Route Planning for the Bermuda Leg

Today, Karen and Dean took the kids out to see Chickeniza Pyramids and then to a Cirque du Soliel show for Halloween and I decided to let them do this as a family and stay back and work on some things around the boat as well as catch up on my blog and emails.  It was a nice day just to sit around the pool and relax for a change.

Hanging out not far from the bar but in the shade.  As much as I like the sun, I do notice that I always find a shady spot to sit.  Still, I pretty much had the entire Wyndham resort to myself.  

Yesterday we all went to a sort of an adventure park in the jungle about an hour away from Cancun where we started out repelling down into a underground cavern (Xenote) and then swimming into a landing area where we got to some cliff jumping into a refreshing pool.  From there we did zip lining (again into pools in Xenotes), more cliff diving, kayaking, swimming and just generally got tuckered out.  It was very fun to watch my 1st cousins once removed doing flips off the cliffs into the water at ages 7-10.  The youngest and smallest, Noah is utterly fearless and would flip off a 30' cliff if allowed to.  The other tourists were completely fascinated by them.  I didn't bring my waterproof camera so didn't get any pictures unfortunately.

I also spent a portion of the day running weather and routing analysis for the trip to Bermuda from Annapolis as the skipper (Alan) suggested that he thought Sunday am would be the best time to leave and asked me what I thought.  I figured that this would be a good topic for a blog post and so will explain the tools and techniques I use to plot our course for an extended voyage on the open ocean (it will also help me memorialize them for myself).

First, I typically use a very handy web based app (they also have iPhone version) called Windy.com that gives me a very good perspective on what will happen over the period with winds, gusts, waves, swells (two types), barometric pressure (useful for seeing what will be happening in the short term and how bad things will get in the longer term), temperatures, precipitation, clouds, and thunderstorms among dozens of options.

Probably the best view on weather on the internet.  It allows you to really see what is happening with the weather systems on a global down to a regional basis.  It acts like an iPhone app and you can zoom in/out for more detail.  Windy.com

The first thing you do is select the area you are interested in by panning and zooming in on it.  I try to stay wider than the actual route I'm using as I want to know about anything that might change - especially at the tail end of Hurricane season like we are in now.  I also like to see what's happening off the west coast of Africa as that is typically where most of the big hurricanes are born.  I'm looking at a wind map at the moment with the blue areas showing calm areas, green decent breezes, yellow is good sailing, orange and red are uncomfortable and deep red and dark areas are dangerous.  The legend is along the bottom on the right.  The arrows over all the maritime locations show where the wind is coming from at that location.  Just above the legend, are three selections that allow you to pick and choose which forecast model to use.  In the Caribbean area there are three to choose from.  The North American Model (NAM) which really just covers the continental US out to a few hundred miles offshore, the European Centre for Medium-Range Weather Forecasts (ECMWF), and the Global Forecasting System (GFS) which is a US model for the globe.  I always look at all three but have found that the ECMWF is often the best for predicting weather in the Caribbean (at least as far as predicting hurricane tracks).

At the very bottom of the screen on the left side is a play button with a timeline that allows you to move the slider (or watch it progress automatically after you press play) and see what will happen to the weather over the next 8-10 days dynamically.  This doesn't show what your position will be at each of those times but stay tuned... we have a way to do that.  You can see on this map that there is a Northwester forming off of the east coast of the USA around the New York/Boston area so you know something is up there and it is worth a closer look.  From the lack of winds in the middle of the Atlantic and over the continental US and fairly strong winds over the Great Lakes and maritimes, I'd guess that there are three major systems at work here:

1. A big high pressure zone over the Atlantic (not much of a stretch here as at this time of year there is usually a Mid-Atlantic High that sits between Bermuda and the Azores which drives the Tradewinds from west to east at Caribbean latitudes;
2. Another big high pressure zone over the Continental US, and;
3. A strong low pressure zone over Quebec being squeezed by those two highs (winds always move from high to low pressure - although they get "spun" by the earth's rotation (the Coriolis Effect - more on this in a bit), clockwise around a high pressure zone and counter clockwise around a low pressure (in the Northern Hemisphere, it's backwards down under).  

