The Offshore Voyaging Reference Site

The Loss of “Team Vestas Wind”

A huge amount of internet ink has been spilled about the shipwreck of the Volvo 65 Team Vestas Wind. But the significant news amongst all of that blather and second guessing is that the navigator Wouter Verbraak has taken complete responsibility for a terrible mistake.

There is also this useful piece that postulates that Verbraak made the classic error of not examining his route at a large enough magnification, and thereby missed the shallow water in their path. In my opinion, that’s probably exactly what happened, particularly since Verbraak himself linked to it.

To me there are three things we can all learn from this accident:

  • There but for the grace of a higher power, or luck (depending on how you look at the world), go any of us.
  • The magnification error is a constant source of danger that can trap even the best navigators and we must all guard against it.
  • Electronic data representation has an intrinsic danger: all of us tend to ascribe a higher level of accuracy to the underlying data than is justified.

Let’s look at each of these lessons in more detail:


Login to continue reading (scroll down)

45 Comments
Oldest
Newest
Inline Feedbacks
View all comments
Toby

The concept of chart detail being deprived by zoom level is known as ‘Scamin’. At least thats the term used at both Taunton Hydrographic and the Royal Navy. As a Specialist Navigator RN its certainly one of the key checks made before any navigation, offshore, coastal or pilotage is approved (by command). Every single route must be checked at a zoom scale of 1:1 in order that Scamin of data does not allow us humans to miss key charted information.

Toby

The electronic navigation terminals on a warship have inbuilt protocols that have as yet not filtered to the pleasure market.

A route or navtrack created by the navigator is not completed for navigation until it has been scanned. This is an electronic process whereby the terminal scans every detail on that electronic chart to a 1:1 detail in every chart availiable to the terminal be it an ENC, RASTER or a simple reproduction of a paper Admiralty. The scan of the navtrack is completed to a pre-set corridor which is generally the distance a junior watchkeeper may leave said route without calling Command (2 nm). The scan takes minutes an produces a list of every detail on a 1:1 scale the computer dezignates as a hazard. The navigator must then manually accept each and every one in tern before the route is approved. The Commanding Officer also signs off on the scanned route before it may be used for navigation.

As yet this doesnt seem to have filtered down to pleasure craft electronics (to my knowledge) but could provide that safety net we all need from time to time.

My wife and I love your articles, we keenly await the next. Thank you and please keep them coming.

Rob Gill

Hi Toby, interesting comments thanks. One software application that is available for private users now is weather routing, which will take the weather date for an area and provide an optimal route given the wind, waves and swell predicted, also taking into account the land.
But does this weather based routing software takes account of all land masses or just the big ones? Did auto weather routing software play a part in the Vesta grounding, following a route change to avoid the bad weather?
Is there a danger that software developers will be plotting our safe courses, not navigators?

Marc Dacey

At this level of racing, my impression is that routing software is used extensively in order to maximize the efficiency of the helming, in which the autopilot plays a role. I would like to know if, as has been the case in other head-scratching incidents of this type, whether the AP and GPS were “slaved” together to drive the boat to a waypoint. I prefer personally to drive the AP to a compass bearing and to chart a hypothetical waypoint, say, 10 NM down that course. If you don’t end up (by GPS) where you thought you should, factors of current and drift and even the boat making lee due to sail balance are revealed. There might be elements of those issues in this crash, assuming it just wasn’t about rolling the navigational dice and losing. The race committee has to bear some responsibility, however,as with such fast boats, I think it was imprudent, bad weather or not, to route the race so closely to a reef with known ship-wrecking potential.

Marc Dacey

Yes, I realized that after I posted it…whoops. That does bring up the point, however, of the importance of adequate offing, given that even the pro-level helmers would (as has been said) have trouble steering to closer than five degrees to a set course. That makes the consideration of offing, which is only prudent distance off a known hazard/shore, even more important, as APs helm better than humans for the most part.

