Using Physiography to Understand Stream Network Expansion and Contraction Across Spatiotemporal Scales

Non-perennial streams (i.e., streams that cease flowing regularly across time or space) comprise ~60% of the global river network and play an important role in the physical, chemical, and biological functions of downstream waters. However, predicting their patterns of network expansion and contraction remains a key challenge across regulatory, practitioner, and research communities, especially given that most investigations focus on high-relief watersheds. To address this challenge, we employed physiography as a lens to investigate the impacts of geology, soil characteristics, topography, and vegetation on spatial and temporal patterns of stream wetting and drying. We instrumented three headwater stream networks located in the Coastal Plain, Piedmont, and Appalachian Plateaus physiographic provinces in the southeastern United States. In each network, we used ≥ 20 water presence/absence sensors over two water years (2023 and 2024) to investigate seasonal and interannual variability in network extent. Across the physiographic gradient, we found that a combination of topographic, geologic, and vegetative drivers best explained variability in stream network persistence. Our results also emphasized the role that sensor placement plays in understanding network-scale patterns, as deploying sensors in areas of greatest hydrologic variability was crucial to capturing the full range of network expansion and contraction. This study demonstrated that low-relief stream networks challenge contemporary perceptual models of network dynamics, and that consideration of other factors such as soils and vegetation can help explain network expansion and contraction in low-relief headwater stream networks.