Prolay Saha
New member
Dear WRF community,
I am performing an SST perturbation experiment using WRF. I have 6-hourly ERA5 SST data and constructed a July 1991–2020 SST climatology.
After metgrid.exe, I regrid the climatological SST to the WRF d01 and d02 grids using xESMF (Python library) and replace the SST variable over ocean grid points in each met_em file. The original land values are retained.
In simplified form:
Python
ocean = LANDMASK < 0.5
SST[0,:,:] = np.where(ocean, ERA5_SST, SST[0,:,:])
The modified met_em files are then used for real.exe and WRF.
I would like to ask:
1. Is this a correct approach for applying an SST perturbation in WRF?
2. Since met_em also contains SKINTEMP, should I perturb SKINTEMP over the ocean to the same SST value, or should I leave SKINTEMP unchanged and modify only SST?
I would appreciate any advice on the recommended practice for SST perturbation using ERA5 6-hourly SST.
Thank you
I am performing an SST perturbation experiment using WRF. I have 6-hourly ERA5 SST data and constructed a July 1991–2020 SST climatology.
After metgrid.exe, I regrid the climatological SST to the WRF d01 and d02 grids using xESMF (Python library) and replace the SST variable over ocean grid points in each met_em file. The original land values are retained.
In simplified form:
Python
ocean = LANDMASK < 0.5
SST[0,:,:] = np.where(ocean, ERA5_SST, SST[0,:,:])
The modified met_em files are then used for real.exe and WRF.
I would like to ask:
1. Is this a correct approach for applying an SST perturbation in WRF?
2. Since met_em also contains SKINTEMP, should I perturb SKINTEMP over the ocean to the same SST value, or should I leave SKINTEMP unchanged and modify only SST?
I would appreciate any advice on the recommended practice for SST perturbation using ERA5 6-hourly SST.
Thank you