What is the difference between surface mass balance and melting?

By Peter Langen, DMI/Polar Portal
May 22 2014

The term surface mass balance is used for the gain and loss of mass at the surface of the ice sheet. This includes melting, snowfall, refreezing of melt water and evaporation of snow (sublimation). Daily maps of this will often look quite different from maps showing occurrence of surface melting.

May 20 2014 is a beautiful example of the difference between surface mass balance (SMB) and melting alone. Below we show the model-derived SMB as posted on Polar Portal compared with the melt extent in the same model. We also reproduce the satellite-inferred melt area from NSIDC. Clear differences between the SMB and melt are obvious, especially in the south: the SMB shows blue areas with mass gain in places where both model and satellite show melting.

Illustration from May 20 2014 of the difference between (left) model-derived surface mass balance used on Polar Portal and (middle) areas where surface melting exceeds 1 mm/day in the same model. To the right is shown the satellite-derived melt area from http://nsidc.org/greenland-today/ (Image courtesy of the National Snow and Ice Data Center, University of Colorado, Boulder).

How is this possible? Recall that SMB takes snow- and rainfall, melting of snow and bare ice, refreezing of melt water as well as sublimation into account. Especially this early in the summer season, the snowpack is still quite cold and can refreeze a lot of water. This means that you can have a situation like May 20 where the surface is at the melting point, but the melt water is refrozen further down in the snowpack. This mass is therefore not lost as runoff to the ocean. At the same time, it can snow or rain on the surface thereby adding mass to the ice sheet.

A couple of other features are very interesting: At the very margin around the southern part of the ice sheet, the SMB shows negative values. This happens either because the snowpack is gone, so the underlying bare ice is exposed, or because the snow pack has warmed up so much that it can no longer refreeze the melt water. In both cases, the melting leads to runoff to the ocean and actual mass loss from the ice sheet. 

Another interesting feature is the area of pink, slightly negative, SMB values in the East and Northeast. This mass loss was not due to melting. In fact, the PROMICE measurements showed that it was still quite cold (down to -10°C). This mass loss is due to sublimation - evaporation directly from a frozen surface.