Facebook

Login

Support Sailonline

If you haven't already - join the SAILONLINE YACHT CLUB!

Please also consider making a donation - all amounts are greatly appreciated!

Board » Technical Discussion » The Weather Research & Forecasting Model (WRF)

Page: 1 2 Next

Posted by
ita10267


Users avatar

Posts
88
Registered at
2008-12-29 10:33:09
2015-02-09 08:28:44 - The Weather Research & Forecasting Model (WRF)
We are proudly introducing the WRF Model as a new source of weather forecast data (GRIBs). WRF is recognized throughout the scientific community as the most accurate and reliable weather forecast model. An extensive (and technical) description of WRF can be found at:

http://www2.mmm.ucar.edu/wrf/users/model.html

The scope of this post, without entering into too much technical details, is to inform the SOL community about the features that this model can introduce and how we can benefit from it.

In SOL we historically use the GFS weather model. According to its definition the GFS model is basically one “of the predominant synoptic scale medium-range models” in general use. Therefore it is good (within its accuracy which is around 30km with the resolution we are currently using) if wind is dictated by pressure/temperature fields. This means that, in shorter races, we rely on a small number of grid points and, in between them, we use interpolation. The result is a smooth wind field that is hardly able to follow coastlines. Of course we miss all the finer features that come within this scale, like funnelling, day/night breeze, swirls etc.

We therefore decided to propose a trial of the WRF model, which is a local weather forecast that takes into account a number of small scale (accuracy in the 10m range can be reached) factors neglected in the GFS model. This is not an exhaustive list but a few of these factors are:

- Landmass elevation profile
- Urbanization
- green fraction
- soil composition (as sand heats differently from clay, concrete etc etc)
- ground heating
- sea heating

So, many new local effects can be visible by SOL users. Of course, for those who are not aware about basic meteorology, there will be no difference in sailing with the old GFS model compared to the new one other than a more “realistic” feeling and less smoothed appearance of the wind field. The experienced sailor, on the other hand, will recognize the patterns he is usually able to see IRL. So, overall, we believe that the realism of SOL can be improved and increased by the new wx source.

We have decided to test drive the new model on a classic run between Boston and Newport. For this purpose a 0.05deg spaced grid will be used. This is 10X more accurate than the most accurate GFS grid we have ever used. As a drawback, given the additional computational time and with the purpose of keeping update times consistent with the GFS model, a shorter maximum time window of only 2.5 days is available for the weather forecast as seen when moving the slider at the bottom of the client.

Please take your time to explore the new wind data and post your feedback here in this thread.
Posted by
bonknhoot


Users avatar

Posts
106
Registered at
2013-08-06 21:35:15
2015-02-09 10:47:52 - Re: The Weather Research & Forecasting Model (WRF)
brilliant! like!!
Clipping along
Posted by
JB


Users avatar

Posts
305
Registered at
2013-03-07 08:54:34
2015-02-09 16:20:21 - Re: The Weather Research & Forecasting Model (WRF)
Viva Andrea.

Absolutely fabulous!
Sail Fair.
Posted by
Vince Nel (WW)


Users avatar

Posts
103
Registered at
2011-11-05 09:30:13
2015-02-09 16:39:02 - Re: The Weather Research & Forecasting Model (WRF)
Smart stuff. Nice. Thanks Andrea
I like to talk a lot... sometimes I even make sense, mostly not...
Posted by
RainbowChaser


Users avatar

Posts
2656
Registered at
2009-06-14 09:44:50
2015-02-09 16:54:48 - Re: The Weather Research & Forecasting Model (WRF)
And here's a quick pic to show you what the difference(s) are and.. how to set them - this pic shows how to chance the arrows, but with this new model the contour-settings will also be worth playing with:




--- Last Edited by RainbowChaser at 2015-02-09 18:44:13 ---
Attachments
Posted by
A2


Users avatar

Posts
347
Registered at
2009-10-26 00:39:42
2015-02-09 22:20:59 - Re: The Weather Research & Forecasting Model (WRF)
Typically with Our normal resolution Wx grids are organized by every half degree of Lon/Lat & High Def by every quarter degree.

With this new model, which sounds even better, will the Wx Grid quadrants be even smaller as well ?
None so blind
Posted by
hmm


Users avatar

Posts
118
Registered at
2011-11-03 13:35:22
2015-02-09 22:35:47 - Re: The Weather Research & Forecasting Model (WRF)
The Boston-Newport HI-RES test uses 0.05 degree grid. We of course need finer resolution for showing the detail.
Posted by
A2


Users avatar

Posts
347
Registered at
2009-10-26 00:39:42
2015-02-09 22:58:01 - Re: The Weather Research & Forecasting Model (WRF)
Well if we can resolve the new model so that our translation displays quadrant boundaries per every 1/8th degree then that would be most welcome.

Also:

My understanding is that the NOAA Wx is extrapolated from a 3 hour time frame. So what is the time resolution of the new model prior to our adaptation ?

