Top 10 Wind Forecast Models
HTML-код
- Опубликовано: 11 сен 2024
- In this video we share our top 10 wind forecast models. We discuss the pros and cons of each model, as well as which wind forecast websites you can access them through. We cover global and regional wind forecast models, the resolutions of each as well as their ability to predict thermal wind and sea/lake breezes (useful for kitesurfing, windsurfing or sailing). Enjoy!
____
Gear we use + discount codes: www.ourkitelif...
![Chlöe - Shake (Feat. Jeremih) [Official Visualizer]](http://i.ytimg.com/vi/aDs_1ufpfv8/mqdefault.jpg)
I continue to appreciate the kiting content that isn't just highlight rolls of pros doing tricks!
Thanks, glad you like our videos!
@@OurKiteLife Totally with him on this! Super thanks for taking the time to do it!
After watching this, I decided to visit my local spot when HRRR was the only model predicting strong wind from thermal effect. Wind was perfect and I had a nice session. If I hadn't watched this video, I would have seen on Windfinder that there was no wind today. So, I'd say watching this video was a good use of 10 minutes.
I have been using IGetWind since I saw your video - nice stuff, well researched and laid out. Turns out where I am in summer (Lac St-Jean, Qc) HRDPS/LAM is pretty much dead accurate around 6 to 8 knots, it tends to overestimate below 6 and underestimate above 8. Once we know that, we know if we are likely to go out or not and what to tote around. Bless you guys for that info !
A timely video! I am working on a model which makes predictions at my local kite beach for the next 24 hours at 5 minute intervals. It is updated every 5 minutes. I am using the Weatherflow network. The goal of the project was to use machine learning algorithms to produce a more granular prediction than existing models. Excellent video! Cheers from Santa Clara University
We just got our Tempest - do you have one? If so, what are your thoughts?
@@OurKiteLife The Temptest looks like a great embedded device. I hope to purchase two in the future! The dataset I am using to train the LSTM RNN came directly from WF. I don't know what hardware systems are deployed by WF commercially. I have kited near one of their stations and the sensor does not look ultrasonic like the Temptest so I suspect they are different. A huge hurdle to this research project was the non-uniform time intervals of observations and missing missing observations which required significant interpolation. While this was not ideal in my case it should not impact your use case (getting a wind observation at 11:59 instead of 12:00). Enjoy the new gadget!
Thanks for the quality of your videos! Here in Quebec city, HRRR is definitely the one we check the night before. For 1-2days I use NAM3, NAM12 for 2-3 days and GSF for 3+ days. We don't look much at the prediction in altitude since we use short line and ride on small kites (7-9m) most of the times. HRRR comes in very handy for S-W winds which are powered by a major thermal effect. N-E winds are normally a no brainer, its always good if it stays for more than 1 day.
In Montréal I find hrdps the most accurate igetwind.com
I'm from the Netherlands, and I usually use and compare the GFS and ECWMF for longe range expectations, and the Arome and Harmonie (which is a Dutch bureau of meteorology's short term highj resolution version of the HIRLAM) for the next day or 2 days. They tend to be the most accurate, though the WRF is not too bad either most of the time. One thing I find is that gusts are exaggerated with all the models a lot of the time. ICON is pretty crap imo.
Agreed on gusts. Those the tend to show the worst possible scenarios due to thunderstorms - at least that seems to be the case in our area
Excellent video! I knew a bit about the models but the details you go into and the quality of the presentation is great.
Thanks :) glad you found it useful
As usual, you and your group are far more advanced in your technical understanding of this sport. I use WF-WRF for all my local forecasting in south western Ontario (Windsor area). I'd estimate it correctly predicts our "micro-climate" winds, for the day, 90% of the time. I rarely find that it over estimates wind speed - in fact, there's been many days where its predictions on wind speed have been underestimated
Good to know! For us, WF-WRF tends to over estimate due to the fact that we have doming
Very useful 👍 I didn’t even know all of the different ways to see the forecast..... nice work 🤙
Thanks! Glad you found it useful
Wow, amazing. Came here looking for ocean sailing forecasting but this is a wonderful comparison and explanation of the models and the 40m tip is quite relevant to a yacht with a tall mast.
