What curve does a kayak make...

[djlewis]

PS

Funny how Dr. Lewis get's these off-topic discussions (arguments?) going. Sometimes he throws a grenade and then runs away!

I'm here.

And I'm confused by Jason's example. Assuming we are in a craft that can handle 12 kts of cross-current, I don't see why a straight line from launch to destination (at 0 degrees) is not the fastest course. Of course, 0 degrees is not the heading you would take. That's 270 + ArcTan(4/12) -- about 298 -- for the first 3 hours, then 90 - ArcTan(4/12) for the final 6 hours -- about 72. But the resultant course would still be 0 throughout, assuming your boat is not destroyed when you cross the 24 kt zipper!.

Perhaps the issue is the heading you would steer vs the resultant course your boat would describe given the current. Jason apparently used the phrase "correct direction" as if it were the heading, and "GPS arrow" to mean resultant course. Of course those are not the same in the presence of any cross-current -- you don't need an exaggerated example to prove that.

And the original disagreement between me and Leon was whether you could use a compass to stay on the correct course (the unique straight line from launch to destination), or is that something only a GPS can do. I still maintain you could use a compass, assuming it had adequate resolution to go up against the almost instant precision of a GPS, which it certainly wouldn't in Jason's case. But that's a practical issue, and as the great Yogi said -- in theory, practice and theory are the same. In practice they aren't!

[JohnHuth]

Just to amplify Suz and Jason's comments: if you take a direct GPS line in the presence of current, the GPS is likely to continually tell you to alter your heading, as you will not be moving as the GPS "thinks" you should. This in itself probably generates an interesting 'pursuit curve', and I'd wager takes a lot more energy than a course that uses knowledge of current.

That Excel spreadsheet can be downloaded to an iPhone or equivalent, and takes a lot of the pain out of adjustments for current, so it's not so much of a headache.

I do know some airplane pilots who have apps like this to calculate headings in the presence of wind, so it's not too difficult to take the pain out of it.

[lhunt]

if you take a direct GPS line in the presence of current, the GPS is likely to continually tell you to alter your heading, as you will not be moving as the GPS "thinks" you should.

As I said before, and Leon repeated, I'm not trying to say that a straight line is the best way to go, but if a straight line is what you want, the GPS gives it to you with very little extra effort. Maybe I'm not understanding the arguments about "zig zagging" or, in this case "continually telling you to alter your heading", but...

The GPS tells you what heading to use and it doesn't change all that much. Here is how I use the GPS: I do a find on a waypoint, then start paddling. Once I get up to speed (in a few to several strokes), I check the GPS for a heading. I do this every couple of seconds until I get to a point where most of the time when I look down I'm on the heading - usually within 15 seconds or so. Then I establish a point on the horizon (if there is one), and use that. Every minute or so I look down to make sure the heading is still good and adjust if necessary. The adjustments tend to be, as David has pointed out, linear. That is, the ferry angle lessens as I approach the endpoint. I can also see the ferry angle go up for extended wind gusts and when I cross a current, which is what I want.

I don't think the adjustments I have to make every so often, or the intermittent swinging of the pointer when a wave pushes me around, vary negatively from what a compass does during a crossing, so I'm discounting the small confusion that causes.

If you take a step back, using a GPS is not terribly different from using a compass, except that the GPS corrects you based on course, not heading. That is a big difference, but the actual "in practice" way you go about following the pointer is about the same.

The fact that the GPS keeps a track of where you went, and can display it on a map afterwards, gives very good feedback as to how straight the line was.

This in itself probably generates an interesting 'pursuit curve'

If you want to see the "pursuit curve", check out the straight lines here, particularly the long one on the southeast side.

Anyway, again, I'm not trying to talk people into using a GPS, but this topic might be of interest to someone thinking about it, so I want to be clear. Also, if readers have made a conscious decision not to use a GPS, it is worth at least making the decision based on real-world data. The theories I hear about how it "must" work are often wide of the mark.

[leong]

Just to amplify Suz and Jason's comments: if you take a direct GPS line in the presence of current, the GPS is likely to continually tell you to alter your heading, as you will not be moving as the GPS "thinks" you should. This in itself probably generates an interesting 'pursuit curve', and I'd wager takes a lot more energy than a course that uses knowledge of current.

That Excel spreadsheet can be downloaded to an iPhone or equivalent, and takes a lot of the pain out of adjustments for current, so it's not so much of a headache.

I do know some airplane pilots who have apps like this to calculate headings in the presence of wind, so it's not too difficult to take the pain out of it.

