web analytics

Computer-assisted groundings? Bad Navionics charts?

Ben Ellison

Ben Ellison

Panbo editor, publisher & chief bottlewasher since 4/12/2005, and now excited to have Ben Stein as a very able colleague. Panbo is going to the next level in 2018 and beyond.

31 Responses

  1. Xavier Itzmann says:

    «Navionics SonarCharts should be used, side by side with the company’s regular navigation charts»
    While we like Navionics due to ease of use, to the point it is our “primary”, we always run two other chartsets simultaneously, and at least one must be raster (paper-like). Currently:
    * MaxSea TZ – like Ben says, these are official government raster charts at a very reasonable price. If you are in Canada, you get a CHS chart; in France, a SHOM chart, in Italy, an IIM one, etc. Highly recommended.
    * Imray Navigator – very clear raster charts with excellent readability. But we probably won’t be buying any other areas because of lack of detail.
    I don’t think anyone should rely on only one chartset when in unfamiliar waters, and especially not on a vector-based one (such as Navionics).

  2. Dave Evans says:

    Ah, that shallow spot! My old Garmin BlueChart has led us neatly around it to stop at the Dolphin Marina this and past summers.
    http://images.sailonset.com/PottsRock.JPG

  3. RickR says:

    Last week I came by the location of the law suit grounding, and I can tell you that Garmin (I assume BlueChart) are not the only ones that are wrong. I used C-Map NT on a Furuno MFD, and NOAA charts on an iPad (iNav-X), and they were both wrong. In fact the whole waterway from the Roosevelt bridge to the St Lucie lock is incorrectly charted.
    ActiveCaptain and CruisersNet are wrong too, but at least they make no attempt to conceal that they use NOAA charts. Only ARGUS data on CruisersNet shows the correct location of the channel, but ARGUS is now gone, so there are no more updates.
    The first rule of navigation is follow the markers and not the charts. Chart plotters are useless in this area.

  4. Walkabout08q says:

    A few summers back I had a similar accident a bit further up the coast, using b&w cartography on an ancient Raytheon chart plotter. Unexpectedly striking a rock on your keel at 6.5 kts will get your heart racing in a hurry. My lesson learned: nav aids are there to mark a hazard (duh) and since then I have made it a religion to identify where the hazard lies in relationship to where I am and where I’m heading. In MAINE ESPECIALLY there are too many hazards to mark on more than one side. Caveat Navigator!

  5. Patrick says:

    Hi, here in Sweden we have had our own experiences with Navionic charts. An insurance company tested a bunch of plotters for some reasons, but during the tests they found that in some zoom-in levels the whole blue area over a ground disappeared! Sometimes a whole island disappeared! In one case it led to a speedboat went up on the ground. Navionics had the wrong answer (not our fault, not our problem, zoom in and you will see what you want) but they had to eat it all up and finally re-do all their charts so nothing was lost, at least not island and depth colors when zooming out.
    I myself have Garmin and BlueChart and went nearly on ground out in the open sea where it was supposed to be 18 m (55 ft). Suddenly it was only 4 meters (12 ft) so I quickly checked my chart. The small rock was so tiny on the display that Garmin didn’t bother to put out a depth beside it. I had to zoom in to nearly insane level before the true depth of the ground showed up. 1,5 m (5 ft). Puh – I missed it with a few meters.
    I discussed it with the man from the insurance company that had made the test mentioned above and showed him my experiences. He was very interested so I went home and went through (nearly) the whole chart after more places where the depth disappeared when I zoomed out. I found a whole bunch so I made a spreadsheet with some 20 places only on my West Coast. But what was most appealing was that a depth figure could disappear when I zoomed out and the show up again when I zoomed out once more! It seemed very inconsistent.
    I managed to present my findings for the Manager for Garmin Marine Europe and we had a chat for some hours about this. The problem, he said, was that there was more info on other layers, like prohibited areas due to birds, that – however – did not show up on that zoom level but anyway prohibited the depths to be presented. There was not much to to about it and they got the source material from the Swedish Maritime authorities so if they changed anything, next time they got an update they had to change everything again. Never the less they changed the appearance of their BlueChart till next release anyway so it should be better now. I haven’t had the chance to look into this because I haven’t updated my own chart the year.
    With all the respect to the birds, I find it more important to show the depth than to present info on the chart when the birds are laying their eggs.