So let's have a look at the pressure map and see what's really happening here....

Well look at that!  The red zones are high pressure and the light blue/white zone is the low pressure zone.  You can even see a fairly strong trough between the two high pressure systems which is causing the winds to be opposite to each other... more below on that.

The wind vectors (short arrows indicating direction) still show up on this chart but the colours don't mean speed, but rather barometric pressure at that location.  You can see very clearly that the wind flows clockwise around the high pressure zones (which is why when you get two high pressures against each other the winds are in opposite directions) and counter clockwise around the low pressure zones - as well as from high pressure into the low pressure zones (which is really the root cause of the rotation).  Putting it more clearly, the wind is the movement of air from high pressure zones into low pressure zones and it appears to be rotated across the ground by the earth's rotation.

So why does the wind turn around these pressure systems?  Well it really doesn't... "What's that you say?".  Well remember I talked about the Coriolis Effect above? The Coriolis Effect (aka the Coriolis Force) is the apparent deflection of objects (winds, airplanes, and even ocean currents) moving in a straight line relative to the earth's surface.  Hang on a minute - it will get clearer.  As I said, wind is the movement of air from zones of high pressure to zones of low pressure just like when you let the air out of a balloon or a car tire.  The problem is that we aren't actually sitting on a non-moving constant reference plane - the earth we are sitting on is rotating (once a day) as it travels around the sun.  The strength of the effect is proportional the the speed of the Earth's rotation at that latitude - much stronger at the equator than it is at the North Pole.  The easiest way to picture this is to imagine an airplane flying straight north from the equator to the North Pole.  You would think that it would just be a straight line but it actually will appear to be following a curved path when viewed from the ground below.

I won't devolve this into a whole explanation of the root causes of weather but those rules of high to low and clockwise/counterclockwise are worth remembering as you can tell a lot from just those two rules - for instance if you put your back to the wind, the high pressure zone will be to your right and the low to your left (in the northern hemisphere... backwards down under and damn near useless at the equator where there are typically no winds except in squalls - that's why they call those latitudes the doldrums)

Anyway, back to the pressure chart.  This is a sea level chart but to see what is going to happen in the future you should really see the high level (roughly 18,000 foot in altitude) 500 millibar constant pressure charts as surface low pressure zones track parallel to the isoheight (constant height) contours on the 500mb chart.  This is actually how they predict the path of hurricanes.  Surface wind speeds are usually 1/3-1/2 of the winds on the 500mb chart as well so you can learn a lot just by looking at that chart. Unfortunately Windy doesn't show this but it is easy to get from the US National Oceanic Atmospheric Administration (NOAA)

Despite the similar look these lines do NOT mean constant pressure, they are constant ALTITUDE (or height) at a constant pressure of 500mb (1/2 of the earth's atmospheric pressure - roughly 16-22K feet).  They are expressed in metres with the last zero dropped so the 588 line around the H ALTITUDE (not pressure) zone around Cuba means that that line resents that 500millibars of atmospheric pressure will be found at 5880 metres (19K feet).  

Remember that I said that low pressure zones will track parallel to the isoheight contours?  That means that nasty low over Quebec is going to be driven northeast out of our path so the nice high pressure zones will play over our area of sailing (Annapolis to Bermuda) for the next few days at least.... very low risk that it will turn and cause us issues.   So let's play the animation on windy and see if we are right.

That low is well off to the north east now two days later and a new low is developing over the Great Lakes but it is very weak and not likely to be a short term factor.

Yup, two days later, we see the low has moved off north east of Labrador and is no longer a factor and that trough between the two highs has broadened into a very weak low with a new one popping up over the Great Lakes.  Not a lot to worry about here.  Let's run it a bit more.

True to form, that low between the two highs has gotten squeezed and moved along those parallel isoheight lines on the 500Mb chart and is starting to merge with the new low pressure zone over the Great Lakes. 