John,

I paddle about 5 days per week all year long (going between 10 and 25 miles and averaging about 13 miles). Mostly I don’t use a GPS because I don’t need one. But when I have used the GPS I seem to follow a straight line (very little wiggling around, probably less than when I use ranges).

What’s the point of using a computer to calculate ferry angles? I don’t know the speed and direction of the local currents and wind as a function of time. But even if I knew all of this, how would that help without a fast on-board computer? Because my kayak is pushed by each wind gust, I would then need a model of the kayak’s wind drag curves as a function of angle and a computer to make the real-time corrections. My GPS is not much bigger than a cell phone and is waterproof. It does the job very well when I use it. It doesn’t need to know the wind speed, current, etc. It just knows how the kayak is moving with respect to the earth and that’s enough to do a good job.

All of this reminds me of using an automobile navigational GPS. Sure, if you know that a road is washed out (and the GPS doesn’t know) it’s better to go navigate manually. But would you give up your automobile GPS because of those rear occasions?

Gotta go now, the kayak is on the car and waiting to go to Lanes Cove.

Respectfully,

Leon

[JohnHuth]

Actually, I don't use GPS'es for cars or anything. I don't know if that's a bad idea, but it makes me feel disconnected from my surroundings.

[djlewis]

As I said before, and Leon repeated, I'm not trying to say that a straight line is the best way to go, but if a straight line is what you want, the GPS gives it to you with very little extra effort. Maybe I'm not understanding the arguments about "zig zagging" or, in this case "continually telling you to alter your heading", but...

Below is an explanation of zig-zagging.

The GPS tells you what heading to use and it doesn't change all that much. Here is how I use the GPS: I do a find on a waypoint, then start paddling. Once I get up to speed (in a few to several strokes), I check the GPS for a heading.

There is your zig-zagging! You start with a heading directly on your destination waypoint. That means you immediately drift down-current a bit -- zig #1. Then you look at the GPS and correct by turning upcurrent -- zag #1

I do this every couple of seconds

Say you do that three times more -- zig #2, zag #2, zig #3, zag #3, zig $4, zag #4

until I get to a point where most of the time when I look down I'm on the heading - usually within 15 seconds or so. Then I establish a point on the horizon (if there is one), and use that.

At that point, you are presumably setting a heading and distant point upcurrent from your actual destination. You have finally set a ferry angle! You now won't zig-zag any more, until...

Every minute or so I look down to make sure the heading is still good and adjust if necessary.

Each adjustment will be a small zig-zag, perhaps inconsequential, especially if you use the linear reduction method, as you say.

The adjustments tend to be, as David has pointed out, linear. That is, the ferry angle lessens as I approach the endpoint. I can also see the ferry angle go up for extended wind gusts and when I cross a current, which is what I want.

I don't think the adjustments I have to make every so often, or the intermittent swinging of the pointer when a wave pushes me around, vary negatively from what a compass does during a crossing, so I'm discounting the small confusion that causes.

Of course, if you stubbornly insisted on using the initial method -- drift, adjust, repeat -- all the way to the destination, never setting a ferry angle, then you would zig-zag all the way. You don't do this -- great. But then why not start with a guess at the ferry angle and eliminate or reduce even the initial zig-zags. It's not that hard to make a good guess to start and then use the GPS to correct, but less correction and zig-zagging will be needed.

If you take a step back, using a GPS is not terribly different from using a compass, except that the GPS corrects you based on course, not heading. That is a big difference, but the actual "in practice" way you go about following the pointer is about the same.

Precisely! That's what I have been trying to say for a week. But I will grant that with a GPS the zig-zags -- initial or all the way across -- will be less than with a compass. That's because the GPS tells you quicker and more accurately than a compass when you are off the ideal straight-line course, so your zigs and zags are smaller than with a compass.

[jason]

Below is an explanation of zig-zagging.

There is your zig-zagging! You start with a heading directly on your destination waypoint. That means you immediately drift down-current a bit -- zig #1. Then you look at the GPS and correct by turning upcurrent -- zag #1

Say you do that three times more -- zig #2, zag #2, zig #3, zag #3, zig $4, zag #4

At that point, you are presumably setting a heading and distant point upcurrent from your actual destination. You have finally set a ferry angle! You now won't zig-zag any more, until...

Each adjustment will be a small zig-zag, perhaps inconsequential, especially if you use the linear reduction method, as you say.

Of course, if you stubbornly insisted on using the initial method -- drift, adjust, repeat -- all the way to the destination, never setting a ferry angle, then you would zig-zag all the way. You don't do this -- great. But then why not start with a guess at the ferry angle and eliminate or reduce even the initial zig-zags. It's not that hard to make a good guess to start and then use the GPS to correct, but less correction and zig-zagging will be needed.

