r.watershd & topo index

[(jane rossing)]

In Message Thu, 1 Apr 1993 11:42:11 -0500,
  riker at hydro1.geo.duke.edu (Dan Riker) writes:

>I have a quick question about the accumulation map for finding the topographic
> index ln(a/tanB)
>For my accumulation map, I find many lage negative values, and I always get a
> message that overflow occurred in the watershed.  Is this okay, or can the
> problem be readily solved.
>Any help would really make my day.
>Dan Riker
>Duke University

The negative values in the accumulation map mean that there MAY be
accumulation from outside the area on which GRASS has computed.  I
understood that to mean that negative values are basically unavoidable
unless you somehow had a mask one cell bigger than the watershed.  If you
mask the watershed, you will get negative values around the border, where
GRASS doesn't know if water runs in or not, but they should not be large
values.  So, I chose to ignore the negatives. (Make them positives, if they
are within your watershed).