Effects of model spatial scale on soil water movement and ecosystem fluxes

Abstract:

Data availability through the advances in remote sensing technology along with increase in computation capability are making it possible to model the land surface at increasingly high resolutions over large areal extents. This should allow us to overcome the existing tradeoff that current land surface process models face in choosing between resolution and extent. Large scale models can enable continental to global simulations, but only at coarse resolution. Hyper-resolution models can resolve landscape heterogeneity and processes at the meter scale but can only practically achieve local or regional coverage. In addition, land surface processes vary in scale and may not match that of the model. Therefore, the effects of model spatial resolution on simulation results are important to understand in land surface modeling.

Richards Equation, frequently used to model unsaturated subsurface flow, has a characteristic length scale on the order of meters based on the physics of Buckingham-Darcy flux. However, subsurface processes modeled using Richards Equation are often applied at much larger spatial scales. To understand the effects of model scale, we develop MLCan3D, a high resolution, topography aware, ecohydrologic model that includes a multilayer canopy model (MLCan) coupled with 2D overland flow and 3D subsurface flow. Using data collected by NEON (National Ecological Observatory Network), we apply our model to several NEON sites. We examine how changes in model scale affect soil water processes, such as infiltration-runoff partitioning and lateral subsurface moisture distribution, as well as ecosystem level energy and water fluxes by modeling at increasing grid scales from 1 meter to 500 meters. By comparing field measured data and simulation results at multiple scales for different ecosystems, we identify which processes are affected by model resolution and how they change with scale.

Through this work, we aim to better understand the effects of spatial scale on surface-subsurface soil water processes and associated ecosystem responses. The results of this study can contribute to the study of the interaction of model spatial resolution with process scales and heterogeneity and help in bridging the gap between high-resolution high-fidelity land surface models and larger scale models.