Metservice f@ck up again

[Fogg 427]

They need to be disbanded. TOTAL waste of money.

 

This weekend is yet another example of complete failure of service. Forecast for Auck region was doom-laden wind & rain whereas the reality has been opposite ie mostly sunny and no winds.

 

Anyone who tries to rely on their forecasts for planning purposes simply cannot any longer.

 

No excuses any more they have f@cked up multiple time during the first few months of this year and have now lost all credibility.

 

In the circles I operate in I will now actively campaign and influence for Metservice to be disbanded.

[armchairadmiral 411]

Agreed. Seems they all forecast the absolute worst and the when it doesn't eventuate they are in the clear whereas if their forecast is even marginally under what happens they get well and truly caned by the media. Perhaps you should also campaign for the media to lay off the forecasters  who are,afterall only human. As an afterthought  remember the days of manned lighthouse keepers who used to send in weather updates and how much more reliable that system was !

[Fish 0]

Predict Wind was fairly much spot on.

I spotted a major difference between PW and Metservice early on for this weekend, and followed PW.

 

To be fair, its 'troughy' weather. Classic situation, forecast for blowing dogs of chains with lots of rain, just infront of it is a band of nothing-ness, glassy calm etc. The forecast will still have a gale warning, but until the trough moves (as little as five miles) its glassy calm.

 

The old rule, assess all sources of information, use your own judgement...

 

Relying on one source of information, or specifically just one publicly funded, inherently conservative, forecast is a bit naive really.

 

Oh, and as my missus regularly tells me, 'just look out the window'.

[Kevin McCready 83]

Aleana, Fish is right (except for the irrelevant comment about public funding). I'm happy for MetService to err on the side of caution. Better Safe than Sorry.

[markm 30]

Does also pay to be aware of the details not the headlines.

 

In fairness to Metservice, reading the details they said conditions were conducive to thunder storm formation which is not precisely what most headlines said.    That at least to me translates to "be aware of the possibility" not "there will be".

 

 

Thought they got the details on both the ex cyclones right, ie, they wouldn't be overly horrendous for most places and they certainly nailed the storm that clobbered Auckland a month or so back.

As others have said though, multiple sources is better than one and paying attention to forecasts and how they're changing over time again gives a much better feel than picking on a single forecast and assuming it gospel.

[Fogg 427]

Disagree about better safe than sorry. Because result will be people become conditioned to ignore them so they lack credibility to warn about genuine problems.

 

Better to be accurate than not. And NZ Metservice are not. And getting worse in my view from last decade experience.

 

Most general public (who by the way fund Metservice) won’t have the time or inclination to trawl multiple other weather sources for forecast - that might be a wise choice for a sailor about to embark on a trip - but for land-based general public it is entirely fair and reasonable for them to expect their national weather service to be accurate.

 

But NZ Metservice is not.

 

Even as late as Friday eve it was showing terrible weather for Sat & Sun which turned out to be bollocks.

 

They’re just not up to the job and better to stop deluding the NZ public that they are. Save the money and source our forecast from more reliable European & US forecasters instead.

[Zozza 293]

Accu Weather appear to be more accurate that Met

 

Sometimes I look at Met and it looks like the next five days are hang yourself weather, but on Accu it will be just an odd shower in places and mostly sunny, which often proves to be the case.

 

If Met are erring on the side of caution, then what happened to them missing the strength of that massive blow a month or so back?

 

I'm with the OP...I think they are just a bit useless.

[Fogg 427]

Forget boating. Just general weather forecasting you can’t rely on NZ Metservice. Plus we have NIWA which is a double waste of money because they are now overlapping & competing - but adding no extra value to end users ie ordinary NZ citizens who just want to trust the forecast they see on the TV news or their daily NZ news website.

 

But they can’t.

[wheels 543]

A weather forecast should NEVER be influenced by "safer than sorry" bias. It should be cold hard fact.
It is diabolically hard to predict weather in NZ. But I don't believe it should be that hard, to result in forecasts to be so wrong.
The first point, all the weather predicting sites should drop the 10 day forecast ideas. Apart from the very very ruff view of a low or high or inbetween being in the ruff area of NZ, they are totally a waste of time. 7 Days gives you a little greater accuracy, but still nothing to bank on. 5 day is about as far out as any of them can get with any real accuracy and even then, it is only an idea that NI/SI will be Wet or Dry. 3Days give reasonable good accuracy for individual areas. But that brings in the next issue. What those areas are can be chalk a cheese. For instance, the term "Auckland".  The place is huge and one end of Auckland can have very different than the other. Picton can be cold and miserable and Blenheim fine and sunny. Marlborough covers a huge area of vastly varying topographical difference. (wow that is my big word usage for today We can have rain here where I live, right at the foot of the hills to the Sounds and just a few Kms up the road is fine and sunny. So area's need to be better defined.