Rikki

Cut to the bone… any sailors worst nightmare.
I would put it down to inferior tools of the trade. Reliance on electronic charts brings with it an understanding of their shortcomings. Non navigate-able areas should be apparent on all zoom levels. Dah…
An indication that would incourage closer scrutiny. It is long overdue that the maritime authorities draw up some guidelines.

This reef shows up on the chart on Virtual Race game, zoomed way out. This is not a game for a lot of people. Let’s make sure they have the proper tools.

Navigators today are aware of the lack of detail and should scrutinize the area in their path. Not always an easy task in view of sudden unplanned course changes. Jibes and tacks means course alterations of 90° at any given time. A very difficult task trying to foresee your path at max zoom.
Routines on board should include the skipper in the task of navigation in the traditional sense seeing that the job of the “navigator” is often saturated with tactics involving other boats and weather routing. A task which often leaves them dazed, blind and exhausted.
Watch captains should visit the nav station before their watch for an overview of their situation ahead. In a perfect world this reef should have been spotted. In a perfect world non navigate-able areas should be apparent on lower zoom levels.
This fact has been the frustration of all who use electronic charts as a tool. About time for the industry to address the issue, and do something about it.

Shit happens.. in this case it couldn’t have happened to a nicer more competent guy. In the spirit of his response, I hope lessons will be learned on all (zoom) levels.

Jim Patek

John
Coincidentally, I had just exchanged emails with an Australian sailing friend about the incident relating to him, I hate to admit in such a public way, the number of near misses (and impacts) I have had while sailing in poorly charted waters or due to just plain lack of attention to detail (or in one case, failure to have my reading glasses on my face). So, obviously, I was nodding in full agreement with your conclusions.

I wish that I could give C-Map credit for the accuracy of their charts in the islands of the South Pacific but you cannot trust them…..ever. If approaching any island and associated reef at night, you need radar to confirm your position relative to the chart.

mark

i d say the boat designer is to blame. a plotter behind the wheels and one at the front of the cockpit. i have a couple at the front of mine, one with the screen split to see up close and far for the whole crew to see. silly me but i don’t like to take chances like real sailors do.

Ben Tucker

Thanks John for writing a nice and balanced take on the vestas incident. I too have had a fright with SCAMIN scaring the crap out of me (SCAMINed!)… About 50 miles SSE of Warnambool, Victoria (Australia) is an oil rig. it doesn’t show up on navionics on a normal scale. you have to zoom in a fair way to reveal it. Anyway needless to say, I didn’t expect any nav dangers, and had not properly validated the route. And I got a nasty surprise when it popped up directly ahead of me! Just by chance I zoomed in at exactly the right time as we where just crossing the shelf and I wanted to get a feel for the contours. up popped the well about 10 miles ahead on the plotter (AKA Iphone), giving me quite a shock. The old unupdated charts I had borrowed also didn’t show it, predictably. Hrmm.

In Vestas case I must admit to being very surprised that such a large chunk of land and reefs was SCAMINed out. There is a video somewhere that shows the reef popping up when the scale is dropped. I do think this is a good wakeup call to everybody, and ultimately the lessons learnt from it will in the future save lives, plus improve our systems, both procedurally and technology wise.

Anyway, a couple of ways I use the electronics. I always set an appropriate safety depth ie 20m offshore, this helps highlight any shallow water, and on many systems alarms can be set to warn if you are about to stray inside any preset limits. This isn’t foolproof, I notice my navionics charts don’t show all shallow water on all scales, but it helps highlight the shallow areas at a glance. I also always run trails. This shows up multipath errors or other sudden short term errors very quickly, and also show any change in COG it a glance. It can also be followed back if needed, eg MOB, or escape routes. If I have radar I regularly overlay or compare distances to check datum or GPS accuracy, and check depths as well. And I often check HDOP and sat strengths. RAIM, HDOP or EPE alarms are handy. What happens if the GPS looses signal? Disconnect the antenna and see. Does a nice loud alarm sound as it should? Some systems just stop, or even worse quietly revert to DR mode. Scary!

