I wrote this post last year for the National Journal, but it also relates to the way I think about Petermann Glacier’s ice islands. There are now at least 4 larger ice islands that formed from last year’s single calving: one is the tourist attraction off Labrador and Newfoundland, a second has left Petermann Fjord last week, a third was grounded off Ellesmere Island for much of the year and is now where #1 was Nov.-2010, while the fourth … I do not know. Last I heart, it was grounded off central Baffin Island. With this much variation of where pieces of the ice island went, how can we possibly claim any skill in predicting anything?
Neither climate nor weather is linear, but this neither makes them unpredictable nor chaotic. The simple harmonic pendulum is the essence of a linear system with clear cause and effect relations. Oscillations are predictable as long as the initial forcing is small. Furthermore, a linear trend will show the pendulum to slow down due to friction. Corrections are straightforward.
Unfortunately, climate is not a simple, harmonic, or linear system. While this does not make it unpredictable or chaotic, it means that our “common sense” and loose talk of “totality of events” can easily fool us. We know that CO2 emissions for the last 150 years changed global temperatures. We also know that our current climate system has been very stable over the last 10,000 years. What we do not yet know is how small or how large a perturbations the last 150 years have been. If the pendulum is forced too much, if the spring is stretched too far, the system will find another stable state by breaking. Climate dynamics can find an adjustment less tuned to the areas where people presently live. This is what “tipping points” are about. Only numerical experimentation with the best physics and models will suggest how close to a different stable climate state we are. The IPCC process is one way to do so.
Ice cores from Greenland contain air bubbles 100,000 years old, which clearly demonstrate that our present climate state is the “anomaly of quiet” in terms of temperature fluctuations. The absence of large fluctuations for about 10,000 years made agriculture and advanced civilizations possible. The ice cores show that abrupt climate change has happened and may happen again, not this election cycle, but it is one possibility perhaps as likely as the possibility that climate change is mundane, linear, and follows trends that we can easily correct or mitigate later. Both are excellent hypotheses.
For scientists, these are exciting times as we conduct a massive, global experiment to see how much CO2 we can add to the atmosphere to perhaps find a different climate state. Dr. Terry Joyce, Senior Scientist at Woods Hole Oceanographic Institution once said: “I’m in the dark as to how close to an edge or transition to a new ocean and climate regime we might be. But I know which way we are walking. We are walking toward the cliff.” I agree with this sentiment, but add that we do not know if this cliff is a 1000 feet fall or a 2 feet step. Can we affort to wait until we know for sure? As a scientist I do not care. As a citizen, however, I think the time to act responsibly is now.
Icy Seas 
Ice sheet loss is much more than a 2-foot drop. The bottom of the drop that would be constituted by loss of the East Siberian Shelf methane is entirely out of sight. To the extent that we aren’t already committed to both, there’s good reason to suspect that the forcing associated with the rapid melt of the shallow permafrost (irretrievably on the way out, apparently) will be enough to do one or both jobs. Time to abandon that second hypothesis, I’d say.
The time scale over which changes happen matters. For example, remove all of Greenland’s ice, that’s 20 feet or more of global sea level, but this takes extreme and sustained forcing over a period of ~1000 years, a very unlikely scenario. I am unsure on the time scales of the methane that is released when permafrost melts. The 2-feet increase in sea level is at the upper end of what Greenland may do at the end of this century based on what we do (and do not) know now. Greenland’s current contribution to sea level rise are closer to 0.4 mm per year as per IPCC, even doubling these would give 2-3 inches.
Regarding the height of the “cliff,” I was speaking methaphorically, thinking of myself as sitting in a canoe hearing a water fall in the distance. A 2 feet drop will hurt, but may not swamp the boat, if I brace myself. The 1000 feet drop will kill me, as it did with the hikers in Yosemite. To me a hypotheses is never a statement of fact, but a provisional idea that can be tested with an experiment. And the experiment is still ongoing, right? All we know for sure is the direction that the water flows at the moment.