Precisely! That's what I have been trying to say for a week. But I will grant that with a GPS the zig-zags -- initial or all the way across -- will be less than with a compass. That's because the GPS tells you quicker and more accurately than a compass when you are off the ideal straight-line course, so your zigs and zags are smaller than with a compass.

If you know the currents like in my above example, you won't have to correct at all with a compass bearing. In the above example using a GPS will put you on a tread mill never getting to your destination.

The correct way would have been to paddle with your compass at 0 deg after the first 3 hours you would have paddled 12 NM north and drifted 36 NM East. In the next 3 hours you would paddle another 12NM while drifting 36 NM back to the west ending up on course.

Paddling in locations that have currents (we have them within a day trip range) and you will see why current tables are useful.

[jason]

I found a few pages that might help.

http://www.bask.org/...5_currents.html

http://www.kayarchy....03chartwork.htm

http://www.amazon.co...s/dp/0762738340

-Jason

[djlewis]

The correct way would have been to paddle with your compass at 0 deg after the first 3 hours you would have paddled 12 NM north and drifted 36 NM East. In the next 3 hours you would paddle another 12NM while drifting 36 NM back to the west ending up on course.

I see your point -- using the opposing current to allow you to paddle from A to B even with a cross-current that would ordinarily preclude that if there were not an opposing current. In fact, this applies to any situation where the current is not uniform, that is it varies during your crossing.

Well, that's still a ferry angle kind of situation. You set the ferry angle by calculating the total eastward (say) drift on the destination shore during the time of a hypothetical driftless crossing. Then aim at a point on the far shore that drift distance west of your destination. In your exaggerated case with opposing currents, there is no total drift east or west, so your ferry angle is zero, which says, aim directly at your destination.

So with non-uniform current, the course set by a ferry angle is not a straight line at all -- good point. And in fact, your course with a ferry angle is a straight line if and only if the current is uniform!

So the upshot of your example is that a GPS will not do anything for you in the case of significantly non-uniform cross-current. You will need to fall back on actual knowledge of currents and calculate the ferry angle based on the total drift you can expect over the crossing, taking into account the non-uniform current. You will also need to vary that ferry angle as you approach your destination, except in a case (like your) where the total drift is zero and the ferry angle is zero -- just leave it at zero.

Leon, Lisa -- what say you?

(Of course, this method assumes that the current varies only in one direction, the direction from the start to destination) and not at right angles to that direction, that is, it's not one speed in the middle and a different speed off to the side. If that isn't true, then things get really hairy. Also, if your crossing is long enough that current varies significantly over time, then it will also be more complex.)

[leong]

Gentlemen,

I worked in a radar systems design laboratory for over 30 years. Modern radar systems are extremely accurate. But not nearly as accurate as GPS. That’s why the “NextGen” air-traffic control system, which uses GPS satellites to pin-point every plane’s precise position in the sky once a second, plus onboard radios that let each aircraft continually see (and be seen by) all others nearby, is to be rolled out in 2012 and fully implemented by 2022.

Using a GPS on a kayak is a closed loop system with the paddler in the loop. The GPS tells you if you’ve left the minimum distance straight line. I don’t think any other way suggested approximates this with nearly the same accuracy (try paddling at night in heavy conditions and you’ll see what I mean). I've done it several times without a GPS.

Instead of all your theoretical hand waving about the straightness of a GPS track why don't you 1. read the responses by Lisa and me, and 2. look at a real GPS track that goes through currents and winds of varying speed and direction and see if it looks straight to you. In fact, feel free to compute the R-Squared goodness of fit coefficient. Then try it any other way you want to (inertial navigators are not allowed), to see how close it comes to a straight line. Only if you can show the line is straighter without a GPS I'll carry on this conversation. Otherwise, I've had enough of this. And I have nearly certain confidence that the conversation will end if the above experiment is performed.

Respectfully signing off,

Leon (who had a nice paddling trip today with four other NSPN-ers)

[leong]

I found a few pages that might help.

http://www.bask.org/...5_currents.html

http://www.kayarchy....03chartwork.htm

http://www.amazon.co...s/dp/0762738340

-Jason

Jason,

I read the two articles and own the book (it's sitting on the desk next to me). All of the suggested approaches to ferry angle are just approximations (you have to approximate your paddling speed, the current speed and direction and then do some trigonometric approximations to compute the ferry angle). Yes, John, you can do the trig exactly with your program or with my obsolete HP programmable calculator. But the problem is the very poor accuracy of the inputs (especially the wind speed and how it affects the kayak at different aspect angles).