  6. Richard C says:

    For awhile I used Vector charts exclusively on Coastal Explorer but last summer started to appreciate the full picture offered in an old fashioned raster chart. I do still have some paper left aboard, just in case there’s a lightening strike.

  7. Crystal Blues Digital says:

    Ben, thanks for the analysis, appreciate the work. You suggest that Sonar Charts are constructed from a range of unidentified unofficial sources – whereas Navionics clearly state that they start with the Government chart and then add additional data. I certainly didn’t expect them to completely lose so many well known major reefs and ledges as a part of their own production processes.
    So I did learn my lesson about trusting just one data source, specially since like most cruisers we have several good ones onboard. My favoured Transas system showed that ledge perfectly well … as did the NOAA charts.It’s kind of painful to watch our course replay within the Transas software, including the rear-guard action with the dinghy towing us backward.
    I also take onboard your suggestion about using the Sonar Chart data set side-by-side with the official data. Certainly would have saved me on the day.
    Finally, another well-seasoned cruising friend chided me for not doing a thorough route analysis before we departed, using Government sourced data. That would also have saved me plenty of angst….cheers, Neil Langford

  8. Al says:

    Too bad Raymarine chartplotters will not support NOAA charts and Navionics on the same card! If you try to download NOAA charts to a Navionics card, it won’t work because the MFD will not put the ID file on the Navionics card.

  9. This brings up another situation that I have been pontificating about with many other boaters while underway.
    With the advent of Navionics dock-to-dock autorouting, are we going to see an increase in collisions because the software is choosing the same, or similar routes for everyone?
    I swear that in the last 2 years of using that feature, I have been on collision courses more often than before, and I have been wondering if it is because the skipper of the other boat has used autorouting as well, but for the reverse course, and we are on a computer generated identical track.
    It is an interesting thing to consider, and perhaps something Navionics/Garmin needs to consider – adding some randomness into the autorouted courses. Wouldn’t want to see lawsuits over that sort of thing either!

  10. It’s fun to finally see a place we know featured here :). We went thru there both last year and this, including the rather tighter section East of Potts Harbor with no issues – using Navionics in our Raymarine e95. Yes, I/we were paying close attention to the chart, and since I have the NOAA Raster charts in the plotter this year, referring to both charts (tho I didn’t see any discrepancies).
    The rock that got us this summer isn’t on ANY of the charts we have (off the Onset channel entrance in the main channel just South of the Cape Cod canal entrance). ActiveCaptain does have a warning for it, but thats the only one I’ve found.
    I found a couple years ago that the Navionics SonarCharts are NOT to be trusted in tight spots, especially in areas where folks DON’T cover the areas much – for example, I doubt ANYONE actually cruised over the top of that shoal in a vessel that was recording Navionics soundings – so it never showed up on anyone’s file, and the computer didn’t fill it in from the “normal” chart?!
    Al, I’ve never tried to load additional charts onto the Navionics card, but our MFD has two slots, so with a 32GB card, we have quite a few other charts available (NOAA Raster & Vector, some Bluecharts from MX, etc.)
    Steve, a similar situation occurs in the Bahamas – in quite a few places, EVERYBODY is using the same guidebook/chartbook-supplied courses, and you see a LOT of other boats (close!).

  11. Larry Hall says:

    Hi Ben,
    Very well done and I agree with your conclusions. The first order term in the equation is “why traverse a known shoal with no or limited detail when there is plenty of deep water on either side?” On the continuous improvement front, I would like to see Navionics eliminate the “depth box” that pops up when you slew the cursor over a shoal like that implying the depth at the cursor location as it is almost always wrong, no data is better than wrong data.
    Larry Hall
    F/V Gunsmoke

  12. billknny says:

    I have been on a bit of a rant recently against using “crowd-sourced” data for any kind of application with real life safety issues. That comes partly from the time last year when I as close to losing my boat on a reef as I ever have. Definitely a case of “live and learn.”
    We are cruising between out islands in the eastern Bahamas. We are trolling and hoping to drag up a wahoo for dinner. To follow the underwater ridges and drop offs, I have my Navionics chart displaying the “SonarCharts” data. We are off soundings in >1000 feet of water.
    We hook a fish, and as it came to the boat at first I thought it was a wahoo, but no such luck… just a big barracuda. I think: Odd to catch one of those in water so deep. But while we are clearing the fish, we can see bottom! Our fathometer has found bottom–at 40 feet. while the chart on display is showing us 400 with safe water for some distance all round.
    Lets have a look at the “crowd” data which was what I was looking at at the time:
    https://fetchinketch.files.wordpress.com/2017/12/shot-60.jpg
    Now with the cursor at exactly the same spot, lets have a look at the straight Navionics data:
    https://fetchinketch.files.wordpress.com/2017/12/shot-61.jpg
    Granted, this is a remote place, and even the “official” data is a bit thin of survey points, but it is RIGHT as to the depth and location of the reef, and the crowd data is dangerously WRONG for the depth and location of the ENTIRE REEF, not just some isolated rock. I know because I was there.
    I am sure there will be people who will say, “This could never happen with my C-Map/Garmin/Whoever chart.” But… are you SURE????
    It is important to realize that the “official” chart might have its shortcomings, but I could SEE on the display the data available, and just as important, the places where data is sparse. The crowdsourced display might (or might not!) have had data uploaded for this spot. No matter, the algorithm just fills in and makes the rest of it up… Not a huge deal if you are looking for a likely spot to catch fish, but if you are using it to pilot your boat, you can be in a world of hurt.
    When cruising in water densely populated by boats (which we try to avoid) the Sonar Data could be accurate and helpful, if only they left the parts blank where they had limited data, or none at all! I had been lulled into trusting the SonarData because I had used it only in such well traveled places.
    For what it is worth, our paper charts (which are our primary backup) present data that looks exactly like the Navionics display.

  13. Anonymous says:

    In addition to much of what has been stated I have seen in many cases truncation of depths. Ie. 759’ may appear as 59’ at the next higher zoom level. 83’ may apoear as only 3’. It doesn’t work in the reverse – a shallow depth won’t ever appear deeper. There indeed are many idiosyncrasies in using eCharts. In my 225 page x 600+ image Enhanced Guide to Understanding & Using (Navionics) eCharts I found much to learn. Eg, in 2014 two Norwegian and one British researcher won the Nobel Prize for defining the brain’s GPS. After reading their study and putting its findings in perspective with boating I gained a much healthier perspective on navigation, charting and GPS enabled technologies, selecting crew members and just who can handle the helm in which conditions.
    Chuck Cohen, DVM
    Marina Del Rey, CA

  14. Neil of Crystal Blues, thanks for being such a good sport!
    When Navionics says that SonarCharts begin with official data, I don’t think they mean the soundings. The point of SonarCharts is that the bathy data is from alternate sources.
    In fact, SonarCharts used to be called Fishing Charts — actually “Fish’n’Chips” at one point ( https://goo.gl/b6aQvZ ) — and they are a lot like the fishing bathymetry all the other chart makers offer as extra layers. I gather that the underlying data comes from all sorts of places — often the government, since they tend to give it away — but it has not passed the strict requirements of hydro offices like NOAA.
    It doesn’t get talked about much but SonarCharts are largely the old fishing charts, not crowd sourced data (billknny, please note). And it’s impossible to tell what’s what, unless you actually submit sonar logs and later see them merged into the massive SonarCharts database, examples here:
    http://www.panbo.com/archives/2014/07/navionics_sonarcharts_now_via_sonarphone_or_gpx_file_plus_the_lowrance_autopilot.html
    That doesn’t necessarily mean “SonarCharts bad, NOAA data good.” NOAA requirements seem so strict that data updating is slow and expensive, and worthy projects like Argus ( https://goo.gl/XfFah8 ) fail.
    At any rate, what I always recommend is to use many sources, but trust none. I really do believe that great navigators are moderately paranoid 😉

