Hi Moritz, I eventually solved it like this:
t.rast.aggregate -s input=cla granularity="1 years" method=minimum output=yearly_min basename=yearly_min t.rast.aggregate -s input=cla granularity="1 months" method=minimum output=monthly_min basename=monthly_min YEAR_I=2003 ; YEAR_F=2013 MONTH_I=1 ;MONTH_F=12 for YEAR in `seq $YEAR_I $YEAR_F` ; do for MONTH in `seq -w $MONTH_I $MONTH_F` ; do r.mapcalc expression="month_min_${YEAR}_${MONTH} = if(monthly_min_${YEAR}_${MONTH} == yearly_min_${YEAR}, ${MONTH}, null())" --o done done r.series input=`g.list type=raster pattern=*_ag sep=,` output=month_of_min_mode method=mode Alternatively, 1. t.rast.aggregate -s input=cla granularity="1 months" method=minimum output=monthly_min basename=monthly_min 2. t.rast.aggregate -s input=monthly_min granularity="1 years" method=min_rast output=yearly_min_index 3. r.series input=`g.list rast pat=yearly_min_ind* sep=,` method=mode 4. r.reclass with 0 = 1 1 = 2 2 = 3 3 = 4 4 = 5 5 = 6 6 = 7 7 = 8 8 = 9 9 = 10 10 = 11 11 = 12 Meanwhile, I still think t.rast.algebra should do the job in one command... I just cannot figure it out (yet) from the manual page: https://grass.osgeo.org/grass71/manuals/t.rast.algebra.html I keep studying it, though Thanks a lot! Vero 2015-09-24 3:58 GMT-03:00 Moritz Lennert <mlenn...@club.worldonline.be>: > On 18/09/15 17:38, Veronica Andreo wrote: > >> Hi Soeren, list >> >> I need some help with t.rast.algebra. Well, if it is that my problem is >> really a t.rast.algebra problem... >> >> I have 2 strds with different granularities. A is an annual time series >> that contains the min annual temperature, B is a monthly time series >> containing the min monthly temperature. I want to compare the two of >> them, and when a pixel of B equals that of A, I need to get the >> start_month(B) but with the temporal resolution of A. So, what I want as >> output is a yearly time series containing the month in which the minimum >> temperature of each year occurs. Does that make any sense?? >> >> I been studying t.rast.algebra all morning now and trying different kind >> of relationships (with no success), and I got lost in the different >> relationships and the different ways to express them. >> >> Can someone point me to the right way of using t.rast.algebra for this >> problem? >> > > No help for t.rast.algebra, but wouldn't looping over the years and > applying t.rast.series with method=min_raster get you what you want ? > > Moritz >
_______________________________________________ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user