Search for: All records

Abstract In the western United States, water supplies largely originate as snowmelt from forested land. Forests impact the water balance of these headwater streams, yet most predictive runoff models do not explicitly account for changing snow‐vegetation dynamics. Here, we present a case study showing how warmer temperatures and changing forests in the Henrys Fork of the Snake River, a seasonally snow‐covered headwater basin in the Greater Yellowstone Ecosystem, have altered the relationship between April 1st snow water equivalent (SWE) and summer streamflow. Since the onset and recovery of severe drought in the early 2000s, predictive models based on pre‐drought relationships over‐predict summer runoff in all three headwater tributaries of the Henrys Fork, despite minimal changes in precipitation or snow accumulation. Compared with the pre‐drought period, late springs and summers (May–September) are warmer and vegetation is greener with denser forests due to recovery from multiple historical disturbances. Shifts in the alignment of snowmelt and energy availability due to warmer temperatures may reduce runoff efficiency by changing the amount of precipitation that goes to evapotranspiration versus runoff and recharge. To quantify the alignment between snowmelt and energy on a timeframe needed for predictive models, we propose a new metric, the Vegetation‐Water Alignment Index (VWA), to characterize the synchrony of vegetation greenness and snowmelt and rain inputs. New predictive models show that in addition to April 1st SWE, the previous year's VWA and summer reference evapotranspiration are the most significant predictors of runoff in each watershed and provide more predictive power than traditionally used metrics. These results suggest that the timing of snowmelt relative to the start of the growing season affects not only annual partitioning of streamflow, but can also determine the groundwater storage state that dictates runoff efficiency the following spring. 

Abstract Instream barriers, such as dams, culverts, and diversions, alter hydrologic processes and aquatic habitat. Removing uneconomical and aging instream barriers is increasingly used for river restoration. Historically, selection of barrier removal projects used score‐and‐rank techniques, ignoring cumulative change and the spatial structure of stream networks. Likewise, most water supply models prioritize either human water uses or aquatic habitat, failing to incorporate both human and environmental water use benefits. Here, a dual‐objective optimization model identifies barriers to remove that maximize connected aquatic habitat and minimize water scarcity. Aquatic habitat is measured using monthly average streamflow, temperature, channel gradient, and geomorphic condition as indicators of aquatic habitat suitability. Water scarcity costs are minimized using economic penalty functions while a budget constraint specifies the money available to remove barriers. We demonstrate the approach using a case study in Utah's Weber Basin to prioritize removal of instream barriers for Bonneville cutthroat trout, while maintaining human water uses. Removing 54 instream barriers reconnects about 160 km of quality‐weighted habitat and costs approximately US$10 M. After this point, the cost‐effectiveness of removing barriers to connect river habitat decreases. The modeling approach expands barrier removal optimization methods by explicitly including both economic and environmental water uses.