Hydrologic Process Synthesis across Diverse Landscapes: Towards a hierarchical classification of North American hydrology

Hydrologic processes can be notoriously place-specific and much of our understanding about these processes originates from intensively monitored but small research basins.  This project is motivated by a need to connect this small-scale hydrologic process understanding with large-scale model applications, to strengthen the theoretical underpinnings of the models used for water resources planning and prediction of water-related risks across large geographical domains. Here we describe a community-driven synthesis effort that convened multiple virtual workshops, in-person community engagements, and online interactions that bring together water science experts working in various regions across North America. The community expertise was used to develop a hierarchical division of the North American continent into distinct hydrologic domains and provinces, and to describe the dominant hydrologic processes of each landscape.

At the highest level, we recognize five distinct domains: (1) the east, characterized by complex surface-groundwater interaction; (2) the west, with complex topography and resulting climate patterns as a dominant feature; (3) the central domain covering the prairies and plains across landscapes with extensive agriculture; (4) the north, primarily characterized by complex cold-region processes; and (5) the tropical islands, where large gradients in hydrologic drivers occur over relatively short distances. At the second level, we divided the domains into 35 hydrologic provinces. We then developed perceptual models of the hydrologic behaviour of each province using a combination of expert knowledge, literature reviews and data-based quantification of hydrologically relevant landscape characteristics. We envision further development of a third level in the classification that includes progressively more local detail. In parallel, the current perceptual models can be mapped onto computational models, modules and individual equations to support a theory-based large-domain effort to develop appropriate hydrologic models for any location in the wider North American continent. The procedures used in this work are general and could be applied to any geographical domain where expert knowledge of local conditions is available.