From the UK, but love to YYZ 😊
Thank you for that first sentence. Making it all clear to people that only ask "which app?"
was looking for the maths models behind them :)) still good to see this high level pic!
I am kiting in the Netherlands and use GFS and Arome on the apps windy and windguru. On the day iitself it is also useful to check out the aviation forecasts of the surrounding airports. Most airports make reliable 24 hour prediction for wind. They call it METAR/TAF. (Meteorologic Aerodrome Report/Terminal Aerodrome Forecast)
I use Predict wind and its excellent. Its the cost of two cups of coffee a month. totally worth it. I am in Australia.
Wow, you guys keep getting better and better. Loved the analysis, the background graphics were awesome and the foreground spokesperson was articulate and crystal clear. I am a proud PATREON!
By the way, I tend to rely on WF-WRF on the subscription version of iKiteSurf. And yes, I have noticed it tends to be optimistic.
Thanks Jeff - we appreciate your support!
I am from the East Coast of Australia, I like using the local ACCESS G and ACCESS C models. I think the ACCESS C is 3km resolution, super helpful in major Aus cities such as Sydney and Brisbane, it is great when forecasting rain, wind changes, and storms. ACCESS G is a country-wide model, that is great for everything across the continent. Great video btw
Thanks!
Thanks for this great summary video. From France I usually look at WRF, ICON and AROME, have to say my personnal 'feeling' is that AROME 1.3km has a tendancy to be optimistics on wind speed, while the 2.5km model is smoother, but this would need to be factually investigated.
from our research, the downside of high resolution models, like AROME 1.3 km, is that they can produce artifacts which don't represent reality. That may explain why it is too optimistic sometimes. TDLR: maximum resolution doesn't always mean better forecast
Thanks, that was a super presentation. I'm in Australia and we just started using a new model in windy which is a local 13km model called Access. it's great!
My method is I’m off weekends so I go windsurfing. If not enough wind i paddle board. 😂
Great info though!
Your content is great. Love to watch some foiling & windsurfing videos, but somehow I watch yours and get more informed. Thanks.
Thanking you very much for the information
Last night a storm hit here in ibiza but all the modules was different
Living in my Sailing boat can be beautiful but some times stressful when to module’s are tell different
Thanks again from John 👍
High-quality content as usual! 👍
As a French kitesurfer I use GFS for long term forecast, WRF and AROME for short term forecast. These help to schedule the next session, but webcam and live wind are the most powerful tool to determine which kite and board I will bring to the spot!
Sounds like you’ve got it figured out! Thanks for sharing :)
i fly hang gliders around the coasts of Cape Town South Africa and found this fascinating . insightful and well explained , you have a new sub :)
Thank you!
I usually use GFS for long term and AROME as soon as data are avaible for my next session.
AROME is just so good !! And I am not saying that because I'm french, I actually learnt in this video that it is a french model haha
Nice - so far all the reviews we've heard for AROME have been very positive!
Great video! thanks for sharing this!
also the WF WRF is best for the great lakes but you are right about it being optimistic.
we use it the most lately
Great video,
I've been using HRRR and the other Windguru models at Longpoint, Lake Erie with limited success on all but the windiest days.
currently we rely most on WF-WRF from Ikitesurf, it works like a charm as long as the skies are clear
Wow thank you so much for putting in the time to make this video. Really well done and great information.
Thank you!
This was a great, and what you said about thermals is dead on in my experience.
Outstanding video! I’m always delighted with your content and good advice... bravo from Italy 🇮🇹
Thanks so much, this one is awesome!!! Best from Munich Daniel
You're welcome - glad you enjoyed it!
