Nearshore Met Office vs Offshore NOAA - Surfcore: Surf Forecasting and Community

Surfcore»forums»Surf technology and equipment»Nearshore Met Office vs Offshore NOAA

Previous topic Next topic

NOAA data comes from the Wave Watch III global wave model, this is freely available data and can be downloaded by anyone. It is considered to be a cutting edge wave model and is constantly being improved and worked on.

We use NOAA WWIII data to produce our animated charts and it is the same data that is used by Magicseaweed.com, Buoyweather.com and Wetsand.com for forecasting. It's very good for spotting long range weather patterns that might produce good waves - but is limited for near shore forecasts.

The data the we use for our beach forecasts come from a different source - the Met Office. This is a high resolution Near Shore Wave Model that is only run for the UK waters and we're the first and only people to use it for surf forecasting. The only other people are the British Marines who use it to work out if it is safe to do beach landing exercises - Military grade data!

The main difference is that the Near Shore data that we use from the Met Office is much closer to the beach (maximum of 7.4 miles from the beach) than the NOAA data point (max of about 200 miles offshore).

17th Nov Saltburn on Surfcore A screen shot of the Surfcore forecasts

This is a good example, you can see clearly that the pattern of the period arriving tomorrow, is much smoother showing a pulse hitting then dissipating in an expected way.

Saltburn on Magicseaweed 17th Nov

Here on Magicseaweed at Saturday night 6 pm the period of the swell leaps from 4 secs to 12 secs then disappears on Sunday morning. What is happening here is that it is not a forecasts for Saltburn but for a virtual buoy (Buoyweather uses the data more accurately) approximately 200 miles offshore.

With the strong offshore winds - the local sea state created by the wind which has travelled 200 miles offshore - will have developed a 4ft with 4 sec period wave. But this is not what is happening back on the beach.

The Met Office data is a maximum of 12km offshore and therefore is much more accurate in exposing pulses of swell.

Anyway just an interesting observation that I thought you might find useful.

NOAA experiments

Been doing a bit of work today with NOAA data developing a zoomed in version of our animated charts for the UK and thought this demonstrated the resolution of the data quite well.

UK map with NOAA dataEach pixel/square shows a NOAA data point around the UK(from the Culture gallery)

Each square represents a NOAA data point - the actual distance between each point is large which leads to discrepancies when it comes to forecasting local conditions.

A real world example is that you can be looking for some surf - check a beach and it's onshore drive 30 miles down the coast and the wind direction is cross/off. NOAA data would give a blanket forecast for that stretch of coast saying that it was all onshore.

--
sam@surfcore

forecasting

some interesting info on forecasting makes it a bit more clearer to fathom out all the info.find it all very interesting , especially with the conditions here near lands end , nearly allways working somewere being south or north coast if you get it right.a fantastic site by the way its my daily work companion.one thing , was your article on forecasting prompted by it being all a bit wayword over the last two weeks.still the best site about keep it up , taz

The differences

The weather has been all over the place for the last couple of weeks. This post was really prompted by the experiments that we've been doing with the NOAA data while we've been building the animated charts.

I'm going to try to put together some good clear articles with diagrams which will be on the chart pages, so you can quickly reference them, when the information conflicts.

I'm not trying to say don't use the other sites (we link to them because they're useful resources) - I'm just trying to explain the differences so that people get what we're on about.

Hope it doesn't confuse you more :)
--
sam@surfcore

links

more info the better. your links are unselfish and time saving .tide times are a good addition .do you get much feedback of the conditions after its hit the beach to compare with your forecasting? ........confused ,my wife thinks i'm more than confused when i'm sitting on the beach waiting for first light to get in ,disturbed i think she calls it :)

Testing

Hi Taz - sorry I posted a reply to you yesterday - but we had a minor hicup on the server and some stuff was lost (oops).

Funnily enough I actually got some feedback emailed to me about this report that we've talking about today!

Thought you might like to know that your forecast was spot on today! Had 2 sessions with nobody in; just got back from #@!%^&{ with mark and we had it to ourselves (bar some learners in the middle who stayed well away) for 3 hours; like a right hand pointbreak mate, shouler-head on the sets and purrfect! also scored it to myself for half an hour on thurs; proba one of the best sessions I've ever had out there! Better hope everyone keeps checking magicseaweed and not your forecast; I'll tell them all its rubbish just in case!!!!!! ;-)

It looks like I wrote that myself :) - seriously that's from a mate who lives up in Scarborough.

