Partitioning of water and energy budgets at the land surface is controlled by complex interactions between climate, soil, geology and vegetation types. Integrated groundwater-surface water-land surface models that simulate the hydrologic cycle behaviour between the subsurface and atmosphere are valuable tools to characterize catchment functions and identify how feedback processes between the land and atmosphere impact catchment behaviour. Despite the complex structure of an integrated hydrologic model, model predictions are still impacted by conceptual model design, model parameters, and uncertainty in model states, such as the initial soil moisture and groundwater storage. In this study, we used an integrated groundwater–surface water–land surface model (ParFlow.CLM) to explore the impact
of model state uncertainty on simulated land surface fluxes across a 208 km2 catchment in western Denmark. Using one year of forcing data, long term recursive simulations was performed until the model equilibrated using multiple spin up measures on various model state and diagnostic variables. Our analysis illustrates that it takes longer for the subsurface storage and discharge to equilibrate based on various spin up measures, and improper initialization can have large impacts on a catchment’s water balance. Further, we evaluated the impact of model state initialization on subsequent simulations under various equilibrium scenarios (thermal, unsaturated zone and saturated zone equilibrium) and climatic conditions such as wet versus dry periods. Our results reveal sensitivity to initialization
conditions and highlight the role of equilibrium state on surface water-groundwater exchange and water balance partitioning at the catchment scale.
This page is maintaind by Jason Evans |
Last updated 31st January 2013