Concerns about health effects due to emissions from ships have magnified international policy debate regarding low-sulfur fuel mandates for marine fuel. Policy discussions center on setting sulfur content levels and the geographic specification of low-sulfur fuel use. We quantify changes in premature mortality due to emissions from ships under several sulfur emissions control scenarios. We compare a 2012 No Control scenario (assuming 2.7% or 27 000 ppm S) with three emissions control scenarios. Two control scenarios represent cases where marine fuel is limited to 0.5% S (5000 ppm) and 0.1% S (1000 ppm) content, respectively, within 200 nautical miles of coastal areas. The third control scenario represents a global limit of 0.5% S. We apply the global climate model ECHAMSSy-MESSy1-MADE to geospatial emissions inventories to determine worldwide concentrations of particular matter (PM2.5) from ocean going vessels. Using those PM2.5 concentrations in cardiopulmonary and lung cancer concentration-risk functions and population models, we estimate annual premature mortality. Without control, our central estimate is approximately 87 000 premature deaths annually in 2012. Coastal area control scenarios reduce premature deaths by approximately 33 500 for the 0.5% case and approximately 43 500 for the 0.1% case. Where fuel sulfur content is reduced globally to 0.5% S, premature deaths are reduced by approximately 41 200. These results provide important support that global health benefits are associated with low-sulfur marine fuels, and allow for relative comparison of the benefits of alternative control strategies.