Okay, I lied about the last post being my last. It was my penultimate post because I didn't notice Jason's post until after hitting post.

Goodbye folks.

Respectfully,

Leon

[JohnHuth]

I suspect, perhaps, we might actually all agree....although perhaps not? Let's see -

1.) A GPS gives you a highly accurate position fix

2.) A path given by your typical hand-held GPS from one point to another doesn't factor in conditions like current, etc.

3.) A straight line ferry angle in the presence only of a uniform current, factoring in the current is probably the most efficient

4.) You can calculate this ferry angle even with a piece of paper and a pencil

5.) Non-uniform currents, winds etc will create deviations from the idealized ferry angle that assumes a uniform current

Places where perhaps we don't agree:

1.) In the 'real world' are GPS'es just as good or better than using a compass and calculating a ferry angle assuming a uniform current?

I'd maintain that there's something of an 'art' that's lost in all this. I might know that there's a 2 knot current in the middle of a crossing and start out figuring out a ferry angle, but then realize that the current near shore is less than that, and back off the ferry angle by a bit. I'd then look at the wind conditions, and use some figure of merit about the sideways slip of the kayak due to windage.

Some of the stiffest tests of this is are crossings in the fog, where you lose visuals of the far shore, and you only learn how well you did once you reach the shore - no room for fudging. In some cases, the more foolhardy of us will make a crossing to a relatively small off-shore island in the fog in downeast Maine (mea culpa) where the tidal currents are large. In cases like that, I often don't have a table of currents, but just know the phase of the tide and some notion of what the max current is based on the topography and bathymetry. I'll still do a correction rather than follow a GPS heading, and in my experience you can do rather well, but a person has to throw in all the knowledge they have about the conditions. The straight-line ferry correction is just one factor that informs your decision about heading.

[leong]

I suspect, perhaps, we might actually all agree....although perhaps not? Let's see -

1.) A GPS gives you a highly accurate position fix

2.) A path given by your typical hand-held GPS from one point to another doesn't factor in conditions like current, etc.

3.) A straight line ferry angle in the presence only of a uniform current, factoring in the current is probably the most efficient

4.) You can calculate this ferry angle even with a piece of paper and a pencil

5.) Non-uniform currents, winds etc will create deviations from the idealized ferry angle that assumes a uniform current

Places where perhaps we don't agree:

1.) In the 'real world' are GPS'es just as good or better than using a compass and calculating a ferry angle assuming a uniform current?

I'd maintain that there's something of an 'art' that's lost in all this. I might know that there's a 2 knot current in the middle of a crossing and start out figuring out a ferry angle, but then realize that the current near shore is less than that, and back off the ferry angle by a bit. I'd then look at the wind conditions, and use some figure of merit about the sideways slip of the kayak due to windage.

Some of the stiffest tests of this is are crossings in the fog, where you lose visuals of the far shore, and you only learn how well you did once you reach the shore - no room for fudging. In some cases, the more foolhardy of us will make a crossing to a relatively small off-shore island in the fog in downeast Maine (mea culpa) where the tidal currents are large. In cases like that, I often don't have a table of currents, but just know the phase of the tide and some notion of what the max current is based on the topography and bathymetry. I'll still do a correction rather than follow a GPS heading, and in my experience you can do rather well, but a person has to throw in all the knowledge they have about the conditions. The straight-line ferry correction is just one factor that informs your decision about heading.

"Just when I thought I was out... they pull me back in." Godfather Leon

Try reading http://seakayakphoto...s-and-gpss.html to get a better idea of using a GPS to go to a waypoint.

"Non-uniform currents, winds etc will create deviations from the idealized ferry angle that assumes a uniform current"

Well, the GPS arrow (or line in some models) will account for non-uniform current and wind. Although the GPS doesn't directly compute a ferry angle, it's readily available. Just take difference between the heading you're paddling at (assuming that you're on track according to the GPS) and the bearing to the waypoint. That is the correct instantaneous ferry angle. Of course the ferry angle can change with wind and current velocities and the kayak’s velocity too. No matter what technique (other than using a GPS) you use to compute the ferry angle, you have to know the kayak’s speed, right?

[JohnHuth]

Yes, that will do it. There are many ways to skin a cat, no doubt. Yes, you'll need to know the kayak's speed, clearly.

As I said, I personally don't use GPS'es, but to the extent that it acts like a compass *and* gives you progress to a waypoint that strategy in the link is certainly viable.

This makes me think that perhaps GPS'es should have a little mini-calculation thrown in to give a ferry angle on-the-go, with some inputs from the user. It seems like a useful feature. I would imagine that some enterprising GPS programmer must have done this by now. If not, a great opportunity!