Predictwind is the only way to go most accurate and reliable from Australia Cheers😎
Another great video by you guys! I (based in NL) use a paid version of the windy app that has the option to display most of the models together in a graph. Next to that I use Windfinder (paid) super forecast. I am happy with both. I get super frustrated however if none of the models forecast any wind 😆
Yeah - that's the worst, haha
Very helpful 👍👍
Happy to hear that, thanks! :)
Arome particulièrement performant pour la prévision du vent
the real answer is it depends. where you are. when in europe I use different models then when Im at the great lakes. I prefer pessimistic/conservative models since I invest considerable time in my session setup (getting to and from, etc.). thanks for the cool vid.
Forecasts age really quickly, i.e. become obsolete with the passage of time, the best forecasts refresh frequently, ideally within 3h.. resolution is that not as important as local features in gauging what the actual wind will be like once you get there, such as doming (a prevalent problem with Lake Erie) or wind tunneling effect features i.e. Venturi effect (hills, etc.). Each model should be consulted based on its historic accuracy for the spot that you are considering so as to not waste your session time. The Venturi effect alone can make the difference between rigging a 9m kite and a large flysurfer and thus affecting your session quality. When in Europe I'll go to one beach to rig a 9 and another beach if I'm feeling lazy and rig a 14, all because of wind compression around cliffs, etc.
Thanks for the valuable information! Are you willing to share any spots in Ontario that get thermals? We get them in Toronto but inconsistently. Some unverified places we’ve heard of are Kettle Point (Huron) and Beaverton (Simcoe)
I first check wind guru, then windy..
For the region of Munich I have a hit rate of 97% over the last 5Years for Ammersee and Walchensee.. Pretty perfect..
Nice! Sounds like you’ve got it figured out
Btw: for thermal spots the sail plane reports are most times more relevant than the wind reports.. For Germany : www.dwd.de. For austria: austrocontrol ( pilot license needed.. I do have one..) Just my 2 cent..
great review, thanks!
Thanks, glad you liked it!
Thank you,great review. I'm from the Russian and I use Red Windy ( ECMWF, GFS) for long term forecast. I look at the actual information about the wind from sensors online tied in a telegram channel, wind strength, direction and select the appropriate spot for this wind direction.
thanks for this, spared me quite some time 👍 It was never clear to me what all those forecast types meant. Was just about to do some research of my own.
No problem - glad we saved you the research time 😀
Really terrific and informed video, thanks so much.
Thanks 😊
Hi, i use windfinder pro and windy. I live in the Netherlands near the beach. The combination of both apps gives me a pretty accurate prediction
Nice! We also use Windfinder Pro version on our phone
Sailing 25 to 50ft boats in San Francisco, I will not change plans or be disappointed if the forecast is off, so I want ease of use-thus Super Forecast with Windfinder is my choice on a mobile device.
Hello I'm from Mauritius, the wind is kicking very hard in winter. We use ECMWF on windy more often.
Nice, thanks for sharing :)
I am from Norway. Fishing in the Barents sea. Using ECMVF. Finding it usually more accurate than AROME
Thank you for your great synthesis! I mostly kite in France and I am quite happy since I use Arôme from Windy at least for short term decision😎, and still use the other long term models for general planning, or preliminary alert. I have been so often very disappointed by the Forecast of Windfinder (using GFS) and never received any apologize from them!😜. I would highly appreciate that models and apps will give accuracy statistics between actual vs their previous 1 or 2 days (up to 5 days) forecast on key spots where measurement is accurate e.g. airports🤓... Unfortunately none seems to give those data!🥴 May be your influence will push them to include those data😇
Hey Nicolas, unfortunately you can't compare the forecast to actual readings, as most time it wont add up. See our video on why wind blows 14% stronger in US than in Europe.
I saw again the other vid and must admit your great expertise in wind forecast science! 👍. My poorest experiences were on lake surrounded by trees, and close to sea or large field areas, so certainly the zone average was overestimating the actual value on those lakes. If wind is largely over my lightest limit, I am ok with several sails, but when it is borderline, I must admit that I have always been disappointed and wondered why it was always to the lowest 🥴. Now I start to understand! 😇. Wind direction lead probably also to more or less difference between model average and actual on those lake spots! 🤓. Thanks a lot for your great vids and clever answers to the comments 😎
ILTVVM, oh, sorry, that's my acronym for I Liked This Video Very Much......cheers! (Columbia River Gorge)
Haha, thanks!
