bug in passive tracer loading in river runoff code

[rubash]

When adding tracer to river runoff using a NetCDF forcing file ROMs crashed when the tracer loop in the file get_data.F tried to load data for the first tracer after temperature and salt. idRtrc(3) was zero instead of the 152 that was set in mod_ncparam.F for the tracer.

Most things seemed fixed when I added idRtrc(inert(i))=varid just after idTvar(inert(i))=varid on line 1825 of mod_ncparam.F, except that stream inflow directions are all east-west (0) regardless of the values read from the forcing file. I'm trying to figure out what to do about that now.

[arango]

Yes, good catch. I guess that users are nowadays are using more complex applications with inert passive tracers (T_PASSIVE). Notice that we still don't have source and sink right-hand-side terms for the inert tracer equation as we have in the sediment and ecosystem models. It is very difficult to come up with a generalized formulation for this. I will probably will add a simple decay term in the future so user can customize their applications.

I updated the code, see the following ticket. Notice that the metadata definitions was also missing in varinfo.dat so this file was modified

Thank you for reporting this issue...

[rubash]

Thanks for updating the code.

I think I solved the problem in the second part of my posting--the part about river directions--by adjusting my u-mask and v-mask.