Even warmist Revkin doesn't sound completely convinced that CO2 is driving fluctuations in Arctic sea ice

The Arctic Ice Watch - NYTimes.com

Q. Ignatius Rigor was among those who pointed me to a particularly important one-time “flush” (my term) of thick old ice around 1989-90 that had an enduring impact on the proportion of older ice from then on. See this animation by Rigor:
...
Is that kind of non-linear event replicated much by models?

Also, given the short time scale of satellite observations, does such pulse-style ice behavior add uncertainty to efforts to link greenhouse-driven warming to recent (post 2000) sea-ice behavior?
...
[A] 2. Regarding the attribution: As shown in Fig. 1 of my 2009 PNAS paper, sea ice usually recovers within one year or so from extreme loss events [but if the feedbacks are so positive, why would this happen?] (same conclusion can be drawn from figure 3 of the RealClimate post).