Checking potential collision course?

[Keen_Ed]

Have to admit that I use the compare agains the background method for the last 10 years while racing in the Solent. In discussions with an American, she called it "making trees". If the boat is moving forward compared to the bank - i.e., the scenery is being revealed from behind the other boat's jib, you're "making trees" - and will/(should!) pass in front.

A few notes. It works best when you're both beating. The time you really want to be in front is when you're trying to cross a starboard tack boat when you're on port. So what's critical is the last 30 seconds leading up to the cross.

And if you have any sense, you want to leave a bit of a gap if you're crossing on port (At least half a boat length in ideal conditions - more in windier)

If you're sailing at 5 knots, your VMG is going to be something like 3.5 knots to windward. 3.5 knots is 6 feet per second. So in thirty seconds, you travel 180 feet directly to windward (and 180 feet sideways). If the shore is 4 miles away (easily done in the Solent), then the bearing change with a fixed point ashore is half a degree.

 

[DJE]

Where the "movement relative to the background" method never fails is the case when the other boat appears not to move in relation to the shore. In that case the two boats and the fixed point on shore are staying on the same straight line (albeit that line is rotating as you move) so if you're both stuck on the same line you're either moving directly away from each other or you're going to hit.

 

[timbartlett]

Where the "movement relative to the background" method never fails is the case when the other boat appears not to move in relation to the shore. In that case the two boats and the fixed point on shore are staying on the same straight line (albeit that line is rotating as you move) so if you're both stuck on the same line you're either moving directly away from each other or you're going to hit.

If you are moving past a fixed object, then the bearing of that fixed object from you must be changing.
If the other boat appears to be staying on your line of sight, then the bearing to the other boat from you must also be changing.

Only in quite unusual circumstances is a changing bearing indicative of a risk of collision.

Yay! Tim ... we missed you ...

I hope you spotted the in my previous email!

Seriously, though, I note that everyone who advocates this "lining up with a landmark" method does so in the context of racing -- particularly inshore racing, where distances are very short, and the decision to duck, tack, or stand on is being made over a matter of seconds and at a range of a few yards. Under those conditions, it's fine, because the bearing of a landmark (even one that is only a mile or so away) cannot change very much at sailboat speeds over a few seconds.

But for collision avoidance against a large vessel in open water, where a CPA of less than a mile might be regarded as uncomfortably close and many skippers would want to take a decision at a range of 3+ miles, it would be flawed. You wouldn't, for instance, want to use it to dodge the ships through the Dover Straits ! (would you?)

 

[jimi]

Seriously, though, I note that everyone who advocates this "lining up with a landmark" method does so in the context of racing -- particularly inshore racing, where distances are very short, and the decision to duck, tack, or stand on is being made over a matter of seconds and at a range of a few yards.

Wrong! I use it to assess the risk of collision whenever there's land on the background.... and theoretically it works better the further the land is away.

IE the ultimate comparison against land (as I said earlier) is when using a comapss bearing, because the magnetic pole is a fixed point on the earth. In that case you're just using a compass instead of your eyes.

 

[timbartlett]

Wrong! I use it to assess the risk of collision whenever there's land on the background.... and theoretically it works better the further the land is away.

IE the ultimate comparison against land (as I said earlier) is when using a comapss bearing, because the magnetic pole is a fixed point on the earth. In that case you're just using a compass instead of your eyes.

Generations of professional seafarers have learned that a compass bearing is the only reliable test. Every textbook mentions the compass bearing test -- somewhere between none and very few of them mention the "moving against land" test. The collision regulations specify the compass bearing test.

But of course, this is all just book learning. None of it is half as reliable as an expert opinion on ybw.com

 

[Twister_Ken]

You wouldn't, for instance, want to use it to dodge the ships through the Dover Straits ! (would you?)

Actually, I think I would. If France (or Britain) was emerging from the bows of the ship, I'd know I was going in front of it*. If the ship was 'eating' land, I'd know I was going behind it.

*But if land was emerging only very slowly, I'd probably conclude that a dangerously close bow-crossing was in prospect, and take appropriate action. I might also resort to the HBC as well as seeing what Alfie the AIS thought.

 

[jimi]

Generations of professional seafarers have learned that a compass bearing is the only reliable test. Every textbook mentions the compass bearing test -- somewhere between none and very few of them mention the "moving against land" test. The collision regulations specify the compass bearing test.

