Spatial Data Analysis in R [was: Basic Question about use of R]

Hi David,

A general answer to your question is: yes, R would be useful for such
analyses - particularly when interfaced with a GIS.  For an
introduction to the types of spatial tools available in R see the Task
View for spatial data:
http://cran.r-project.org/web/views/Spatial.html

Below are a few more specific comments:

On Fri, Jan 2, 2009 at 12:12 PM, David Greene <greene107 at ntelos.net> wrote:

Dear Sirs:
I am not yet a user of R.  My background includes the use of (Turbo) Pascal
for scientific analysis of underwater acoustics problems (e.g. sound ray
tracing and array gain in directional noise fields.)
I have become interested in the following type of problem:
(1) select , say, 1000 random locations within the continental United
States;

This could be as simple as using the runif function, but more likely
you'll want to look at sp::spsample, or for more advanced tools see
the spsurvey and spatstat packages.

(2) characterize (statistically) the probabilities of:
  (a) distance to the nearest paved road;
  (b) elevation above sea level;
  (c) (?) ownership (public or private); etc.

R is outstanding for the types of 'statistical characterization' I
guess you are interested in.  It also has excellent capabilities for
importing and manipulating spatial data (e.g. see the "Reading and
writing spatial data" section of the Task View).  However for doing
things like calculating geographic distances using objects of varying
types (points, lines, polygons, grids) it's generally easiest to use a
GIS (such as GRASS, SAGA, ArcInfo, ...).  You can then use the
available tools for importing the GIS results into R for statistical
analysis, and if you wish, exporting back to the GIS.  However if you
do not want to put the effort into learning a GIS, it is usually
possible to work out a solution using only R.  As you run into
specific problems the R-sig-geo list is a good place to get helpful
answers to well formulated questions.

Would R be useful , perhaps in combination with Google Earth, to carry out

As far as I know Google Earth is designed for visualization rather
than analysis.  R in combination with a GIS is really the way to go.

Here is a current book that covers many of the spatial tools available in R

http://www.springer.com/public+health/epidemiology/book/978-0-387-78170-9

hope that helps,

Kingsford Jones

this kind of study?
Thank you for your consideration.
David C. Greene
greene107 at ntelos.net