[GRASSLIST:5648] sampling methods

[Trevor Wiens]

I wondering what is the best method to sample areas around survey sites. I'm using the version 6 beta.

I've tried two approaches with similar results

General method
a. Import point files
b. use v.buffer to create circular buffers around my survey sites (bird point counts)
c. use v.type and v.patch to re-attach the attributes
d. use v.to.rast to create a raster.

>From here I diverged into two methods
Method 1
a. convert back into a vector, but this time as points.
b. use v.what.rast to extract the actual values from my dem derived, and MODIS derived layers.
c. run my basic stats in postgres and export to R for more involved analysis

Method 2
a. create a MASK from the raster buffer using r.mapcalc (in the hope to speed processing)
b. run r.neighbors to calculate min,max,mean,variance, diversity, and mode using a window the size of my buffers, so that the center value becomes a measure for the entire buffer.
c. extract to postgres using v.what.rast using the original points file

Most of my stuff is floating point so r.statistics doesn't work, and r.stats returns ranges (as far as I can get it to function) instead of actual values.

If any of the experience GRASS users has an opinion as to which is best and why or if there is an easier way to do this, I would appreciate the input.

Thanks

T
-- 
Trevor Wiens 
twiens at interbaun.com

The significant problems that we face cannot be solved at the same 
level of thinking we were at when we created them. 
(Albert Einstein)