Vegetation growth depends on climate and land-use conditions; this directly affects groundwater recharge, because plants compete with shallow aquifers for water through evapotranspiration. Most recharge models use greatly simplified representations of vegetation that do not capture feedbacks between vegetation growth and soil moisture - they typically assume plants to be static and neglect their impact on the timing of water table impacts. We address this gap in the state of Minnesota by developing an ecohydrological model that can produce spatiotemporal estimates of recharge that incorporate the effects of vegetation dynamics on seasonal to interannual time scales.
Minnesota serves as an ideal test-bed for evaluating how dynamic vegetation growth impacts recharge over a range of conditions: the state contains: (1) a sharp east-west precipitation gradient, (2) diverse ecosystems (including grassland, evergreen forests, mixed forests, and cropland), and (3) extensive hydrogeologic observations for model calibration.
To evaluate how changing vegetation affects our groundwater resources, we are implementing NCAR's land-surface model with the prognostic vegetation module (CLM4.5-BGC). Such ecohydrological models require a many hydrogeologic and vegetation input parameters that are highly uncertain, and so a major part of this work is developing a data assimilation framework for calibrating the model to the extensive subsurface and land cover data available throughout the state.
PhD student Harsh Anurag has compiled statewide input data, set up the implementation of CLM4.5 over Minnesota, and is developing ensemble Kalman filter (EnKF) data assimilation framework for calibrating simulations. Through preliminary sensitivity tests, he has demonstrated that taking into account vegetation feedbacks impacts recharge predictions, confirming the importance of incorporating vegetation dynamics into groundwater evaluations.
(Funding: Legislative-Citizen Commission on Minnesota Resources (LCCMR))