PredictWind Changes How Superyachts Predict Weather

Jon Bilger founded PredictWind, an advanced weather modeling technology that was the official weather forecaster for the 37th America’s Cup, providing athletes and fans with elite-level weather forecasts and observation data. Bilger’s journey to creating cutting-edge technology that he hopes will be used by every member of the superyacht industry started in the same place he hopes to help his platform’s users — the ocean.

Bilger’s love for the ocean began after he followed his father, Jochum “Jock” Bilger’s lead. Jochum participated in the 1972 and 1976 Summer Olympics in The Flying Dutchman, a two-person heavyweight dinghy sailing event, sparking Jon’s interest in the world on water.

“I basically followed my father around during his Olympic campaigns, and one day I overheard him talking to his crew, and the crew said, ‘Jon should do sailing,’ and [my father] said, ‘Well, he’s too young and he can’t afford it,” Jon said. “The next day, I went to him and said, ‘Look, I’ve got $300 saved up. Can you build me a boat?’ That’s sort of the start of my sailing career.”

Jon began sailing in his home country of New Zealand, eventually winning national titles and following in his father’s footsteps to compete in the 1992 Olympics in the Sailing 470 event. Jon then began working for Alinghi, the Swiss America’s Cup team, for 10 years. Jon ran Alinghi’s weather program with some of the best weather forecasting technology at his fingertips.

“That was the basis of me thinking, ‘Imagine if we could actually make this technology available for the average sailor?’” Jon said. 

Although Jon ran Alinghi’s weather program for 10 years, he never had a meteorologist background — he is a sailor and a mechanical engineer. To continue his dream of bringing advanced weather technology to the everyday sailor, Jon had to go out to the marketplace, assemble a team of amazing meteorologists, and connect with different weather stations.

“I quickly realized that 90% of what meteorologists use is the model, so if you have a good suite of models, like we do at PredictWind, and apply a bit of common sense to those models, you are 95% of the way to giving a top meteorologist forecast,” Jon said.

Like any top meteorologist, PredictWind doesn’t rely on one model. Some websites use one to two models or inundate users with a long list of models to impress. Jon’s suite of models uses six carefully picked global models and six high-resolution regional models. PredictWind’s six global models include the European Centre of Medium-Range Weather Forecast’s (ECMWF) model, which consistently rates as the top global weather model and offers record-breaking nine-kilometer resolution; Spire’s model that uses radio frequency to create a 3D profile of atmospheric temperature, pressure, and humidity more 10,000 times per day; the U.K. Meteorological Office’s (UKMO) model; Global Forecast System’s (GFS) popular and free model; and two exclusive models created by PredictWind that offer exceptional short-term accuracy. 

Obtaining data from these different models isn’t cheap, and PredictWind spends a quarter million New Zealand dollars each year for it.

“There’s no point in having a forecast and you don’t know how certain it is,” Jon said. “So, this way, if you got most models showing the same thing, the forecast is pretty confident it’s going to happen.”

While most people think of only precipitation with a weather forecasting program, PredictWind  can accurately forecast wind, gusts, CAPE, waves, air and sea temperature, solunar, clouds, and isobars. PredictWind offers users a daily briefing of user locations as well.

“In the America’s Cup, the coach to New Zealand, Ray Davies, called [the daily briefing] the Jon Bilger forecast,” Jon said. “So, if you asked me what the forecast was, I would get the data and get rid of the outliers and look at the trends. It’s basically converting that data into a sentence.” 

While PredictWind’s models offer elite forecasting, Jon believes they are the pioneers of weather routing as well.

 

“It used to be the domain of top yacht racing navigators and now all of our cruising customers use the weather routing tool — we made it very simple,” Jon said. 

Like Google Maps, users select their start and end location and PredictWind creates the optimal route, taking into consideration speed, comfort, and weather conditions. Knowing what the weather will be in the future is one thing, but knowing how your vessel will react is entirely different. Jon considered this too, and users can record their vessel’s data using PredictWind’s data hub device to create a picture of the boat’s performance. PredictWind takes the length, beam, draft displacement, motion, vertical acceleration, and more into consideration while creating a weather route. 

Looking into the future is what most users look for in PredictWind, but the platform also allows users to investigate past weather as well, making it easy to look at location’s seasonal averages. Users can also use “Local Knowledge,” which is a community-based feature where users can share information on passages, marinas, fishing spots, and more with others. 

“If you go to a new place, you can almost be a local by actually looking at local knowledge and getting some of that information,” Jon said.

Users can share their location and conditions with family and friends hundreds of miles away too. Using their relationship with Spire, PredictWind is also able to pick up AIS signals, relaying to users if a vessel is ever coming their way. 

Jon points to sourcing of high-quality data, a user-friendly interface, accuracy, and a dedicated support team as factors that separate PredictWind from others. PredictWind is supported on all platforms, including Mac, PC, iOS, Android, and can connect to a vessel’s bridge as well. While some captains might rely on forecasting systems they’ve used for years, Jon promises users won’t regret using his platform. 

“We’re happy to give captains a trial of our service, and once you’ve tried it, you’ll never go back,” Jon said. 

Topics: