Monday, July 18, 2016

Tropical Storm and Hurricane Advisories using FTPmail from NWS

In a recent note I explained how to get these crucial Hurricane Advisory reports underway using email request to saildocs. For completeness, we should also look at the primary source for getting the same information directly from the NWS using ftpmail.

I am reminded to do this because I just discovered that there is an excellent presentation of the ftpmail service in the back section of the NWS publication on radio fax schedules (rfax.pdf), and you would not guess that from the title—in fact, you would not guess it from the Table of Contents, either, because it is in Appendix B, and there is no Appendix B in the Contents. Nevertheless, it is there and it is very helpful.

So with this document you not only get the latest HF frequencies and broadcast schedules for conventional HF FAX weather map transmissions, you also get the ftpmail instructions with interactive links, as well as a long list of other marine weather resources.  It appears they have updated this pdf  with regard to the HF schedules and it does have a table of contents for the fax schedules, but other parts of the publication are not fully updated.

To help focus on the FTPmail part, we have pulled out that section and added a table of contents for the ftp products, and removed some outdated and redundant sections. You can get this document here:

This is an important program to know about and test using.  It includes essentially all NWS products, both maps and text—for local waters as well as oceans.  For some graphic map products, saildocs still has the advantage as they reduce them by a factor of 2 before sending, so this is a big factor for sat phone communications.  But saildocs do not offer all the maps that are available, so we need the ftpmail service.  We have made another convenient pdf for getting either the larger size maps from NWS directly or by email request to saildocs for the reduced file size.

Sunday, July 10, 2016

Tropical Cyclone Advisories by Email Request

If you sail in hurricane zones or their broader waters, you might eventually see something like this:

The unforgivable red arrows on this map mark two entirely different things. The horizontal ones show the directions  of motion of the Lows; the down bound ones are associating the text notice with the systems they refer to.

The NWS tells us to see the Advisory, because they know that no map they make, nor any map that anyone makes, will tell us what we need to navigate in the vicinity of the path of these storms. (We do have a cyclone danger area map available by rfax, but it does not layout the winds).  We also know that the grib formatted GFS forecasts for these systems are not dependable, especially in the early days of formation... and we have to interpret "not dependable" as including "very wrong." 

So it is crucial that we know how to get these Advisories underway.... and we can, once again, be very grateful for the wonderful services of, as they will send them to us promptly with an easy request—though we should keep in mind that saildocs is just tapping into the NWS's long established FTPmail program, which is the primary source.

Send an email to and in the body of the text have these 6 lines

send met.12
send epac.disc

send epac.disc4
send epac.tech_advis4

send cpac.disc4
send cpac.tech_advis4

Then in a few  minutes you will get 6 emails back, all small text only.

The first is an overview with some TC data covering all of Metarea 12.  The second is the forecasters discussion of what they forecasted.

The data for Metarea 12 is given in 3 separate parts in the text file.
The "4" in the next requests identify the 4th Pacific storm of the season, which is Celia.  Blas is #3, and the next is Darby,  #5.  Change this number in the request to the number of the storm you are following.

The last two requests (Central Pacific) are only needed if you and the storm are west of 140W.

It is worth sending that email now to see what you get. You can cut and paste the above request lines, then remove any signatures that come up.

Here is what you get in the tech_advis report looks like for Celia, storm #4 of the 2016 Eastern Pacific season.  Note that each Advisory has a number so it is easy to keep track of what you have. They are issued every 6h.


WTPZ24 KNHC 100232

0300 UTC SUN JUL 10 2016




50 KT....... 30NE   0SE   0SW  30NW.
34 KT....... 80NE  60SE  60SW  80NW.
12 FT SEAS..140NE  60SE  60SW 110NW.


FORECAST VALID 10/1200Z 14.5N 120.6W
64 KT... 20NE   0SE   0SW  20NW.
50 KT... 40NE  30SE  30SW  40NW.
34 KT...100NE  70SE  70SW 100NW.

FORECAST VALID 11/0000Z 14.6N 123.0W
64 KT... 30NE  20SE  20SW  30NW.
50 KT... 50NE  40SE  40SW  50NW.
34 KT...110NE  80SE  80SW 110NW.

FORECAST VALID 11/1200Z 14.6N 125.3W
64 KT... 35NE  25SE  25SW  35NW.
50 KT... 60NE  50SE  50SW  50NW.
34 KT...120NE 100SE  90SW 120NW.

FORECAST VALID 12/0000Z 15.0N 127.3W
50 KT... 60NE  50SE  50SW  60NW.
34 KT...130NE 100SE  90SW 120NW.

FORECAST VALID 13/0000Z 16.7N 130.7W
50 KT... 60NE  50SE  50SW  60NW.
34 KT...130NE 100SE  90SW 120NW.


OUTLOOK VALID 14/0000Z 19.0N 134.5W

OUTLOOK VALID 15/0000Z 21.0N 138.5W




Our job is to

(1) Plot these positions on the chart and also sketch in the quadrants marking the 34 kt wind zones for the next 2 days or so. 

(2) Plot the extended 34-kt radii using the Mariners 1-2-3 Rule that expands the radii by 100 nmi for each 24h of forecast. If the radius is 20 mile now, it should be extended to 120 nmi for tomorrow at this time, and 220 miles for the next day. 

(3) Then try to maneuver to avoid this extended regions (Mariners 34-kt rule).

Note that the 34kt-wind Rule was designed for ships, so they obviously apply to small craft as well. It is not marking unmanageable conditions—experienced sailors can sail in 34 kt winds—but it is just a way to mark that region wherein the storm can change quickly either in direction or intensity. Staying out of that extended region (Mariners 1-2-3 Rule) is the recommended policy of both the NWS and the US Navy.  We have seen tragic loss of life at sea in recent times because that rule was violated.


Here is a video example of applying these Advisories in the echart program Expedition.


Deep back up  for all of this are the HF time tic broadcasts that include storm warnings every hour. These can be received by a simple shortwave receiver.

...AND, of course, the NHC has a Twitter feed for the E. Pacific @NHC_Pacific, but i do not know yet how we might use that.  It should be a useful way for fast info.

Friday, July 8, 2016

Wind Model Comparisons for Start of Pacific Cup

Before the start any data available are legal, but after the start only free public data are legal. We compare here a couple numerical predictions using Buoy 46026, located 18 mi west of San Francisco, of interest, for example, for the start of the Pacific Cup, San Francisco to Kaneoe Bay, Oahu.

Here is the data from Buoy 46026 followed by several numerical  forecasts. [When this post first went up, these were embedded links that updated automatically so we could follow the forecasts. Now we have replaced these with the complied data up till July 11, and these are static.] 

We are basically testing here the forecasts made at 18z on July 7 to see how they pan out for a day or so, with a main goal to simply illustrate one way to test models when preparing for a race.

These are both pictures captured from the NDBC site for Buoy 46026.

To follow are the forecasts presented as metorgrams at the location of Buoy 46026 from Expedition. Note that this function defaults to presenting the full time span of data from the latest download, but you can then zoom it for better comparisons.

So the procedure is, look at the wind speed and direction from the auto-updated pictures above and then look below to see how the models forecasted that back at 18z on Jul 7.

GFS at 28 km (0.25º) goes out >96h at 3h steps, updated every 6h

HRRR only goes out 15h at 1h steps, updated every hour
HRRR ony goes hour 15h, so very good for that period, but then need something longer.


NDFD  every 6h, 3h steps, 3 km, out

PW-GFS every 6h, 1h steps,

Note that several models call for 20 kts by 02 to 08z July 10, so it will be interesting to see who has this the closest... [Note added 7/1; we see the wind did indeed go to 20 kts at about the right time, though a bit later earlier or later than some predicted—but these are the details one might want to evaluate.  We did not know this at the time of the original post at 18z, July 7.]  

The hi res regional models do not go out far enough to see this back at 18z on the 7th.  They are intended to be run more frequently.  

Apply this for other cases you might care about. You can look at a buoy of your choice in the waters you care about at the NDBC site, and then load the various models in Expedition and just plot a meteogram at the buoy location and watch how the forecast pans out.

Note that you can check buoy data directly by sending this to

Find buoy IDs at the ndbc site.

...and you will get back the data below, from which we see the models are pretty good, but NAM and GFS are off more than the others, and HRRR has timed out.

Station 46026
37.755 N 122.839 W

8:50 pm PDT 07/08/16
0350 GMT 07/09/16
Wind: WNW (300°), 11.7 kt
Gust: 13.6 kt
Seas: 3.3 ft
Peak Period: 12 sec
Pres: 30.03 steady
Air Temp: 56.8 °F
Water Temp: 56.3 °F

Wave Summary
9:00 pm PDT 07/08/16
0400 GMT 07/09/16
Swell: 2.6 ft
Period: 12.1 sec
Direction: SSE
Wind Wave: 2.0 ft
Period: 4.2 sec
Direction: WNW
Enter a station ID:

Or you can subscribe to that and get it back every hour, or 3h etc. Best to check with saildocs on that procdure.

Sunday, July 3, 2016

Light Lists—A Modern Look

The Light List (LL) is a crucial USCG publication for navigation for several reasons. The primary one is it includes information about the aids to navigation (ATONs, ie buoys, lights, day marks, radio aids) that is often not available on nautical charts, such as the nominal range and height of some lights, and other descriptions. This is especially true when relying on paper charts or using echarts in the RNC format, which are just graphic images of the paper charts.

Cover of the US Light List

A partial exception comes into play when using official US vector echarts (ENC) as these charts do include ATON data that are updated when the charts are updated, so in principle with automated ENC updates, you have all the latest ATON information as of about 10 days. They are updated every week, with about a 3 day processing time.  Thus an updated US ENC is much like having the latest Light List on board—but we are then at the mercy of our echart program that we count on to alert us that a new edition is available. 

But even with updated ENC, we should at least download one copy of the LL so we get all the extensive support tables and information that appears in the front part of the book.  We have long held that your best lesson on lights is to just read the 25 or so introductory pages to an US LL, including especially the Preface and Glossary included. There are also many helpful illustrations on these pages.  So the second important reason to have a LL is to learn how to interpret the lights, buoys, and other aids we run across on the waterway.  The LL is especially value for interpreting daymarks and it is growing increasingly important to understanding AIS aids. Those using USCG DGPS will also find crucial data there

But the point of this note is not the value of the LL, but how to get it onto the boat or your phone, and to keep it up to date.  The printed version is the same as it has been for years. Namely, once a year they publish an annual edition that is updated as of week 52 of the previous year.  The US government no longer prints these, but private companies have picked it up and do so.  They sell for about $53. The Canadian counterparts cost $30 printed, and the British Admiralty versions cost about $68. Both US and Canadian books are available as free pdf downloads, but not the Admiralty products.

We have then these options for getting the US LL into our computers or phones:

(1) US annual editions 

The point to stress here is this link gets you the full book, introduction and data, but it is not updated. This download will be the version of the LL that was up to date at the end of the previous year, ie week 52 of 2015. You get the same product here downloading in April as you would in November.

You have two ways to keep this Annual LL up to date. You can download just the changes that have occurred this calendar year since the annual edition was published, or you can download the full book as of the week of the download, but this file will not include the important front matter mentioned above.

(2) Summary of changes only

Using this solution you would have to have a copy of the annual edition as these are just the changes. These are small files (~250 kb). The practical application would be look up your light in the annual edition then do a quick check in the updates file.  You can do this fairly quickly because every ATON in US waters has a unique number, and that number is listed on the left of every entry, ie for ebook editions we just search on that number.

(3) Updated editions with optional front matter.

This seems a good approach. Just download the front matter once (~3 Mb) to have this important info somewhere available to you, then download the full set of data as needed to keep your LL up to date in one file. Then when underway you have just one file to look at. These files are about 800 kb.

The date of each of these is printed next to the download link in terms of weeks. The one I am looking at now is called Week 26 of 2016. Week 26 is June 27 to July 3, so this being the 4th is right. Again, the weekly updates can be up to 3 days late. It is also shown on each page of data as shown below.

When leaving for a trip, go to link (3) download both files (front matter and latest contents) and you are done. You have the front matter and you have the LL up to date at the time you leave.  That is the best we can do.  

Broadcast Notice to Mariners
If a crucial light does change once we are underway, then we can hope that it is reported by the USCG on their twice daily VHF broadcasts of the Notice to Mariners. They announce this on channel 16 then give the reports on 22A.  Publication 117 Radio Aids to Navigation tells when these are broadcasted if we want to focus on that without relying on the channel 16 announcement.

Or, for the navigator who has everything, you can request that all warnings and significant changes to ATONS be sent to your email directly at this page:

This is actually a wonderful service that you can try now, and then see how it might work for you with your onboard email service. It will also tell you when chart updates are available or other maritime notices... ie Navy bombing practice.

Canadian Light List
The Canadian books are also available with front matter and index separated from the main data, plus they offer the option to choose just the regional parts you need.  Thus when traversing Canadian waters as in the R2AK, you would select the regions needed and combine them for a custom version.

The Canadian LL is in principle updated monthly, but it is sometimes delayed as it must be released by the Canadian Coast Guard, which is a different agency than makes the actual updates.

When it comes to deciding the actual dates, we are almost certainly safe by just downloading those files, but there is some apparent inconsistencies in the date descriptions.  The individual sections are annotated with a date as shown below...

...but the actual pages in this section are each marked as last updated 5/16.
And when we check the front matter section, we get a bolder statement, with a still different date of 6/16.

In short, I think this means:  this LL was updated with all corrections as of June, 2016, and within the Straits of Georgia section, the last change (whatever it was) was made in May, 2016. This does not explain the date on the download page (2/16), which might be just an error. (They are checking that for us and I will learn shortly.)

Thus for our "truth meter" for deciding which light data were right in the short course, we can assign June, 2016 to the Canadian LL data.

But back to the ENC point raised at beginning...
If you own the latest official Canadian ENC charts for your region of interest,  the ATON data on them is the latest available and can be relied upon as being same as the LL or better.  This is also true for the US ENC. The official Canadian ENC come with 2 years of updates for the cost of $199 for all of BC.  Canada also has an email subscription service for chart and ATON updates at

It is another matter to track down the  date of latest ATON data included with commercial echart programs that offer Canadian vector charts such as Navionics, C-map, Garmin BlueChart, MapMedia, and others.  These companies have a license to reproduce the charts, but no obligation to keep them up to date. When they do an update, we do not know what has been updated or what was the latest official chart was used. Some of these companies are known to be slow on updates, and in many of their apps you simply cannot tell what the latest edition actually is.  I have earlier addressed an issue related to that: Don't Blame eCharts for Anything.

Side note on ENC vs RNC.
We have several places pointed out the value of having ENC charts installed, even if you prefer routine navigation with RNC, because they are more familiar looking. The interesting thing I have learned is Canadian printed charts (now all Print on Demand) are also updated weekly, just as the US ones are, but the Canadian RNC made from these printed charts are much delayed, being at most once a month, and often longer.  They are not updated until the Canadian Light List is updated, which is on a monthly schedule, but often delayed as noted above. Thus they hold off on updating the RNC until the Light List is issued.

Put another way, with Canadian charts we have the unusual circumstance where the printed chart could be up to a month or more newer than the RNC that represents it, which is one more argument to have both the ENC and the RNC. Not only does the ENC sometimes add data to the description of the aids, it is almost certainly up to date if latest chart has been installed (which is not the case for the RNC)—this does not forgive the  ENC protocol used by the IHO for the terrible job it does on rocks, but that is a different topic.

Latest chart editions
To be sure you have the latest chart editions at departure, you can download these pdf files. For the US, we can assume that all formats are updated at the same time; for Canadian charts as noted above the printed charts can be earlier than the echarts.

            US Charts Dates of Latest Editions

            Canadian Charts Dates of Latest Editions


For an extended discussion of how to use the Light List see Inland and Coastal Navigation. The Light List is structured in a unique way, separating main waterways and secondary, and sometimes a light on the corner of these is tricky to find if we do not know its name, or it does not have a name. What we do know is, all lights and buoys are in there, somewhere, and each has a unique number.

A special thanks to the Canadian Department of Fisheries and Oceans, CHS, NOTMAR Division in Victoria for their useful information on details of charting and chart publication.

Navigation Lights—The R2AK Short Course

No matter how sophisticated our navigation electronics might be, safe navigation at night boils down to keeping track visually of where you are relative to navigation lights—on buoys, lighthouses, or posts in the water.

Recall that the lookout, Maureen Jones, warned the third mate in charge of the Exxon Valdez that the crucial Bligh Reef Light was on the wrong side of the bow as they sailed forward, onto the rocks. In fact she warned him twice. [*]

Efficiency is also at stake when you have limited navigation aids on the boat or aids to navigation on the waterway. You might turn into the wrong channel or miss the one you want  if you misjudge when certain lights should or should not be visible.

We have more in-depth treatments of this important topic in the book Inland and Coastal Navigation, but we present here the short version that we prepared for the R2AK team for a quick review, which accounts for examples from Canadian waters.

Define the terms.

Visible range (VR) is how far off in nautical miles you can see the light from your elevation.

Nominal range (NR) is the brightness of the light expressed as how far you can see it in nautical miles when the atmospheric visibility is 10 nmi

Geographic range (GR)  is how close to the light in nautical miles you must be for it to be over the horizon, as viewed from your specific elevation above the water

Luminous range (LR) is the effective nominal range when the atmospheric visibility is not 10 nmi.

h = our height in feet above the water level.

H = charted height (ft) of the light above MHW… not often do navigators account for tides in these computations but in extreme cases they could dominate the answer.

Figure the values

• VR is the smaller of GR and NR to the light.  We can look up NR, but we must compute GR as it depends on h and H.

• NR is on the chart, or failing that we find it in the Light List, which is available for US and Canadian waters in pdf format. US Light List for AKCanadian Light List for Pacific Coast (need several parts)... doesn't every navigator want to have the Light List in their phone?

• GR (nmi) = SQRT (h) + SQRT (H).  

GR Notes.  There are tables in the Light List that list GR for various values of height, from which you find the value for you and for the light and add them, but the height steps are generally too large to be useful for small craft. Sample at the end here.

The CND Light List has a nomogram to determine GR based on h and H, but again, this is an easy computation for better results.  Both of these solutions are presented at the end here, but a phone calculator is best solution.  Sample at the end here.

Also, when comparing those published solutions you will note that our formula gives a smaller value, i.e. the official solution has a factor of 1.17 in front of it, so our GR is 17% shorter.  Our value is easier to remember and usually more accurate. The slightest chop on the water will reduce this value by that much, not to mention that the entire analysis is not better than ±10% or so, not even counting the effect of Luminous range. On the other hand, omitting the effect of tide can have a large error for cases where H is relatively low (under say 50 ft) and the tide range is large (over 10 ft or so). Refer to the "long course" notes for sorting that out.


Assume your h = 9 ft, so GR will always be SQRT (H) + 3 nmi.  A light that is 25 ft high has a GR = SQRT (25) + 3 = 8 nmi.

If the NR to light in question is 6 nmi, then you will not see it till you are within 6 nmi.  In other words, the light is well over the horizon (on a clear day with binoculars you might be able to see the lighthouse in daylight), but the light is not bright enough to see it at night till you get to 6 mi off.

On the other hand, if the NR is 13 nmi, then you will not see it till you are within 8 nmi.  That is, it is shining out into space a distance of 13 nmi, but is below the horizon till you are within 8 nmi of the light.

From the book Inland and Coastal Navigation

Look up the NR form chart of Light List; then compute the GR from h and H, and then VR (when you see the light), will be the smaller of the two.  You will see it closer than this visible range, but not farther.  Your h is not always constant, depending on your boat.

You can look from the cockpit, or from the deck, or from standing on the boom, etc. It is beyond this “short course” to investigate what it means when you see the light standing on the boom, then jump back in the cockpit to tell your mates, and you can no longer see it. Back up on the boom, and it is there again. Hint: it is called bobbing a light.

Example 1

Lights in question viewed on Canadian RNC #

Light on Lasqueti is only described as "FL," meaning flashing.  (See page 33 of the AK Light List for the definitions of characteristics, shown later here.)

With no color given we can usually assume white.  But to figure how close we have to be to see this takes more work because we do not know how high or bright it is from the chart, i.e. we do not know GR or NR from the chart in this case.

Go to Canadian Light List, sample below.  If you search or check the Index on Lasqueti  you will not find it, which is a good exercise in this business.  So we have to guess a name, i.e. try looking under “Prowse,” and that finds it, and now you know the name of that light is False Bay Light.  The beauty of the pdf version is it searches better than an print index.  There you will see this:

From the Canadian Light List

We see for this light (using h = 9 ft… i.e. a sailor standing on the deck of M32x on the way to Ketchikan)!

GR =  3 + SQRT(5.5 * 3.28 ) = 3 + SQRT(18) = 3 + 4.3 = 7.3 nmi, or

GR = 7.3 nmi

Note our formula uses feet, but the Canadian tables are in meters, which accounts for the 3.28 factor.  A little practice with a cell phone calculator makes this very fast.

The NR of this light is listed as 5 nmi, i.e.

NR = 5.0 nmi,

which is smaller than the GR, so the VR of this light is limited by its brightness;  you will not see it till you are some 5 mi off.   These are ±10% type numbers, maybe ± 15%.

Answer:  Visible range = 5 nmi.

Now you can go on the chart, but a waypoint at that light, and then a range ring on that waypoint with radius of 5 nmi to show clearly when you will see it.  This is illustrated in the video below.

Example 2.

We also see Sisters Is light in the raster chart (RNC) above marked as:  Fl (2) 15s 21m 21 M.   In the Light List we learn that this group flashing light has a period of 15s, same as on the chart, meaning that is how long it takes the pattern to repeat, not the time between flashes  This would imply the light is 21 m high (69 ft) with nominal range of 21 nmi. 

However, it appears that this latest version of the Canadian raster chart (RNC) is not correct.  We have to assume the Light List is correct. It says 24.3 m high = 80 ft, with a nominal range of 16, not 18.  That is the virtue of the Light List. Refer to the Truth Meter at the end here.

Note that the latest official Canadian vector chart (ENC) shown below has it right, or at least consistent with the Light List.

Lights in question viewed on Canadian ENC #

It is much easier for an agency on a budget to keep the vector charts (ENC) updated compared to the RNC, which are copies of the printed charts.  We compare these to the latest Navionics charts below.

This vector chart, however,  has the False Bay NR wrong, i.e. shows 6 but Light List says 5. Back to the Truth Meter.

So for Sisters Is. Light  GR= 3 nmi + SQRT(80) = about 12 nmi.  Its NR = 16, so in this case the visibility of this light is limited by our low perspective.  For most small boats looking at bright lights it will be the GR that is the limitation.

Answer: VR =  12 nmi.

Below shows how me might plot these on a paper or echart to keep track of VR.

The red circle shows range of False Bay Light; blue circle is range to Sisters Is Light. So we should see Sisters Is Light not far  beyond the Ballenas Is waypoint, headed NW, and if the route ended up along the southern shore, you would not see the False Bay Light at all. The faint colored lines are the tracks of a few boats from the 2015 R2AK.

End of Lesson. To follow are a few background pics.

*   *   *

Here is a table of light characteristics from the  US Light List.

And here is a table of GR values from the US Light List
The notes above propose computing directly, but some navigators (in more comfortable conditions than on an M32x) might prefer to just have this table handy and use it.
Here is the Canadian solution to the GR computation, but this definitely requires a nav station.

Here are the displays of these lights in the Navionics app called Boating US and Canada, which is quite a good bargain and the charts seem pretty good. It sells for $16, whereas the full set of Canadian echarts that can be run in any echart program costs $199 each for RNC and ENC. Rose Point has a nice set for $99 each that run in Coastal Explorer. For this race we made the routes in Coastal Explorer and then transferred them to both the Garmin 78s and the Navionics App... both of these processes takes some hokey pokey that I will write up in the near future.

Lights in question viewed in Navionics charts (PAC493). The pictures  are out of order. Sisters should be shown on the left. The team used two Android phones and one iPhone 6

The only thing left to add is the "Truth meter" table that compares dates of all these resources, which agree fairly well, but not precisely.

coming soon...

[*] It is not clear if this fundamental observation was early enough to have prevented the grounding had he responded to her first alert, or if maybe had they all looked more closely to lights on the horizon earlier it might have been prevented, but it was indeed awareness of where the light should be by one person that told everyone they were in trouble.

She was sailing as an AB, but actually had a Third Mate License of her own. These notes are from published testimony. I have no special knowledge of this incident nor of the the people involved.

Saturday, July 2, 2016

Using Polarized Sun Glasses as a Viking Sunstone

This note is an annotated, illustrated version of a Chapter 8 section of the book Emergency Navigation, called Finding the Sun as a Viking Would. This note comes to be written now because we had made a (cryptic) Youtube video on this topic some years ago for the students of our online course in emergency navigation, and frankly it was not very clear what was going on in the video without the background of the course or that book.  But the video gained some attention and several commenters ask for clarification, so this is an attempt to do that. It is frankly a pretty neat technique, so I can appreciate the frustration of not knowing what was going on in the video.  Something seemed to be working (according to the narrator, me), but it was not obvious what!


Leif Karlsen, in his book Secrets of the Viking Navigators, gives a good argument that the Vikings used crystals of Iceland spar to find the direction of the sun when just below the horizon or when obscured by fogbanks on the horizon, common at latitude 60° N where they did much of their ocean sailing. The technique he proposes works very well. We have done it many times using these Viking sunstones, which are readily available from gem shops online (you need a clear crystal, 1 to 2 inches on a side). This technique can pinpoint the direction to the sun to within a few degrees, and it was in fact used for many years in a more sophisticated arrangement as the basis of the Kollsman Sky Compass in the early days of polar exploration by air.  Leif’s original instrument is still on display and the Nordic Heritage Museum, just around the corner from Starpath HQ. I had the pleasure of working with Leif on the project for many years. His study is the definitive work on the most likely way the device was actually used in Viking times, but we had to publish a special peer-reviewed article to confirm his right to this claim.

You can simulate the sunstone method with about the same precision using any polarized film, such as the lenses of  polarized sunglasses or some polarized camera filters. Some older sextant models also have polarized filters on the sunshades.

In addition to that, you need a small piece of cellophane. Many clear packaging tapes, the transparent windows of a CD sleeve, or the protective packaging of many products are often cellophane. On the other hand, some clear plastic products that look like cellophane, are actually another type of plastic and will not work for this application.

To prepare the lens, attach a piece of cellophane diagonally across the lens as shown in the left inset in the graphic below. Look through the lens with the cellophane on the far side (sky side) of the lens. The lines with the arrowheads in the figure represent the edge of the cellophane, crossing the middle of the lens.  

Figure 8-4 from Emergency Navigation. The bottom left inset shows a cellophane tape on the glasses; the bottom right inset shows the direction to look. The top part shows 3 examples of applying this technique.  The middle one is what you see when looking in the right direction. The three images of that group would be on top of each other when actually yawing the orientation to the right and left of the true sun direction.   The groups on either side of that show what it looks like if you are not quite looking in the right direction when you yaw the device to the right and left.  You still find the direction to the sun, but it is not as easy to locate that on the horizon.   Start by looking in the direction that is opposite to your best guess of the sun’s bearing. You will not know exactly where to look, as that is what you are trying to discover. Once facing that best-guess direction away from the sun, angle your view up from the horizon by an amount that you estimate would put you about 90° from the line straight to the sun. Referring to the right inset, if you point to where you think the sun is, your thumb will point in the best direction to start looking for maximum polarization. If the sun is just on or below the horizon, you would look straight up. If the sun is about 30° above the horizon, you would look in the opposite direction at about 60° above the horizon.

In the video example, we used inexpensive polarized sunglasses that had no frame at all (Bartells in Ballard), it was just one piece of plastic formed into glasses. The we just cut them in half.  I think we used the clear packing tape we get from U-line, and taped a strip across the glass diagonally as shown.

Then you want to hold this up and look at the sky through the glasses and tape (tape on the opposite side you are looking at), but you do not look in the direction you think the sun is—even without seeing it (already set or obscured) you know very roughly the direction to the sun. Rather, you want to hold this up in the direction that is 90º from that direction. In other words, it the sun had just set (roughly the case in the video), then you would be holding this up, looking directly overhead. For cases when the sun is overcast or behind clouds but not near the horizon, then you would look in a direction opposite to the sun at an angle from the sun as illustrated in the picture.

An important requirement is that this direction you are looking, i.e. 90º from the estimated sun direction, has to have some clear sky. If you look 90º from the sun and it too is covered by clouds, this will not work.  Thee can be some level of overcast, but there must be some level of clear sky in that direction.  The closer you are to looking at 90º to the actual sun, and the clearer the sky is in that direction, the better it will work.

The three center group of drawings in the picture above show what you will see when you are facing the proper direction of maximum polarization, directly opposite the true bearing to the sun. When the cellophane edge is perpendicular to the horizon, the lens will be the same shade on both sides (cellophane side and no-cellophane side). When you rotate it slightly to the right and left (that is a yawing motion), the sides will change brightness, as shown in the two figures adjacent to the center one.  There are three groups of pictures here, we are talking now about the middle group, which is what you see when you are looking in the proper direction.

The two rotated views are shown to the right and left (in the center group), but they would actually be in the same position as the middle one, just rotated to the right or to the left. The video will illustrate what we actually see.

On the other hand, if you are not facing opposite the true direction to the sun, you will see what is shown on the right or left of the center figures. If you are facing to the right, the edge of the cellophane will lean right when the sides have the same brightness, as shown, and when facing to the left of the proper direction, you will see what is shown on the left.

In short you can find the direction to the sun even when pointed slightly wrong, but you will not get a good vertical line as shown in the middle group, which takes more judgement on locating the bearing on the horizon.

This is a subtle process that takes some practice… and patience. But once the procedure is grasped, it can then be repeated much more readily. The sequence is: Look in the approximate right direction and then in the approximate right elevation. Then, starting with the cellophane edge perpendicular to horizon, rotate the lens to find the orientation of that edge line when the two sides are about the same brightness. When you are in about the right direction, the two sides will switch in brightness quite prominently—providing you do indeed have a polarized filter and true.

I have always found when using the sunstone or these glasses in a hand held mode—as opposed to the device Leif invented for using a mirror—that it is easier and more accurate to hold a ruler along side the crystal or in this case the polarized glasses. The ruler enhances your perception of the yawing motion for better precision.  Also in the video I also used a mirror.  I just held the ruler and glasses up in one hand overhead and looked at it with a mirror in my lap. These modifications are (very crudely) sketched in the pictures below. (I will improve these when we can.) Also shown below is an illustration of the right direction to look.

You can try both methods to see what works best for you. Just looking up works fine and is likely best. We used the mirror just so someone could make the video.

Here is the video showing an actual measurement.  The screen caps preceding it below show what is meant by "right" and "wrong" heard in the video. 

Here is the explanation that is printed in the video description:

This is just after sunset, with a bright patch on the horizon to locate sun direction as a test. It is using a mirror facing straight up to view the glasses that have a piece of cellophane tape pasted diagonally across the glass. This can be done without the mirror by looking straight up to the taped glasses, but i could not take a video of it that way. The video was taken with an iPhone.

This is a modern version of using polarized film (cheap clip-on sunglasses, cut in half) as a Viking sun stone (Icelandic Spar). It is a way to find the direction to the sun after it has set or when it might be behind a cloud bank. Maximum polarization is in a direction 90º from the sun direction, which is straight up with the sun near the horizon.

I am holding a meter stick along the edge of the cellophane tape to enhance the rotation angle, which helps pinpoint the direction.

When you watch the video, I am rotating the ruler and glasses to the right and left and noting which side is dark or light, and when the two sides are the same brightness, the the edge of the cellophane and in this case ruler are pointing to the sun... which was set below the horizon for some period of time as i was doing this.  I was using a mirror in this case only so i could make the video, which would he hard to to looking up. In this case a friend was taking the video over my shoulder looking down at the mirror. Doing this on your own there is no need for a mirror. It just gets in the way.  By the way, the obvious way to practice is do it (without a mirror as you can then see where you are pointed, and do it just at or after sunset so you know indeed what is the right answer.

Too far right. One side is darker than the other.

Just right, both sides same color. The Ruler is pointed to the sun, even though the sun is not viable at the moment. It could be set (as in this case) or obscured behind a fogbank or clouds.

Too far Left. One side is darker than the other.
 ----------------------------------------------- Here is the video -----------------------------------------------

A couple details  (my guess is, no one wants more details at this point!)
This technique works because scattered sunlight is polarized. If you imagine a line straight up to your zenith and another line straight toward the sun, then these two lines define a plane. The electromagnetic oscillations within sunlight that have been scattered by air molecules are perpendicular to that plane, whereas in direct sunlight these oscillations are randomly distributed in all directions. Sunlight that has been reflected from a surface (glare from the water, for example) is also polarized, which is the motivation for polarized sunglasses. This light is horizontally polarized parallel to the reflecting surface, which means that sunglasses designed to block this glare are vertically polarized. An easy way to test that a pair of glasses is polarized is simply to look at such a reflective glare and rotate the glasses (looking straight through the lenses to the glare, roll the lenses, without pitch or yaw), and you should see the intensity of the glare change quite noticeably. On land you can do the same with bright glare from a window or car hood or bumper. When the glasses are working properly, you can see though the glare-producing windshield into the car; without them, or with your head turned 90°, you see only glare.

If you are using a lens of a pair of sunglasses, chances are the polarization axis is parallel to the bottom of the glasses, which means you will find the brightest or darkest light transmission with the lens parallel or perpendicular to the horizon. If  this is not the case, find the best orientation and draw a line across the bottom of the lens when rotated into the darkest or brightest orientation—or mark the horizon with a piece of tape if you don’t have the right pen for the job.

When you attach the cellophane it should go diagonally across that axis.  Once you have this "polarization compass" assembled, look again to the correct direction and rotate the device. You will notice now that the right and left sides alternate in brightness; when the two sides have the same level of brightness, the edge of the cellophane is pointing to the sun. The trick is to hunt around for the best direction; rotate the compass, note the sharpness, turn left a bit, try again, turn right and try again. When the angle is optimum, look up a bit and then down, etc.

As shown in Figure 8-4, you are finding a great-circle arc that points to the sun, so when looking to the left of the proper vertical plane, the line will point right, and when looking to the right of it, the line will point left.

Once you find it with access to clear sky at 90º, the location of the brightness transition becomes very sharp. In short, this really works well.


Final note.
After coming back to this video because of the several notifications we were getting here, I looked around at other videos online about the use of the sunstone itself.  There are quite a few. They vary from almost right, to totally bizarre.  I will try to get help and make a quick video of how to use an actual sunstone for this purpose.  It is easier than this.

Vic-Maui Race Day 1, 2, (3) Winds

The Vic-Maui race in 2016 has 4 starts, each at 930 am on July 9 to 12.  This note is about available wind data during the first few days of the race.  This would also, of course, apply to anyone sailing out the Strait, headed to Hawaii on any date.

We have gained fresh experience recently with regional wind models while working on the R2AK planning and analysis for our fast friends on MAD Dog. The best wind data available for the start are commercial products, and as such are only legal up to the starting gun. These winds are the main topic at hand, but once underway we must rely on free pubic data, which would include:

GFS (28 km),  available all the way to HI, but not too valuable till out in the ocean.

NAM (12 km), available from the Continental US run (CONUS) that goes about half way across and then again in HI region, but the NDFD would likely be better for that.

NAM CONUS coverage

NDFD (3 km). This is almost by definition the best free, public data, but it only goes out to Lon 133º W, and then is available again in HI, as shown below. It can be useful in the Strait as well. See The National Digital Forecast Database.  This is not strictly "model forecast," as any one of several models might have dominated its production on any given run, but rather it is the digitized forecast of the NWS using whatever model or input they chose to use.

NDFD HI coverage

All three of the above are available from saildocs underway.

For the start the free hi res data are the NDFD (above), but the best data is likely to be one of these:

HRRR (3 km).  This (High Resolution Rapid Refresh) model is updated hourly,  but only extends out 15 hours, and to get all 15h you must download 5 files. It is available from the Ocens WeatherNet app. Ocens offers short term accounts for this access as well as a 3-day free demo.  I am proud to have taken part in convincing our friends at Ocens to offer the grib format of this important data.  In the R2AK planning, there were several cases where this was hands down the best forecast. For leg one, it remained correct, despite all other models being wrong, as well as the NWS and CND text and VHF forecasts being wrong.

HRRR data. 15h of forecasts in 5 files. Goes north to Cambel River... ie perfect for the R2AK run to Seymour Narrows.

With that said, that is not the only data we used and benefited from.  We also used the

PredictWind PWC/G (1 km, 8 km). The smaller regions in the Strait are 1 km; the coastal regions are 8 km. See

This product has several advantages. First it is super easy to access in Expedition or by email, and the 1 km hi-res models extend out 36 hours,  compared to 15h with HRRR.  They also have a convenient (Mac or PC) app called Predictwind Offshore that lets you either request and look at the data online, or what is often more convenient, it will prepare a template request that you just email to them when you need an update. The larger regions of 8 km are 390 kb each, the smaller regions of 1 km data are 250 kb each.  The 8 km (also hi res) goes out a 7 days—like the lower res GFS, more or less into the realm of the unknown. The navigator is of course going to say that in the ideal world you have both the hi res from PredictWind, and the HRRR from Ocens.  Maybe someday saildocs will offer the HRRR.

Predict Wind also has a weather routing/optimizing feature that seems to work pretty well. You can also use custom polars for the analysis.  For inland waters, however, it does not account for currents, which leaves Expedition at an advantage for that— but I should add here, if you do not have good current data, then you could be better off without using currents at all!  PW does offer and use ocean currents for the ocean routing.

UW WRF (1.3 km).  Unique to our local waters, we should always remember the UW WRF model, which is likely as good as any, but only available as a graphic format;  there is no grib format. The  model is run every 6h and data would typically be 8+ hr old when we get it, and indeed they might not be there when you want them. This is a public service of the UW Atmospheric Sciences Dept.,  with no guarantees.

UW WRF model, run every 6h extends out 60h.

 In fact, this model could be the best model for the region, but it is not updated often enough to compete with the HRRR, which is updated with all the latest actual wind observations every hour.  The HRRR model actually has wind forecasts every 15 minutes, but Ocens has not included that option.

Nevertheless, it is worth printing out the UW data before the start, and with a good connection offshore, perfectly legal underway.  You can figure out how to request the image forecast underway using saildocs—to be added in a future post.

Looking ahead, two articles on the horizon are (1)  the use of a plugin for OpenCPN that *very neatly* lets you load weather maps into your navigation program. We have  been struggling on ways to do this, and simply did not know about that option.  Expedition can do this nicely. And (2) also still working with OpenCPN, which we use in our classes, there is another plug in for OpenCPN that does weather routing!  I have no idea how this works yet; we need to study that and report on it.

Monday, June 13, 2016

Can exceptional behavior be considered ordinary?

The answer is Yes, according to Justice Wilmer in the 1955 case of Velox vs Viking Monarch, wherein one anchored vessel dragged anchor and collided with another when both were sheltering from a severe storm, when there were maneuvers and procedures that might have avoided it if taken.

The nicely put conclusion was:

"I have been reminded, and quite properly reminded, that no seaman can be called upon to exercise more than ordinary care (see Rule 2a below); but I think it is necessary to observe that when a seaman is called upon to face wholly exceptional conditions, ordinary care of itself necessarily demands that exceptional precautions may have to be taken."
In other words, it is the ordinary practice of seaman to take exceptional precautions in exceptional conditions, just as it is to take normal precautions in normal conditions.


RULE 2 Responsibility

(a) Nothing in these Rules shall exonerate any vessel, or the owner, master or crew thereof, from the consequences of any neglect to comply with these Rules or of the neglect of any precaution which may be required by the ordinary practice of seamen, or by the special circumstances of the case.

(b) In construing and complying with these Rules due regard shall be had to all dangers of navigation and collision and to any special circumstances, including the limitations of the vessels involved, which may make a departure from these Rules necessary to avoid immediate danger.

See also our very convenient (free) presentation of the Nav Rules and all related  documents we call Pocket Nav Rules Handbook.

Friday, June 3, 2016

Inside Passage and R2AK Weather

Here is a peek at an old page we had for our onboard training trips to Petersburg. We ran across it today by accident and realized this could be valuable to those doing the R2AK at the end of the month, or anyone headed north this summer.  So we checked and updated the links and added a couple new ones. You can see the page more clearly at the link below. This view is just to show what is there.  To view the links here you have to right click and choose back to return.

Saturday, May 28, 2016

Chart Plotting Tutorials

We are updating several plotting videos we use in our online courses, which were pretty good twenty years ago, but pretty clunky by modern standards. We start out here with a compilation of text descriptions to short videos.  For the time being, these are just YouTube videos of the old Flash movies. We will redo them from scratch in the near future.

Measure Distances
1. Short distances with the miles scale
2. Short distances with the latitude scale
3. Long distances in steps

Measure Directions
4. Magnetic direction using parallel rulers
5. True direction using Weems plotter
6. True direction using a string plotter

Plotting positions and courses
7. Using Lat - Lon scale
8. Plotting COG from a plotted position
9. Using the Starpath card plotter method

1. To measure a short distance using the miles scale...
(1) Set divider tips to span the distance.
(2) Move the dividers to the miles scale with one tip on 0 miles to get the approximate distance.
(3) Then slide the other tip to the nearest integer on the scale, and read the fractional part from the tip that was on 0.

General Notes
Miles scales will appear on most charts with scales of 1:80,000 or larger. When no miles scale is given, use the Latitude scale for the miles measure. Each minute of Latitude equals 1 nautical mile.
When using electronic charts, the display software will typically include an option on the menu bar for measuring ranges and bearings. With these functions, you just drag the mouse cursor from one point to the next and the distance between points and the bearing between them show numerically on the screen.

Video Notes
We will measure the distance between the two points marked by these fingers. The dividers used are a speed bow, which we adjust with the knob between the two points. We are measuring the distance from a channel marker buoy to a daymark at the entrance to a channel.

Once the dividers are accurately set to these points, we move the dividers — with out changing their separation — up to the miles scale, which in this case was at the top of the chart. It could be anywhere on the chart.

Notice that we first put left tip on the 0 and look to the right tip to see that we have here a distance of just over 2 and one half miles (nautical miles). Next slide the dividers to the left until the right tip is exactly on the 2, and read the fractional part from where the left tip hits the tenths scale. Count out the number of tick marks from the 0 back to the left to see that the fractional part is about 0.64. The total distance is then 2.64 nautical miles.

Whether or not the 0.04 part has any significance at all depends on how carefully we have done the other parts. Normally, such a measurement would only be done to 0.1 mile precision, so this answer would be stated as 2.6 miles. Top

2. To measure a short distance using the Latitude scale...
(1) Set divider tips to span the distance
(2) Move the dividers to the latitude scale on the left or right side of the chart, roughly due east or west of the segment you are measuring.
(3) Set one divider tip on some latitude of a whole number of minutes, such at 00' or 05' or 40'
(4) Let the other divider tip fall onto the latitude scale wherever it might. The second tip can be at a higher or lower latitude that the first since we care only about the distance between them.
(5) Figure the distance using the rule that each minute of latitude equals one nautical mile. Always double check the latitude scale to be sure you understand what the tick mark or graduation marks mean. Sometimes the latitude scale is marked in tenths of minutes, other times in various units of arc seconds, such as a mark every 5".

General Notes
The latitude scale can be used for a miles scale on all charts, even if they happen to have a specific miles scale printed on them. Remember to go more or less straight left or right to reach the latitude scale — and that this must be measured from the latitude scale (sides of the chart), not the longitude scale (bottom and top of the chart).

That is, do not set dividers to some distance near the top of the chart and then put them on the latitude scale near the bottom of the chart. On a small scale chart which covers a large area, if you make a shift like that you can introduce significant errors in the distance measurement.

A note on the reason for this: Nautical charts are made in what is called the Mercator projection. This type of projection is very convenient for navigation since North is always the same direction on a chart — usually toward the top of the page — and all bodies of land and water have the proper shape. This latter point is important since it lets us determine what the compass bearing is from one point to another from the chart.

This seemingly obvious task is not so simple as it might seem. Near the North Pole for example, or at any very high latitude, this type of chart (Mercator) is not possible and consequently it is difficult to determine bearings or directions from a chart in these areas. Remember if you are standing at the North Pole, every direction is South! — a good hint that we have a charting problem in this region.
In any event, away from the polar regions, Mercator charts serve us well for navigation, but there is a price to pay for the bearing convenience. Although the shapes of lands are right, their relative sizes are not. Greenland is not larger than the US as it appears on a Mercator projection, it is in fact less than one quarter the size of the US. So it is with all land masses on Mercator charts. They are bigger at higher latitudes. Put another way, the number of inches per degree of latitude or per nautical mile is higher at higher latitudes.

It is always true that 1 minute of latitude = 1 nautical mile, but the physical size of this unit will vary with latitude. That is why when using small scale charts (those that cover a large area) we must use the nearest latitude scale to measure distances.

Video Notes
This is the same measurement explained in the previous example (Short line with miles scale), but this time we use the latitude scale to read the miles. If you have not done so, please read through the previous example first.

After setting the separation, we move the dividers straight left (or right) to the nearest latitude scale along the side of the chart. Here we set the top tip to the latitude of 37° 20' N and let the bottom one fall where it will. We chose this one simply because it was the nearest whole value due west of our measurement. This particular latitude scale is marked with broad bands every 1' of latitude. Although the numbers are not shown, the bottom tip is at latitude 37° 17.4' N.

What we care about here, though, is only the separation, which we can count down from the 20' mark to see that it is 2.6' = 2.6 nautical miles.

In this very common operation of reading a latitude, or in this case a latitude interval, it is very important that we first check to see what the tick marks mean along the scale. Don't guess this, always double check it. Different charts use different conventions on what is bold and how many minutes or seconds there are per tick.

We usually check this carefully once, and then actually write it on the chart. That is, had we used this chart before, there would be a "19' " label and an "18' " label hand drawn in along the border. On this chart, the next printed label is located out of sight here at 37° 15'. Top

3. To measure longer distances by walking the dividers...
(1) Set divider tips to a convenient whole unit such as 1.0 or 2.0 miles
(2) Start at the first point and "walk" the dividers along the route, counting the total number of steps.
(3) At the end, squeeze the divider tips together to measure the last, smaller step.
(4) Take dividers back to the lat scale or to a miles scale to measure the size of the last step and add this to the total number of steps for the full distance.

General Notes
This is the standard method for measuring longer distances. In most cases, you must balance out the best choice for the step size. If it is too large, you will lose accuracy going around corners, but if it is too small it takes longer to measure.

On electronic charts, this measurement is typically done by setting up a route of multiple waypoints between the two end points. Then the route properties can be displayed which among other data will include the total distance from start to end. Such a route could be temporary or saved. The properties will include the bearing or course between waypoints as well as the time of transit based on some input speed.

This procedure can also be used to measure or estimate the time a trip will take. If traveling at 7 knots, set dividers to a spacing of 7.0 miles, then walk off the route. Each step is 1 hour of travel time.

Video Notes
This example shows measuring the distance between two buoys some distance apart. It starts down and around a peninsula, up a channel, and into a bay. Dividers are first set to a separation of 2.0 miles using the latitude scale and then walked along the path.

There are 5 whole steps along the route, and then the dividers are squeezed together to span the last step. When moved to the latitude scale, we see it is about 1.1 miles. The total distance is then 5 x 2.0 + 1.1 = 11.1 miles (that is, nautical miles). Top

4. To measure the direction of a line on the chart...
(1) Align parallel rulers with the line in question.
(2) Move the parallel rulers without slipping to the nearest compass rose and align one edge with the center of the rose.
(3) Read true orientation of the line from the outer ring of the compass rose, or read the magnetic orientation from the middle ring of directions. 
General notes
Take care to read the rose in the direction of interest, not its reciprocal. These are all obvious matters, of course, but when tired or in a hurry, it is easy to overlook the basics unless we train ourselves to check and double check even the simplest matters in all applications. (Remember, as a navigator, you always run the risk of having the helmsmen do exactly what you tell them to do!)

"Compass rose" is an unusual term for this diagram on a chart, but it is used universally, even on high latitude charts that do not include any compass directions at all on the diagram. On typical compass roses, however, you have 3 scales, an outer scale showing true directions relative to true North at 000 and a middle scale which is rotated by the local magnetic variation so it reads magnetic headings, and then a third inner ring that shows compass directions marked off in compass points (each point is 11.25°).

If you want to be double sure that you did not move or slide on the way to the compass rose, you can walk the dividers back to the line after reading or marking the direction on the rose. If the parallels do not line up when you return, you will have to repeat the whole process, since you do not know if you moved coming or going.

Video Notes
The hand traces out the line in the direction we wish to measure. The first step shown is just a trick to use dividers to help align the parallel rulers.

Place a tip on the line and then slide the parallels up against the tip. Then rotate the parallel rulers back down to align their edge with the line on the chart. This method is a good way to get a careful alignment even when bouncing around in a seaway.

Then use one hand to hold down the aligned ruler, and use the other hand to move the other part of the rulers away from it. This is taking the first step with the rulers. Then change your pressure to hold down the one you moved, and bring the back one up to it.

When the ruler edge will finally reach the center of the compass rose, align it with the center of the rose and read the appropriate ring where the parallel ruler edge crosses it. If the light is poor or when bouncing around, it may be necessary to draw a line along the ruler edge crossing the scale and then read the scale.

To read the scales, we must do just as when reading the latitude scale or any other scale. Count the tick marks in between the labeled ones to be sure we know what they stand for. A common mistake can be to count them as 1° each when they are actually 2° each, or vice versa.

In this example, there is 1° per tick mark on the True scale and the direction we want is 320° T. Notice that it is difficult to read along the edge of the clear plastic rulers. Drawing a line will help in a case like this. Top

5. Measuring directions with a rolling plotter...
(1) Align parallel plotter with the line of interest
(2) Roll plotter to nearest meridian and place center of plotter's scale on that meridian.
(3) Read line direction from the plotters scale, where the meridian crosses it.

General notes
This example is the same as last example, but with a Weems plotter, rather than parallel rulers. Read in that section about the trick of using dividers to get the alignment, or in this case, a first step toward the alignment, since we can fine tune that one using this tool (see Video Notes, below).

Meridians are the lines of longitudes, vertical lines on most charts. This measurement can also be made on a parallel of latitude (horizontal lines on the chart) using the diagrams on the two edges of the plotter. These plotters work very well when you have good chart table area. It is sometimes difficult to get to use them near the edge of a chart, so we always have parallel rulers at hand if needed.

These plotters are very convenient, but you must use true bearings — as opposed to using parallel rulers and the compass rose, in which case you can choose which scale, true or magnetic, you prefer. When using these tools, you must have the local magnetic variation in mind at all times and make the corrections whenever you are relating to compass courses or bearings.

Video Notes
As in last example, the finger traces out the direction we wish to measure. And, as explained in the last example, we use the dividers to align the plotter with the charted line.
In this case, though, we fine tune that alignment using the thin black line printed on the plotter itself. You could use the edge of the plotter or this line, but we have found that using this line is more accurate. To do this, look through the plotter and micro-adjust the plotter orientation until the line on it coincides exactly with the line on the chart.

Then carefully roll the plotter to the nearest meridian line and align the center of the plotter's compass rose scale with the meridian as shown.

Read the heading from where the meridian crosses the scale. Note that the diagram reminds you with small arrows that if looking NW use the inner scale and when looking SE use the outer scale (see last frame of video). The bearing we want in this example is 320 T, in the NW quadrant. The reciprocal is 140 T, which is in the SE quadrant.

As cautioned in the last example, we must be even more careful here to get the right direction. Generally keeping some rough numerical estimate in mind will solve this problem. Top

6. Measuring true course with a string plotter...
This is a method found useful for many paddlers and other small craft operators that have little room for tools or surfaces to use them on.

Drill a hole (if needed) in the center of a 0-360° protractor and insert a string about 18 inches long. Tie knots at both ends and use as illustrated.

For some applications, the center of the protractor can be placed on your position or other reference mark, but in other examples, as the one shown here, the center is offset so the string can pass through both points of interest.

The edge or some other vertical or horizontal lines on the protractor must be aligned with some straight lines on the chart, but it is rearley a problem to make this orientation. The bearings or directions are read from the protractor where the string crosses the scale. These directions will always be true directions which must be converted to magnetic if used with a compass, which is the usual case.

Video Notes
There is no video here, just slides.
Slide 1 shows general use of the tool, aligning the protractor with some vertical or horizontal reference and stretching out the string to read a direction.

Slide 2 shows the alignment and the string in more detail. In this example, the plotter is used to measure the course from a plotted position to the pass between two islands.

Sldie 3 shows the string crossing the protractor scale and going through the plotted position. The true direction of this course is 307 T. If the local magnetic variation were 20° East, then the magnet course would be 307 - 20 = 287 M.

The compass conversion rule is "correcting add east," where "correcting" means going from Magnetic to True. We are going the other way, so we subtract. If there is ever any doubt about this, just go to a compass rose and draw a line from center to either true or magnetic course and read the "converted course" from the other scale on the rose. In short, a compass rose is just a graphic table of compass conversions, from True to Magnetic or vice versa.

This type of string plotter could be called "Sutherland plotter," named after kayaker Chuck Sutherland who, to our knowledge, is the inventor of this idea for a measuring device. Top

7. To plot a position using Lat / Lon scales...
(1) Double check the Lat value from original source.
(2) Identify this Lat on the Lat scale on the side of the chart, double checking the tick mark spacing before hand.
(3) Use plotter or dividers to transfer this latitude to chart and draw a short line in your approximate position.
(4) Repeat the process using the Lon scale at the top or bottom of the chart. Draw another short line to intersect the Lat line, and the intersection is your position.
(5) If this is a position fix, draw a small circle around the intersection and label this position with the time.

General Notes
There are numerous ways to carefully plot a Lat / Lon position. The key issue is doing it precisely. In these days of GPS, much of our position navigation has been reduced to simply plotting a position on the chart. Needless to say, we don't want to do a bad job of all we have left!

But, jokes aside, this is a crucial part of navigation regardless of how we got the info in the first place. And it is one of those things that seems so easy, we might not treat it seriously enough. Which is a mistake. It cannot be done too carefully in most cases.

If the plotter won't reach the region of your position as shown in the video — which it certainly won't in many cases — then we have a bit more work to do. First, on the latitude scale itself, set the dividers to the distance from the given latitude to the nearest parallel shown on the chart, and then mark that distance on the chart near the approximate longitude. Then use plotter to draw the short Lat line at that position, rolling the plotter down from that nearest parallel. Then do the same for the longitude.

Video Notes
In this example, the first step shown is the double checking of the latitude tick mark spacing, by counting off the steps between the two labeled latitudes. Here the bold bands mark off 1' intervals from 36° 50' on up. One divider tip is then placed at the 52.7' mark and the plotter is rotated till it is parallel to a latitude line. In this case, this was performed by aligning the vertical line in the center of the plotter with a meridian on the chart.

Once the plotter is parallel, it is slid up to the divider tip to mark the proper latitude, and a short line drawn in the region of the position. We know where this should be from looking at the longitude scale and the longitude we must plot.

Then the process is repeated for the longitude.
The conventional plotting symbol for a position fix is a full circle about the intersection. Other positions on a chart are marked with other symbols, such as a half circle for Dead Reckoning positions, and a square or triangle for estimated or projected positions. Top

8. To plot a course line from a position...
(1) Orient a plotter or parallel rulers with the desired direction using the compass rose.
(2) Move the plotter back to the position and draw the course line.

General Notes
This process is essentially the reverse of the exercises shown in the Directions section. There we had a given line and used the compass rose to measure its orientation. Here we start with a given orientation, and wish to draw a line through a specific point in that direction.

Generally the line is our intended course away from a known position. The course could be specified in True or Magnetic terms, but in either case it is just one unique line on the chart. The only difference is which ring we use on the compass rose.

As with plotting positions, this is one of the key operations in chart plotting. We have a known position which we might have obtained from electronic equipment or from some piloting technique, and we have a known course we wish to sail which will take us to our destination. Or, we could be under sail, and we are simply plotting out our course over ground (COG) as given from the GPS.
Generally it is the COG that we need to plot here since this is the way we are actually moving. The compass heading of the boat is the way the boat is pointed but is not necessarily the way the boat is moving.

Video Notes
In this example, we have a Course Over Ground of 238M, which we wish to plot away from our 1352 position, which is already plotted.

We use the dividers to help align the plotter. First use the divider tip to carefully count off the numbers on the magnetic ring on the compass rose. Again, we wish to confirm what the tick marks stand for since the bearings are only labeled every 30°. Once we find 230, we count out the next 8 degrees and then plant the point at 238M.

Next slide the plotter up to that point and then rotate it till the center is aligned along the edge. Then carefully roll it back down to the 1352 position and draw the line. Top

9. Position plotting with card plotter...
First you have to make a plotter as described at the end of these notes. Next you must prepare your charts with a few extra meridians and parallels over the region you will be traveling, also described at the end.

There is obviously some time involved in this preparation to use this method, but once completed it lasts for a long time and will speed up your accurate position plotting by a great deal on the charts that apply. For tricky passages or for racing navigation where quick navigation is essential, this simple method has proven extremely valuable many times now since the advent of digital read outs, back in the old LORAN days. With it, and very little practice, you can plot an accurate position on a chart about as fast as you can read them from the electronics.

Video Notes

In this example, we plot: 48° 23.25' N, 122° 33.82' W on a chart with scale of 1:25,000. This method requires a separate card, or at least corner of a card, for each chart scale it will be used with (explained below).

Step 1. Align the right side of the card with the nearest meridian of whole minutes of longitude of your position, in this case 122° 33'. (This meridian has to be on the chart. If it did not come that way, then this is one that we had to add ourselves, as was done in this case, in red.)

Step 2. Slide the top edge of the card up to the nearest parallel of whole minutes of latitude of your position, in this case 48° 23'. Then slide it farther up to match the decimal part of your latitude, in this case 0.25'.

Step 3. Count left, to the west, from the meridian along the top edge of the card to mark the decimal part of your longitude, in this case 0.82' and mark your position with a point. Then it is a good idea to immediately label this position with the time it was valid.

How to make a "Starpath card plotter"
We call this a Starpath plotter so that this particular method is easy to refer to. There are numerous plastic plotting aids on the market, but we have not seen any like this, and we have not seen any that work as well. Furthermore, it is a valuable exercise to construct the device and plot the extra chart lines as it gets the user a bit more involved with the actual scales being used. This technique was developed for racing navigation back in the days when we first had electronic positions, but no electronic chart plotting—to be used when we had to plot fast in dangerous conditions, as in rounding a corner closely at night or transiting a narrow channel in strong current.

Use any piece of cardboard. Index cards work well — we have many times cut the back out of a notepad to make the cards. Next study the chart you will be using for this to see what the decimal scales are for latitude and longitude. Generally this method works well for the common scales of 1:40,000 and 1: 80,000 charts or 1:25,000. For larger scales it works even better, and for smaller scales there is not much need for this type of plotting.

Note from slide 2 that you mark the latitude scale going down the right side and the longitude scale going across the top to the left. The scale should be marked in decimal degrees, which in turn means it is most convenient when the GPS output is in decimal degrees — as opposed to degrees, minutes, and seconds, which is awkward to plot.

A separate card or corner is needed for each chart scale used. Also the latitude and longitude scales must be divided into tenths, which is not given on all charts. The nautical miles scale will do it for the latitude, but you will have to interpolate the longitude on some charts, with 6" = 0.1' etc.
Since most charts do not have parallels and meridians drawn every 1', we usually need to add extra lines to use with this plotter. Once added, it is helpful to label them in the vicinity of your travels. Needless to say, we should double check these labels with the chart scales themselves.

Once prepared, this is a fast, convenient, and accurate way to plot. Time spent in preparation and practice should prove rewarding. Again, though, this is intended for fast plotting in somewhat special circcumstances. For most routine navigation, simple use of parallels and dividers, covered elsewhere in this plotting tutorial, should do the job just fine.

PS. If you are in the Navy or on any large government vessel, then this might be something to keep in mind, as they still have to plot positions frequently on a paper chart. Maybe that will change with new ECDIS rulings in 2018, but there remains sound logic for the paper plotting. Top