r.stats and FF layers

[Agustin Lobo]

I still suspect an error in r.stats in the grass5.0beta3.
Could the developing team consider this observation:

r.stats reports lines like the following for a float layer (spc1NW.av )
stratified by a clump map (spc1NW.peps.clump):
r.stats -c in=spc1NW.peps.clump,spc1NW.av > a
one line in a:
441 122.76745-123.614312 49

That is, according to r.satst,
for clump 441, the 49 values in spc1NW.av are
within the range  122.76745-123.614312

Nevertheless, I kmow that all 49 values labeled
as 441 in spc1NW.peps.clump are identical and equal to
123.938774 in spc1NW.av, which I can check with d.what:

d.what.rast map=spc1NW.peps.clump,spc1NW.av,spc1NW

609(E) 827(N)
spc1NW.peps.clump in eps  (441)
        spc1NW.av in eps, quant   (124)
        spc1NW.av in eps, actual  (123.938774)


Actually, spc1NW.av was generated by r.average:
r.average base=spc1NW.peps.clump cover=spc1NW out=spc1NW.av

Instead, r.volume is correct
r.volume -f data=spc1NW clump=spc1NW.peps.clump > spc1NW.vol

with the line for 441 in spc1NW.vol being:
441:123.94:6073:49:610.00:829.00:6073.00

which indicates that the average value in 441 is 123.94

(by the way: could the output from r.volume include more
significant digits? Only 2 gives problems in some ulterior analysis)


Thanks


Dr. Agustin Lobo
Instituto de Ciencias de la Tierra (CSIC)
Lluis Sole Sabaris s/n
08028 Barcelona SPAIN
tel 34 93409 5410
fax 34 93411 0012
alobo at ija.csic.es
http://pangea.ija.csic.es/alobo