Anyway the system has been in testing for 2 years and we're always tweaking and improving it. We can see the sea from the top of our road (Aberystwyth) we also use web cams and actually go to the beaches as well. We've really stuck our neck out by forecasting for the breaking wave height as no one has done it before and it is calibrated to give the best results when the waves are surfable. When it's a huge onshore storm we expect that most surfers understand and are able to interpret the conditions for themselves.

Well as for being disturbed - I must be disturbed as well - for the first time in a while last weekend I was having porridge at 5:30am and walking across reef in the dark at 6:15am - but we got perfection and it was bloody great! Heaps of waves just me and a mate...must be mad/sane?
--
sam@surfcore

feedback

awesome feedback he sounds stoked . good to know all your work does'nt go to waste ....by the way your forecast for this afternoon breaking wave etc is spot on keep up the good work cheers sam

No brainer

Is there gonna be waves next week?

waves

laugh

no i heard its going to be like a millpool lol

Magicseaweed

Hey Guys,

Just so you know most of what we're doing now revolves around our own nearshore model that operates to a 2km resolution - only time and testing will tell how much more accurate this is, or more to the point how much more useful for surfers (we all know that conditions on the same beach on the same day can change dramatically in an hour as the wind shifts even slightly and the tide shifts from sandbar to sandbar - these little variations are always going to be nearly impossible to model - both the Met Office model and our new model are still very coarse by this standard).

For anyone interested our new nearshore charts are here.

Magicseaweed Nearshore Charts

Best of luck with it,

Ben
magicseaweed.com

P.S. By the way none of our forecasts are from 'virtual buoys' 200 miles offshore as you mention above.

Fight

fight, fight, fight...... heehee

Nesting models?

Magicseaweed hhmmm I think I've heard of that site...... Just joking :) Wow this is great - Ben. You're and expert in forecasting and have been doing it for longer than us this is a fantastic opportunity to discuss the pros and cons of what we do.

As far as I understand any forecasting you do is based on the NOAA WWIII data? The only way to produce higher resolution forecasts is to nest another wave model within the NOAA data set? There are two commonly know models to do this swan or Funwave.

This is a handy little diagram showing what I mean by nested models - www.metocean.co.nz are a small New Zealand marine consultancy who provide the data for www.swellmap.com - another good surf forecasting service doing near shore forecasts. My mate Dave...

Nesting wave modelsFrom www.metocean.co.nz(from the Culture gallery)

Most people who watch the forecasts realise that 5 to 7 days out you can get some pretty outrageous readings as this is the experimental limits of the model, 3-5 days out you can start to get some more accurate readings if there are well defined weather patterns about, 1-3 days usually produces quite accurate results.

Now if this is how you are producing the near shore forecasts -

don't inaccuracies get magnified with each nesting of the model?
So that it is really only useful on the same day or tomorrow? I'm not sure but it seems to make sense to me? Have your experiments shown this to be true?
Do you show any of this Near Shore data in tabular format on the site as that is the only way to have a close look at the results?

If you've written your own NearShore modling software I'll eat my hat :)

This is all very timely as Tony Butt has been in contact with me to gather information about an article that he's writting for Surfers Path titled "Second Generation Forecasting Systems". This discussion could clarify many of the questions that Tony and many surfers have about how modern forecasting is done?

I don't want to challenge you and scare you off - but I do hope we can have a discussion. I'll keep this forum clear of any pointless posting so that we can stay on the subject. Loads to talk about. Gary's alright though... Bit of a rambling post as I had a few Stella's last night, sloppy storm surf at Borth helped a bit.
--
sam@surfcore

nearshore

The complexities of modelling the atmosphere are enormous of course and quite as anyone would expect the longer range the forecast the less accurate it's likely to be. I'd dispute "some pretty outrageous readings" only because they are more likely to be wrong - not more likely to be unrealistic or unusual.

As I've said modelling the atmosphere, winds and swell generation is very very complex - modelling swell propgation is less so. The NWW3 in general performs well validated against the local wave buoys:

http://polar.ncep.noaa.gov/waves/validation.html

The issues you raise with our forecasts are the results of modelling the propogation of these swells close to shore - specifically most shortcomings in UK waters tend to be as a result of the complexity of our coastlines and the shallow coastal shelf and simplifying the data into significant height and dominant period. All the nearshore model does is to anticipate the interaction of these with the incoming swell and make the surfers job of forecasting on the beach conditions easier. There is no reason that nesting one model inside another will magnify errors in the original forecast - it'll just propogate those errors onto the beach and of course the Met Office model you use works in exactly the same way nesting a higher resolution nearshore model inside a coarser global model. The only difference is the models are slightly different and our nearshore model runs at a significantly higher resolution (which is, however, absolutely no guarantee of greater accuracy).

