Verification for a passive tracer setup

[l_fr]

Hello,

I've a run already set up that works fine, with custom BC, forcing and IC. I'd like to add a passive tracer at one point (basically to illustrate, after a few days run, the error of the position of an object due to an error in the velocity field).
I fetched informations where I could find it, in the wiki and in the forum, and I'm still running through the error:

DEF_HIS - inquiring history file: ocean_his.nc

NETCDF_OPEN - unable to open existing NetCDF file:
ocean_his.nc
call from: def_his.F

after the time step 0. I already had it when I forgot some elements, but this time I can't find what.
So, I just wanted someone to check if I have forgotten something. Here is what I've done so far:

Compiled the executable with
T_PASSIVE
ANA_BPFLUX
ANA_SPFLUX
added to the header file, for analytical fluxes at bottom and surface for the passive tracer.

Set in the ocean.in file:
NPT=1

Added to the ocean.in file (last line)
LBC(isTvar) == RadNud RadNud Clo Clo \ ! temperature
RadNud RadNud Clo Clo \ ! salinity
RadNud RadNud Clo Clo ! dye_01

Added in my bc.nc file:
double dye_west_01(time, s_rho, eta_rho) ;
dye_west_01:long_name = "dye concentration western boundary condition" ;
dye_west_01:missing_value = 9999. ;
dye_west_01:time = "time" ;
dye_west_01:field = "dye_west_,scalar,series" ;
dye_west_01:units = "kilogram meter-3" ;
double dye_south_01(time, s_rho, xi_rho) ;
dye_south_01:long_name = "dye concentration southern boundary condition" ;
dye_south_01:units = "kilogram meter-3" ;
dye_south_01:field = "dye_south_,scalar,series" ;
dye_south_01:time = "time" ;
dye_south_01:missing_value = 9999. ;
All values are zero, dimensions are correct, and north and east conditions are wall (see above).

Added to the ic.nc file:
double dye_01(time, s_rho, eta_rho, xi_rho) ;
dye_01:missing_value = -99999. ;
dye_01:time = "ocean_time" ;
dye_01:field = "dye_,scalar,series" ;
dye_01:units = "kilogram meter-3" ;
dye_01:long_name = "dye concentration" ;

Time has only one value of course, like my other variables (temp,salt)...
It is a 3D array full of zeros, except at ONE point in surface (not in land mask of course). (The resolution is 1/6th degree).
I didn't change the varinfo.dat.
Besides here are some of my output parameters (those are correct for a run without tracer, and so are DSTART, NTIMES...):
NRREC == 0
LcycleRST == T
NRST == 0
NSTA == 1
NFLT == 1
NINFO == 1
LDEFOUT == F
NHIS == 288
NDEFHIS == 0
NTSAVG == 1
NAVG == 0
NDEFAVG == 0

Any suggestion appreciated.

Cheers,

François

[kate]

The error says it can't open an old history file, so tell it to make a new history file by setting LDEFOUT to true.

[l_fr]

Thanks Kate!
Now it is working properly.
It is strange however, since I let this LDEFOUT to false with the setups without tracer, and whitouth HIS or AVG file present, and they worked properly as well. Ah! I get it. I guess it's because I don't ask any HIS output (NHIS==0) with the other setups, but only RST.

François