Looking below the ground: analyzing the processes that drive spatiotemporal variation of wet channels in dynamic river networks using a physics-based hydrological model

Non-perennial streams, i.e., rivers that periodically cease to flow, are the focus of increasing research attention. Understanding how the spatiotemporal dynamics of runoff generation drives expansions and contractions of their active stream network is still challenging, due to the complex interplay among climate, topography, and geology. In this context, experimental data on the spatiotemporal variations of the wet channels of a river network are very valuable to study the joint variations of active stream length (L) and discharge at the catchment outlet (Q) and to analyze the processes driving such complex L-Q patterns. However, experimental data usuallydo not provide insights on what happens below the catchment surface; therefore, important insights can be gained by integrated surface-subsurface hydrological modeling (ISSHM), whereby the spatial configuration of the wet channels, the corresponding catchment discharge and the processes that drive the wetting and drying of different portions of the stream network can be simulated across the whole surface-subsurface continuum. In this study, we used CATHY (CATchment HYdrology) to simulate the stream network dynamics of two virtual catchments with the same, spatially homogeneous, subsurface characteristics (hydraulic conductivity, porosity, water retention curves, depth to bedrock) but different morphology (shape and slope).By running simulations under transient and steady-state conditions for different levels of antecedent catchment wetness, we investigated the role of topography, climate, and morphology on the resulting L-Q relation and on the processes that lead to the emergence of wet channels while comparing the numerical results with corresponding outcomes from simplified analytical formulations. Overall, we show that ISSHMs are useful tools to identify the main physical drivers of non-perennial streams, thanks to their capability of accurately describing the spatiotemporal variations of the storages and fluxes across the landscape, which eventually control network dynamics.