Abstract
Vegetation is a critical component of ecological and hydrological systems, but rising temperatures, worsening droughts, and new disturbance regimes are changing the composition of many ecosystems. The connections between vegetation, climate, and streamflow are complex and often require long-term data to characterize, but sites with sufficiently long datasets are limited. At the same time, many modeling frameworks used to predict future streamflow are not designed to capture long-term, nonstationary relationships between vegetation, climate, and streamflow. While it is well-known that ecosystems affect runoff, significant research is required quantifying how vegetation dynamics mediate rainfall-runoff relationships and how disturbance regimes modify these relationships through time. Resolving connections between vegetation, climate, and streamflow includes understanding mechanisms contributing to ecosystem change (e.g., drought or cavitation), timescales at which ecosystems adjust to disturbance (e.g., seasonal or decadal), and the various spatial templates imposed upon ecosystems (e.g., soil heterogeneity or topography). This dissertation resolves climate-vegetation-streamflow interactions in forested ecosystems by exploring long-term physiologic signals of vegetation in hydrologic states and fluxes (e.g., soil moisture and streamflow, respectively). It uses decades of data from research stations across the US East Coast to demonstrate that streamflow and forests are ‘competing’ for the same soil water resources and that feedbacks between forests and soil water resources are spatially and temporally generalizable. Recession analyses applied to 30 years of observed and modeled streamflow revealed transpiration as one of the primary mechanisms contributing to lower storm-event flows during the growing season (Chapter 2). Subsurface measurements of soil moisture and groundwater during these storms indicated flow generation originating primarily from the rooting zone (Chapter 3) and that the spatial template of subsurface soil moisture was controlled by groundwater upwelling related to topography and vegetation (Chapter 4). Enhancing an ecohydrologic model to capture critical vegetation-climate interactions revealed new feedbacks suggesting future forests comprised of more drought tolerant species may have greater subsurface moisture and streamflow (Chapter 5). Overall, forested ecosystems across the US East Coast control storm-event flows due to shallow subsurface throughflow dominating runoff, but that these dynamic ecosystems may serve to lessen the hydrologic impacts of future disturbance like drought through management.