On charts, the Echarts we used in the beagle channel had datum errors up to 3 miles out in some places! The peninsula was also very erratic. We used waypoints that had been previously marked. And radar overlay to ‘adjust’ the datum. sometimes we had to ‘adjust’ it every ten minutes or so, and once we ran out of adjustment. I think there’s alot of benefit to having two separate types of Echart. I notice CMaps seems to have better SCAMIN settings than Navionics. But even the proper approved ENC charts aren’t perfect. Anyway I have certainly been more carefull since my own little wakeup call, and I guess thats a good thing.

Cheers

Ben

Ben Tucker

Hi John, On the ships we always had a formal Night orders book that was signed by each of us watchkeepers. Similar to your written limits. I also use a similar system for anchor watches, with set limits of wind speed, direction and ice to be called. So far its worked well for me, with untrained crew (and passengers) being able to be very useful.

But the safety depth feature I was referring to was the ability of most modern vector based electronic chart systems to show depths of less than the safety depth (or whatever jargon they use) as a blue color, and leave the Safe deeper depths white. So offshore I’ll usually set it to 20 meters (the most my Navionics will allow). This is a safe depth to avoid breakers in most normal swells. As I approach land I’ll wind it back to 10 meters and in harbours maybe 5m or less. I find it very useful, but like everything it has limitations and still wont show isolated wreaks or rocks, which can be scamined out. The Proper ENC charts (with ECDIS) enable very precise safety depths to be set, much better than my experiences with CMAPS and Navionics.

Marc Dacey

That situational “debriefing” at the turn of the watch happens regularly on some boats and not often enough on others. It’s an excellent habit to recap the game plan with the watchstander and to mention little things like “there’s an occasional weird tapping noise under that locker…let me know if you hear it” as trivial stuff can become important quite soon and while many people still keep a watch log, there isn’t always the opportunity to read it if one must helm by hand.

Dick Stevenson

John, nice article and some very interesting comments. Most of the comments seem to assume one navigator, and I suspect that is the way it goes on many/most boats. On Alchemy, I always ask Ginger (and vice versa) to check out the proposed route. This we attempt to do cold, not sharing our assumptions ( Ie., I know the west side of xyz island is safe) until after the work has been independently done. This is harder than it seems and sometimes feels contrived, but has held us in good stead.
Dick Stevenson, s/v Alchemy

Stedem Wood

We carry many, and even different-sized tools because each has a limitation. It’s up to us to choose the right one, to understand how to use it and what it can’t do.

I learned about electronic chart’s “scamming” issue simply through using them. I was horrified after hundreds of miles to learn what was happening at different scales. Since then, I too am adamant about plotting and scanning specific routes with waypoints and setting limits for cross track errors. At least twice a day I check 36 hours ahead of travel at a scale that I know will show hazards. I make a specific plan if passing too close to “The hard bits.” Radar is great, but won’t always show what you need to avoid, so I leave lots of room.

I use several electronic charting systems. iNavX, MacEnc, Furuno, and Maxsea. It’s different on different systems. iNavX on iPad has to be under a 10 mile scale before bottom contours and hazards show up. (By the way, different chart groups don’t overlap when buying contiguous chart packages from X-Traverse for iNavX. You can buy New Zealand, and Australia and Pacific Island packages and the charts are great, but if you’re navigating from one to the next, there are holes in coverage from one chart “package” to the next. I asked and was told there was “nothing in that part of the ocean to worry about.” No kidding!)

For landfall, especially when unfamiliar, three systems have to be up and running and I compare each to the other and to what I can see. In some areas they’re all spot on, in others, one is, and in some places the one that was good earlier is off, and another is closer to reality. For reef entrances and tight spots, it has to be eyeball, sounder and a mental offset when the charts are obviously wrong. Sometimes it’s best to just punt and ask if someone can help you in. Going very slowly, I nearly crawled right up a reef until I could see the entrance was 150 yards (meters) from where the GPS put it.