Maybe you already know Temira's site: thegorgeismygym.com/forecast.
Your video are really interesting and professional. Continue your good work! We appreciate it.
Eric
Quebec City
Thanks Eric 🙂
Thanks a lot for the information.. !! Best content..
Thanks!
Awesome video guys!
Thank you!
I was expecting a quiz at the end hahaha :)
Locally in Vancouver the most important forecast we use is the TAF forecast for CYVR airport. It's a text based forecast updated about 4 times a day by professionals and it's used by all the planes landing and taking off so it's rather important. It only goes out about 36h. You can find Pearson and Hamilton here: wtfbc.ca/taf?station=CYYZ wtfbc.ca/taf?station=CYHM
We also rely on the University of Washington Atmospheric Sciences model that takes GFS and refines it for Washington/Oregon and Southern BC. This model is only any good outside of summer though. wtfbc.ca/gspot.php
On top of that we also use the EC forecasts for Georgia Straight, Howe Sound and Lillooet.
What were you using to visualize the map with WR-WRF for Squamish?
Thanks for sharing! For Squamish we used PWG & PWE from Predict Wind
So good infos. Thanks
FYI, The UK Global Model resolution is 10Km not 17Km. Also, the UK has a 1.5Km model, the UKV, with forecasts updated hourly that cover the N.Sea, the Irish Sea, and the northen part of the Bay of Biscay. Model resolution is always a moving target.:-)
thanks for Info :)
Good job on this vid!
Wow what a great video. I try smashing the like button in super fire rapid presses and it doesn't do more than one like.
Haha - it's kind of like refreshing the forecast multiple times... still the same but there's always hope for more wind :P
I really appreciate this video, I wish there was more weather reading information.
Thanks Danny!
You guys must have a television studio in your house. How do you make such professional productions.
believe it or not, we do it in our living room with a small green screen :)
Nice Video!
Thanks!
Noi usiamo NICO!!!
Piantala pagliaccio
windfinder (superforcast ( GFS + NOAA) + windguru
usually works fine for me
Thanks for sharing!
Cool vid. Is there publicly accessible data to confirm the oft-repeated refrain of "this model has a better resolution therefore its predictions are better"? Every claim in the video and in the comments seem to be nothing but anecdotal. I'm thirsty for some numbers behind the claims. While it makes sense to assume better resolution = better predictions, how better is better? Is it academically better or noticeably/functionally better? Plus, I don't really care how good a model is overall; I only care if it's good for the spot I want to go.
your last sentence is the answer. You care about forecast being good for the spot you go, and that totally depends on the resolution of the model. Watch our previous video in full to understand how weather measurements and forecast work, and it will answer your question ruclips.net/video/HJ2Wp8B8Uns/видео.html
@@OurKiteLife Dude!!! Awesome stuff! Thanks for the reply and the other video reference. I don't kite surf but am fascinated by wind and I don't like to SUP when it's too windy. Thanks again for your efforts.
Thank you for the review. HRW 4km seem to be the most accurate around SFO, available on Windguru
Thanks for sharing!
Also available on igetwind.com
Hello, thanks for detail video. Could you clarify how to open PWE or PWG forecast for Europe on iKitesurf? Unfortunately, I haven't managed how to do it (
I'm curious which is the highest resolution model that does the Caribbean? Specifically Turks and Caicos. Also would be nice to have a link list in the description or blog post summarizing your findings (resolution, coverage area, etc.) for each model. I am also confused which companies offer which models. For example HRRR, you showed 3 different companies in the background, but it's unclear if these are the ones you recommend to acess data from that given model?
