Recent studies suggest that climate change is causing rising solute concentrations in mountain lakes and streams. These changes may be more pronounced in mineralized watersheds due to the sensitivity of sulfide weathering to changes in subsurface oxygen transport. Specific causal mechanisms linking climate change and accelerated weathering rates have been proposed, but in general remain entirely hypothetical. For mineralized watersheds, a favored hypothesis is that falling water tables caused by declining recharge rates allow an increasing volume of sulfide-bearing rock to become exposed to air, thus oxygen. Here, we test the hypothesis that falling water tables are the primary cause of an increase in metals and SO4 (100-400%) observed since 1980 in the Upper Snake River (USR), Colorado. The USR drains an alpine watershed geologically and climatologically representative of many others in mineralized areas of the western U.S. Hydrologic and chemical data collected from 2005 to 2011 in a deep monitoring well (WP1) at the top of the USR watershed are utilized. During this period, both water table depths and groundwater SO4 concentrations have generally increased in the well. A numerical model was constructed using TOUGHREACT that simulates pyrite oxidation near WP1, including groundwater flow and oxygen transport in both saturated and unsaturated zones. The modeling suggests that a falling water table could produce an increase in metals and SO4 of a magnitude similar to that observed in the USR (up to 300%). Future water table declines may produce limited increases in sulfide weathering high in the watershed because of the water table dropping below the depth of oxygen penetration, but may continue to enhance sulfide weathering lower in the watershed where water tables are shallower. Advective air (oxygen) transport in the unsaturated zone caused by seasonally variable recharge and associated water table fluctuations was found to have little influence on pyrite oxidation rates near WP1. However, this mechanism could be important in the case of a shallow dynamic water table and more abundant/reactive sulfides in the shallow subsurface. Data from WP1 and numerical modeling results are thus consistent with the falling water table hypothesis, and illustrate fundamental processes linking climate and sulfide weathering in mineralized watersheds.