[Black Panther 1,581]

I'm always surprised how different it can be north and south Whangaparaoa

[Fish 0]

 (except for the irrelevant comment about public funding).

 

I don't think the public funding aspect is irrelevant - it is key to understanding the purpose of the forecasts.

 

The role of a public forecaster is to forewarn of impending extreme weather events. Floods, cyclones, damaging wind etc. I don't believe the core purpose of publicly funded forecasting is to enable planning of recreational activities in terms of 'will it be nice', but only in terms of safety.

 

If you want to know if it will be 10 knts or 15 knts, so you know if you should run the #1 headsail or the blade jib, don't use Metservice, its not designed for it. Use something like predict wind, which is designed with the accuracy to get that type of decision right.

 

 

They need to be disbanded. TOTAL waste of money.

 

You may be at risk of doing a Jerry Brownlee here. He spat the dummy at GNS (or who ever it was) - the people responsible for warning of Tsunami after the Kaikoura quake. GNS couldn't hit the side of a barn. Turns out they don't fund siesmologiests 24 / 7. The duty siesmologist has to wake up, get out of bed, get their laptop out, log on, decipher a whole lot of highly technical information, and make a decision that could impact the lives of 1,000's, all without even having a coffee.

 

The lesson from that is, to increase warning accuracy, you need to INCREASE funding.

 

I don't know, but I wouldn't be surprise if the NZ Metservice forecasts between 5 pm Friday and 8 am Monday are run out of an office in Chennai. Basically, if you want more accurate forecasts, I suspect Metservice will need more funding, not less.

 

And this business with NIWA (and MetVuw et al) starting to compete with Metservice - that IS a waste of money and needs to be knocked on the head.

 

 

Now, I can see how you could be aggrieved by the forecast for Sunday, it was warm and sunny when it was forecast to be wet and windy - but from a technical point of view, the forecast wasn't that far out. Based on two elements, the forecast weather was not far away, and it arrived not long after forecast.

All afternoon there were dark clouds over Little Barrier, and I couldn't even see Gt Barrier (which I can in clear weather). The rain radar showed serious rain around that area all afternoon. Additionally, the satellite photos showed the position of the weather feature, both infrared and visible.

The forecast weather did hit here about 8:30 pm, roughly 5 hrs late.

As I said earlier, the feature was a trough. The major clue, and it is in this morning's forecast is "A Complex Low". This is forecast code for "we can't tell what this will actually do". Basically the only part of the forecast they got wrong was the timing of the weather.

 

Metservice is very valuable for real time data, rain radar, visible cloud maps etc. I use those to assess what is actually going on.

I really think you are asking too much if you want the little 1 cm square icon on your phone to give an accurate forecast of weather (or any TV headline or icon over your town), especially with the complexity of this current weather. Its mid May and there were ladies sunbathing and kids swimming at the beach I went to, this is not normal weather. Normal blocking high with Sou-Westers dieing out to light and variable, yes you will get accurate forecasts, but with a complex low, a couple of occluded fronts, more than one band of rain, and a trough - no, you can't forecast that. That is why its called a 'forecast' and not a 'statement' or a 'fact'.

[Fish 0]

This is the link to the Auckland rain radar for the last 24 hrs. (Another handy tool Metservice provides for free).

It shows where the rain was, how close it was to Auckland, and when it did actually hit. This shows the forecast wasn't actually that far off, there was plenty of rain about, it was moving towards Auckland, and it did hit Auckland, just the timing was out.

 

http://www.metservice.com/maps-radar/rain-radar/auckland-24hr

[Fogg 427]

just the timing was out.

 

And so if the timing was out then the forecast was wrong!!!

 

Making it useless for people who were trying to plan the weekend (or even the day).

 

I get that NZ is a tough place to forecast for obvious reasons but I'm not criticizing the accuracy of their 5-10 day forecasts. I'm criticizing their inability to forecast the next day accurately.

 

They don't even admit their shortcomings - they get super-defensive which makes critics like me even more angry with them - if they came clean and actually said something like "The next 2-3 days is looking especially unsettled and hard to forecast and so the following forecast only carries a 50% accuracy." that would buy them some understanding.

 

The trouble is they sound equally confident about their predictions regardless of whether we're in the middle of a large, stable, slow-moving settled period or something more rapidly changing.

 

In my world, if you don't know the answer to something confidently then admit it - people will respect that far more than bluffing - and in my view NZ Metservice is guilty of bluffing on a grand scale for years without being held accountable.

