"Imagine that the ablation zone of Greenland – the area where melting exceeds accumulation so you have a negative surface mass balance – is only a hundred kilometres wide, and the output of global circulation models is of the order of 200 or 300 km," Marco Tedesco of the City College of New York told environmentalresearchweb. "You cannot really study these kinds of processes using such coarse spatial-resolution output. You need a much higher spatial resolution, like the [regional] model we used."

According to Tedesco, satellites can tell us where and when Greenland is melting but not how much water is produced by the melting. "The model can actually provide us [with] an estimate of that," he said.

For the first time, the scientists analysed Greenland's ice by splitting the country into six major drainage basins according to topography. "Once the ice melts, the water will go on one side or the other depending on the slope," said Tedesco.

The researchers found that the surface mass balance of basins along the southwest and north coast of Greenland is most sensitive to temperature rise. In these locations a global temperature rise of 0.6–2.16°C would lower surface mass balance beneath the 1980–1999 average, when the ice sheet was near equilibrium. Basins along the northwest and northeast of Greenland, in contrast, would require a temperature rise of 1.50–3.40°C for the same effect.

"We tried to relate the global surface-temperature anomaly projected by the different Earth-system models with some of the parameters that we extrapolate at very high spatial resolution for the different drainage basins," said Tedesco. "This is very important. [It is] one way we can provide useful information to the community."

Tedesco and colleague Xavier Fettweis used the regional climate model Modèle Atmosphérique Régionale (MAR), which has a spatial resolution of 25 km. They forced the model with outputs from the CanESM2, NorESM1 and MIROC5 Earth-system models, using greenhouse-gas scenarios with either 850 ppm or more than 1370 ppm of carbon dioxide equivalent by 2100.

The pair stress that the results are likely to be conservative as they do not include iceberg calving, ice-sheet dynamics or account for changes in the ice-sheet topography as melting occurs. Once the top of the ice sheet begins to melt the remaining ice is at a lower altitude, where temperatures are warmer.

"The numbers we are providing in terms of global temperature anomaly are probably too high," said Tedesco. "We do not know by how much, it could be by 10, 5%, 20%. It is something that we need to figure out."

The figures for surface mass balance the team calculated indicate that the basin along the southwest coast could contribute up to 8 cm of sea-level rise by the end of the century, and the basin along the north coast up to 4 cm, not accounting for ice dynamics, routing of surface water and elevation changes.

Now the pair would like to combine the MAR model with models for ice-sheet hydrology and ice dynamic. "When the MAR model generates liquid water from melting, the water is automatically removed from the ice," said Tedesco. "It is basically like somebody withdrawing money from a bank   it goes out straight away. In reality the water flows on the ice and there is some latency from the time it is generated to the time it leaves the ice. There are some specific routes that the water takes, depending on the topography."

According to Tedesco, coupling the current model with a surface hydrology model would provide more information on the actual dynamics – how much water stays at the surface, and how much percolates beneath the surface and goes through the ice. It would also "better focus the spatial location of the places that are most affected by the increasing surface melting". This could pinpoint "those parts of the ocean surrounding Greenland where you can expect a change due to increased run-off, changes in salinity, changes in temperature, changes in current and so on".

Tedesco now plans to collaborate with specialists in ice-sheet flow models. He and Fettweis reported their work in Environmental Research Letters (ERL) , as part of the ERL Focus on Cryospheric Changes in a Changing Climate.