There's a big difference between average temperatures and peak temperatures. On Larsen C, we regularly measure localised temperatures in excess of 10˚C for days at a time. That's enough for some serious melt.
Also, surface melt is only half the story. The base of the ice is in contact with the ocean, which is also warming. Water in contact with ice is (more or less by definition) at the freezing/melting point - if it warms, the ice melts a bit, if it cools, the water freezes. So a change in ocean temperatures can have a big effect on melt rates.
Finally, it's not enough that the shelf be in equilibrium, melt-wise. The ice is constantly thinning, as it stretches and flows under its own weight. There needs to be enough extra mass added (mostly as snow) to compensate for that, if the shelf is to remain a constant thickness.
Also, surface melt is only half the story. The base of the ice is in contact with the ocean, which is also warming. Water in contact with ice is (more or less by definition) at the freezing/melting point - if it warms, the ice melts a bit, if it cools, the water freezes. So a change in ocean temperatures can have a big effect on melt rates.
Finally, it's not enough that the shelf be in equilibrium, melt-wise. The ice is constantly thinning, as it stretches and flows under its own weight. There needs to be enough extra mass added (mostly as snow) to compensate for that, if the shelf is to remain a constant thickness.