[GRASS-user] t.rast.aggregate

GRASS GRASS user
1

Hi all,

I have this MODIS time series of 8-day composite of chlorophyll
concentration from 2003 to 2013. I've already aggregated it by month
(average), so from 506 images I now have 132 (12 x year, 11 years)

but what I need doing now is to aggregate months over the years, what
i want are 12 summary images (one for each month over the 11 years).
How do i do that with t.rast.aggregate?

Thanks a lot in advance! And CONGRATULATIONS on GRASS 7 release!! :slight_smile:

Vero

2

Hi Veronica,
you can use t.rast.series for this kind of task. Use a SQL expression
to select only specific months of a year. Please try this code
(untested, may contain errors):

# January averages
t.rast.series input=monthly_aggregates \
    output=jan_average method=average \
    where="start_time = datetime(start_time, 'start_of_year', '0 month')"

# February averages
t.rast.series input=monthly_aggregates \
    output=feb_average method=average \
    where="start_time = datetime(start_time, 'start_of_year', '1 month')"

# March averages
t.rast.series input=monthly_aggregates \
    output=mar_average method=average \
    where="start_time = datetime(start_time, 'start_of_year', '2 month')"

...

Best regards
Soeren

2014-03-31 17:43 GMT+02:00 Veronica Andreo <veroandreo@gmail.com>:

Hi all,

I have this MODIS time series of 8-day composite of chlorophyll
concentration from 2003 to 2013. I've already aggregated it by month
(average), so from 506 images I now have 132 (12 x year, 11 years)

but what I need doing now is to aggregate months over the years, what
i want are 12 summary images (one for each month over the 11 years).
How do i do that with t.rast.aggregate?

Thanks a lot in advance! And CONGRATULATIONS on GRASS 7 release!! :slight_smile:

Vero
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

3

Hi Soeren,

2014-03-31 13:05 GMT-03:00 Sören Gebbert <soerengebbert@googlemail.com>:

Hi Veronica,
you can use t.rast.series for this kind of task. Use a SQL expression
to select only specific months of a year. Please try this code
(untested, may contain errors):

# January averages
t.rast.series input=monthly_aggregates \
    output=jan_average method=average \
    where="start_time = datetime(start_time, 'start_of_year', '0 month')"

Thanks so much for your help! Worked perfectly just removing the
underscore in 'start of year'

Have one question though: is there a way to check which maps were used
to estimate the average (or whatever operation t.rast.series perform)?
Is that information stored somewhere? or is there a way to check prior
to t.rast.series??

Thanks again!
Best,

Vero

4

Hi Veronica,

2014-04-01 16:07 GMT+02:00 Veronica Andreo <veroandreo@gmail.com>:

Hi Soeren,

2014-03-31 13:05 GMT-03:00 Sören Gebbert <soerengebbert@googlemail.com>:

Hi Veronica,
you can use t.rast.series for this kind of task. Use a SQL expression
to select only specific months of a year. Please try this code
(untested, may contain errors):

# January averages
t.rast.series input=monthly_aggregates \
    output=jan_average method=average \
    where="start_time = datetime(start_time, 'start_of_year', '0 month')"

Thanks so much for your help! Worked perfectly just removing the
underscore in 'start of year'

Ooops, sorry for the underscore's.

Have one question though: is there a way to check which maps were used
to estimate the average (or whatever operation t.rast.series perform)?
Is that information stored somewhere? or is there a way to check prior
to t.rast.series??

You can use the same SQL expression in any other temporal command that
support SQL where expressions. For example t.rast.list:

{{{
t.rast.list input=precipitation_1950_2013_monthly_mm@soeren
where="start_time = datetime(start_time, 'start of year', '0 month')"

precipitation_monthly_mm_0 soeren 1950-01-01 00:00:00 1950-02-01 00:00:00
precipitation_monthly_mm_12 soeren 1951-01-01 00:00:00 1951-02-01 00:00:00
...
precipitation_monthly_mm_732 soeren 2011-01-01 00:00:00 2011-02-01 00:00:00
precipitation_monthly_mm_744 soeren 2012-01-01 00:00:00 2012-02-01 00:00:00
precipitation_monthly_mm_756 soeren 2013-01-01 00:00:00 2013-02-01 00:00:00
}}}

Best regards
Soeren

Thanks again!
Best,

Vero

5

Great!!

Obviously I didn't think about that :stuck_out_tongue:

THANKS!!!

Vero

2014-04-01 11:13 GMT-03:00 Sören Gebbert <soerengebbert@googlemail.com>:

Hi Veronica,

2014-04-01 16:07 GMT+02:00 Veronica Andreo <veroandreo@gmail.com>:

Hi Soeren,

2014-03-31 13:05 GMT-03:00 Sören Gebbert <soerengebbert@googlemail.com>:

Hi Veronica,
you can use t.rast.series for this kind of task. Use a SQL expression
to select only specific months of a year. Please try this code
(untested, may contain errors):

# January averages
t.rast.series input=monthly_aggregates \
    output=jan_average method=average \
    where="start_time = datetime(start_time, 'start_of_year', '0 month')"

Thanks so much for your help! Worked perfectly just removing the
underscore in 'start of year'

Ooops, sorry for the underscore's.

Have one question though: is there a way to check which maps were used
to estimate the average (or whatever operation t.rast.series perform)?
Is that information stored somewhere? or is there a way to check prior
to t.rast.series??

You can use the same SQL expression in any other temporal command that
support SQL where expressions. For example t.rast.list:

{{{
t.rast.list input=precipitation_1950_2013_monthly_mm@soeren
where="start_time = datetime(start_time, 'start of year', '0 month')"

precipitation_monthly_mm_0 soeren 1950-01-01 00:00:00 1950-02-01 00:00:00
precipitation_monthly_mm_12 soeren 1951-01-01 00:00:00 1951-02-01 00:00:00
...
precipitation_monthly_mm_732 soeren 2011-01-01 00:00:00 2011-02-01 00:00:00
precipitation_monthly_mm_744 soeren 2012-01-01 00:00:00 2012-02-01 00:00:00
precipitation_monthly_mm_756 soeren 2013-01-01 00:00:00 2013-02-01 00:00:00
}}}

Best regards
Soeren

Thanks again!
Best,

Vero