This is mid day on Sunday when we should be working our way down the Chesapeake and getting a nice push by those clockwise winds around the continental high.  Looks great!  Let's run it a bit more.
(Remember that dark orange on these Pressure charts doesn't mean high winds.. it means high pressure (usually clear skies and not much wind at the centre)

Not a lot of wind and we are very clearly in a good spot of high pressure for the next few days.  There is a low developing over Lake Superior that we want to keep an eye on but it might drive north east if that high behind it manages to squeeze south.  
 

So as we leave the Chesapeake, the winds will die right down and then by Tuesday they are going to be either slightly on our nose or from the port side beam but I suspect we'll be motoring then with the winds so light anyway.  Nothing out there causing big swells so I expect that the seas will flatten as well.  That low concerns me a bit but I really do think that High behind it is going to squeeze down so let's let it run another day and see what happens.

You can see Bermuda as that dot to the just down and right from the centre of the screen.

Well, look at that... that high did squeeze down (pushed by the big high behind it) and push that low north east... glad we aren't in that!   Things look good for us in the middle of the Atlantic we should be about half way from the coast to Bermuda at this point.  It's about 700 miles to Bermuda from the Annapolis area and because I don't ever expect to do a straight line on a sailboat, I'll say 800 miles and so at an average boat speed of say 6 knots, I would expect that our passage would take 5 and a half days from our departure on Sunday morning so lets run the forecast out to Friday afternoon and see what it will be like as we arrive in Bermuda.  The only thing concerning me at this point is that little low between the two highs right over the Great Lakes... it might deepen the trough between the two highs and give us a bit of excitement but let's see...

Just as I thought.

Yup, it did deepen the trough into a very weak low pressure but that should just give us some winds to make it into Bermuda.    So it looks like we are going to have a pretty sweet trip into Bermuda at least using the ECMWF.  I would also run the same scenarios using the GFS but I don't expect much discrepancy here as the patterns look pretty established into the winter scenarios now.

At this point, I'm thinking Alan is right and Sunday am looks very good for departure (if not even earlier to take advantage of the winds) so let's transfer over to our routing software.  I use a free package called qtVLM which was written by some French guy to help the Vendee Globe (those crazy, predominately French, men that sail nonstop around the world on Open 60's by themselves).  It's a bit complicated to understand at first but is very powerful plus they have iPhone/iPad and even Raspberry Pi versions so I can transfer from my Mac to the iPhone and see where we are compared to Alan's forecasting tools - he uses a paid subscription to PredictWind (like I will if I ever get my own Amel).

The qtVLM package is available on Mac and Windows over the internet and quite safe to download at  https://sourceforge.net/projects/qtvlm/. It does have some strange conventions, waypoints are called "Mark" or "Point of Interest", a "Route" is what results from running a "Routing", and a "Pathway" is a blocked off area that you can not sail in.  In practice, all you really need to understand is how to Mark, create routings and get the weather so you can get it to create your routes for you.

After you install it, the first thing to do is create a "Mark" or "POI" (Point of Interest) at each end point of your trip (so I chose Annapolis and Bermuda).  You simply move the mouse over those points (zooming in as necessary), left click and select "New Mark", name it and move on to the next one.  If your system isn't connected to a GPS source then it will ask you if you want to move the boat to the first location so try to select your departure point first.  Then you select (View->Selection Mode and then drag your mouse to expand the selection area over your intended route) the entire area you are interested in (I go a bit wider that that to keep an eye on the edges) and then select some weather and current forecasts to check best route options.

The weather comes in very tight little packages called GRIB files.  GRIB stands for "General Regularly distributed Information in Binary form" and is a WMO (World Meteorological Organisation) standard format for archiving and exchanging gridded data.  The files are sent by a free service called Saildocs which sends them to you when you send an email in the right format.  You download them and then upload them into the program.  I get two forecasts, one GFS (Global Forecasting Service) and the other called RTOFS which provides the current.  You load one into GRIB slot 1 and the other into GRIB slot 2.  You can add a third GRIB slot but I don't bother as it will try to average them all together and I think it is actually worse.

