Box Counting Method

Michael Schulz schulzm at perm.geologie.uni-freiburg.de
Mon Sep 6 09:33:00 EDT 1999


Dear List!

Together with a diploma student, we are trying to use GRASS to 
perform calculations with the Box Counting Method. The box method is used 
here to measure the fractal dimension of the spatial and scaling 
distribution of fractures. 

A sequence of grids, each with a different cell size is placed over maps 
of fracture traces and the number of cells intersected by the fractures 
is counted. 

We used scanned maps of the fracture traces, imported them into GRASS and 
tried to run r.reports on the same mapset with varying resolutions. this 
did not work. We thought about using r.mapcalc to do some kind of 
neighborhood analysis, but the problem seems to be the varying cell sizes.

Is it possible to use r.mapcalc within one location but different 
resolutions of raster maps ? Could it be done with several locations ( 
r.mapcalc takes input from location 1 with res 1 and analyzes this grid 
with neighborhood-sizes of location 2 with res 2 and writes the result to 
location 2 ) ?

Any comments or ideas would be appreciated !

Cheers, Michael


==============================================================
Michael Schulz

Christoph-Mang-Str.5       Geologisches Institut
79100 Freiburg             Universitaet Freiburg 
                           Albertstr. 23b
                           79104 Freiburg  




More information about the grass-user mailing list