In that case the extra certainty of the electronic plotter might have put me closer to a mistake than I would have made with paper, but probably not, there were no references for normal plotting. Seas were “flat-calm” and a 30 mile atoll, one or two feet under the surface was detectable only where the decades-old wrecks were still visible.

Because I’m so reliant on electronic navigation, I carry two iPads, two laptops, three “hockey puck” USB GPS’s, plus the boat’s integrated navigation system, all with appropriate charts. In addition to relying on the gadgets, we have to rely on a means to charge them, so I have small USB charging batteries (toped up each trip) small, portable stand-alone solar panels, small portable inverters, 12-volt USB power supplies and a 2KW gas generator as back ups to the boat’s several systems for charging. I’m considering another iPad with charts as additional backup.

Gadgets alone will not make you safe, I have landfall checklists, and, though I often travel alone, I have watch-change checklists that I use whether I’m handing off the watch, or not. Check list items include route and 36-hour route check, alarms for cross track, AIS and radar targets, navigation and other boat’s systems checks.

Since I’m the system that has to integrate it all, I have personal checklists for food, water, sleep schedule, etc.

We have more information than ever before. Just considering charts, I know I have more coverage than I ever carried 24 years ago when using the paper version. I previously purchased Admiralty charts and pilots, but knew I hadn’t purchased every harbor and every large, or even small-scale chart.

The only reason we’re using the new systems is that they are easier, promise more information and present it in a more useful manner. Who would ever go back to carrying a mountain of government conversion tables for tide and current information when you can push a button from your bunk and see more almanac information about tides, currents, twilight, moon phase, SHA, sight reduction (I’m tempted to ask who uses those, but I can predict the answer I’d get) than you would ever spend the time to interpret before? Cost might be a consideration, but paper charts are extremely expensive now and often only available on a print to order basis.

By the way, I do have two sextants and the tables, but I’d use the iPad to do the reductions these days, even if I ever get around to unearthing either of them.

The trouble is, the new systems also come with new ways to mess it up. Our job is still the same, though. —–Not to do that.

Stedem Wood
M/V Atlantis

PS

—John, I’m with you, I miss the fun of dividers, parallel rules, hand-bearing compasses and the satisfaction of plotting all that stuff on fine paper charts, I really do. I have all of those tools on board, I just don’t have the fine paper charts to do it on.—

Svein Lamark

Hi John, I think you have made an interesting point here. It is not only possible to do the scale reduction mistake (scamin), several common navigation programs has functions that makes it possible to remove details from the map. If you look at one of the most common navigation programs (it is free to download), the openCPN, you will see that it has many opportunities to remove details from the screen. A ship with two professional captains on the bridge (both of them sober) wrecked in South Norway because of this mistake: They had removed several sea marks form the map and forgotten about it. Removal of details can also be a problem to the next watch captain if he does not know about it. Even with the correct scale he can not see all details. Even Tobys strict procedure could not prevent this mistake. When approaching land I prefer to navigate on a German system (Chart Navigator Pro) with NV-digital maps. On this system the scale- or detail-mistake should not happened. I also have the correct paper chart at hand. A paper map does often give me a better understanding of the land scape ahead. Another trick is to study a satellite photo of the landing area before arriving. And in Norway I use the Norwegian Cruising Guide. It has helped me several times.

Richard Marsh

I think that the points about having information available to the helmsman in the cockpit is a good one. Perhaps we should aim to have a display of the electronic chart, at a large enough scale to show all dangers, visible to the helmsman even on a long ocean passage where no problems are anticipated. It is unrealistic to expect the navigator to take all the responsibility.

scott flanders

Anyone who has traveled in the South Pacific know the charts aren’t accurate. Most are at least 200 meters off up to a mile in less commercial areas. For example, C-Map doesn’t show the Duke of Gloucester Islands north of the Gambier Island Group. There is a notation written in script well off the position but the DofG islands are a very real N/S chain with a few inhabited islands.

