Well, none really. The 2°C limit which has become the fallback position isn't something to take lightly -- after all it's about 1/4th-1/3th of an inverse ice age, which is hardly something to scoff at, and with the warming and melting it will ultimately raise sea level by 25-50
But let's play along. Let's assume atmospheric carbon levels are increasing exponentially (which they are, once you look beyond a 10-15 year window), with a doubling time of D.
And let's assume the radiative forcing (and hence temperature change) is proportional to the logarithm of the atmosphere's carbon-equivalent level.
Then just a little algebra shows that the time t to reach a certain temperature T is
t = DT/S
where S is climate sensitivity.
Using greenhouse gas forcings since 1979 from NOAA, I find that CO2 is increasing at an annual rate of 0.46%, but CO2-equivalent is increasing at 0.66%, which gives a doubling time of D = 105 years.
The Otto paper reduces transient climate sensitivity (TCR) from a value of 1.6°C (based on data through the 1990s) to 1.3°C (based on data up through 2009).
So if we want to limit T to 2°C, the difference between these two TCRs gives us an additional 30 years.
3 decades. That's a fair bit of time.
Will we use those 30 years wisely and take climate change seriously, to finally do something about carbon dioxide, to build a clean, sustainable, noncarbon world for the many future generations who will live after us?
N.b. a lower climate sensitivity doesn't change the ocean acidification problem.