  15. Tom says:

    Ben, why not use the Raster Charts by Raymarine, “LightHouse Charts”. Wonder if they would show the missing data, since they are Raster charts derived from NOAA?
    To the person who said thay cannot use Noaa charts on the Navionics card…….That would likely be a result of an agreement with Navionics.

  16. Tom, the Lighthouse NOAA Raster charts certainly do show the shoal in question – even Navionics’ primary charts do, but the SonarCharts did not, hence the problem.
    Lighthouse charts are encrypted (so you can’t just copy them and give ’em to everyone) and require a “fileID” file to be present on the card they were downloaded onto that tells the MFD how to decrypt the chart for display. Since the MFD will only place this file onto the card if it’s empty, then you can’t store Lighthouse charts (of any kind) on that pre-filled Navionics card and have them readable. In general, you don’t want to, anyway, as there is usually very little spare room on the Navionics cards.
    You also can’t move your Navionics charts over to a larger card – they are tied to the physical card they’re installed on, and if you copy the files to another card, they won’t be recognized.
    There are ways around some of this, but that’s beyond the scope of this discussion 🙂

  17. Tom says:

    In all fairness, what person completely ignores the fact the card warnings and acknowledgement screen TELLING YOU THEY ARE NOT FOR NAVIGATION?
    If that’s the case, then I have ZERO sympathy for people that inappropriately used the wrong chart type to navigate with.
    Where is the personal responsibility?

  18. billknny says:

    Ben,
    You wrote:
    [QUOTE]It doesn’t get talked about much but SonarCharts are largely the old fishing charts, not crowd sourced data (billknny, please note). And it’s impossible to tell what’s what, unless you actually submit sonar logs and later see them merged into the massive SonarCharts database[/QUOTE]
    It really doesn’t matter if the SonarCharts (which are still called Fish’N Chip on my MFD menus, BTW) are all, mostly or only partially, “crowd sourced.” They contain crowd sourced data, and that is reason enough for me to mistrust it.They are not vetted to the degree that NOAA charts are–if the data are vetted at all!
    I do not consider them a safe and useful source of navigation data. I have found no places where they have given my information that I trust more than the official charts. You might feel differently, and you are certainly free to do so. I do think they are a great help for fishing.
    Soundings aside, I disagree that NOAA charts are not updated quickly. I have several cases where I reported things out of whack on the charts to the local USCG office in charge of such things, and they were very responsive in confirming my report and the changes were quickly brought into the update system… usually within a week of confirmation. Their email address is in the LNTM, if anybody actually knows what that is anymore!

  19. Thanks, Bill, but I wasn’t asking you to trust any source. It just seems unwise to completely dismiss crowd sourcing when it’s just gotten started.
    It happens that I’m running a portion of the ICW right now and I’m fairly confidant that the “magenta line” channels and the popular anchorages are very well covered in SonarCharts with actual boater collected soundings data. Heck, it’s my 5th passage thru here collecting sonar logs and I don’t go south every winter.
    So while I have no way of knowing for sure, I suspect that I’m looking at good quality crowd sourced data where it counts. Which is to say that a lot has been collected, some of it quite recently, and Navionics algorithms have been smart about throwing away the anomalies and favoring the newer info.
    On the other hand, I presume that almost all the SonarCharts data outside the ICW was originally fishing baythy and probably came from one source, who knows when or how. My wild guess: crowd sourcing is 5% of SonarCharts at most, maybe more like 1%.
    As to NOAA data, please see what NOAA itself has to say about how current much of their soundings data is:
    http://www.panbo.com/archives/2016/07/noaa_introduces_zones_of_confidence_mind_your_zoc.html
    I’m very appreciative of all the work NOAA does for us, and thankful that the cost gets spread among all taxpayers, but I did once document a case where it took many years to chart a large breakwater that was built and precisely surveyed by another branch of the government:
    http://www.panbo.com/archives/2010/10/why_we_need_crowd_sourcing_downeast_2.html

