As mentioned in a previous thread, I have managed a successful compile of the w2 generic under gfortran. This is a record iof my attempts to verify the install. Lacking a pre-existing verified 3.6 model control file, I have been attempting to validate this with the files from the 3.7 version download, using the "De Gray Reservoir" example.
Running De Gray - execution completes without reporting any issue, and the message:
JDAY = 78.708 DLT = 3600. DLTAV = 3600. NV = 0
Normal termination at 16:31:32on 09/04/12
Step 2: NaN errors observed
Upon inspection of outputs, hydrodynamic execution seems to fail early in the simulation, with the following appearing from the file "cnp.opt":
Time Parameters
Gregorian date [GDAY] = March 4, 1980
Julian date [JDAY] = 64 days 13.00 hours
Elapsed time [ELTMJD] = 0 days 1.00 hours
...
Outflows
Structure outflows [QSTR]
Branch 1 = 22.01
Total outflow [QOUT] = 22.00 m^3/s
...
Time Parameters
Gregorian date [GDAY] = March 18, 1980
Julian date [JDAY] = 78 days 17.00 hours
Elapsed time [ELTMJD] = 14 days 5.00 hours
...
Outflows
Structure outflows [QSTR]
Branch 1 = 64.13
Total outflow [QOUT] = NaN m^3/s
Going to take more frequent snapshots to see if things fail immediately or in between the two snapshots.