As surfers in a sense ALL we are actually interesting in IS offshore swell. All decent surf producing swells that are relevant to surfers will be showing 100-200 miles offshore (to pick a distance you've used). To focus on these isn't a bad thing - the issues (dominant swells running away from the coast, swell shadowing, swell decay) are ones that a good many surfers experienced with their local breaks can adjust for manually, particularly if they're using the spectral output - the nearshore nested model just makes this even easier and more accessible. Moreover the problems that exist with using gridded model data still exist, albeit some less frequent or obvious, namely:

The data is averaged over a grid square.
The presentation shows only the significant swell height and period (all though the model offers a lot more information if an easy way to present it can be thought up).
The model can just be plain wrong - we can run the nearshore model at 5 minute intervals, to a 1cm precision etc - this additional detail doesn't equate to additional accuracy.
Two beaches next to each other can have different bathymetry, different sand bars, slightly different directions and completelty different breaking waves at the same time on the same day, to model to this degree of accuracy is somewhere between not possible and not necessary anyway for the average surfer.

It'd be good to talk with you and Tony Butt on this and work to both improve both our services and clear up some of your misconceptions about how ours works.

Ben
magicseaweed.com

Life!

Sorry Ben - I've been really busy, my chickens are ill, the weathers been shit and someone picked my car up and rolled it onto it's roof the other day so I got fix that!. But do me a favour just to get into the 'Surfcore' swing of things for a bit of fun - vote on one of the vids for us, as you've got an account. Which one do you like? And I promise that I'll sit down this afternoon and write a proper reply this afternoon :)
--
sam@surfcore

No worries mate :) Best of

No worries mate :) Best of luck with the car and chickens and probably best to catch up over a beer sometime and talk it through like that...

Not much difference...

I can't really spent my full time on the site at the moment. So that's why there has been a bit of a gap.

Ben do you mind putting a bit in your profile about how you ended up doing magicseaweed.com - it might make it easier for people who are coming to the site for the first time to work out what this conversation is about. You can have a read of my profile here. You just click on 'My Account' then 'edit' and then 'In The Real World'. (You'll have to excuse my sense of humour - but just for a little practical joke I put up a member picture for you - you can of course change it in your account, but I think that it's funny!)

I pretty much agree with you on most of the points above - you're right a 2 km resolution is probably going to by only as useful as a model that runs at 12km resolution and at the end of the day it's off shore swell that really concerns surfers. It's just the rarer spots that might work with a very narrow swell window and specific wind conditions - but really if you're a surfer and you've become that anal about predicting the surf you completely missing the point! At the end of the day if you look at a couple of forecasts and it looks alright you should go and check it out, as you never know what little suprise you might find, it's good to get out the house!

The first example right at the top of the forum, demonstrates the near shore model interpreting the local wind conditions more accurately, which is why our forecast was right. We spend most of our time at surfcore thinking about presenting the information in a clear a meaningful way? For example that 'Wave Tracker' application that you developed - it's difficult to present spectral data usefully, I think that you should have persevered with that, have you dropped it completely?
--
sam@surfcore

virtualspectralsateliteboys

Ben tell me more. How's it all work, who's da man/men/ladies at MSW? The site does forecasts for everywhere.. how many people are employed by MSW, is it a franchise, do you have to be good at oceanography/physics/metereology/surfing/programming/disclaimers to qualify for your post, is there a MSW office, are you having a christmas party? (don't really wanna know that last one) It's a massive site... Answers please. Oh and stick a photo up, that one makes you look silly.

For example that 'Wave

For example that 'Wave Tracker' application that you developed - it's difficult to present spectral data usefully, I think that you should have persevered with that, have you dropped it completely?

It's still there, we didn't drop it - just it's more interesting to take the spectral output from the nearshore forecast and build something from that so that's what we'll do next.

No time on the profile thing right now - luckily the photos an almost perfect likeness anyway - what are the odds..... ;)

Ahh you're not too bad

By the way Ben have you noticed that we link to you on all our forecast pages to make it easier for people to use both our sites. We would do this any way because it helps surfers, so this isn't a link trade but perhaps you would consider repaying the compliment on your 'Sites we like' page?
--
sam@surfcore

enjoyable read

well that was an enjoyable conflab you shared with us . kept me from going mad between customers on a saturday when 3 miles up the road praa sands is going off ......that link both ways would be a good idea..