End of the SMB Season summary 2017

12 Sept 2017

As the surface mass budget (SMB) year ends, it’s time for our now traditional look-back at the season since the 1st September 2016. There has been quite some discussion about Greenland in the climate blogosphere this year. Heavy snow and rain in winter with a relatively short and intermittent summer melt season have left the Greenland ice sheet with more ice than has been usual over the last twenty years – in fact we have to go back to the 1980s and 90s to see a year similar to this one in terms of snow fall and ice melt, though perhaps not for iceberg calving.

At the end of the SMB year that started on the 1st September 2016 and went through to the 31st August 2017, and not including the loss of icebergs, about 544 Gt more snow fell on the surface of the ice sheet than melted and ran into the ocean, compared to an average for 1981 to 2010 of about 368 Gt at the same point in the year. Note that this figure and the surface mass budget in general do not account for calving losses and the submarine melt of glaciers. Together calving and submarine melt have averaged around 500 Gt of ice loss per year over the last two decades. The end of year surface mass budget is therefore rather high for the 2000s and the 1990s, though it’s not the highest in the record and 2016-2017 has therefore not been a record year for Greenland ice sheet mass balance. If we rank the annual surface mass balance since 1981 from low to high, the lowest on record was 2011-2012 (38 Gt) and this year is the 5th highest out of the 37 year record. The highest on record 1995-1996 had an end of year SMB of 619 Gt in our records.

Graph of daily (top) and accumulated (bottom) surface mass balance for the 2016-17 season


So far we’ve only discussed the SMB year, what about the other component of ice loss in Greenland, calving icebergs?

Since at least 2002, the total mass budget has been substantially negative (on average from 2002 onwards it has lost -200 to -300 Gt per year). This year, thanks partly to ex-hurricane Nicole’s snow and partly to the relatively low amounts of melt in the summer, we estimate the total mass budget to be close to zero and possibly even positive. Greenland on average loses around 500 Gt of ice each year from calving and submarine melt processes. If we subtract this from our figure of 544 Gt for the SMB it would suggest Greenland gained a small amount of ice this year. However, compared to the approximately 3600 Gt of ice, corresponding to 1cm of global average sea level rise that Greenland has lost since 2002 this year’s slightly positive balance does not add much extra.

We should emphasise that this is a preliminary estimate of the Greenland total mass budget. We will have to wait for further data to become available from satellites to estimate the actual calving losses. Unfortunately, the ageing GRACE satellites have finally expired, up to now have been the workhorses monitoring the trend in ice sheet total mass balance (see below). Their replacements are planned for launch later this year. 

So what has contributed to this high SMB this year? As we have written previously, Heavy rain and snow in October in especially eastern Greenland gave record totals of precipitation in the main east coast town of Tasiilaq as the remnants of former hurricane Nicole passed by and, much as with Harvey in Houston this year, got lodged over eastern Greenland for some days. However, after Nicole’s extreme precipitation, the rest of the winter was actually pretty average in terms of the amount of snow that fell, and indeed parts of north western Greenland had less snow than usual, as can be seen in the accumulated SMB anomaly map below.

The other important process this year has been a relatively short and below average melt and runoff season. The summer melt season is extremely important for the annual SMB and the overall health of the ice sheet. The date of onset of the melt season was the 3rd earliest on record but it was then followed by a cold and snowy period so that the onset of high rates of ice loss (more than 1Gt of ice lost per day, known as the ablation season) was actually average. Going into summer with a large amount of accumulated fresh snow means that the albedo feedback took longer to kick in. Albedo, the reflectivity of the surface, is a very important control on the rate of melt of glacier ice. Fresh snow is bright and reflective and slower to melt, but when the older and darker snow and glacier ice is exposed underneath, surface melt accelerates. The large amount of fresh snow in June substantially dampened this feedback.

After the average start to the ablation season there were some substantial pulses of melt particularly in July. Summer snow and rain over the ice sheet added more ice but did not significantly reduce melting in the lower parts of the ice sheet and maximum melt occurred, much as usual in late July with daily SMB values well below the average due to the high melt and runoff rates in this period. The large swings in the daily SMB reflect the weather in Greenland. Very unusually, July 2017 managed to set both a new cold record (-33.0 °C on the 4th July) and a new warm record (+1.9 °C on the 28th at Summit station on top of the ice sheet (3216 m elevation). July and early August were also dry especially in the south and west, with conditions proving perfect for the spread of two wildfires adjacent to the ice sheet. These are unlikely to have had much impact on the ice itself but may be a sign of changing climate and vegetation conditions on the land in Greenland.

 From mid-August there was one final surprise in this unusual year. Although it was still warmer than average in much of Greenland, a series of storms swept over from the west and northwest, bringing more rain and snow to balance the melt losses (and extinguish the wildfires). As a consequence the ablation season ended roughly a week earlier than usual over the last couple of decades. Melt continues, but the ice sheet started to gain more ice than it lost from 17th August this year.

So, is this good news for the ice sheet? Well, it’s not bad news, the year-on-year decline of the ice sheet has seen a small check this year, and the 2016-17 season demonstrates in part how important weather conditions in Greenland are in determining the annual SMB, however, when seen against the long term trend it quickly becomes apparent that there is not much reason to celebrate this year.

GRACE data showing ice mass changes in Greenland since 2002 up to January 2017.
Accumulated surface mass balance for 2016-17



Each year glaciers gain ice from snow and freezing rain and lose ice by melt that runs off. Adding these together gives the surface mass budget (SMB) in Greenland, the ice sheet typically gains mass from around September to May and loses more mass than it gains in the ablation season of June, July and August. In our work at DMI we use output from a weather forecasting model every day to calculate how much ice Greenland is gained by snowfall and how much is removed by melt. This is accumulated over the year to give us the annual surface mass budget, which is something like a health indicator for the ice sheet. We use the period from the 1st September to the 31st August as our SMB year as this reflects the annual cycle in snowfall and snowmelt. We don’t consider calving of icebergs in this and the surface mass budget at the end of the balance year should be much higher than at the start as the snow that falls during the year has to also compensate for the loss of ice by calving icebergs. Including this calving we get the total mass budget that has, since 2002 been calculated using the GRACE satellite. The GRACE results show that for almost every year since 2002the Greenland ice sheet has lost more ice than it gained from snow,. Analysis of model results and satellite measurements of velocity show that much of the increase in ice loss since the 1990s has occurred due to increases in melt and runoff in Greenland, including the famous 2012 summer event when much of the ice sheet had surface melt. 2012 is the lowest year in the record since at least 1980. 

Polar Portal is a collaboration between DMI, GEUS, DTU-Space and DTU-Byg with funding from the Danish Energy Agency (Energistyrelsen).

Polar Portal is also on Twitter: @PolarPortal and Facebook: https://www.facebook.com/PolarPortal/.