-------- Original Message --------
Subject: Re: "Natural borders"
Date: Thu, 3 Nov 2011 21:43:22 +0000
From: Mark Jackson <[log in to unmask]>
To: Maps, Air Photo, GIS Forum - Map Librarianship
<[log in to unmask]>
Chris,
This is really one of those can of worms where the patron doesn't really
realize how many variations must be considered. In addition to the scale
issue (illustrated here by overlaying the ESRI supplied dtl_cntrys_ln
file on the Topographic basemap) is just the identification of what is a
natural boundary. I suppose in the patrons initial question she
indicates a natural boundary as a mountain range so this boundary
between Turkey and Iran may be considered natural to her. To me it is
still a pretty arbitrary line - following ridgelines where possible but
in other places cutting across valleys wherever convenient.
Looking past that, the method I would do (and tested out briefly was to
add the line version of the detailed country boundaries (I'm sure it is
on ArcGIS online, I just got it out of my ESRI data folder). I started
editing and just cut the lines where they transitioned from "natural" to
"unnatural". Then deleted the "unnatural" sections. The result is a
layer of just natural boundaries. Add a field to the table for length
and calculate the geometry. Here is another problem. Length cannot be
calculated (in ArcMap at least) on a Geographic Projection. It must be
projected, and depending on the projection the length changes slightly.
This next part works manually but I didn't attempt a model. Add the ESRI
dtl_cntry (this one is polygons), then add to it a new field for
Perimeter and do a Calculate Geometry operation to populate it. In this
example I then selected Turkey, then used a Select by Location to select
the lines from my Natural Boundaries layer that intersected Turkey. Then
use Selection -> Statistics and you can see the Sum of the line
segments. Use that with the Perimeter value to determine the % of the
total perimeter is natural vs. unnatural.
Perhaps it would be faster to use Joseph's manual method. I was just
trying to come up with a method that would result in a data set at the
end. Which reminds me - don't forget to record the percentage in another
new field on the polygon dtl_cntry layer. Then you could make a
choropleth map based on percentage of border that follows natural earth
features.
--
Dr. Mark Jackson
Brigham Young University
Geography, Geology, Civil Engineering, & Technology Librarian
2420 HBLL
Provo UT 84602
801.422.9753
[log in to unmask] <mailto:[log in to unmask]>
On Nov 3, 2011, at 2:26 PM, Angie Cope, American Geographical Society
Library, UW
Milwaukee wrote:
> -------- Original Message --------
> Subject: RE: "Natural borders"
> Date: Thu, 3 Nov 2011 13:13:54 -0700
> From: Joseph Kerski <[log in to unmask] <mailto:[log in to unmask]>>
> To: [log in to unmask] <mailto:[log in to unmask]> <[log in to unmask]
> <mailto:[log in to unmask]>>, [log in to unmask]
> <mailto:[log in to unmask]>
> <[log in to unmask] <mailto:[log in to unmask]>>,
> [log in to unmask] <mailto:[log in to unmask]>
> <[log in to unmask] <mailto:[log in to unmask]>>
>
>
>
> Chris, what I would do is go to www.arcgis.com <http://www.arcgis.com>
> <http://www.arcgis.com>
> and zoom to the Middle East. Then use the measure tool to measure the
> boundaries of interest, using the world topographic map or satellite
> imagery to determine which boundaries follow natural boundaries and
> which ones follow artificial ones.
>
> Or, for a more precise figure, use ArcGIS desktop version 9 or 10, add
> the desired national boundaries as a geodatabase feature class or a
> shapefile (available from the Esri Data and Maps DVD and also from
> www.arcgis.com <http://www.arcgis.com> <http://www.arcgis.com> and
> other data depositories
> online), and then measure each of the boundaries, tallying up the length
> of the boundaries that follow mountains, rivers, water bodies, etc and
> also the length of the boundaries that do not.
>
> Perhaps there is a better way and I look forward to hearing what others
> have to say too.
>
> Joseph Kerski
>
> *Joseph J. Kerski, Ph.D.| Education Manager*
>
> Esri | 1 International Court | Broomfield CO 80021-3200 | USA
>
> Tel 303-449-7779, ext. 8237 | Fax 303-449-8830
>
> [log in to unmask] <mailto:[log in to unmask]> | esri.com <http://esri.com>
>
> Twitter: @josephkerski
>
> *From:*[log in to unmask]
> <mailto:*[log in to unmask]>
> [mailto:[log in to unmask]] *On Behalf Of *Christopher Thiry
> *Sent:* Thursday, November 03, 2011 8:29 AM
> *To:* [log in to unmask] <mailto:[log in to unmask]>;
> [log in to unmask] <mailto:[log in to unmask]>
> *Subject:* "Natural borders"
>
> All,
>
> I need help answering this email I received from a patron. Thanks in
> advance.
>
> Christopher J.J. Thiry
>
> Map Librarian
>
> Colorado School of Mines
>
> 1400 Illinois
>
> Golden, CO 80401
>
> p. 303-273-3697
>
> f. 303-273-3199
>
> [log in to unmask] <mailto:[log in to unmask]> <mailto:[log in to unmask]>
>
> http://library.mines.edu/
>
> ------------------------------------------------------------------------
>
> *From:*Ashley Pryor
> *Sent:* Wednesday, November 02, 2011 5:01 PM
> *To:* Christopher Thiry
> *Subject:* Map Question
>
> Christopher,
>
> I am wondering if you can assist me to compile a statistic for %
> “natural border” for all countries in the Middle East. Natural borders
> for instance would be coastlines, mountain ranges, large rivers, etc. Of
> course sometimes borders are built along these natural boundaries, and
> sometimes they are arbitrary of natural boundaries. I am wondering if
> you have a set of maps that clearly show natural boundaries against
> national borders, or better yet, already have access to a data set that
> has already compiled the percent national border that follows a natural
> boundary. If you have such a data set, any or all regions would be
> beneficial to my research. If you do not know of any particular data
> set, perhaps you have some time to meet with me and let me know which
> map sets would be most helpful in compiling such a statistic.
>
> Please let me know if you have any availability in the next couple of
> days to perhaps discuss.
>
> Thanks so much for your time and your work.
>
> Regards,
>
> Ashley Pryor
|
