out_dt in hydro.namelist

[Sara B]

Dear WRF-Hydro users,

I would have a question regarding the temporal resolutions defined in FORCING_TIMESTEP (in namelist.hrldas) and out_dt (in hydro.namelist).

I am running WRF-Hydro forced by WRF (FORC_TYP = 3) for one day (from 2026/04/30 00:00 until 2026/04/31 00:00). 

Time frequency of forcings is one hour, i.e. FORCING_TIMESTEP = 3600 and NOAH_TIMESTEP = 3600 (seconds). Time frequency of hydrological outputs is defined in out_dt (in minutes). I wished to get outputs at a resolution higher than one hour, so I tried to set out_dt=15. However, the result was not correct...
- the time written in the names of the output files was every hour

- I obtained 24 output files instead of 24*4, as I thought

- nevertheless, the timestamp inside the output files was every 15 minutes...

For example, the time inside the last file (202603310000.CHRTOUT_DOMAIN4) was 2026-03-30 06 (ncdump -v time -t 202603310000.CHRTOUT_DOMAIN4).

I have the feeling that the outputs refer to hours (every 60 minutes), although out_dt=15.

I would like to ask you if the minimum time that can be defined in out_dt is maybe the time of FORCING_TIMESTEP and why (since the integration time of the model is of the order of seconds, I thought it was possible to save the output files at the desired temporal resolution).

Thank you very much!

Best regards,

Sara B

[Alfred Zhong]

From some of the earlier threads in this forum it seems that setting out_dt to something smaller than one of the other timesteps is not well supported:

https://groups.google.com/a/ucar.edu/g/wrf-hydro_users/c/I2cGawB78zs/m/4Zw9RRNkAwAJ

https://groups.google.com/a/ucar.edu/g/wrf-hydro_users/c/-HKaieEyTlA/m/aGmCrHJpAAAJ

[aubrey]

Hi Sara:

Thanks for alerting us to this issue. If you are comfortable adding this to our github repo as an issue we can track it there, otherwise we can migrate it over for you.

https://github.com/NCAR/wrf_hydro_nwm_public/issues

We'll start to look into this on our end.

Thanks!

Aubrey