But of course, this is all just book learning. None of it is half as reliable as an expert opinion on ybw.com

I'll bear it in mind next time I read your posts

 

[timbartlett]

Here (you'll need to click on the thumnbnail to look at it full size, I expect) is a diagram of a yacht steering approximately SE at 5knots and a ship steering approximately NE at 20knots, towards a potential collision at T=0.

I've plotted their positions at three minute intervals from T=-18 (18 minutes before the collision, at which point they are a little over 6 miles apart.

The yellow triangle represents land, and the dotted lines represent the yacht skipper's line of sight at T=-18, T=-12, and T=-6.

You can see that the three lines of sight are all parallel -- as one would expect, given that the classic test for a risk of collision is the "steady bearing".

Equally, you can see that they all strike the coastline at different places. To the yacht skipper, the ship would appear to be "eating up" the coastline, leading him to (wrongly) assume that he was going to pass safely astern of the ship.

 

[timbartlett]

Wrong! I use it to assess the risk of collision whenever there's land on the background.... and theoretically it works better the further the land is away.

IE the ultimate comparison against land (as I said earlier) is when using a comapss bearing, because the magnetic pole is a fixed point on the earth. In that case you're just using a compass instead of your eyes.

The fallacy in this argument is in assuming that just because the N mark of the compass points to a "fixed" point, all the other marks also point to fixed points, when they clearly do not.

If I was near Alderney, for instance, the N mark would point towards the N magnetic pole, but the E mark would point towards the Cherbourg peninsula. If I was off the Scillies, the N mark would still point towards the N magnetic pole, but the E mark would point towards Lands End.

 

[alant]

Here (you'll need to click on the thumnbnail to look at it full size, I expect) is a diagram of a yacht steering approximately SE at 5knots and a ship steering approximately NE at 20knots, towards a potential collision at T=0.

I've plotted their positions at three minute intervals from T=-18 (18 minutes before the collision, at which point they are a little over 6 miles apart.

The yellow triangle represents land, and the dotted lines represent the yacht skipper's line of sight at T=-18, T=-12, and T=-6.

You can see that the three lines of sight are all parallel -- as one would expect, given that the classic test for a risk of collision is the "steady bearing".

Equally, you can see that they all strike the coastline at different places. To the yacht skipper, the ship would appear to be "eating up" the coastline, leading him to (wrongly) assume that he was going to pass safely astern of the ship.

Can this be done as a printable version?

 

[fireball]

Can this be done as a printable version?

Without wanting to circumvent copyright statements Tim might want to state - just right click on the image and open in a new window ...
The text you can copy/paste ...

Don't forget to add who it was written by ...

 

[alant]

Without wanting to circumvent copyright statements Tim might want to state - just right click on the image and open in a new window ...
The text you can copy/paste ...

Don't forget to add who it was written by ...

Had already tried that & after your post, retried, but doesn't do it for me, when pasting onto a word doc.

 

[ChrisE]

Here (you'll need to click on the thumnbnail to look at it full size, I expect) is a diagram of a yacht steering approximately SE at 5knots and a ship steering approximately NE at 20knots, towards a potential collision at T=0.

I've plotted their positions at three minute intervals from T=-18 (18 minutes before the collision, at which point they are a little over 6 miles apart.

The yellow triangle represents land, and the dotted lines represent the yacht skipper's line of sight at T=-18, T=-12, and T=-6.

You can see that the three lines of sight are all parallel -- as one would expect, given that the classic test for a risk of collision is the "steady bearing".

Equally, you can see that they all strike the coastline at different places. To the yacht skipper, the ship would appear to be "eating up" the coastline, leading him to (wrongly) assume that he was going to pass safely astern of the ship.

Can't deny your geometry here, perhaps the thing that we have all missed out is the aspect of the approaching ship. In the case you have outlined the aspect would remain the same, I suspect but don't know, is that what we do is correlate both movement relative to shore plus change in aspect to adjudge risk of collision.

 

[ancientsailor]

So all in all, happy conclusion, we're all right. ;-)

I’ve been away for a couple of days – I thought that comment had ended this thread – I thought it was a good place to stop!

Tim Bartlett’s diagram shows when the method’s especially weak – a fast-moving target close to the shore.

I’ve been thinking about a couple of scenarios to illustrate the strengths and limitations ...

First: Own vessel heading (say) north with land object and another vessel on the starboard beam. Land object bearing is therefore moving “astern” onto the quarter. This change in bearing is most marked because the fixed object is on the beam. The higher own vessel's speed the less relaiable the outcome. Vessels' tracks will cross at some point:
Other vessel:

• moving astern against land object: - will pass astern: monitor with care
• steady on land object – probably pass astern: monitor with care
• moving ahead slowly against land object – may pass ahead or collide: monitor with even greater care
• moving ahead rapidly against land bearing – may pass ahead: monitor with care.