Departing Tonga for Opua, NZ as hundred of boats do a year, we drew a single line to the head pin off Opua. I checked the route and there were no obstacles. My wife zoomed down and came across Ada Island exactly where the red line crossed. We left the route stand and one morning at daybreak there was Ada Island. No person I have talked to before or since has heard of Ada Island but I promise it is a vertical volcanic rock with no beach. I wonder how many cruisers over the years have happened into Ada? Scary.

S.

Ben Tucker

Ada or Ata? remember heaving to under Ata on my way north many year ago, certainly a foreboding chunk of land right on the rhumb line, but for us it was a welcome sight of green and some flattish water after a long trip north in a 28 footer many years ago, thanks for digging up that memory . Interestingly Ata pops right up on my old version of CMaps, but Minerva Reef is nowhere to be seen?

Myles Powers

According to the latest software release notes for the Simrad NSS (July 2014) multifunction displays, the capability to automatically reroute a course based on chart data depth information is available. However, for some reason this capability is deactivated for US versions of the equipment. Most likely this is a corporate fear of liability issue.

richard s. (s/v lakota)

had a rueful chuckle reading this as just the other day making for anegada from n gorda sound in near perfect conditions i was headed a little upwind of my entrance waypoint to abide by that rule of thumb…i know there is a reef upwind from the waypoint so as i approached i kept falling off gradually not only to ultimately make my waypoint but also to stay clear of the reef which wasn’t showing on my plotter, but i know it’s there…suddenly i saw the small breakers a little closer starboard than i had expected…then i saw breakers even closer prompting more falliing off even though there was no real danger…a few mins later i zoomed in the plotter which then revealed the reef showing my track skirting right along side of it…that was a bit of a surprise esp as i thought i was doing good staying upwind of my destination waypoint…overall i think i was acting correctly for the most part…just needed a little more ‘correctly’…for me this is a timely post, and yes i had a bit of luck on my side with this…what a joy these last couple of days have been here in anegada

richard (home based in tampa bay)

Svein Lamark

Hi John, your post has trigged me to do some more thinking on the problems of zooming in and out on the scale of an electronic chart system. Recently I nearly did the same mistake as Verbraak, but in a slow cruising yacht. The low speed saved me. This yacht has a modern C-map type CM93 version 3. Most yachts have the older type of C-map called CM93 version 2. I find version 2 easier to read and understand because it has less details. Of course a more detailed map is better, but sometimes harder to read. This made me zoom into a smaller scale with less details and made me overlook a rock. When a super navigator as Verbraak can do this mistake, we ordinary sailors sure can do it. Our advantage is however that we have more time and can study the map longer. I guess that soon the CM93v2 will be replaced by the more difficult CM93v3 also in yachts using C-map. This can be a challenge to the navigator. I have seen that the German map company NV-charts use a critical team of old navy captains to discuss new maps before printing. This captains are used to sail war ships at high speed trough narrows. Then the map most be easy to read and understand. This will reduce the Scamin-risk. A map is only a generalisation of reality and more details can be helpful to the navigator. But the presentation of the details is important.

Svein Lamark

Hi John, I agree 100% with you. A friend of mine who is a flag captain of a large cruise line told me the same experience as your friend. He was entering the archipelago of Stockholm which has about 100 000 islands and he could not understand the map. He also used his private map to understand the very complicated land scape of Stockholm.
Rosepoint has a good navigation program called Coastal Explorer. It can be downloaded free for 14 days. If you download it and install the C-map CM93version 3 of Stockholm (which is easy at hand in the test version), you can see an example on how difficult navigation can become. My friend the flag captain is famed among professionals for his ability to navigate in difficult waters. When professionals like this are in trouble, what about us yachters?

