hi,
When you specify uwnd[l=1:12:2@ave], you are defining a dynamic grid
with a time axis that has every second point, and regridding to that
grid. So you're probably doing more smoothing with that operation
than you expected. There's some discussion of the weights that are
used for a regridding with @AVE, here:
http://ferret.pmel.noaa.gov/Ferret/faq/regridding-a-time-series-with-ave
To take (point1+point2)/2, how about using @SHF:
yes? use monthly_navy_winds
yes? list/L=1:5 /x=180/y=0 uwnd,uwnd[L=@SHF:1],
(uwnd+uwnd[L=@SHF:1])/2
DATA SET:
/home/users/tmap/ferret/linux/fer_dsets/data/monthly_navy_winds.cdf
TIME: 01-JAN-1982 14:45 to 02-JUN-1982 19:15
LONGITUDE: 180E
LATITUDE: 0
Column 1: UWND is ZONAL WIND (M/S)
Column 2: UWND[T=@SHF:1] is ZONAL WIND (M/S)(shifted by
1 pts on T)
Column 3: EX#3 is (UWND+UWND[L=@SHF:1])/2
UWND UWND EX#3
16-JAN-1982 / 1: -2.738 -2.331 -2.534
16-FEB-1982 / 2: -2.331 -3.251 -2.791
18-MAR-1982 / 3: -3.251 -3.124 -3.188
18-APR-1982 / 4: -3.124 -1.818 -2.471
18-MAY-1982 / 5: -1.818 -2.007 -1.912
! Then you could keep every second point
yes? let uwnd_ave2 = (uwnd+uwnd[L=@SHF:1])/2
yes? list uwnd_ave2[L=1:`lmax`:2]
On 6/29/2012 3:11 AM, Steeven Paul
Yerraguntla wrote:
Dear ferret_users,
I wanted to average the data of every two consecutive
time points. For example, I want to average Jan-Feb, Mar-Apr,
May-Jun, Jul-Aug, Sep-Oct etc.,. in monthly_navy_winds
climatological dataset.
I tried using the ferret command:
yes? list/x=87/y=15 uwnd[l=1:12:2@ave]
but the list output averaged value is wrong (of course, i might be
used list command improperly) when compared with the calculated
one. Please let me know, how can i do the same with @ave.
regards,
- paul.
|