Abstract. We present a new method of modelling the growth of supraglacial lakes at the western margin of the Greenland ice sheet, based on routing runoff estimated by a regional climate model across a digital elevation model (DEM) of the ice sheet surface. Using data acquired during the 2003 melt season, we demonstrate that the model is 19 times more likely to correctly predict the presence (or absence) of lakes than it is to make incorrect predictions, within an elevation range of 1100 to 1700 metres above sea level (m a.s.l.), when compared with MODIS satellite imagery. Of the 66% of observed lake locations which the model correctly reproduces, the simulated lake onset day is found to be correlated with that observed with a Pearson correlation coefficient of 0.76. Our model accurately simulates maximum cumulative lake area with only a 1.5% overestimate. However, because our model does not simulate processes leading to lake stagnation or decay, such as refreezing or drainage, at present we do not simulate absolute daily lake area. We find that the maximum potential lake-covered ice sheet area is limited by topography to 6.4%. We estimate that this corresponds to a volume of 1.49 km³, 12% of the runoff produced in 2003. This can be taken as an upper bound given uncertainty in the DEM. This study has proved a good first step towards capturing the variability of supraglacial lake evolution with a numerical model. These initial results are promising and suggest that the model is a useful tool for use in analysing the behaviour of supraglacial lakes on the Greenland ice sheet in the present day and potentially beyond.