 

Valid to: Midnight Friday 9 Mar 2018

Issued: 11:02am Thursday 8 Mar 2018

Warning: Wind warning Nil

Situation: A low over the upper the region moves away to the east late Friday. A ridge spreads over the North Island Saturday. Cyclone Hola approaches from the north late Sunday into Monday. The track of the cyclone still remains uncertain,but current forecasts expect a southeast track near northeastern parts of the North Island.

[muzled 140]

Would have been much more fun if they'd called it cyclone 'how the f*ck are ya' than just Hola.

[John B 106]

The value to a cruising yotty of seeing which forecast model ends up being accurate a week out is you add it to your quiver of similar observations and develop a trend as to which model is more right than wrong in any given season.

 

 Over the years, some seasons  I've seen metvuw bang on and metservice  either just plain wrong or so conservative they may as well be wrong.

 Then another year , olliversudden, metservice is right and metvuw drags the chain.

Add to that the generic overseas forecasts and start figuring out which models the various outfits are using and apply that to the trend .

 So  I'm not looking at one event , Hola, but I would look at that plus the cyclones that have been plus the ones to come  and  by end of may ,  hopefully be able to conclude one forecast model  is more reliable this year. ( if we were going again this year but lets not discuss that)

 

 Last year as an example  going to and coming back from the Islands for us it was GFS for serious stuff like gales and cyclones, and ECMWF or one of the PW models for day to day  0 to 25 knot type stuff.( and especially if you could get them all to generally agree) Plus local knowledge of course.

[Black Panther 1,581]

Hola is just a friendly greeting. It's a nice cyclone.

[Vorpal Blade 89]

Hola sounds and looks a bit too much like Bola for my liking.

I hope not, Bola kind of had two goes at us - it came down one the west coast a bit then bounced up a bit and then came down the east coast.  I was hiding at Bon Accord watching trees get blown out of the harbour... definitely don't want that again. Although the HUGE waves we got sailing home from Kawau were a bit of fun.

[Black Panther 1,581]

I was at bon accord too. We were in a 22ft variant and came back pregnant. Moral of the story: get a boat with standing headroom.

[wheels 543]

Absolutely nothing like Bola. This one has far too much cold air sucking into it and that cool air will be cooling down the Ocean surface temp around us. It'll fall apart once it gets close to us. Don't get me wrong, we will still get Rain, especially the Far north and some wind, but nothing like Bola. Not even like Gita. In fact, possibly not even as bad as you guys have had today by the looks of the news.