[RegCNET] Model not working after a particular time
Dilip Ganguly
dganguly at prl.res.in
Wed Dec 27 14:03:19 CET 2006
My appologies to everyone. Instead of decreasing the time step, I had
actually increased it to satisfy the rule dt=3*ds
However, now the model seems to be running without any problem so far with
the following combination of time steps
radfrq = 30.,
abemh = 18.,
abatm = 600.,
dt = 150.,
Thanks everyone.
Wishing you all a Happy New Year ahead.
Dilip.
On Wed, 27 Dec 2006, Dilip Ganguly wrote:
> Hello,
>
> I am trying to perform my simulation with a horizontal resolution of 90
> km. Initially I used the the following time steps as mentioned in the
> regcm user manual
> radfrq = 30.,
> abemh = 18.,
> abatm = 900.,
> dt = 225.,
>
> Now I increased the time steps dt to 300 and restarted my simulations.
> However still I am facing similar problem. Simulation starts and then it
> hangs. I got the following message on my screen
>
> ******* OPENING NEW OUTPUT FILES: 2004020100
> OPENING NEW OUT FILE: output/ATM.2004020100
> OPENING NEW BAT FILE: output/SRF.2004020100
> OPENING NEW RAD FILE: output/RAD.2004020100
> BATS variables written at 2004020103 180.000000000000
> at day = 92.1562, ktau = 11950 : 1st, 2nd time deriv of ps =
> 0.30840E-04 0.12518E-06, no. of points w/convection = 143
> OUT-history written date = 2004020106.00000
> BATS variables written at 2004020106 0.000000000000000E+000
> Writing rad fields at ktau = 11976 2004020106
> BCs are ready from 2004020106 to 2004020112
> at day = 92.3299, ktau = 12000 : 1st, 2nd time deriv of ps =
> NaN NaN , no. of points w/convection = 0
>
> --------------------------------------------------------------------------
>
> Can anyone suggest me proper time steps (for horizontal resolution of 90
> km) to avoid this kind of problem possibly due to convective instabilities
> as told by Jagadish.
>
> Thanks.
>
> Dilip.
>
> On Wed, 27 Dec 2006, jagadish karmacharya wrote:
>
> > Dear Dilip,
> >
> > I used to face similar problem. I think this error arise due to convective instability. A common work around is to decrease the time step and restart the model. If run goes well you can stop the model after few days run and restart it with original time step if you want to lose computational time.
> >
> > Wish you Happy New Year!
> > Jagadish
> >
> > Dilip Ganguly <dganguly at prl.res.in> wrote: Hello everyone,
> >
> > My Greetings to you all on Christmas!
> >
> > I wanted to run RegCM3 for the year 2004 over the south Asian domain. I
> > choose an initial spin up time of two months. That means I started my
> > simulations from 1st Nov. 2003. Simulations started normally and worked
> > upto 1st February 2004. However after this it started showing problem in
> > calculating 1st and 2nd time deriv of ps and I was getting following
> > message on the screen
> >
> > at day = 93.0286, ktau = 12300 : 1st, 2nd time deriv of ps =
> > NaN NaN , no. of points w/convection = 0
> > BATS variables written at 2004020203 180.000000000000
> > at day = 93.1589, ktau = 12350 : 1st, 2nd time deriv of ps =
> > NaN NaN , no. of points w/convection = 0
> > OUT-history written date = 2004020206.00000
> > BATS variables written at 2004020206 0.000000000000000E+000
> > Writing rad fields at ktau = 12384 2004020206
> > BCs are ready from 2004020206 to 2004020212
> > at day = 93.2891, ktau = 12400 : 1st, 2nd time deriv of ps =
> > NaN NaN , no. of points w/convection = 0
> >
> > ---------------------------------------------------------------------
> > At this point I stopped the model and restarted it. However the problem
> > is not solved I am getting following message again
> >
> > ******* OPENING NEW OUTPUT FILES: 2004020200
> > OPENING NEW OUT FILE: output/ATM.2004020200
> > OPENING NEW BAT FILE: output/SRF.2004020200
> > OPENING NEW RAD FILE: output/RAD.2004020200
> > at day = 93.0286, ktau = 12300 : 1st, 2nd time deriv of ps =
> > NaN NaN , no. of points w/convection = 0
> >
> > --------------------------------------------------------------------
> >
> > Can anyone tell me why is this problem arising and how to overcome it?
> >
> > Thanks in advance.
> >
> > Dilip.
> >
> > ---------------------------------------------------------------------
> > DILIP GANGULY
> > Postdoctoral Fellow
> > Space And Atmospheric Sciences Division
> > Physical Research Laboratory
> > Ahmedabad 380009
> > Gujarat, India
> > Phone: +91 79 26314558 (Laboratory)
> > +91 79 26300505 (Hostel)
> > Fax: +91 79 26314659
> > Alternate Email: ganguly.dilip at gmail.com
> > Visit me at http://www.prl.res.in/~dganguly
> > ---------------------------------------------------------------------
> >
> >
> >
> >
> > _______________________________________________
> > RegCNET mailing list
> > RegCNET at lists.ictp.it
> > https://lists.ictp.it/mailman/listinfo/regcnet
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam? Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
>
> Dilip.
>
> ---------------------------------------------------------------------
> DILIP GANGULY
> Postdoctoral Fellow
> Space And Atmospheric Sciences Division
> Physical Research Laboratory
> Ahmedabad 380009
> Gujarat, India
> Phone: +91 79 26314558 (Laboratory)
> +91 79 26300505 (Hostel)
> Fax: +91 79 26314659
> Alternate Email: ganguly.dilip at gmail.com
> Visit me at http://www.prl.res.in/~dganguly
> ---------------------------------------------------------------------
>
>
> _______________________________________________
> RegCNET mailing list
> RegCNET at lists.ictp.it
> https://lists.ictp.it/mailman/listinfo/regcnet
>
Dilip.
---------------------------------------------------------------------
DILIP GANGULY
Postdoctoral Fellow
Space And Atmospheric Sciences Division
Physical Research Laboratory
Ahmedabad 380009
Gujarat, India
Phone: +91 79 26314558 (Laboratory)
+91 79 26300505 (Hostel)
Fax: +91 79 26314659
Alternate Email: ganguly.dilip at gmail.com
Visit me at http://www.prl.res.in/~dganguly
---------------------------------------------------------------------
More information about the RegCNET
mailing list