My 2 kids were singing the rain rain go away rhyme during last weekend’s epic rainfall in Copenhagen and it reminded me that I have not yet put up a post about a paper I was a co-author on this summer related to late summer/autumn rainfall and the effects on the Greenland ice sheet, so here goes….
Mostly when we think of precipitation in Greenland we think of snow in the winter, but it does rain quite a lot, as I know from personal experience (see photo taken as the clouds started to clear one September field season in Eastern Greenland…). This paper in Nature Geoscience by Sam Doyle and co-authors including myself shows that when rain falls on the ice sheet at the “wrong” time of year it can have a very far-reaching effect, causing the speed up of a large area across the ice sheet.
Exploratory mining camp tents in the foreground.
The important caveat is that rainfall during the main part of the melt season is more or less evacuated away quickly. Glaciers – and the Greenland ice sheet is basically a very big glacier – develop a drainage system more or less analogous to large underground sewers during the melt season. These tend to close down during the colder accumulation season and reopen by the sheer pressure of water running through them when the melt season starts. Rainfall during that crucial late summer/early autumn period when the drainage is closing down and therefore less efficient at evacuating surplus liquid water is therefore not able to move away from the glacier very easily and forces its way through any way it can find.
During this period, most of the snow will have melted off the surface, leaving vast areas of bare ice. By contrast, rain on snow in the early part of the melt season when there is a thick snow pack is more likely to refreeze inside the snow. In late summer however, there will be a relatively short period between rain falling and accumulating in the glacier drainage system.
In practice this means the water makes its way to the bed of the glacier through moulins and englacial channels, where it more or less hydraulically jacks up the glacier over a large region, allowing the ice to flow to the margins faster. There may then also be a knock-on effect with increased calving of icebergs at outlet glaciers. in 2011, the field team were able to measure both the rain fall and the following cascade of processes in a range of different datasets as shown below:
My contribution to the paper was in the form of some HIRHAM5 model runs for Greenland which show the last decade has seen a significant increase in rainfall events in the summertime compared with the previous decade. We chose as a study region the K-transect of weather stations in western Greenland. These are operated by Utrecht University and have a long time-series of data which previous work has shown our model can replicate quite nicely. The model is forced by the ERA-Interim reanalysis, a data set based on weather forecast models with real observations included in it run for the whole world so we are pretty confident the rainfall patterns are realistic. There are actually two interesting points illustrated in the picture below taken from the paper. Firstly that there is more rain falling and secondly that this rain is falling at higher elevations on the ice sheet, potentially causing a much wider area of the ice sheet to be affected by late-summer rainfall events.
The decadal change in rainfall events is partly due to a persistent North Atlantic Oscillation anomaly which has funnelled storms over the western edge of the ice sheet. There is also some evidence that the stratospheric Rossby waves have become more “wavy” over the same period, due to the increasing warming and vanishing sea ice in the Arctic. This hypothesis was articulated in a very nice paper by Francis and Vavrus but it remains a very open area of research as we just don’t have a lot of evidence right now.
We do know that the Arctic is one of the fastest warming regions on the planet and this will certainly have a knock-on effect on the Greenland ice sheet both in terms of melting and, perhaps, in the frequency of storms bringing rain over the ice sheet in the future. I am now preparing a new study to see if we see a signal along these lines in our future simulations of the Greenland domain.
Sterna Paradisaea 
Wonderful explanation. In the figure above it is clear that the rainfall event induced a velocity increase, which would be expected. However, in the big picture is this increase offset by a post speed up slow down that offset the short term acceleration? This has been observed to be the case on some alpine glaciers.
LikeLiked by 1 person
Thanks for your comment! The speed-up recorded here is unusual because of the timing, as you point out, on alpine glaciers there is often a slowdown afterwards, which is also true in Greenland, there is definitely a seasonal cycle in velocity at lower elevation locations at least which is driven by hydrology. In this case the late summer slow-down as melt declines, was already starting, the rainfall reversed this trend temporarily. I don’t think there was any overshoot in slow-down but I will check with Sam Doyle who analysed the GPS data.
LikeLike