I should note that Pacific Islanders had some ingenious ways of dealing with current. In some cases, they would paddle or sail away from an island some distance, then stop and monitor their drift to get some idea of the current and adjust the heading appropriately. In other cases, they would line up two landmarks on a near shore, and then adjust the heading so that they were moving in a direction that kept these in line. The anthropologist Raymond Firth noted that some islanders in the Santa Cruz Reef Islands had specific names of landmarks on islands that were used to find the proper heading in the presence of currents. With a technique like this, no calculations are necessary.

[leong]

Yes, that will do it. There are many ways to skin a cat, no doubt. Yes, you'll need to know the kayak's speed, clearly.

As I said, I personally don't use GPS'es, but to the extent that it acts like a compass *and* gives you progress to a waypoint that strategy in the link is certainly viable.

This makes me think that perhaps GPS'es should have a little mini-calculation thrown in to give a ferry angle on-the-go, with some inputs from the user. It seems like a useful feature. I would imagine that some enterprising GPS programmer must have done this by now. If not, a great opportunity!

Using a modern GPS you don't need to know the ferry-angle. The track (defined by the GPS’s line or arrow that you’re supposed to follow) has the correct ferry angle incorporated (although the GPS doesn't actually calculate the ferry angle explicitly). See one of Lisa’s posts for more information on how the GPS does it. Here’s a thought experiment to see why ferry-angles don’t need to be calculated: Suppose there was always a visible line from your current position to the waypoint. Always head the kayak in the direction to keep it on the line. So you’re picking a heading that keeps you on the straight line to the waypoint. If you do a good job following this line, it’s always the same line. If you move off the line a new line is drawn from your current location to the waypoint. Anyway, always keeping on the line means that your heading is consistent with the correct ferry-angle, even though you never need to know the ferry-angle explicitly or compute it (except for curiosity).

I should note that Pacific Islanders had some ingenious ways of dealing with current. In some cases, they would paddle or sail away from an island some distance, then stop and monitor their drift to get some idea of the current and adjust the heading appropriately. In other cases, they would line up two landmarks on a near shore, and then adjust the heading so that they were moving in a direction that kept these in line. The anthropologist Raymond Firth noted that some islanders in the Santa Cruz Reef Islands had specific names of landmarks on islands that were used to find the proper heading in the presence of currents. With a technique like this, no calculations are necessary.

Yes, David Burch’s book “Fundamentals of Kayak Navigation”, (third edition which I’m now holding) calls this technique the “Set Angle Technique”. Quoting from the book, “One convenient way to judge your set looking back is to use the angular span of the hand. A typical outstretched hand at arm’s length is approximately 24 degrees …” Ok, what Burch is saying is to compute your ferry-angle by estimating your set angle. The ferry-angle is just the set angle in the opposite direction. So there is no need to know the wind or current velocity with this technique, although a constant paddling speed is implied, otherwise the set angle (ferry angle) doesn’t remain constant. In a practical sense here’s what I think is wrong with this technique. 1. It’s not that easy to look back over a kayak in heavy bump. 2. At best, the method is only a crude approximation. 3. The set angle changes with changing paddling speed and drift speed due to changing wind and current. 4. You have to have something to look back at so it only works for a short time after you paddle away from the put-in. Need I say more?

In the book’s third edition that I have (copywrite 1987) the chapter on GPS is quite obsolete although the author does covet GPSs. But the modern GPS with lines (or arrows) to point to the correct heading are not discussed because they’re probably weren't available yet.

[djlewis]

Hey, Leon -- I'm mostly staying out of this now, partly because I am confused by your responses. Seems to me that the status of the things we are discussing is pretty clear...

• with uniform current...
  • the least-distance, least-energy course is a straight line from launch point to destination
  • as Lisa said fairly clearly, you can use either a compass or a GPS to stay on that line
  • you can also compute a ferry angle and do the linear reduction trick; again as Lisa explained
• with significantly non-uniform current (thanks to Jason's wake-up call)...
  • the optimum course is not a straight line
  • the correct headings depend on the current patterns and boat speed
  • therefore, a GPS is not much use, except perhaps to help determine those patterns
  • the total-ferry-angle method with reduction is required (but linear reduction would only be an approximation)
  • that (potentially changing) angle can be monitored with a compass or a GPS being used only as a compass

I think that's the basics. There are a lot of details and tricks in the total picture, but they mostly involve executing the ferry-angle strategy.

Bye --David.

[leong]

Hey, Leon -- I'm mostly staying out of this now, partly because I am confused by your responses. Seems to me that the status of the things we are discussing is pretty clear...

