upwelling test force

[redhat007]

dear
i know that in upwelling test, analytically files such as ana_wind, ana_humid are used for wind forcing and relative humidity. now i have 2 questions:
1- why do not those analytical files define in upwelling.h file? as i understand user should define all test's necessary analytical files in header file?

2- if we consider that in this test requested forces are humidity, wind, pair and tair but why NFFILES in ocean_upwelling.in file is equal to one. should not be equal to 4 ? (because of ana_humid, ana_tair, ana_wind and ana_pair)
thanks in advance for your attention

[kate]

The wind force can be computed from the winds, Tair, etc., or it can be provided directly as surface momentum flux (sustr, svstr). The upwelling case depends on the latter by not defining BULK_FLUXES. The sustr,svstr fields can be read from one or more files or computed in ana_smflux.h - in this case the latter by specifying #define ANA_SMFLUX.

By providing everything analytically, there is no need for ROMS to read any forcing files at all. The NFFILES line in ocean_upwelling.in is merely a place holder.

Based on the options chosen, ROMS will decide whether or not to look for forcing files, boundary files, etc. Some of this logic takes place in globaldefs.h.

[redhat007]

dear kate
thanks. but while all forces in this case provided by analytically file why NFFILES is equal to 1. you say that NFFILES is just for number of netcdf files.
also base on manual and your answer it is true that there are 3 ways for providing surface forcing (wind forcing)
1 computed wind tress directly.
2-input wind speed components in ocean_xx_.in and define bulk-fluxes.f in header file
3-using ANA-SMFLUX.h
and for suface meteorological forcing (fresh water and heat fluxes) there is two ways, too.
1- by required file (tair,Pair, lwrad,swrad,qair)and using bulk-fluxes.F for converting them to heat flux.
2-using analytically routines . for example ANA_SSFLUX for fresh water flux.
thanks again

[kate]

You ask why NFFILES is one when really it should be zero. If you try setting it to zero, you might find that ROMS will complain.

[redhat007]

thanks
i have got it