[R-sig-Geo] Hex GIS, (was) Moving-window ...

Frank Hardisty HardistF at gwm.sc.edu
Tue Jan 25 20:23:41 CET 2005


Sander and All,

For Dan Carr's work:

Most important is the R package mantained by Nicholas Lewin-Koh,
including documentation, which you mention. This is another area where R
is the leading edge!

Also, online at http://www.galaxy.gmu.edu/~dcarr/lib/v8n2.pdf

and 

Carr, D. B., A. R. Olsen, and D. White. 1992. "Hexagon Mosaic Maps for
Display of Univariate and Bivariate Geographical Data." Cartography and
Geographic Information Systems, Vol. 19, No. 4, pp. 228-236,271.

Carr, D. B. and L. W. Pickle. 1993. "Plot Production Issues and
Details: Smoothed Cancer Rates and Hexagon Mosaic Maps." Statistical
Computing & Graphics Newsletter, Vol. 4, No. 2, pp. 16-20.

Additionally, I have some Java code for generating hexagonal
spatio-temporal patterns with various properties which could potentially
be used with R spatial packages using rJava, SJava, or the like. I'm
happy to share this code with any interested parties.

regards,
-Frank Hardisty 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Frank Hardisty
Department of Geography
University of South Carolina
hardisty at sc.edu
Office -- 803-777-5729 
Fax -- 803-777-4972
http://people.cas.sc.edu/hardistf/

>>> Sander Oom <slist at oomvanlieshout.net> 01/25/05 01:46PM >>>
Hi Dennis and Frank,

Could you en light me on possible references (papers, books) about the

use of hexagons in spatial analysis? Did your search give any
direction?

I am using the hexbin package (find it at bioconductor.org) and can 
highly recommend it. Nicholas (the maintainer) is currently updating
the 
package!

Although hexbin and R are not a GIS, much of the 'GIS' functionality is

available in R!!

Cheers,

Sander.

Frank Hardisty wrote:
> Denis and All,
> 
> The best thing out there is probably the "hexbin" package in R,
based
> on Dan Carr's work. It's not a GIS, however. Hexagons do have some
very
> nice properties for spatial analysis.
> 
> -Frank Hardisty
> 
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Frank Hardisty
> Department of Geography
> University of South Carolina
> hardisty at sc.edu 
> Office -- 803-777-5729 Fax -- 803-777-4972
> http://www.cla.sc.edu/geog/faculty/hardistf/ 
> 
>>>> <White.Denis at epamail.epa.gov> 12/17/04 12:18PM >>>
> Does anyone know of a GIS based on a regular hexagon tessellation
> rather
> than on squares or rectangles?  (There was a project twenty some
years
> ago called System X developed for the Navy by Dean Lucas that
started
> into commercialization but never finished as far as I know.)
> 
> r-sig-geo-bounces at stat.math.ethz.ch wrote on 2004-12-16 12:17:28:
> 
>> Stephane, have you tried how well this scales to, say, 1000x1000
>> matrices? Indeed, most basic raster GIS functionality is not
> available
> 
>> in R packages, and would be welcome. R code is efficient for small
>> problems, but for cases like this scales badly as Marius noted.
>>
>> We are actually working on an interface from R to the functionality
> of
> 
>> PCRaster (http://pcraster.geog.uu.nl -- yet another raster GIS),
and
>> expect to have this working within a few months. PCRaster is not
> open
>> source, so it is unlikely to become a CRAN package straight away,
> but
>> still it may fill a need.
>>
>> This all depends on availability of sp, the long awaited package
> with
> S
>> classes for spatial point, line, grid, arc/node and polygon data.
> We're
>> working very hard on this, and most likely will announce a beta in
>> January. I expect a PCRasteR package to follow that soon.
>> -- 
>> Edzer
> 
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/r-sig-geo 
> 
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at stat.math.ethz.ch 
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo 
> 

-- 
---------------------------------------------------------
Dr. Sander P. Oom
Animal, Plant and Environmental Sciences
University of the Witwatersrand
Private Bag 3
Wits 2050
South Africa

Tel (work)      +27 (0)11 717 64 04
Tel (home)      +27 (0)18 297 44 51
Fax             +27 (0)18 299 24 64

Email   sander at oomvanlieshout.net 
Web     www.oomvanlieshout.net/sander




More information about the R-sig-Geo mailing list