To format the note to get the two GRIB forecasts to Saildocs, you simply select Grib->Grib Saildocs.  You can select the forecasts you want (GFS and RTOFS) and select the features you want to see and the period of the forecast as well as the resolution (I usually use .25 to get the highest resolution if I'm on a good internet connection).  This will open up your mail program and create an email to Saildocs with something like:
GFS:31.00N,40.00N,77.00W>64.00W|0.5,0.5|0,3,6..288|WIND,GUST,PRMSL,WAVES
RTOFS:31.00N,40.00N,77.00W>64.00W|0.08,0.08|0,3,6..192|CURRENT

As you can see, I selected the area of 31N to 40N latitudes and 77W to 64W longitudes and asked for Wind, Gust, Pressure at Mean Sea Level, and Waves from GFS and Current from RTOFS.  Saildocs will immediately send you back an email for each of the two requests with a GRIB file attached.  You simply save that and then point qtVLM at it by loading one into each slot (e.g. Grib->Grib slot 1->Open and then navigate to the appropriate GRIB file you saved).  Again, I typically use slot 1 for the GFS GRIB and slot 2 for the RTOFS GRIB.

Once you load in the GRIB files, you can select View->Zoom on Selection or GRIB File and that will zoom in on the area you selected and show you what the currents and winds will look like at the start of the period you selected.  It should look something like this:

After adding in the GRIB files, you can see the area you have selected and what the weather will look like at the beginning of your requested forecast period.

Now it is time for the fun stuff, if you know exactly when you want to leave, simply select Routings->Create a Routing.  In the dialog you will need to select the two Marks you created.  Typically the first mark asks you to move the boat there but if you happen to be sitting on the boat, it will do that automatically for you.  The second mark is your destination.  You also need to set the departure time and date and then press OK.  It will then create a bunch of turns to follow the winds and course changes required.  You then will likely be presented with a couple of dialog boxes to "optimize" the route which basically gets rid of any extraneous turns and usually results in shortening up the trip somewhat.  At the end, you should have a Route that looks something like this:

The route qtVLM generated based on the winds, waves and currents.  As you can see, we have a fair bit of motoring to do now! (This is the latest forecast I just got this am)

The little symbols beside the sections of the route show the wind direction you can expect at that time/location.  You can zoom in on the route now and look at it in greater detail.

A zoom in on the Chesapeake portion of the generated route.

You can see from this that we will start with the wind on our starboard beam moving to our starboard tail about halfway down the Chesapeake when it dies and we will have to motor (I have set the vessel parameters such that it will have us motor if the boat speed falls below 4 knots and that we motor at an average speed of 6.5 knots).  You can also input the polar charts for the boat (which you should do before you create a routing) by going to Boat->Settings.

If you select the route, it pops up a little dialog box giving you the particulars of that particular route. You can create as many routes (by creating routings) as you want using different forecasts, different start times, changing the boat parameters, etc.  qtVLM also has a neat feature that allows you to create a multi route using steps of days, hours, minutes or wind variations from forecast.  If I am trying to find the best day to leave, I will do that and run them every 12 hours for a few days to see which one I like best.  It is on the Advanced page under Routings->Create a routing.

The popup box you get when you select the route using your mouse.  As you can see, it shows us taking 5 days, 1 hour and 15 minutes to get to Bermuda under these conditions.  The ETA to the point on the route I selected was 1d6h40m.  It also shows the parameters for that particular point on the route including the COG (Course Over Ground), SOG (Speed Over Ground), TWD (True Wind Direction), TWS (True Wind Speed), Apparent Wind Angle & Speed (what you really feel on your face in the boat), Current Direction and Speed, the maximum wind gust expected and any rain.  

Finally, like with Windy.com, you can tell it to play an animation showing the boat moving down the route so you can watch what the conditions will look like at each point along the route.  You do this by selecting Grib->Animate the Grib.  You will see the screen flash as it updates the GRIB files with the requested interval (I usually go for every 3 hours) that you requested in the forecasts.  

