[GRASSLIST:2316] r.neighbors variance calculation

[Izak Thomas Henry Deyzel]

Good day everyone,

I would like to know if the method of variance calculation in
r.neighbors has changed from Grass 4.2.1 to Grass 5pre1.

I am trying to remove contamination in instantaneous radar rainfall
fields by looking at the variance fields. I did most of my investigation
using r.neighbors of Grass 4.2.1. When I tried to duplicate the fields
in my own C-program I used the source code of Grass5pre1 for r.neigbors
(c_var.c), but the Grass 4.2.1 field differs drastically.

Any ideas where my problem might lie?

Thank you.

Izak

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Izak T.H. Deyzel
Senior Meteorologist            
METSYS
South African Weather Service

Tel: +27 58 303 5571
*******************************
Call our Weatherline at 082 162
Visit SAWB website at   : www.weathersa.co.za
Visit METSYS website at : metsys.weathersa.co.za