Climate warming impacts on streamflow in mountainous mid-elevation watersheds are commonly evaluated as changes in snowpack dynamics directly affecting the seasonal timing and magnitude of downstream flows. From an ecosystem perspective, changes in vegetation structure and productivity due to climate shifts are commonly interpreted as responses to changes in phenological triggers and resource availability. Water and vegetation systems, however, are intrinsically linked, particularly in water-limited environments. In arid and semi-arid landscapes, this tight coupling can lead to feedbacks between climate-streamflow and climate-vegetation processes that may ameliorate or exacerbate climate change impacts on ecosystem health and downstream water resources. Integrated ecologic-hydrologic process models can be used to explore these linkages in ways that are difficult to do using empirical evidence alone. To this end, we focus on two coniferous forest watersheds in semi-arid Northern New Mexico spanning ponderosa pine, mixed conifer, and spruce-fir ecosystems. Our objectives are to use the spatially-distributed, eco-hydrologic model RHESSys to simulate vegetation productivity response to warming and evaluate how these vegetation dynamics interact with direct climate controls on hydrologic processes. We first test the model’s ability to capture observed climate-driven variation in productivity and mortality for 3 plots that span a 1-km elevational gradient in Bandelier National Monument. We then apply RHESSys to the nearby Santa Fe Municipal Watershed to evaluate how these model-predicted dynamics scale across a broader elevation range and species distribution. Finally, we use a set of model scenarios to examine how direct impacts of climate warming on snow and evapotranspiration interact with changes in vegetation productivity and mortality to ultimately affect streamflow.
Results/Conclusions
Our coupled eco-hydrologic model is able to accurately reproduce spatial differences in productivity and mortality across the 1-km elevational gradient during the 2000s drought. This productivity-climate relationship is sensitive to temperature, as this same drought under 2° C warmer temperatures shifts the predicted upper mortality bound into the mid-elevation range. Applying this field-validated process model to the broader forest gradient in Santa Fe, we confirm the robustness of our productivity estimates by comparing to local tree ring measurements. In this watershed, while temperature effects on snowpack remain the dominant control on streamflow, incorporating vegetation dynamics in response to climate variation improves annual streamflow predictions by 20-30% during the early 2000s drought years. We show that while predicted climate warming impacts on snowpack lead to lower streamflow, productivity changes due to warming can counterbalance this effect in some scenarios.