• with uniform current...
  • the least-distance, least-energy course is a straight line from launch point to destination
  • as Lisa said fairly clearly, you can use either a compass or a GPS to stay on that line
  • you can also compute a ferry angle and do the linear reduction trick; again as Lisa explained
• with significantly non-uniform current (thanks to Jason's wake-up call)...
  • the optimum course is not a straight line
  • the correct headings depend on the current patterns and boat speed
  • therefore, a GPS is not much use, except perhaps to help determine those patterns
  • the total-ferry-angle method with reduction is required (but linear reduction would only be an approximation)
  • that (potentially changing) angle can be monitored with a compass or a GPS being used only as a compass

I think that's the basics. There are a lot of details and tricks in the total picture, but they mostly involve executing the ferry-angle strategy.

Bye --David.

David,

I disagree with almost everything that you said, even when you attributed it to Lisa. What Jason said was not a wake up call for me. Of course, there are situations where a straight line is not the best way to get somewhere on the water. Lisa and I know this well and discussed it years ago. You set multiple waypoints in those situations.

Apparently you’re not convinced that a modern GPS does a much better job than charts and a compass.

Ok, let’s settle this with a $100 bet. Here’s the bet (you can have your lawyers propose changes to the wording that I may or may not accept):

A third party will choose some random lobster pot about 5 miles from the put-in in Manchester Harbor. You and I will be given accurate lat/long coordinates of that “waypoint” 15 minutes before we launch together at night. The task is to paddle to and touch the chosen waypoint (it’ll be marked below the waterline with “you win”). I’ll use a GPS and you’ll use a compass and chart. My GPS will be continually illuminated and you can do the same with your charts and compass. We have two hours for the task. If I find and touch the waypoint and you don’t, I win. If you find and touch the waypoint and I don’t, then you win. It’s a tie if neither of us touch the waypoint in the allotted time. We can both carry flashlights and/or other lamps and others can go along for safety.

If it’s windy and/or there is a strong current my track (once leaving he harbor and clearing any islands) will be much straighter to the waypoint than yours will be, but we won’t count that. For simplicity, this bet is just to show how good a GPS is for finding a location using a GPS, not that it’s the optimum track. But that could be our next bet.

Wanna take this bet? Yes or no, enough with all of the pedantics.

Leon who’s sorry that he didn’t go paddling today (some errands had to be done)

[jason]

David,

I disagree with almost everything that you said, even when you attributed it to Lisa. What Jason said was not a wake up call for me. Of course, there are situations where a straight line is not the best way to get somewhere on the water. Lisa and I know this well and discussed it years ago. You set multiple waypoints in those situations.

Apparently you’re not convinced that a modern GPS does a much better job than charts and a compass.

Ok, let’s settle this with a $100 bet. Here’s the bet (you can have your lawyers propose changes to the wording that I may or may not accept):

A third party will choose some random lobster pot about 5 miles from the put-in in Manchester Harbor. You and I will be given accurate lat/long coordinates of that “waypoint” 15 minutes before we launch together at night. The task is to paddle to and touch the chosen waypoint (it’ll be marked below the waterline with “you win”). I’ll use a GPS and you’ll use a compass and chart. My GPS will be continually illuminated and you can do the same with your charts and compass. We have two hours for the task. If I find and touch the waypoint and you don’t, I win. If you find and touch the waypoint and I don’t, then you win. It’s a tie if neither of us touch the waypoint in the allotted time. We can both carry flashlights and/or other lamps and others can go along for safety.

If it’s windy and/or there is a strong current my track (once leaving he harbor and clearing any islands) will be much straighter to the waypoint than yours will be, but we won’t count that. For simplicity, this bet is just to show how good a GPS is for finding a location using a GPS, not that it’s the optimum track. But that could be our next bet.

Wanna take this bet? Yes or no, enough with all of the pedantics.

Leon who’s sorry that he didn’t go paddling today (some errands had to be done)

Leon,

In your bet location, it doesn't have any interesting currents. Doing it i woods hole, Popham etc on springs would be a lot better. Your example wouldn't give David a fighting chance....

[leong]

Leon,

In your bet location, it doesn't have any interesting currents. Doing it i woods hole, Popham etc on springs would be a lot better. Your example wouldn't give David a fighting chance....

Okay, so what you're saying is the compass is better in real conditions than the GPS. Let's find out.

Can you choose a location from Nahant to Kittery Point. (perhaps crossing tidal race of Plum Island during max flood or ebb tide). Pick a place and I'd be pleased to lose the $100 to you. I don't want to drive to the Woods Hole area or farther north than Kittery.