--- Last Edited by A2 at 2015-02-09 23:12:24 ---
None so blind
Posted by
ita10267


Users avatar

Posts
88
Registered at
2008-12-29 10:33:09
2015-02-10 07:39:06 - Re: The Weather Research & Forecasting Model (WRF)
We have also higher time resolution with WRF. GRIB frames are spaced 1hr instead of 3hrs. This can help with faster gradients.

@A2: I don't really get your point regarding quadrants. Can you explain?

--- Last Edited by ita10267 at 2015-02-10 07:41:17 ---
Posted by
A2


Users avatar

Posts
347
Registered at
2009-10-26 00:39:42
2015-02-10 14:34:20 - Re: The Weather Research & Forecasting Model (WRF)
I really do appreciate this opportunity to have these types of discussions.

First let me explain that I have little if any real sailing experience and lack the usual vocabulary that would seem natural to those that do. Also I do not employ any routing software to analyze in greater detail weather data as a whole. Let me be clear and state here & now that I have no complaint for those that incorporate such methods as that was the original purpose for the inception of this virtual sailing venue.

So over the time I have spent here using just the information available in the browser client and the process of careful observation along with my attempt to visualize how to plan my navigational strategy I have identified rectangular areas that I call "quadrants" bounded by, most often, half degrees of lon/lat depending on the Wx resolution. These boundaries are where the isotachs have their pointy edges and if you follow either a 270-90 horizontal or 360-180 vertical line that the wind speed will exhibit its maximal or minimal transitions.

--- Last Edited by A2 at 2015-02-10 14:40:31 ---
None so blind

Page: 1 2 Next

Please login to post a reply.

Races

Next Race: 00d 00h 00m

Current Races:

SSANZ Triple Series 2025 - Race 3

Welcome to the grand finale of this year's SSANZ Triple Series! The stage is set for the Lewmar 50, a thrilling 39nm challenge tracing the southern shores of the stunning Waiheke Island, pushing out to Tarahiki before the sprint back to Islington Bay. Expect fast legs, tactical battles, and no shortage of drama as our SOLers take on the waters in their Young 88s, lining up once again against the talented IRL members of the Young 88 Association. With the series on the line and everything still to fight for, it all comes down to this final race!
Race #1934
INFO by brainaid.de
Young 88 PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking:
SSANZ - SYC
Race starts: Aug 29th 21:15 Registration will open soon
▶ Flash
GO TO RACE

Red Eye - Myth of Malham 2025

In 1947, Hugh McLean & Sons of Gourock, Scotland built a radical racing yacht for John Illingworth, soon-to-be Commodore of the RORC, to a design by Laurent Giles. With a masthead rig, giving her ‘free’ sail area per the rules of the time, and short overhangs reducing pitching, Myth of Malham went on to win The Fastnet that year and again in 1949, and to compete in numerous Admiral's Cups. In 1958, in honour of the yacht, the Myth of Malham race was conceived - a mini-Fastnet to the first lighthouse on a rock west of Cowes, Eddystone, rather than the second, to be held in non-Fastnet years. IRL this year it has been raced already, but this stops us not from “Red Eye” racing its online 230nm in Fareast 31Rs, leaving how to deal with the Isle of Wight open to your own discretion.
Race #1847
INFOby brainaid.de
Fareast 31R PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking: SYC - RED
Race starts: Aug 25th 12:00 Registration Open!
▶ Flash
GO TO RACE

Africa by Sea 2025 - Maputo to Mogadishu


The fifth leg of our exploration of Africa By Sea takes us north along the length of the Continent’s succulent sub-tropical and tropical Indian Ocean coast. Much fought over by traders and colonists both well-before and after a Portuguese expedition under the command of Vasco da Gama first sailed there from Europe back in the 16th century, it is yet to be properly explored as a yachting destination. With plenty of islands off- and inshore, small and very large, there is much to interrupt your 1900nm voyage and visit in your Swan 65, from Maputo to Mogadishu! Race #1858
INFOby brainaid.de
Swan 65 PARTICULARS
WX Updates:
0430 / 1030 / 1630 / 2230
Ranking: SYC - ABS
Race starts: Aug 20th 17:00 Registration Closed
▶ Flash
GO TO RACE

Go to race archive

SYC Ranking

  1. Sailonline Yacht Club Member vida
  2. Sailonline Yacht Club Member WRmirekd
  3. Sailonline Yacht Club Member CriticalHippo
  4. Sailonline Yacht Club Member rafa
  5. Sailonline Yacht Club Member KaSToR
  6. Sailonline Yacht Club Member Kipper1258
  7. Sailonline Yacht Club Member Sax747
  8. Sailonline Yacht Club Member BRENTGRAY
  9. Sailonline Yacht Club Member FreyjaUSA
  10. Sailonline Yacht Club Member CollegeFund

View full list

Series

Mobile Client

SYC members have the benefit of access to our mobile/lightweight web client!

The mobile client