Selection of Spin-up Duration for the CLM Model

[kylin]

I lack long-term spin-up forcing data and have only prepared 21-year meteorological forcing data for the spin-up of the CLM model. Therefore, I plan to reuse it for a certain number of cycles until the model state reaches equilibrium. However, I have a doubt regarding the selection of the number of cycles: How many cycles should be chosen to ensure that the spin-up reaches equilibrium? Are there any methods to determine spin-up equilibrium?

The figure shows some of my settings for the spin-up case. Among them, ./xmlchange RESUBMIT=? is for selecting the number of resubmissions. I want to know what number is the best for this.

 

[slevis]

The length of spin-up depends primarily on whether you run in bgc mode or sp mode. I recommend reading this part of the User's Guide:

1.5. Running Special Cases — ctsm release-clm5.0 documentation

Also searching these Forums, I think you will find a few (possibly many) threads discussing spin-up questions. Some of them will likely mention an ncl tool found in clm's /tools/contrib directory, SpinupStability, that you may find useful.