[JohnHuth]

I have to say, this is quite an entertaining thread. I find that in message boards things frequently devolve this way.

I realized how you can use a GPS to advantage to find a heading in the presence of currents (partly through that URL Leong posted). This, in part uses the GPS as a compass, but also as an aid to measuring currents. If you are able to track your path toward a destination, you can try an initial heading directly toward the destination. The current will make that heading shift, and, in fact, if you have a display of your actual track, you can estimate the angle off, and then adjust the heading until you're heading in a straight line toward the destination. OR, you can go into the current, let yourself drift and measure that drift and adjust accordingly - then switch on the compass feature and it's pretty much the same as an ordinary compass, but you've also measured the local current. So "GPS as a current measuring device" seems possible in principle. I don't use one, so I don't know if it's practical, but it strikes me as possible.

As far as the $100 bet is concerned, I think it misses the point David was trying to make.

Let's go back to the more garden-variety use of a GPS, where it is more of a hiker's device and it gives you a 'straight line' to your destination, and is not employed in a slightly more sophisticated way to measure current, either through drift measurement or adjustment of heading. In that 'garden variety' case, you will be changing your heading constantly and taking a slower path than if you had figured current initially and figured a heading based on a compensation for that.

A 'bet' along these lines would be something more like a crossing to an island in the presence of a current where person A uses a compass and figures in current and person B rigidly adheres to a projected GPS line that tries to always move toward a target, assuming no current. Then, the bet is whether person A or person B reaches the target island first. The target island should be big enough that it presents a minimum 10 degree spread from the starting point. This takes out the meter-ish precision of the GPS as a factor in actually locating a very small target.

The lobster-buoy finding exercise only uses the enhanced position accuracy of a GPS and isn't really pertinent to the question of the original 'pursuit curve' idea of the thread, namely, is the straight line path that factors in current faster than one where the heading is constantly pointing toward the target.

Having said that, I embarked on a rather fun exercise with Mark Schoon. We were crossing from Swan Island to Placentia Island in a thick fog with the current draining out of Blue Hill Bay. The 'game' was to compensate for the current as precisely as possible an hit a buoy halfway across (I think the full crossing was two miles and the buoy was halfway). It was all map-and-compass. We figured a 10 degree ferry angle. After paddling maybe 15-20 minutes, damn if we didn't hit right on that buoy! I was a pretty happy camper that we were able to do that. I wouldn't have expected that precision, but the angular precision was pretty darn high. It might have been luck that we did that well, but it made me convinced you can do quite well with a standard map-and-compass and seat-of-the-pants estimate of current.

[lhunt]

I have to say, this is quite an entertaining thread. I find that in message boards things frequently devolve this way.

...

Let's go back to the more garden-variety use of a GPS, where it is more of a hiker's device and it gives you a 'straight line' to your destination, and is not employed in a slightly more sophisticated way to measure current, either through drift measurement or adjustment of heading.

I'm enjoying it too, but the frustrating part of the conversation is simply this: I suspect the two most vehement detractors don't know how the GPS works. I think probably Jason knows how a GPS works and likes his way better, which is nobody's business but his own, so I'm setting Jason aside (no offense, Jason) :-)

The previous statement by John that "This makes me think that perhaps GPS'es should have a little mini-calculation thrown in to give a ferry angle on-the-go, with some inputs from the user.", and by David that "So the upshot of your example is that a GPS will not do anything for you in the case of significantly non-uniform cross-current." make me think they don't understand how it works. I'm probably not explaining it well, and I don't know that one more time will make the difference, but here goes:

The GPS gets your position. One second later, it gets your position again. It doesn't know or care which direction you are pointing in. It knows which direction you are moving in. If you are moving too far to the left, it tells you to turn right. Variations in current make it tell you to move more or less to the right. No user inputs are needed other than the location you were a second ago vs. the location you are in now. It has no sophisticated way to measure current, it measures current by knowing your direction of travel. It is a garden variety hikers' GPS (at least that's what I use - I bought it before I ever started paddling).

With all respect to Leon, maybe the point would be better made by getting the protagonists together, giving each a GPS, and taking away their compasses just for fun, on a nice day where everybody knows exactly where they are and there is one good non-uniform current to cross. No danger to muss up the experiment. Within 5 minutes you will get the main point of how the thing works.

After that, whether one uses a GPS or not is a completely different question. I think if I were going on a long and dangerous trip of any type, I would want a chart, compass and a GPS (or two). Getting back to Jason for a moment, yes - of course if you have two opposing currents of equal ferocity, a straight line is a bad idea, so your GPS would only be to help you if you have some unforeseen problem that messes up your trajectory significantly.

