[Talk-us] Maui, Lanai, and Molokai (was Re: Characterizing Systematic Errors in Hawaii and US Pacific Island Data)

Scott Atwood scott.roy.atwood at gmail.com
Mon Oct 5 09:05:37 BST 2009

This is a small island with a relatively limited road network, concentrated
mostly in a single small, central town.  It looks like there has been
extensive manual editing of the road network, and it seems to correspond
fairly well to the satellite photos.


The coastline data seems to be fairly accurate.  The road network on this
island seems to have had little or no manual editing since it was imported
from TIGER in 2007.  The roads appear to be displaced approximately 300m to
the west and 400m to the north, based on on the satellite photos.


The coastline data seems to be fairly accurate.  I spot checked a number of
locations in the road network, including Lahaina, Kahului, Wailuku,
Haleakala Highway, and Hana Highway.  Every placed I checked had not been
manually edited since it was imported from TIGER in 2007, and seems to have
the same NNW displacement as seen on Molokaʻi.

Next steps for Lanaʻi, Molokaʻi, and Maui:

I think Lanaʻi can be left as is.  Given that there has been little or no
substantial manual editing of the road networks of Molokaʻi and Maui, and
these islands both suffer from a substatial offset (presumably due to an
incorrect datum in the initial import), I suggest that we remove all
roadways from the previous TIGER import, and re-import from the latest TIGER
data, ensuring that it uses the correct datum.

I don't have the necessary expertise to do the removal and re-import.  Is
there anyone on the list who can take on this task?


On Fri, Oct 2, 2009 at 12:39 PM, Scott Atwood <scott.roy.atwood at gmail.com>wrote:

> Lehua:
> This is a tiny, uninhabited islet off the coast of Niʻihau.  The shoreline
> data could use some cleanup, but it is roughly correct.  The boundary data
> seems to come from the same low resolution source, but the 1km east
> displacement error seems to have been manually corrected.
> Molokini:
> This is a tiny uninhabited islet between Maui and Kohoʻolawe.  The
> shoreline data is roughly correct, but could use some cleanup.  There
> doesnʻt appear to be any boundary data.  I suspect that this is an
> accidental oversight, since this island is officially a part of Maui County.
> Niʻihau:
> This is smallest of the inhabited Hawaiian islands.  The shoreline data
> appears to be roughly correct. The boundary data is low resolution, and
> appears to have been manually shifted to correct the 1km eastward
> displacement.
> Next steps for Lehua, Molokini and Niʻihau:
> Once again, if the political boundaries are supposed to correspond exactly
> to the shoreline data, the whatever is done for Kohoʻolawe and Lanaʻi should
> also be done for these three islands to correct the political boundaries.
>  The rest of the data can be easily cleaned up manually.
> -Scott
> --
> Scott Atwood
> Cycle tracks will abound in Utopia.  ~H.G. Wells

Scott Atwood

Cycle tracks will abound in Utopia.  ~H.G. Wells
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk-us/attachments/20091005/ed61fe83/attachment.html>

More information about the Talk-us mailing list