Carl Stutzner

Electronic charts are an immense help. Sailing here in the Great Lakes and navigating through marked channels I’ve found that when you navigate these channels to not solely rely on them. Numerous times I’ve found where – when zoomed in – the buoys are located in the middle of the channel! By sight – common sense prevails – electronic charts are not infallible. By far I still use paper charts as they give me an idea of what to expect.

Richard Dykiel

Once again you did a public service to our community. Thank you for this post, your balanced analysis, and attracting our attention on the “scamin” issue. I learned navigating in the pre-GPS era, and confess that I let the use of the chart plotter erode those traditional skills. But in my mostly coastal sailing I’m still having a paper chart spread out and still doing my route planning on paper, if only because the chart offers more “breadth” in your vision of the area (I find that screens are always too small). I’m wondering how many of the boats day-sailing around me still carry a hand bearing compass?

Westbrook

From an article in the New Yorker: “At a press conference on Monday, Team Vestas admitted that the crash appears to have been caused by “a simple human error”: the navigator, Wouter Verbraak, did not zoom in enough on his charts. Race organizers have pointed out that it was a high-stress situation caused by approaching bad weather.”
http://www.newyorker.com/news/sporting-scene/twenty-first-century-shipwreck?utm_source=tny&utm_medium=email&utm_campaign=dailyemail&mbid=nl_121114_Daily&CNDID=29788290&spMailingID=7354090&spUserID=NjEyMzM0NzE4ODUS1&spJobID=581431873&spReportId=NTgxNDMxODczS0

Dave

Even if you are in an area where you cannot see something doesn’t mean there isn’t something there. While sailing Casco Bay in Maine last summer, I found out I needed to reduce the size of the boat symbol on the chartplotter. At the magnification I was displaying, a small “rock” indicator was overwhelmed by the boat symbol. What the heck was that? As I looked behind me to see grass waving in the water. How I missed the keel and bumped the bottom the rudder, I’ll never know. Winter project. Heck, the foam inside the rudder probably needs refreshing anyway.

Marc Dacey

Good point. On most of the plotters I’ve used, the “boat” symbol stays the same. Of course, this can cover up things you are within a boat length of. It can’t be hard to program: I can have about 100 mouse pointers on this laptop…

Paul Browning

My first real personal experience of navigating with electronic charts was chartering on the west coast of Scotland in 2013, using both the chart plotter on the boat, and Navionics on my ipad and I couldn’t get over how easy it was for near misses (fortunately there were all misses, albeit some a bit too near for my and the crew’s comfort). I simply couldn’t believe this system was being so widely used as it seemed catastrophically dangerous. Fortunately the boat also had excellent paper charts in great little yachtie’s packs in plastic, so it was really easy to always have the appropriate scale chart/s close at hand. I still think, 5 years and several charters and ocean crossings later, it is just pure luck there are not many more Team Vestas type groundings caused from over-reliance from electronic vector charts.
As most of my navigation is actually on land, driving a 4×4 either across country or on remote tracks in outback Australia where I live, I often use an app on my ipad called Memory Maps, for which really great raster maps are availlable, including, I was delighted to discover the complete marine charts for Australian and NZ waters and much of the south west Pacific. Brilliant! No hidden hard stuff and it either automatically chooses the largest scale map available for your position, or at least gives you the option of choosing it. I used it very successfully chartering in NZ’s Bay of Islands in 2016 and navigating along the west coast of Australia in a 10 day passage from Fremantle to Bali in 2017. I thought it was a very happy medium combining moving map navigation and electronic “paper” charts.

Brian Renwick

Just to be a bit different, here’s an error I made drawing up the courses on a paper chart at night. A red light was used to illuminate the chart table. Later, while faithfully following the course, I received a radio call from the authorities to clear off from the marine reserve I had strayed into. My AIS would have alerted them to my transgression. Close inspection of the chart indeed revealed a dashed magenta line across the chart, which was difficult to see under a red light. Since then I have avoided a red light for chart work, preferring a natural light suitably dimmed.