Anyway, I'm off to the Solstice paddle. Hope to try the experiment someday!

Lisa

[JohnHuth]

Let me try again.

OK, a GPS has many features. You will get your present position, either displayed on a map, or as a latitude or longitude. I will also give you your 'true' heading - that is to say, the direction in which you're moving and also your speed, as measured against a "fixed" Earth. It will also point you in the direction of a pre-determined waypoint. It will also give you a heading to that waypoint, assuming that fixed reference frame. It will also give you a compass showing true north and your 'true' heading with respect to true north. It will also keep a record of your travels and you can upload them to a computer etc.

Now, what a GPS typically doesn't have is the ability to download apps like you might with an iPhone. I was only suggesting that the ability to download certain aids to navigation might be helpful.

Yes, I've used a GPS - it was a Garmin and I've burned through 3 of these as the salt water tends to chew up the receivers rapidly and I've just given up on them. They also have a tendency to fail in heavy rain and fog - just an observation.

Now, using a GPS to assess current set and drift is doable. You paddle out into your stream, and just set there. Leave the GPS on, and it will give you, effectively the current set and drift, as this is just the direction you're moving and speed. OR....you can use a heading that's based on your GPS' compass referenced to true north as you cross, but then you'll find your 'true' heading (i.e. wrt a fixed reference) deviates from that due to the effect of current. If you adjust your heading (direction in which you're paddling) so that the 'true' heading is pointing at the waypoint you've programmed in, you'll have just found a ferry angle using the GPS.

Again, to get back to the pursuit curve point of the thread.... the issue as I see it is that if you program in a waypoint that's at the other end of the crossing, the GPS will keep you pointed in that direction. If you head in that direction as the GPS "thinks" it should if you were not in a current stream, then you will find yourself continually adjusting the heading from the GPS, as you will get swept downstream and the GPS will want to compensate for that.

All David is saying is that if you, instead, calculate your compass heading, assuming that there is a uniform current, you'll take a straight line path, rather than the pursuit curve that results from following the multiple headings the GPS will give you in trying to direct you toward the waypoint on the far side. This is faster.

What I think Leong is saying is that a) there are other factors, like non-uniform current and waves and wind (true) and there are ways of using the GPS to determine current and even take it into account (true), although those are not "smarts" built into the GPS - it is using the GPS as an instrument to determine current and even help you figure out an optimal ferry angle.

However, the issue is not the inherent precision of the GPS, it is simply the difference of a pursuit curve, using the GPS path as an EXAMPLE of what happens if you continually try to point toward a landmark on the far side of the crossing.

There's a difference between what we (or I) am saying about a highly reductionistic statement of a mathematical problem using the GPS feature of pointing toward a waypoint as a way of putting this idea across as a concept and the real world, where the GPS can help you determine current and adjust for it (again using some native smarts).

Whew.... clearer?

[lhunt]

Now, using a GPS to assess current set and drift is doable. You paddle out into your stream, and just set there. Leave the GPS on, and it will give you, effectively the current set and drift, as this is just the direction you're moving and speed. OR....you can use a heading that's based on your GPS' compass referenced to true north as you cross, but then you'll find your 'true' heading (i.e. wrt a fixed reference) deviates from that due to the effect of current. If you adjust your heading (direction in which you're paddling) so that the 'true' heading is pointing at the waypoint you've programmed in, you'll have just found a ferry angle using the GPS.

OK, yes, I think I can see the problem. If the GPS's you've used had compasses, the compass was messing you up. There is no need to sit still and evaluate the speed of the current. You turn the compass off, and use the heading the GPS gives you. This will give you the correct ferry angle "as you go". You adjust your heading so that the course pointer is at the top of the screen. When you look up, the landmark you are heading for won't be at your bow, because you are already on the right ferry angle. Your GPS will show the waypoint as being straight in front of you, but on the "real earth" the waypoint will be some degrees off that. That's because the GPS doesn't know what direction you are holding it in (it doesn't have a compass). It only knows what direction you are heading in.

I'll check this more tonight or tomorrow - gotta go now...

Lisa

[leong]

Many of you still don't understand how you can use a GPS. Make up the rules for your own bet and I'll accommodate you. I wouldn't make it a speed test because, except for about 10 of the faster kayakers that I know, I'll win that bet easily.

Off to the Marblehead I go.

[JohnHuth]

No need for a bet, a straight line path is faster than the pursuit curve as stated, (using the GPS waypoint following as an example). It just is. No arguing or equivocation. It's like making a bet whether or not the tangent of 45 degrees is 1, you don't need to touch water to settle that (I would've thought).