A lot of interest in the Greenland ice loss study from King et al. nature.com/articles/s4324…
To be clear, this is not good news. Greenland ice sheet is demonstrably out of balance with current temperatures. But it doesn't (necessarily) presage the collapse of the whole thing.
To be clear, this is not good news. Greenland ice sheet is demonstrably out of balance with current temperatures. But it doesn't (necessarily) presage the collapse of the whole thing.
This is an analysis of 35 years of data, not a modeling study, and so while it can do a good job at attributing the current rates of loss to dynamic responses of the ice sheet, it says nothing about where the process would end up in the future under any plausible scenario.
Nonetheless, it is a very valid question (and subject of much research) to ask at what point the Greenland ice sheet is unviable.
From the Pliocene records, we know that a global mean of ~3ºC above the pre-industrial does not seem to be compatible with a substantial GIS.
From the Pliocene records, we know that a global mean of ~3ºC above the pre-industrial does not seem to be compatible with a substantial GIS.
From the history of past interglacials, with differing temperature changes, the ice sheet responded to different extents: pnas.org/content/117/1/…
From the last interglacial, we suspect that an Arctic warming of > 6ºC is compatible with a loss of ~1/3 of the GIS (ie. ~2m of global sea level rise). (fig from royalsocietypublishing.org/doi/pdf/10.109…) 
Note that the polar amplification differs between the Pliocene and LIG because of the importance of the orbital forcings in the latter case.
So what do models suggest for the sensitivity of the ice sheet as a whole? History suggests there's unlikely to be a single 'tipping point', but rather a series of non-linear steps to any new state. There's a summary of this in the IPCC sp. rep: ipcc.ch/srocc/ (ch 2+3)
This modeling study looked in detail at the thresholds involved, suggesting that the threshold for complete loss is somewhere between 1º and 3ºC above pre-industrial (best estimate was ~1.6ºC); nature.com/articles/nclim…
But that was a single (relatively coarse) model and we are still waiting on more comprehensive models to weigh in. Note that to get Arctic temps comparable to the LIG (when 1/3 of the ice sheet was lost), we'd expect global mean changes of ~2 to 3ºC.
None of this should suggest complacency, and indeed there is a lot more work to be done (and please link to any papers I missed below), but the current observations don't on their own imply the inevitability of Greenland collapse - and certainly not this century.
