Modify Command

[Arindam Chakraborty]

Hi Matt,
you were close. Please do
'set x 1 102'
'set y 1 102'
before calling the ave() and modify. This is to define the diurnally
averaged variable on the full grid. If you want to exclude the first 4
time points, then
'set t 5 12'
before the average will do the job.

After you have both the diurnally averaged variables in global grid, you
may area average them for display or further calculation as required:
'set x 1'
'set y 1'
'd tloop(aave(diurnallyaveragedvariable,region...))'

Hope this helps.

arindam

Matt Georgescu wrote:
> Dear GrADS users,
>
> I have 2 separate monthly datasets, at 3-hour timestep, over a
> rectangular domain of 102 points X 102 points, for which I am
> trying to calculate a time-series depicting the diurnal evolution of
> near-surface temperature.
>
> I use the following:
>
> reinit
> open /arch2/matt1/1990bc_02lulc/1.ctl
> open /arch2/matt1/1990bc_92lulc/1.ctl
> #t=5 [1990:7:1:12]
> #t=245 [1990:7:31:12]
> set x 1
> set y 1
> set t 1 9
> define temp90lulc02clim = ave(tempk-273, t+0, t=245, 8)
> modify temp90lulc02clim diurnal
> d temp90lulc02clim
>
> set x 1
> set y 1
> set t 1 9
> define temp90lulc92clim = ave(tempk.2-273, t+0, t=245, 8)
> modify temp90lulc92clim diurnal
> d temp90lulc92clim
>
> I want to calculate the diurnal average over the entire domain, not just
> x=1 and y=1 ... but if i set x and y to my entire grid dimensions (ie
> "set
> x 1 102", etc.), i no longer obtain a time series.
>
> Also, my model output begins at 00Z, but I want to throw out the first 4
> timesteps and only use t=5 to t=245 (12Z to 12Z) in the domain-wide
> diurnal calculation.
>
> Thanks for any help from 'modify' experts.
>
> matt
>

--
----------------------------
 ARINDAM CHAKRABORTY
 Department of Meteorology
 Florida State University
 Tallahassee, FL-32306, USA
 Tel: +001-850-6443524 (Off)
      +001-850-5758550 (Res)
 Fax: +001-850-6449642
____________________________