Hi Haijin
Unfortunately, there isn't a single definition of what being "spun up" means. And you can't necessarily define being spunup simply by the number of years run. And it also depends on what you are going to do with the simulation that you are going to use with this startup case and exactly what model version you are going to be running. And what science you are interested in using with it.
Most of what we save under the *_init spaces are things where we have save initial conditions that are good for general purposes. And for that matter they are the initial conditions that were used for the main CESM simulations for a given model version. So this would be a good starting point for example for CCSM4 to start a historical simulation starting in 1850.
In terms of evaluating if something is spunup, it depends on your science questions. If you are only concerned about faster processes like the atmosphere, you are going to be fine. If you are studing deep soil Temperature, or arctic soil you may need to evaluate if it's spunup enough. Likewise for deep ocean. And if you are going to do something that perturbs the CCSM4 1850 climate you might need to reevaluate if it's spunup for those purposes. And the same for a different model version.
So the answer is "it depends". What science questions are you working on for your simulations?
