The Limits of Homogenization: What Hydrological Dynamics can a Simple Model Represent at the Catchment Scale?

Abstract

Large‐scale models often use a single grid to represent an entire catchment assuming homogeneity; the impacts of such an assumption on simulating evapotranspiration (ET) and streamflow remain poorly understood. Here, we compare hydrological dynamics at Shale Hills (PA, USA) using a complex model (spatially explicit, >500 grids) and a simple model (spatially implicit, two grids using “effective” parameters). We asked two questions: What hydrological dynamics can a simple model reproduce at the catchment scale? What processes does it miss by ignoring spatial details? Results show the simple model can reproduce annual runoff ratios and ET, daily discharge peaks (e.g., storms, floods) but not discharge minima (e.g., droughts) under dry conditions. Neither can it reproduce different streamflow from the two sides of the catchment with distinct land surface characteristics. The similar annual runoff ratios between the two models indicate spatial details are not as important as climate in reproducing annual scale ET and discharge partitioning. Most of the calibrated parameters in the simple model are within the ranges in the complex model, except that effective porosity has to be reduced to 40% of the average porosity from the complex model. The form of the storage‐discharge relationship is similar. The effective porosity in the simple model however represents the dynamic and mobile water storage in the effective drainage area of the complex model that connects to the stream and contributes to high streamflow; it does not represent the passive, immobile water storage in the often disconnected uphill areas. This indicates that an additional uphill functioning unit is needed in the simple model to simulate the full spectrum of high‐low streamflow dynamics in natural catchments.

Streams as Mirrors: Reading Subsurface Water Chemistry From Stream Chemistry

“…Such analyses underscore the importance of horizontal, or landscape, connectivity in shaping export patterns under certain conditions. Given the multiple dimensions (e.g., vertical, horizontal) of flow in catchments, it is possible that different flow paths activated under varying connectivity conditions influence solute export patterns to varying degrees (Wen et al., 2021; Xiao et al., 2019), potentially leading to non‐monotonic CQ patterns. To investigate complex CQ behavior, some authors have suggested splitting CQ plots at threshold discharge values and determining separate CQ patterns and underlying mechanisms at different flow regimes (Moatar et al., 2017; Underwood et al., 2017).…”

Climate Controls on River Chemistry

“…The model can be used broadly as an organizing framework to understand competing processes that regulate mean solute concentrations. More complex reactive transport models at the watershed scale can be used to complement such simple models and to examine detailed processes, variables, and spatial heterogeneities that reflect idiosyncrasies of specific sites (Li, 2019; Wen et al., 2021; Wen et al., 2020).…”

Understanding the Hydrogeochemical Response of a Mountainous Watershed Using Integrated Surface‐Subsurface Flow and Reactive Transport Modeling

“…This indicates that the 30-m depth of the subsurface domain may be insufficient to capture all groundwater baseflow. Our model consistently overestimates the baseflow, which is likely due to the limit of spatial homogenization for hydrological dynamics discussed in Wen et al (2021). Further, the model assumes that groundwater in the subsurface does not freeze, which leads to an overestimation of groundwater discharge in winter.…”