UC Irvine

In order to understand ice sheet response to climate change, it is critical to examine errors associated with ice flow model boundary conditions and forcing. It is also important to understand how these errors propagate through numerical ice sheet models and contribute to uncertainty in model output. Using established uncertainty quantification methods within the Ice Sheet System Model (ISSM), we investigate the sensitivity of ice flow within the Northeast Greenland Ice Stream (NEGIS) to key fields, including ice viscosity and basal drag, and compare them with model sensitivity to climate forcing. In addition, we examine how errors in model input manifest as mass flux uncertainties during a forward simulation of the NEGIS from 1989 to 2010. Overall, we find that mass flux is most uncertain in the main outlets, Nioghalvfjerdsbræ and Zachariæ Isstrøm, and that mass flux is most sensitive to basal drag, though errors associated with basal drag are poorly constrained and difficult to quantify. Given our knowledge of errors associated with the thermal properties of ice, we estimate that in the ablation area, the effects of cryohydrologic warming contribute over 4 times more mass flux uncertainty that do errors in geothermal heat flux. We find that NEGIS total ice discharge is associated with a 0.7 Gt/yr (2.6%) uncertainty due to errors in geothermal heat flux and a 3.3 Gt/yr (11.6%) uncertainty due to the added effects of cryohydrologic warming. In comparison, errors in surface mass balance contribute 4.5 Gt/yr to NEGIS total discharge uncertainty. Key Points Mass flux is most sensitive to basal drag Mass flux is the most uncertain at the 79North outlet Geothermal heat flux uncertainty is less than that from cryohydrologic warming.