B2000 spin up time

[Ting-Hui Lee]

Hi,
I'm running B_2000_CAM5 with CESM1.2.2. It should be a warm start with all initial data except for the "finidat" in CLM. The initial net radiation at TOM (FSNT-FLNT) is about 1.45W/m2, and it decreased to 0.6 W/m2 at about year 140. Then it fluctuated at around 0.5 W/m2.
I used its restart file at year 140 and make a branch run with doubling CO2 concentration. The initial net radiation at TOM jumped to 2.1 W/m2 because of the CO2 (attached figure). In this branch experiment, it took more than 400 years for the net radiation to decrease to about 0.6 W/m2. But, it didn't fluctuate here. Instead, it's still decreasing by about 0.1 W/m2 per centenary from year 500-600. My question is can the output data (year 0500-0600) be used for analysis when the net radiation at TOM has a slight decrease? Or do I need to keep running this model until it reaches equilibrium so I can analyze its output data? How long does it usually take? And in which condition I can tell the branch experiment reaches equilibrium (net radiation at TOM fluctuates at 0 W/m2)?
Thank you.

 

[hannay]

I takes a very long time for the model to equilibrate after doing a 2xco2. This is because the most of the heat is going into the ocean (the atmosphere has a low heat capacity) and it takes a long for the ocean to re-equilibrate.
For instance, Bacmeister et al (2020) shows that 4xco2 coupled runs have not equilibrated after 1000 years. But you can still learn a of lot from your simulation. (I don't know what you are looking at here).



Bacmeister et al (2020). CO2 Increase Experiments Using the CESM: Relationship to Climate Sensitivity and Comparison of CESM1 to CESM2. in JAMES.
DOI: 10.1029/2020MS002120

 

[Ting-Hui Lee]

Hi Hannay,
I see. Thank you so much for your reply and paper.