Of course all of this is just what the forecasts say, you still need to keep a strong eye on the weather and if you have a satellite phone, you can use it to get updated GRIB files and can rerun the scenarios enroute - which we always do on ocean passages.

Anyway, I hope this wasn't too boring but I wanted to write it all down anyway as I always seem to have to do some relearning every time I go out for one of these long passages and I thought this would help speed me along in that process.  I'm currently at the airport in Cancun awaiting my flight to BWI - Baltimore Washington International where I will get picked up by Alan and we'll drive back to Annapolis where we will drop off the Rental and then take a Uber to the marina (Bert Jabin's Yacht Yard).  Alan's wife Laura will meet us at the dock with the dingy and we'll motor out to Ora Pai as they are on a mooring in front of the Marina.  Once there, we'll load my crap and then take the outboard off and store it in the huge rear lazarette (a big sea locker that spans the rear of the boat and is about chin deep on me) and then sling the dingy on to the aft cabin top and tie it down well for the passage.  Alan's boat has davits on his solar arch but you don't want to be using those on ocean passages as big waves can sweep it right off the davits - although it doesn't look like we will see anything like that here.  For fun I did a multi-route using the latest forecast to see what the different times and conditions would look like.

The top route is the one for November 2 at 8pm, the other two are for November 3 at 8am.  One is optimized and one isn't.

Name	Nov 2 at 8pm	Nov 3 at 8am (opt)	Nov 3 at 8am
Start Date	11/02/2019 20:00	11/03/2019 08:00	11/03/2019 08:00
ETA	11/08/2019 00:10	11/08/2019 09:15	11/08/2019 11:30
Duration	5 d 4 h 10 m	5 d 1 h 15 m	5 d 3 h 30 m
Ortho distance	699.40 NM	699.40 NM	699.40 NM
Sailed Distance	840.05 NM	808.73 NM	826.37 NM
Avg BS	6.56 kts	6.64 kts	6.74 kts
Max BS	8.36 kts	8.37 kts	8.49 kts
Min BS	4.00 kts	4.00 kts	4.00 kts
Avg TWS	8.40 kts	9.59 kts	9.60 kts
Max TWS	17.07 kts	19.55 kts	20.54 kts
Min TWS	1.63 kts	2.25 kts	2.15 kts
Nb Tacks and Gybes	6	25	33
Beating time	1 d 23 h 50 m	2 d 16 h 55 m	2 d 15 h 10 m
Downwind time	0 d 15 h 5 m	0 d 5 h 30 m	0 d 6 h 30 m
Reaching time	2 d 13 h 15 m	2 d 2 h 50 m	2 d 5 h 50 m
Motor time	1 d 19 h 0 m	1 d 10 h 50 m	1 d 10 h 55 m
Night navigation	2 d 21 h 50 m	2 d 21 h 10 m	2 d 22 h 55 m
Under rain navigation	0 d 0 h 0 m	0 d 0 h 0 m	0 d 0 h 0 m
Max waves height	0.00 m	0.00 m	0.00 m
Max combined waves height	1.73 m	1.85 m	1.84 m
Max CAPE	0.00 J/kg	0.00 J/kg	0.00 J/kg
Max Reflectivity	0.00 dBZ	0.00 dBZ	0.00 dBZ

The table above was created by selecting Routes->Route Comparator.  The column showing Nov 3 at 8am (opt) was the one I created and optimized.  The multi route function doesn't optimize.  You can see that the optimizer managed to shave 1:45 off the trip and a bunch of tacks and gybes.    By leaving a bit later we will see stronger winds as we approach Bermuda which gives us a slightly shorter trip and almost 9 hours of motoring.  20 knots really means about 25 knots in real life but that's nothing and should make for a very pleasant sail on an Amel!  Even the waves look very reasonable for this trip (max height of 1.85m = 6').  

I'll update this forecast tomorrow morning just before we go.  They are calling my flight so Fair Winds!