Quantifying connectivity patterns in dryland ecosystems is crucial for understanding how water and sediments flow across the landscape, crucial for mitigating the impacts of climate change and land use modifications. This study quantifies the multi-scale water-mediated connectivity within grassland and shrubland hillslopes using a weighted, directed network model. By integrating high-resolution elevation, vegetation data, and modeled event-based hydrologic and sediment transport, we quantify both structural (topology) and functional (dynamical) connectivity under varying rainfall and soil moisture conditions. Our findings reveal a marked increase in local connectivity metrics in shrublands compared to grasslands, with metrics such as betweenness centrality and the weighted length of connected pathways increasing up to tenfold. Despite these substantial local changes, global properties like link density and global efficiency show less drastic variations, suggesting a robust network topology that sustains geomorphic functionality across different vegetation states. This indicates that although local connectivity is highly sensitive to vegetation changes, the overall structure and functional connectivity of water and sediment at the global scale remain relatively stable. Functional connectivity is more strongly correlated with structural connectivity for sediment than for water. This difference is particularly pronounced under high rainfall conditions, yet shows little sensitivity to variations in antecedent soil moisture, highlighting the critical role of event-driven processes in shaping connectivity patterns. The study advances our quantitative understanding of how structural changes affect functional processes in dryland ecosystems. It offers a comprehensive framework for analyzing connectivity at multiple scales, which can inform targeted management strategies aimed at enhancing ecosystem resilience.

This study provides a new perspective on understanding the intricacies of water-mediated connectivity in ecosystems, bridging landscape ecology and geomorphology through network science. We highlight dryland and river floodplain ecosystems as distinct examples of contrasting water-controlled systems. We (1) discuss central considerations in developing Structural Connectivity (SC) and Functional Connectivity (FC) networks of water-mediated connectivity; (2) quantify the emergent patterns in these networks; and (3) evaluate the capacity of network science tools for investigating connectivity characteristics. Connectivity is quantified using seven parameters at both network and node levels. We find that Link Density, Betweenness Centrality, and Page Rank Centrality are highly sensitive to directionality; Global Efficiency and Degree Centrality are particularly sensitive to weights; while Relative Node Efficiency remains unaffected by weights and directions. Our study underscores the potential to transform how we quantify and understand water-mediated connectivity, especially in consideration of the role(s) of weights and directionality. This interdisciplinary review has implications for both theoretical insights and practical applications in environmental management and conservation efforts.