For example - it's possible that tundra methane emissions in many areas could be mitigated relatively quickly by the regrowth of birch and pine scrubland whose increased evapotranspiration will reduce water tables and eliminate methane emissions. And aboveground carbon sequestration in woody tree mass could outweigh losses from belowground oxidation. This already occurs in fenlands in Southern Finland (paper I read)
We don't know enough about methane hydrates to assess their stability.
Many ecosystems will be reasonably resilient to climate change - maybe up to 2/3 of them, although some specific species groups like tropical amphibians are going to have a really bad time.