[GRASSLIST:5819] Subsurface Sediment mapping with s.surf.rst

[John Chesnut]

I need to develope a subsurface sediment map of the Salton Sea, 
California using data generated by the "Quester Tanget" sonar 
classification device.   The device analyses the sonar waveform to 
predict sub sea bottom characteristics.  The sample rate of the 
device produce site data seperated by 20 cm or so along transect 
lines.
Output stream consists of 3 floating point attributes and an class number.

I need to reduce the data to fit an expected  resolution of  10 m x 
10 m, and interpolate between widely seperated  transect lines.

What is the correct sequence to reduce the densely packed transect 
data to a surface.  My test data set shows the sample points within a 
neighborhood may belong to several classes.  I will need to produce a 
calculated proportion within a class for a neighborhood.

This would be similar to the reduction of bathymetry or other closely 
packed transect data to a larger cell size, with the exception that 
simple decimation of the data set is not accurate.