[JK 28]

Aleana - if the timing being out constitutes a wrong forecast then you will never be happy with any service. The forecast was for rain to arrive, it did, you cannot expect a "forecast" to be accurate to the hour, let alone the minute.

[Fish 0]

if they came clean and actually said something like "The next 2-3 days is looking especially unsettled and hard to forecast and so the following forecast only carries a 50% accuracy." that would buy them some understanding.

 

The trouble is they sound equally confident about their predictions regardless of whether we're in the middle of a large, stable, slow-moving settled period or something more rapidly changing.

 

I actually agree with that.

And recently they did do that. It was one of those ex-tropical cyclones that came through late summer. They said "the track is currently uncertain, but is expected to go in x direction..."

I thought it was so note worthy that I made a post about it which will be in a thread around here somewhere, about that storm.

 

The best method for assessing forecast accuracy is the Predict Wind method. Four models, if they all agree, you have a high confidence that will happen, if they are all over the place, it is forewarning of low accuracy.

 

Just out of interest Aleana, which particular forecast were you looking at, and on what platform? i.e. the TV weather, the full marine situation and forecast, or some specialised App?

 

And would you be happy to pay for a more accurate forecast?

[John B 106]

Apart from the serious blows, (how did metservice miss that 115knot blast btw) I look out the window,

 

You just need a boat that can handle a solid 30 on the nose and you can go anywhere 90% of the time.

They lost a one hell of a lot of credibility over that didn't they, despite all the bumble speak fudging that went on after.. we did tell you ,we did.....

 

 Weather forecasting seems to go in cycles , some seasons Metservice is bang on, some metvuw.

I've never really used Predictwind locally, they were great to have offshore but even then it was more the GFS and  EC forecasts that  were on target more for any dramatic stuff  than the PW models.

 But that was last year.

[Fogg 427]

Aleana - if the timing being out constitutes a wrong forecast then you will never be happy with any service. The forecast was for rain to arrive, it did, you cannot expect a "forecast" to be accurate to the hour, let alone the minute.

 

Of course forecasts are time-critical otherwise they are pointless!!!

 

A forecast that is hazy / inaccurate on ting is like a broken clock i.e. totally useless but still correct occasionally (twice a day in fact).

 

I'm not expecting it to be accurate to the minute - you know I never said that - I am simply expecting the to be as accurate as they say suggest e.g. they frequently say things like "Heavy rain will appear overnight but will clear by lunchtime." OK, so that's a useful forecast if the timing is as right as Metservice themselves suggest - but frequently it is wrong and the rain might never come or last well beyond lunchtime or something else. And this can be on the day of the forecast itself not 5-10 days apart.

 

And this weekend wasn't just about the rain - they also forecast 20-30kts wind which never came close it was glassy calm much of the time. And exactly the same thing happened the previous weekend when they forecast NW20-30kts by midday Sunday and it never came close at all.

[ScottiE 174]

I think you'll find that their forecasting is far more accurate than you realise.  Your confirmation bias forgets how often they are accurate for a start.  Secondly you are only at one point in space at any given time, whereas the forecast region covers a pretty significant area.  I was up a ridge near Ramarama this weekend and we had some pretty horrendous wind and rain within the time band of the forecast I read on Friday.  

[JK 28]

I'm not going to try and look up the historical forecast for yesterday, but there was rain yesterday morning & last night & the PredictWind site is showing Bean Rock had 20-25kn around 3am Sunday morning & 11pm Sunday night. Admittedly it was fine & 

 

However MetService & the like are providing a forecast, an estimate of how they predict it will play out, and if it is out by 6hrs here or 50km there then that is not much in the grand scheme of things. I'm of the opinion MetService provide the big picture & if you require detail then paying for PredictWind is the answer (although they do offer a limited free option). MetService have a forecast for the "Waitemata Harbour" whereas PredictWind offer different predictions for Kohi, or each side of Musick Point.

 

Also, MetService have Terms of Use on their website, and in accessing the forecast on their site you are presumably agreeing to them - one is that:

"Uncertainty is inherent in weather forecasts............. while MetService uses all reasonable endeavors to ensure the accuracy of information on the website... MetService makes no warranty or representation (express or implied) that this Website, the Services or any of the Information is accurate, complete, current or fit for purpose."

[Fish 0]

What is worse:

A) a nice sunny forecast, take the kids out on the boat, it rains, kids get wet, cold and shitty, it becomes hard work to interest them going on the boat again, or

B.) forecast is for rain and wind, afternoon turns out to be sunny and warm, you take the kids to the beach or park, have a good time and everyone is home for dinner nice and happy?