It appears that Turks and Caicos fall right on the edge of HRRR and NAM3. If you go to Ikitesurf you can see those two models for Providenciales. We usually look at model maps on ikitesurf, which display most of the discussed models in the video such as gfs, nam3, hrrr, etc... Ikitesurf does not display ECMWF, but it comes with our subscription to predictwind. In the end of the day it is up to you which website you use for wind forecast for any given model, as it depends on personal preference. We use ikitesurf for NAM3, HRRR, WRF-2KM, and Predict Wind for ECMWF, PWG and PWE.
What is the best model for Baja Mex
Always using MEPS@2.5km, the Scandinavian version of AROME, up to +66h and IFS beoynd. Any model can output winds at 40m so I think if it's available or not is more a matter of presentation.
From what we gathered not all of them output 40m, winds as some models output 80m or 100m level winds next. It appears it depends on vertical resolution of the model.
@@OurKiteLife I'd say that all models today has a vertical resolution good enough to interpolate to 40m with good result. However, it's probably true that it's not a typical output level. Thanks for a great video btw!
Question for you all. Where dou you leave bags, your pump, spare gear whilst kitesurfing? I have a bite of a walk from where I park my car and dont want to leave my kite and board. Any suggestions would be great
On the beach, knock on wood nobody steals it
The screen you're reading from is too far from the camera, it looks weird when you don't look at "me" :D
Thanks for the great content!
Thanks for the feedback! We are looking into other options at the moment for this... believe it or not we shot the whole video twice because the first take was even worse than this version! 🤦♀️
I also know a very simple wind forecast application - Cyclono. Try it too
Thanks for the suggestion!
I would be useful to find a model linked to our agenda: good wind when you are available for kiting😎 and no wind when you are too busy🥴 I don’t know why but it seems to be built at the opposite 😜
Haha, so true!
I subscribed to PredictWind for about a year, using PWG/PWE for Vancouver, Squamish and Howe Sound. It was terrible, often overly optimistic by 10 -15 knots.
Thanks for your feedback on PWG/PWE Doug
In south Florida GFS and ECWMF are the worst IMO. My top 3 are NAM 3K, WF-WRF 2K (Both free on the Wind Alert APP) and maybe the most reliable.... The weather channel app! No joke. They don't give gust forecast but the base wind is more accurate than any other I have found.
yeah weather channel and accuweather is a topic of its own, as they combine multiple models, post process and cross reference as a nowcast, meaning that they can actually produce the most accurate forecasts out there.
Windy (ICON) forever and AEMET (ARMONIE-AROME) in Spain
you find ICON accurate? some of the comments we got on this video is that it is not so.
@@OurKiteLife Yes so far, here in mid west of spain is the best, at same level that ARMONIE-AROME (model of the spanish weather prediction organization)
I use an app called windy and I’m in Portugal best wind is in the summer🤟!!
Nice! Windy is a popular one :)
Arome (france)
👏🏽👏🏽👏🏽👏🏽
Im in Argentina 🇦🇷👍🏼
Awesome! What wind forecast model do you use?
WindWilly with data from Australian Bureau of Meteorology
Never heard of WindWilly - will check it out. Thanks for sharing!
@@OurKiteLife www.willyweather.com.au/
Can someone tell me the apps they used?
You mean like windy, wind finder or ikitesurf?
@@OurKiteLife oh okay those are what you used during this video?
@@allisondiaz449 mostly those yes
The HRRR is always way off for the great lakes, wind direction is always wrong, wind speed is way off, much much worse than the NAM 3k. Not sure why. I use the NAM 3k for predicting storms from weather.us significant weather map because of the nice graphical display.
We find HRRR is good for local thermal winds, S or SSW in a 10 knots range, anything else is probably as good as any other model. For instance today all three (WF-WRF, NAM3 and HRRR) were wrong with regards to wind direction as it blew W instead of S
@@OurKiteLife hrrr predicts zero thermals for us. I'm always listening and reading about people using the hrrr but for us it's terrible in every way. Not sure why.
Cabarete, Dom Rep: Windguru and the results are worth less than my dark slide skill...
Wind Alert..............Florida, USA
Thanks!