While I look up programming and technical stuff on Stack Overflow all the time, I figured that would probably be the wrong forum for a question like this.

Some background: I'm a climate modeler working on paleoclimates and data comparison for warm climate modes, and I've been writing ice sheet models in C and C++ for a while. It's been painful but finally I'm happy enough with how the ice is behaving to incorporate it into the GCM we've been using. The question boils down to the following: Assuming you have a fairly decent simulation of the present day state, how long would you expect a simulation to need to run to adjust to a similar warm climate state, in my case the Eemian. I've run it through 2000 years, but my ocean salt is still climbing (very very slowly, but linearly -- 0.05% per century, more or less) Physically that translates to a little more than one full ocean overturning.

  • $\begingroup$ Seems to me that most of the time, this is just figured out by trial and error - try it and see when all relevant variables start to stabilise. Unless you know of the spin-up time of similar models, in which case you can probably just use those. $\endgroup$
    – naught101
    Sep 30, 2014 at 3:41

1 Answer 1


Each of your subsystems -- and likely also each coupling between subsystems -- will have an intrinsic time scale. What that is of course depends on your system. The spin up time will then be a small multiple of the maximum of all of the time scales that are relevant in the transition you are considering.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.