The Arctic is melting rapidly and is more than 75% likely to experience its first ice-free summer by 2060 even if a pathway of 1.5 °C with overshoot is achieved (Dunne, 2019). The loss of Arctic ice could have a severe impact on the climate. Therefore, targeted action in the Arctic is required. The proposed method examined herein is Ice Volcanoes: pumping seawater onto the ice in winter so that it freezes faster and the ice can survive the following summer months. Thickening ice by pumping water over its surface is investigated theoretically and experimentally. The change in thickness of the ice with water flowing over its surface is modelled for channel flow. Short time scales are considered during which the dominant heat transfers are advection from water to the water–ice interface and conduction through the ice away from the interface. At short time scales the rates of heat transfer by advection and radiation to the atmosphere are much smaller and so not considered. Advection of heat by the water is modelled for three flows: an inviscid flow without a thermal boundary layer; an inviscid flow with a thermal boundary layer; and a viscous shear flow with a thermal boundary layer. The three models are assessed in the context of experimental results for water injected at 0.5 °C and 0.8 °C. The viscous shear flow with a thermal boundary layer is found to be the most accurate when compared with experimental results. However, the models do not accurately predict the experimental data for water injected onto the ice at 2 °C. Potential reasons for this are discussed. Finally, the paper concludes with suggestions for further work and some implications for ice volcanoes.