  20. George Lieber says:

    Why sail over that shallow area when there is deeper water a few feet away? If you choose to sail over a knob on the sea floor when there is clear sailing right next to it then you get what you have coming. The main problem here is poor seamanship. Any type of chart is not to blame when the skipper makes poor choices.

  21. Erik says:

    Geez, red “4l , an actual physical,ATON is um, right there on top of it. Just sayin’

  22. Peter Jung says:

    Hi Ben,
    And now a comment from the Pacific Northwest! In 2016, several boats from the Seattle area were cruising north in the Seymour Channel west of the Glass Peninsula in Admiralty Island in Southeast Alaska. As we approached Buck Island (57N-44’W, 134N-06’W), fortunately in CAVU conditions, two boats inquired why the island to port was neither shown nor labeled on their Navionics software (NOT SonarCharts) displays. The only thing showing were the two awash rocks located north and south of the island, but not the island itself. A third boat (mine) was last in line, and reported that both the native NOAA raster and vector charts for the area BOTH clearly displayed the Island, it’s name, and the rocks.
    Inquiring minds want to know how Navionics can lose a 20-acre island, with a height above low water of something like 50 feet? I get it when bathymetry can appear and disappear on vector displays, depending on zoom, and whose data you might be accessing at the time. But I just can’t figure out how Navionics can lose an entire island when zoomed into a practical scale for surface navigation. And more egregiously, when the applicable data is clearly shown on the NOAA cartography for the area. That seems sorta basic.
    Roger, roger for using multiple navigational aides at all times (like, your eyeballs in this case). And this is clearly a higher-level topic than using crowdsourced bathymetry for this area. There just isn’t any!
    Regards,
    Pete

  23. Berne Miller says:

    Neil Langford said: “Finally, another well-seasoned cruising friend chided me for not doing a thorough route analysis before we departed, using Government sourced data. That would also have saved me plenty of angst….cheers.”
    Perhaps the most important point in all the posts. As a sometime cruiser and captain of several Coast Guard Cutters it was always our practice to review our charts and planned track with the “drivers” before leaving the dock. That was before all our fancy electronics but I still think it good practice to review where you are going and how you propose to get there before flipping the switch and leaping off into the dark.

  24. Peter, if it helps any, we anchored behind a clearly charted island – Grog Island, off Rones Bay on Dymer Creek in Chesapeake Bay (37D 40′ 13″N,076D 19′ 47″W) this fall – the only problem was, the island is GONE!
    All the extant (Navionics, NOAA, etc – I checked Active Captain today – still there!) charts show it, but locals told us it disappeared a few years ago, part of the ongoing (all natural) process of turning Chesapeake Bay into a swamp 🙂 Fortunately, the situation is not dangerous – the anchorage is still there, but is no longer protected by an island, just a shallow sandbar.
    Point is, cartographers make mistakes – sometimes its because the update cycle isn’t fast enough, sometimes they just blow it – a prudent mariner has more than one source of information -and still treads lightly!
    I’m sure Navionics would correct their error if you tell them, just as they & NOAA will soon remove Grog Island.

