Assume (Knowing where you aren't)

[Cornishman]

Re: Assume (Knowing where you aren\'t)

Navigation News May/June 2002 edition has an interesting solution to your problem by one Tony Crowley making a Seafarer's Sun Dial. The next item in the same edition might also satisfy those who are regularly asking about SOLAS Chapter V.

 

[NigeCh]

Full answer posted

It's on BYM and is just as contentious as it is trivial:

http://www.boatsyachtsmarinas.com/cgi-bin/datacgi/database.cgi?file=Forum&report=TopicIndex

It's under 'Part 1' 'Part II' and 'Part III' - It's long and it's boring unless you really want to delve into wanting to know.

 

[BrendanS]

Re: Full answer posted

You didn't say anything in the original question about having a supercomputer along to process the equation /forums/images/icons/wink.gif

 

[Twister_Ken]

It\'s amazing what you can get in a Casio...

... wristwatch nowadays!

 

[BrendanS]

Re: It\'s amazing what you can get in a Casio...

OOps! My mistake! forgot about the advances in minaturisation! /forums/images/icons/wink.gif

 

[NigeCh]

I\'m eating humble pie

I only tested it out in my back garden. The tests were made over the past few days using sunrise and sunset times taken from the local paper (trees in the way) and the GPS for more 'accuracy'

On BYM I was challenged about how the method would work at equinox .... and it failed. It also failed as you need to create tables for various latitudes and have them with you ... unless you have a calculator capable of iteration. I also didn't take into account that the various astromomic agencies around the world publish DIFFERENT sunrise and sunset times for a unique Lat/Long for the same day .... My approximated sightings of sunrise and sunset times therefore go out of the window as they (the national agencies) are paid to get them right and all I can do is log sunrise/sunset on a watch ... and the time differences as forecast by them have significant variations.

I've emailed them all to ask them why their times are so different each from the other ... Let's see what transpires.

I now no longer have a clue just what time sunrise is anywhere even if I see it and write the time down.

 

[BrendanS]

Re: I\'m eating humble pie

Bleedin' good try anyway! How long did it take to work out the maths? Haven't tried out anything that complicated since I did stats at Uni, and wouldn't like to attempt it now!

The sunrise/sunset timing is v.interesting. One of those things you take for granted, until someone actually does something like this which highlights the issue.

If you get any response from any of the agencies, please let me know. I'd like to know what they say

 

[NigeCh]

I was a bit premature with the results

You will need a list of daily tables or a programmable calculator ... I'm building tables to test the pragma and hope to turn it into a sort of definitive ... ie, sofar it's working after a fashion. I got hammered on equinoxes (Max error) ... but, overall, it seems to be able to locate yourself fairly accurately as long as you are south of 65 degrees N or 65 degrees south.

When I've worked out the anomolies I'll post post the results ... In theory it's simple. In practice it's become nightmare (and that's the fun part to build in the corrections) - You have to use table printouts and then use interpolations ... BUT if you haven't a sextant then I still think that this visual sunrise timing method will work to give you a fairly accurate Lat/Long of where you are.

 

[trev]

Re: I was a bit premature with the results

I think the whole concept is a load of crap!
You can have a good guess but you need a nautical almanac to determine the latitude of the time of sunrise/sunset. You would also require the declination of the sun to apply to an estimated meridian altitude.

Trev

 

[AndrewB]

Not entirely crap ...

Nige's analemma curve provides declination. In any case, if you have means to draw a circle (round the watch perhaps) and can remember that the top corresponds to 23½ degrees on June 21, its possible to estimate this graphically. Length of day, which is constant from year to year, can be determined approximately from a fairly simple formulae involving date and latitude, given for example in Peter Duffet-Smith's 'Practical astronomy with your calculator', but trig tables would be handy.

Length of day as a measure of latitude works best at higher latitudes (not within artic/antarctic circle), and towards mid-summer and mid-winter. I still prefer lashing up some form of cross-staff to estimate the sun's, or the pole-star's altitude.

 

[trev]

Re: Not entirely crap ...

But you are assuming you already know your latitude !!

Trev

 

[NigeCh]

No. You don\'t know Latitude to start with

Now then Trev,

The time of observed sunrise between say 0 and 65 degrees north/south on the prime meridian for each day of the year has a unique set of values for each degree, minute, second of latitude. The time of sunrise includes the analemma, declination and approx defraction.

I think that if you have a cheap watch such as a Casio F-91W (£9.99) set to UTC telling you what the date is and if you record the sunrise time then you will be able to work out what your Lat/Long is by comparing your visual sunrise time to that day on the prime meridian for that day of the year by using the formula that I've previously posted.

The original question posed is trivial in context of having GPS. It's not trivial if you don't have a sextant. It's a proposal for a method of working out where you are with neither GPS nor sextant .... but you do need a watch and either a set of tables or a progammable calculator as Sir Cloudsey-Shovell didn't have.

I haven't proven the method yet. It may be a load of crap (I don't think that it is - I'm getting some fairly good results) ... but just because we have GPS and chart plotters doesn't mean that we still can't attempt to further simplistic 'lost at sea' navigation by visual/time methods.

Sincerely,

NigeCh

 

[AndrewB]

Sorry, didn\'t make it clear ...

... the formula gives day length as a function of date and latitude (to the nearest 15 secs or so). This can be turned around by normal alegbra to give latitude as a function of date and length of day.

 

[NigeCh]

Nice post

Andrew,

It seems to me from all the replies in this thread that there are a number of people here who question and care about navigation away from shores. Yes it's easy for those who (and I hope that I'm not being arrogant) to agree that it's easy to press buttons on either a GPS or a chart plotter and all is revealed, but I think that part of the fun of sailing well away from land is working out your position without reference to electronic aids. Yes, it dates me, but it's part of my chart and pencil philosophy.

Thanks for the support ... a back-staff or knowing solar noon using a metre stick etc., would work just as well ... My arguement is that sunrise on the sea is easier to determine to, say, the nearest 30 seconds or minute providing that there is no cloud cover than to tilt the sextant on a wobbily boat to take a noonday mark - but then I might be wrong.

Cheers,

NigeCh

 

[bedouin]

Re: Nice post

As you know I'm not entirely sure about what accuracy can be achieved. Atmospheric effects, particularly near the horizon, can have a significant impact on these measurements.

The conventional wisdom is that sights with a declination of 10degrees or less are too susceptible to error for accurate navigation.

 

[NigeCh]

And that\'s a nice post too.

A visual timing of sunrise is an exact timing. The only things (that I can think of) that can be out to affect this timing are the accuracy of your watch relative to UTC, the analemma's daily time shift and atmospheric's humidity effect on refraction.

There is no declination involved in taking a time ... on a boat at sea in the middle of nowhere .... I'm not saying that my solution is correct or even approximately correct ...

We have as consultant editors and contributors to YM both JJ and Tom Cunliffe. JJ has posted a solution from Viceroy in Canada. Tom Cunliffe has written a book on celestial navigation.

Perhaps a measured post from Tom Cunliffe as a response might be as a similititude, ie as the difference between a dog doing it on a tree or a dog doing it in a treeless snowfield vis a vis 'lost in an ocean.'

 

[Stardust]

Re: Assume (Knowing where you aren\'t)

Gawd, Nige, are you still at it!

John Hyatt.