Iridium GO and professional weather routing

[Sos]

Originally Posted by wholybee

After using Predictwind for many years, and the first several of those struggling to get good polars, I have determined that it doesn't matter. More accurate polars do not result in more accurate routing because everything else about the routing has errors as well. Maybe a "less accurate" polar puts you closer to where estimated because the grib was wrong too. That is a 50/50 proposition.

People approach Predictwind the wrong way. They see this nice route and try to follow it, thinking that will be the "best" route, and then when they can't they exclaim it's wrong, and/or struggle to get a better polar. I did this for a long time too. Think about this. One of the most important positions on an ocean racing yacht is the navigator. If *ANY* software came close to giving the best route, the navigator would not be needed. These are million dollar yachts with huge budgets, very complex polars that include every sail combination, lots of testing and tweaking them, and they pay a navigator to not follow the polars.

I did some tests in preparation to leaving on the Pacific Cup(ocean race SF to Hawaii), using both Predictwind, and OpenCPN weather routing. The night before leaving, both were creating a very northern route, adding over 500 miles to the trip, to get better wind. Do I blindly follow it? Well, something missing from both softwares is the ability to draw a route, and say "what if I follow this route?" With a whole lot of work drawing exclusions, adding way points etc. I managed that with both of them. Guess what? Adding 500 miles was FIVE MINUTES faster than the direct course. Added that 500 miles would be dumb, because the grib might be 10% or more off. Or the wind angles 5-10 degrees off. The long route might be slower than predicted, and the short route faster than predicted. The smart course was to ignore the predicted route. And unless you are avoiding a hazard its almost always is best to ignore the predicted route.

That went far in my learning how to actually use Predictwind. Unless you see some significant weather event, day to day you DO NOT follow the predicted route. Ever. The predicted route is probably only minutes faster, and only if both your polar and grib are perfect, and they are not. You sail the wind angles and speeds your boat does best. Even if you have perfect polars (impossible because sea state has such a huge effect on performance) you will be no where near the predicted route after a few days or a week.

IMHO, the most important features of Predictwind are the *textual* reports. A day by day, hour by hour account of windspeed, apparent wind angle, and sea state. And even if you are pretty far off your predicted route, those reports are very close. Because if the grib is off a bit and you are sailing your best course, that will probably move both you and the conditions you see to the new spot.(does that make sense?) When planning, I don't care what the *ROUTE* looks like. I don't care what the route predicts my boatspeed to be, or how long it will take. I look at the text reports, and if by them I am sailing close hauled into a swell, screw it, I'm not going. Or, I look at laying out my own very different route. If the text reports have a swell behind me, and the grib shows no possible storms developing that might later affect me, I go.

The text reports, however, are dependent on creating a route first. So it's there. I do reference it on my voyage (and update it daily). But it is mostly to be ignored. When I was beginning with Predictwind I struggled to follow the course, but now I am seldom anywhere near it.

Thanks for this. Very helpful. It's so easy to look at software output and think it is perfect.
From ECMWF "The weather is a chaotic system. Small errors in the initial conditions of a forecast grow rapidly, and affect predictability. Furthermore, predictability is limited by model errors due to the approximate simulation of atmospheric processes of the state - of- the-art numerical models."

We are all too used to using Google Maps and thinking our boats are the same

[Dockhead]

Originally Posted by stebby

I have an Iridium Go with Predict Wind Offshore professional package for my upcoming North Atlantic crossing on a sailboat. Since it will be our first time crossing the Atlantic, a friend suggested getting a professional weather routing service as well. Does anyone have any opinions/experience on whether this will give me more information than what I will already get from my Predict wind grib files? Do you think having a professional service is really worth the hefty price tag?

Thanks in advance for your input!

Where are you crossing from and to, and what time of year?


You don't need it for East to West in the trade winds.


Could be very helpful if crossing West to East at higher latitudes.


Because weather routers can help you dodge the storm systems which track W to E across the higher latitudes.



Some of them can be nasty.

[Snore]

My comments on the over simplifications and generalities.

Westbound it is handy having good routing. I shaved probably 2 days by riding the curve of a system into the developing Tradewinds last year. Could I have done it without the PW data? Probably. But, having the PW data also let me monitor for late Cape Verdes storms coming off Africa.

As to accuracy of forecasts--PW Pro has EIGHT different weather models. On any given day they will return 3-5 different 48-hour forecasts. There can be such significant variation in forecasts that the full website has a "Validation" option under the Forecasts tab. A prudent mariner would "validate" each model for the intended route. In many areas the GFS is more accurate, others the ECMWF. The models almost always agree on the forecast for the upcoming 24-hour planning period.

On a crossing, a prudent master will download weather twice a day and update routing accordingly. This allows you to react to the consensus forecast, while monitoring the "worst possible scenario" shown in one of the downloaded forecast models.

As to polars? Yes they are critical when racing. But if you are updating your routing twice a day, having polars that reasonably reflect the boat is acceptable.

On a delivery I take the factory polars and degrade them by 15%. This removes the creative aspect from the numbers and is usually close to what we do. Keep in mind on passage, you are smarter to operate the vessel conservatively.

[Roniszoro]

A crossing can be challenging enough, even with all the information you can get. Why not have it all.

Chris Parker "Don't leave port without him!". On a trip to the Caribbean from New England in 2016, a weather bomb appeared out of nowhere, and Chris alerted us via emailed with directions of how to best avoid the worst of it. "Priceless!"

[SV Coronado]

This past fall, I crossed the Atlantic going the other direction, and I used a combo of PredictWind and synoptic charts from NOAA (downloaded via SailDocs). If you're comfortable interpreting weather data, then you'll be fine on your own. If you think Windy provides weather forecasts, you should hire a weather router. However, even if you're skilled at weather routing, hiring a weather router allows you to learn from the pros.

[merrimac]

Here is another approach: share your PredictWind login with a sailor friend that is on-shore or at least has access to excellent internet connections, and is willing to update you.

We do that all the time: when I cross, a friend of mine looks after me, and I've just done this for someone else that crossed the Pacific.

The person on shore can corroborate what you are seeing in PW with limited data: he/she has all models with full GRIB data so can warn you when you are about to miss some development. As it is someone you know, not someone you pay, you can go back and forth using Go SMS messaging without feeling guilty. It's actually quite fun to do this for someone else!

Caveat: you need to know someone willing to do it that has basic weather routing skills and PW skills. You are more likely to know someone like that _after_ your first crossing...

[Seapatience]

Sometimes more opinions are better, sometimes they are worse. I have used both Weather Routing Inc (WRI) and Chris Parker. Both are excellent. In addition to the Predict Wind's routing advice, I would always go with a weather router if crossing an ocean. Too many variables can occur.

[Stmurray62]

I am a long distance delivery captain; on my off shore voyages I always use a professional weather router. I also carry a sat phone. I don’t trust the grub files routing process; would rather have up to date/current data from a professional. Generally, it costs about $300 (flat rate) per trip. I depend on Mr. Dane Clarke (a retired NOAA person) and his wife Jennifer (she is a Gulf Stream Current expert). 240 676 3954. The report/forecast from Dane, usually includes a written narrative and, if relevant, a color and B&W graphic and narrative of the existing Gulf Stream currents, including Cold, Warm and reverse eddys. Usually, I call for updates from time to time, or they just text me with relevant changes/updates.