  25. Thanks, Peter (and Hartley). In fact, Navionics has sort of corrected the error you describe. Buck Island is now clearly charted, although the bar between the island and mainland that NOAA charts as uncovering at low tide looks about 10 feet deep on Navionics regular and SonarCharts.
    That’s a very interesting area cartography wise. I imagine you know this already, but it’s not at all obvious in Navionics or even NOAA vector charts that the best chart scale is over 1:200,000. That’s a very crude chart scale, especially given that the similarly rocky coast I cruise has been charted at 1:40,000 or better.
    There are some 1:40,000 charts in that area, like Gambier Bay, but at least half of Stephens Passage is greater than 1:200,000. I hadn’t realized that any of the U.S. coast was charted so crudely.
    Not that NOAA is hiding anything:
    http://www.charts.noaa.gov/InteractiveCatalog/nrnc.shtml

  26. Berne, I thoroughly agree about the value making and checking routes. There are several PC charting programs like Coastal Explorer and TimeZero 3 that are very good at this, but unfortunately the otherwise very capable multifunction displays are generally awkward at routing. Screens aren’t big or fast enough, touch and/or cursor button interfaces not precise enough.
    Improving MFD autorouting looks like it will help with this problem eventually (though the wise navigator will always check it over), but in the meantime many of us are using tablet apps that can then sync routes to the boat’s primary nav screens, like the Navionics Boating I recently wrote about:
    http://www.panbo.com/archives/2017/10/navionics_garmin_part_1_long_live_plotter_sync.html

  27. Norton Rider says:

    Berne wrote”…That was before all our fancy electronics but I still think it good practice to review where you are going and how you propose to get there before flipping the switch and leaping off into the dark.”
    I agree 100% and this is the reason that I am a big proponent of always planning routes on a laptop using both, raster and vector charts. Once one is comfortable with the routes, they can be uploaded to a chart plotter.

  28. Tom says:

    Looks like the LightHouse 3 software is out for the Es and Gs displays!
    http://www.raymarine.com/software-updates/mfds/lh3/es-gs-series/

  29. Thanks, Tom, but I suspect that page is still under construction and wasn’t meant to be found yet.
    There are hints like “Software Version: v3.3.xx” (not the final number) and “Release Date: Jan 2018.” Also the “3.3” files you can download for Axiom and Es/Gs are the exact same, which isn’t right, and I’m pretty sure it’s really just the 3.2 Axiom update file used as a placeholder as someone builds the 3.3 download pages.
    I say be very careful, and look for an announcement about 3.3 on the main LightHouse 3 page, probably not until January:
    http://www.raymarine.com/view/?id=15032387737
    On the other hand, the page you found does contain the most detail I’ve seen yet on the differences between Lighthouse 2 and 3, at least at the version 3.3 stage we are about to experience. Hopefully people will read it before making the move from 2 to 3, so there are no surprises.

  30. Interesting! Tom Hale reports that Navionics “Sonar Charts’ accuracy has improved to the point that we used the sonar charts for the entire ICW this fall. We never touched bottom or ran aground this fall; we do touch occasionally most years.”
    Tom has been doing great coverage of the ICW on the “SAIL Magazine Secrets of the ICW” Facebook page and I consider him a highly trusted resource:
    https://www.facebook.com/SailMagICWRally/posts/958124004341814
    But I’ll add that the ICW probably has fresher and more redundant SonarCharts crowdsourced depth data than most any place on earth. So while they may work fine for primary navigation in the ditch, that does mean they will work well other places, even right outside the ICW.

  31. Happy New Year, Ben and everyone!!
    Looking at the differences, I think I’ll stick with LH2, even if they produce LH3 for my e95. The loss of the tide charts alone does it for me.
    WRT Sonarcharts, maybe Navionics needs to shade them with some sort of indication of how many tracks contributed to that bit of water – places with 0 or 1 might not get the same respect as a spot with 6 or 8 🙂
    Kinda like those charts NOAA put out a few years ago showing the level of sounding coverage in different areas.

Join the conversation

Your email address will not be published.