[GRASS5] Fwd: s.surf.rst with anisotropy

[Markus Neteler]

A forward from Helena Mitasova:

Thanks to Jaro, an anisotropy option was added to s.surf.rst to support
interpolation of data representing anisotropic phenomena and/or handle
anisotropically sampled data. By providing angle and scale the resulting
surface will be "stretched" in the given direction. Note, that the current
implementation handles only uniform anisotropy (constant direction and scale
for the entire area). So just as a reminder s.surf.rst supports spatially
variable smoothing but only uniform tension and anisotropy. It would be very
helpful to get enhanced s.surf.rst tested with and without anisotropy option
to make sure that nothing is broken.

An example showing how anisotropy works is given by the following images
representing the given data (grey balls - profiles far apart and two
paralell lines - this is beach)

http://skagit.meas.ncsu.edu/~helena/publwork/coast02/lnewfig/
anisdefault.jpg : surface with isotropic tension (s.surf.rst with default
                  parameters, resulting contours are pretty "wild"

anis135.jpg: anisotropy at 135 deg (not quite right)
anis160t200.jpg: anisotropy at 160deg gives nice straihght beach (ignore the
                 segments, they are far away from the mapped area)

thank you for testing and let me know if something does not work (manual has
the updated usage)

Helena

P.S You can also read more about s.surf.rst with anisotropy and using
    s.vol.rst for interpolating precipitation with impact of topography in
    Transactions in GIS,   March 2002, vol. 6, no. 2,   pp. 135-150(16)
http://www.blackwellpublishers.co.uk/asp/journal.asp?ref=1361-1682&src=arc&vid=6&iid=2