Second: Own vessel heading (say) north with land and another vessel fine on one bow. Land bearing is therefore moving toward the beam – but more slowly than when it is abeam. Vessels' tracks will cross at some point:
Other vessel:

• steady on land object – high risk of collision
• moving slowly against land object – probably pass close: monitor with great care
  
  - pass ahead (moving “forwards” against land object)or
  - pass astern (moving “backwards” against land object)​
• rapidly moving against land object – probably pass well clear: monitor with care

With the land object bearing close to:

• ahead or astern in a slow-moving vessel, with the land distant, the method is highly reliable
• abeam in a fast-moving vessel, with the land close, the method is unreliable

... and in between!

I’ve used it closing the land and crossing the shipping lanes ... as well as racing. But I resort to other methods if in doubt. It cannot be denied, there is one proper way, and only one - compass bearings. In reality the method's great when used with care.

When I was deep-sea on a stable bridge 70 feet above the water, monitoring ships as soon as the masthead lights came over the horizon – well that was compass bearings. In those days, we didn’t have the radar on - except in restricted visibility. Bouncing around in a small boat trying to use a hand bearing compass or line up against stanchions or stays is a different matter ...

 

[jimi]

yup!!

Those that are good at these things might find it elucidating to draw out several scenarios where the boats pass ahead and behind as well as collide. It will be beneficial to draw 4 points of equidistant time to and note how the angles start to change rapidly as the the craft approach each other. A comparison of these scenarios may even lead the dogmatic to grudgingly concede that there is merit in this rule of thumb .. but I'm not holding my breath!


If I can find a graphics tool that I can operate I might have a bash (sic) later!

 

[timbartlett]

"ye cannae change the laws o' physics, jimi"

 

[jimi]

"ye cannae change the laws o' physics, jimi"

I'm certainly not even attempting to do that, however the application of a bit of simple geometry and trigonometry will display exactly what I have attempted to say. So rather tthan using your considerable talents to attempt condescending oneliners, why don't you try to model the various scenarios I described earlier, and see how the differences in the rate of change are of value as a RoT? If not, I'll be forced to scan in my sketches and inflict them upon you.

 

[RichardS]

Can this be done as a printable version?

Just right click on Tim's picture (the full size version rather than the thumbnail), Save As to your hard drive, and print it off with MS Paint (or whatever graphics program you have on your PC) or import it into your Word document.

Richard

 

[scotty123]

Help

Here (you'll need to click on the thumnbnail to look at it full size, I expect) is a diagram of a yacht steering approximately SE at 5knots and a ship steering approximately NE at 20knots, towards a potential collision at T=0.

I've plotted their positions at three minute intervals from T=-18 (18 minutes before the collision, at which point they are a little over 6 miles apart.

The yellow triangle represents land, and the dotted lines represent the yacht skipper's line of sight at T=-18, T=-12, and T=-6.

You can see that the three lines of sight are all parallel -- as one would expect, given that the classic test for a risk of collision is the "steady bearing".

Equally, you can see that they all strike the coastline at different places. To the yacht skipper, the ship would appear to be "eating up" the coastline, leading him to (wrongly) assume that he was going to pass safely astern of the ship.

I've been trying to follow some of the more 'technical/geometric' offerings on this thread, a few I understand, many I admit defeat in understanding.
Maybe a tutorial is required.

In the above, could you please explain that diagram, because it seems to me, that the ship is on a constant bearing from the yacht and is seen moving along, but not "eating up" the coastline which I always understood (particularly from some posts on this thread) meant that the 'target' was moving ahead (or behind) a steady bearing against a moving land background.

 

[Angele]

In the above, could you please explain that diagram, because it seems to me, that the ship is on a constant bearing from the yacht and is seen moving along, but not "eating up" the coastline which I always understood (particularly from some posts on this thread) meant that the 'target' was moving ahead (or behind) a steady bearing against a moving land background.

Land doesn't move!

No, seriously. What you have described there is a clear case of the relative bearing of the other boat changing which is a clear (dare I say, undisputed?) proof that you are going to miss each other. I'm not sure the land in the background is necessary in identifying that in this case and not just that the two of you are moving forward relative to it. But, it might do if the land is very distant (and its bearing is therefore changing very slowly) and, if you think it does, then fine.