NavList:

I agree that all coordinates "before GPS" have nothing to do with geodesy.
Coordinates of an isolated island like this could be determined before
1970-s only by
astronavigation, that is by Cel nav.

I inspected this island on the satellite photo provided by Google maps.
It seems uninhabitable; no trace of any human activity. The only
conspicuous feature is a white strip
on the perimeter of the island which looks on the satellite photo like a
road. But I conjecture that this is not a road but a sand beach, and
the photo is taken when the water was low. (It is pity that Google does not
give elevation).

It is not clear whether anyone had ever landed on this island. So when
they determined its coordinates, they probably did this from a ship.
Still, I believe that in XX century they could easily determine
longitude with a chronometer to 1 mile.

It is not clear to which point on the island the coordinates refer.
I don't think anyone ever built a geodetic point on this island (what for?)
The width of the island in the E-W direction seems to be less than 1/3  of
a of a mile, so a change of several miles in longitude seems
substantial:-)

Next comes the question whether the island could have shifted its position
in about 100 years:-)

From the Google photo I can infer that this is a mountain top, so probably
it could not shift much, even if the ocean level changed locally.

Alex.


> What has Geodesy to do with it?
>   ----- Original Message -----
>   From: Gary LaPook
>   To: CHRISTOPHER PRESMEG
>   Sent: Tuesday, December 17, 2013 1:07 PM
>   Subject: [NavList] A question for the geodisists
>
>
>
> ------------------------------------------------------------------------------
>
>   On July 2, 1937 Amelia Earhart was attempting to find Howland island.
> Her navigator, Fred Noonan, was using celestial navigation to fly the
> standard single line of position landfall approach to the island. (see:
> https://sites.google.com/site/fredienoonan/discussions/navigation-to-howland-island
> )
>
>
>   The Pacific Sailing Directions (1885) gives the location of Howland as
> 0° 49' north, 176° 40' west. Bowditch (1920) gives it as 0° 49' 00"
> north, 176° 43' 09" west. Bowditch (1938) gives it as 0° 48' north,
> 176° 38' west. This is still listed as the correct coordinates in H.O.
> 126.
>
>
>   Some have argued that Noonan was working with the old coordinates and
> was aiming for 0° 49' 00" north, 176° 43' 09" west and so missed the
> island by about five nautical miles. However, the correct coordinates
> were given on the June 1937 edition of the local chart, 1198, so we know
> that the correct coordinates had been found sometime prior to Earhart's
> departure from the states so I would think that the updated information
> would have been provided to her, considering that the orders to assist
> her came directly from the White House and the importance of the correct
> information would have been apparent to anyone.
>
>
>
>   I have looked at Howland on Google Earth and it now appears that it has
> moved to the east another mile! And the coordinates found on the most
> recent chart now show it also at 0° 48' 00" north, 176° 37' west, one
> mile east of he published location. How does the island keep moving? If
> this is the true location then Noonan may have been aiming another NM
> too far to the west. There is a note on the newest chart that says:
>
>
>
>   So it appears that the most recent movement resulted from changing the
> datum to WGS 84 but it seems to me that an adjustment of an entire mile
> to the east is very large compared to other such shifts. So, if Noonan
> was using the old coordinates would he have been aiming six NM west or
> just five? And I think that I remember this correctly, that the geodetic
> gratical used for map making places the landmarks where they would be
> placed by celestial navigation. So if the island really is a at 176°
> 37' west then this might help explain why they couldn't find the island.
> I thought that the greatest difference between the geodetic and the
> geographic coordinates, as far as celestial navigation is concerned, is
> in latitude and that changing to WGS 84 wouldn't have such a large
> effect on longitude. Can someone explain how shifting from one datum to
> another would cause this type of error in the celestial navigation.
>
>
>
>
>
>
>   gl
>
>
>
>   Attached File:
>
>   (img/125785.howland pages from 1920
> american_practical_navigator_an_epitome.jpg: Open and save)
>
>   Attached File:
>
>   (img/125785.howland position_page_1.jpg: Open and save)
>
>   Attached File:
>
>   (img/125785.howland position_page_2.jpg: Open and save)
>
>   Attached File:
>
>   (img/125785.howland ge.jpg: Open and save)
>
>   Attached File:
>
>   (img/125785.howland chart 83116.pdf: Open and save or View online)
>
>
>
>   : http://fer3.com/arc/m2.aspx?i=125785
>
>
> : http://fer3.com/arc/m2.aspx?i=125799
>
>
>
>