Timeline for Why don't errors accumulate in climate models when the time horizon increases?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 1, 2022 at 17:01 | comment | added | Deditos | Yeah, I'm coming from the point of view of national met services operational models, which I think is the thrust of the original quotes. As you say, there's a much wider range of approaches across the research arena (I've dabbled in LES at times), but they're probably not what most people mean when they say "forecast model" or "climate projection". | |
| Feb 1, 2022 at 14:35 | comment | added | AtmosphericPrisonEscape | @Deditos: Thanks for the clarification. I am writing from the perspective I got from studying astrophysical climate papers (often using RANS with parametrized turbulence physics, very few spectral codes on planetary scale), and assumed that similar approaches should be taken in earth science. If my view point is entirely untenable, I'll delete this answer. | |
| Feb 1, 2022 at 10:00 | comment | added | Deditos | I don't think your point's clear here. There's not really a fundamental difference between NWP and climate models. They both solve equations derived from the N-S equations (e.g., primitive, Euler's), and in many modelling centers their NWP and climate models use the same dynamical core. Climate models are just as chaotic as NWP models, which is why we have to submit multiple initial condition ensemble members to CMIP for each model. The main difference is in how they are used rather than the fundamentals. | |
| Jan 31, 2022 at 20:32 | history | answered | AtmosphericPrisonEscape | CC BY-SA 4.0 |