Bad result

[Shaivi Shukla]

@kwerner @William.Hatheway @Ming Chen
I have run the WRF model using CESM4 data from NCAR as the boundary and initial condition. To improve simulation accuracy, I experimented with various combinations of physics schemes, including:

• Microphysics: WSM3, WSM5, WSM6
• Longwave/Shortwave Radiation: RRTMG, CAM, RRTM
• Cumulus Parameterization: Kain-Fritsch (KF), Betts-Miller-Janjic (BMJ), New Simplified Arakawa-Schubert (NSAS)

Despite these efforts and allowing a one-month spin-up period, the rainfall output shows significant temporal shifts and poor agreement with observations. I would like guidance on how to further improve the model performance.

 

[Ming Chen]

CESM4 is a climate model, and its outputs are only making sense from the perspective of climatology, i.e., CESM results cannot correspond to real-time weather phenomena. As a result, WRF simulations driven by CESM also makes sense in 'climatology'. It is not unexpected that WRF outputs don't agree with real-time observations.