U.S. Stream Network Expands in Annual High-Flow Periods

Rivers and streams may look permanent, but their lengths can change dramatically with the seasons, according to a new study. It reports that stream networks in the United States expand up to five times their low-flow size during wet conditions. The findings offer the first large-scale insights into how water dynamically moves through landscapes and provide a framework for forecasting climate-driven changes in stream networks, particularly in response to increasing storminess. Traditionally regarded as static landscape features, stream networks exhibit significant seasonal and storm-driven fluctuations in their total wetted length. For example, seasonal rainfall and snowmelt saturate the landscape, temporarily increasing streamflow and extending stream networks. Additionally, changes in surface topography and permeability can cause networks to expand or contract. This responsiveness of stream length to shifts in landscape wetness is termed "network elasticity." Alongside variations in wetness, it governs fluctuations in stream extent across drainage basins that influence sediment transport, nutrient cycling, gas exchange, and aquatic habitats. However, understanding stream network length variability has generally been limited to small-scale studies relying on field measurements – and to a handful of small drainage basins. Here, Jeff Prancevic and colleagues used a semimechanistic model to estimate stream network elasticity for 14,765 basins across the contiguous U.S. Prancevic et al. found that the median stream network is five times longer during annual high-flow conditions than during low-flow conditions, with regional differences influenced by hydroclimate and topographic sensitivity. Mountainous and humid regions tend to have more stable stream networks. The findings suggest that a basin's topography and subsurface structure, which determine its network elasticity, are as crucial as climate factors in predicting fluctuations in stream length. Although the study focuses on the continental U.S., the authors note that the methodology, which only requires digital elevation models and streamflow data, could be applied more broadly.