The fate of the Greenland ice sheet during the next centuries is largely uncertain. While there is high confidence that the ongoing mass loss will continue into the future, it is unclear whether the current rate of melting may accelerate as the climate in the Arctic is becoming increasingly warmer (IPCC, 2007). A shrinking ice sheet will not only result in rising sea level but will also inject meltwater directly in to the deep-water formation areas in the North Atlantic. The current generation of climate models has a spatial resolution that may be insufficient to capture the response of the ocean circulation to meltwater injections in a correct way. Natural experiments during the last deglaciation (meltwater pulse 1A, Younger Dryas, and 8.2-kyr event) provide unique test cases to query the fidelity of ice-sheet models and the response of the ocean-circulation to meltwater perturbations in high versus low-resolution models. ArcTrain will generate new and compile existing meltwater histories from paleoclimate archives and use the resulting information for testing the behavior of ice sheet and climate models.