Subject: CCM3.6.6 on the Cray T3E
From: Jules Kouatchou (kouatch@pecos1.gsfc.nasa.gov)
Date: Fri Mar 31 2000 - 16:54:35 MST
Hi,
I am trying to run CCM3.6.6 on the Cray T3E. I am facing some
difficulties in the initialization process (see the error message
below).
I would like to know if someone has had a similar problem.
Any suggestion will be very appreciated.
Regards,
Jules K.
-----------------------------------------------------------------
SIGNAL: Floating point exception (invalid floating point operation)
Beginning of Traceback (PE 5):
Interrupt at address 0x80000d218 in routine 'ATMINI'.
Called from line 986 (address 0x8001142e4) in routine 'LSMINI'.
Called from line 1264 (address 0x8000e1da0) in routine 'INITEXT'.
Called from line 947 (address 0x8000256c8) in routine 'CCM3'.
Called from line 475 (address 0x800000c98) in routine '$START$'.
End of Traceback.
After debugging, I found that the problem is coming from the following
section of "src/lsm/atmini.F":
k = kvec(l,m) !lsm subgrid vector index
asdirxy(i,j) = asdirxy(i,j) + albd(1,k) * wsg2g(l,m)
asdifxy(i,j) = asdifxy(i,j) + albi(1,k) * wsg2g(l,m) !<--this line
aldirxy(i,j) = aldirxy(i,j) + albd(2,k) * wsg2g(l,m)
aldifxy(i,j) = aldifxy(i,j) + albi(2,k) * wsg2g(l,m)
tsxy(i,j) = tsxy(i,j) + tg(k) * wsg2g(l,m)
lwupxy(i,j) = lwupxy(i,j) + sb*(tg(k)**4) * wsg2g(l,m)
snowxy(i,j) = snowxy(i,j) + h2osno(k)/1000.* wsg2g(l,m)
In fact one value of "albi(1,k)" used in the above calculation is
not a number (NaN).
I tried unsuccesfully to determine where the variable "albi" is first
defined.
This archive was generated by hypermail 2b27 : Thu Jan 04 2001 - 10:01:35 MST