We used the process-based and tracer-aided ecohydrological model EcH2O-iso to assess the effects of vegetation cover on water balance partitioning and associated flux ages under temperate beech forest (F) and grassland (G) in Northern Germany. The model was tuned on the basis of a multi-criteria calibration against an unusually rich measured data set from a long-term monitoring site.. The calibration incorporates metrics of the energy balance, hydrological function and biomass accumulation. It resulted in good efficiency statistics for simulations of surface energy exchange, soil water content, transpiration and biomass production. The model simulations showed that the forest “used” more water than the grassland; from 620mm of average annual precipitation, losses were higher through interception (29% under F, 16% for G) and combined soil evaporation and transpiration (59% F, 47% G). As a result, groundwater recharge was greatly enhanced under grassland at 37% of precipitation compared with12% for forest. The model allowed us to track the ages of water in the different storage compartments and fluxes.In the shallow soil horizons, the average ages of soil water fluxes and evaporation were similar in both plots (∼1.5month), though transpiration and groundwater recharge were older under forest (∼6 months compared with∼3months for transpiration and∼12 months compared with∼10 months for groundwater). Flux tracking with Cl tracers provided independent support for the modelling results, though also highlighted effects of uncertainties in forest partitioning of evaporation and transpiration. This underlines the potential for tracer aided ecohydrological models in land use change studies. By tracking storage – flux – age interactions under different land covers, the effects on water partitioning and age distributions can be quantified and the implications for climate change assessed.Better conceptualisation of soil water mixing processes, and improved calibration data on leaf area index and root distribution appear obvious respective modelling and data needs for improved model results.

We combine physics-based groundwater reactive transport modeling with machine learning techniques to quantify hydrogeologic model and solute transport predictive uncertainties. We train an artificial neural network (ANN) on a dataset of groundwater hydraulic heads and 3H concentrations generated using a high-fidelity groundwater reactive transport model. Using the trained ANN as a surrogate model to reproduce the input-output response of the high-fidelity reactive transport model, we quantify the posterior distributions of hydrogeologic parameters and hydraulic forcing conditions using Markov-chain Monte Carlo (MCMC) calibration against field observations of groundwater hydraulic heads and 3H concentrations. We demonstrate the methodology with a model application that predicts Chlorofluorocarbon-12 (CFC-12) solute transport at a contaminated site in Wyoming, USA. Our results show that including 3H observations in the calibration dataset reduced the uncertainty in the estimated permeability field and infiltration rates, compared to calibration against hydraulic heads alone. However, predictive uncertainty quantification shows that CFC-12 transport predictions conditioned to the parameter posterior distributions cannot reproduce the field measurements. We found that calibrating the model to hydraulic head and 3H observations results in groundwater mean ages that are too large to explain the observed CFC-12 concentrations. The coupling of the physics-based reactive transport model with the machine learning surrogate model allows us to efficiently quantify model parameter and predictive uncertainties, which is typically computationally intractable using reactive transport models alone.

Aaron Smith1, Doerthe Tetzlaff1,2,3, Marco Maneta4,Chris Soulsby3,21IGB Leibniz Institute of Freshwater Ecology and Inland Fisheries Berlin, Berlin, Germany2Humboldt University Berlin, Berlin, Germany3Northern Rivers Institute, School of Geosciences, University of Aberdeen, UK4Department of Geosciences, University of Montana, Missoula, Montana, USACorrespondence to: Aaron Smith ([email protected])