From both social and environmental perspectives, water is the main connection between highland and lowland processes in mountain watersheds: Water flows downhill while human impacts flow uphill. For example, in the Oregon Cascades mountain range, geology, vegetation, and climate influence the hydrologic connections within watersheds. Geology determines which watersheds are surface runoff-dominated and which are groundwater-dominated. In this Mediterranean climate with dry summers, surface runoff watersheds will consistently experience near-zero late summer discharge, so declining snowpacks will have little effect on low flows. This contrasts with groundwater-dominated watersheds, where a shift from snow to rain or a decline in precipitation will reduce recharge, thereby reducing late summer groundwater contributions to streamflow. Earlier snowmelt causes forests to transpire earlier, resulting in decreased springtime streamflow. Reduced snowpacks lead to soil moisture stress, making forests more vulnerable to extensive wildfires and affecting the lifespan and composition of forests. Monitoring and quantifying these complex linkages and feedbacks require appropriate measurement networks. Sampling strategies often use watershed typology to identify where measurements should be focused. Such an approach should include not only established watershed classification parameters such as topology and geology but also interannual climate variability and land cover. As concerns of water scarcity and vulnerability move to the forefront, our watershed classifications should be extended to include ecosystem and social–ecological parameters. An integrated and agent-based modeling scheme called Envision has been developed to simulate alternative future landscapes at the watershed scale. Using fully coupled models of hydrology, ecosystems, and socioeconomics, decision-makers can simulate the effects of policy decisions in conjunction with other climate forcing, land use change, and economic disturbances. To understand the combined impacts of climate change and humans on water in mountain watersheds, researchers must develop integrated monitoring and modeling systems that explicitly include connections across eco-hydrologic and social-ecological systems.