Hi all,
I'm encountering a strange issue with auxiliary history output (auxhist7) during a long-term WRF simulation from 2013 to 2022.
I'm encountering a strange issue with auxiliary history output (auxhist7) during a long-term WRF simulation from 2013 to 2022.
My Configuration:
- Regular output: Daily files
- auxhist7: Hourly output (for specific variables)
- auxhist7 settings:
iofields_filename = "my-file_d01.txt"
ignore_iofields_warning = .true.
auxhist7_outname = "zwrfout_d<domain>_<date>"
auxhist7_interval = 60 ! 60 minutes
frames_per_auxhist7 = 24
io_form_auxhist7 = 2 - Expected behavior: One file per day containing 24 hourly time steps
- restart at the begining of every year
The Problem:
The simulation runs normally until around 2021, then suddenly the auxhist7 output behavior changes:- Instead of one file per day with 24 hourly steps, I get 24 files per day
- The time interval within each file becomes very small (around 4 minutes)
- Similar issue occurred in another case around 2017
Questions:
- What could cause WRF to automatically change the output interval during runtime?
- Has anyone experienced similar issues with long-term simulations?
- Are there any known limitations with auxhist outputs for multi-year runs?
Additional Information:
- WRF Version: 4.7.1
- The issue is reproducible in different cases
- No error messages in rsl files when the change occurs