Improving the representation of the non-contributing area dynamics in land surface models for better simulation of prairie hydrology

2021 
Abstract The hydrology of the Canadian prairie region is complicated by the existence of numerous land depressions that change the contributing area dynamically and result in a non-linear and hysteretic relationship between contributing area and storage. Depressions are represented conceptually in most hydrologic models as simple storage (bucket) units. These conceptual approaches are simplified and might not adequately represent the dynamics of the depressions and the changing non-contributing area either temporally or spatially, and therefore, the simulation of streamflow remains challenging. This study advances a more physically based simulation of the hydrology, streamflow, and spatiotemporal pluvial/nival flooding extents and the associated non-contributing area in the prairies. This is achieved by coupling the MESH hydrology-land surface model with a newly developed surface routing component designed to explicitly deal with the prairie-pothole system (PRIMA) and is referred to as MESH-PRIMA. In this model, MESH handles the classical vertical water and energy balance calculations while PRIMA routes the water over the landscape and quantifies the depressional storage and runoff. The streamflow simulation of MESH-PRIMA is compared against that of MESH with its current conceptual prairie algorithm (MESH-PDMROF) on the Smith Creek Research Basin in Saskatchewan, Canada. MESH-PRIMA shows an improved streamflow and flood simulation compared to MESH-PDMROF and can replicate the non-linear and hysteretic relationship of the basin response. MESH-PRIMA allows for mapping the spatial distribution of water (pluvial/nival flooding) and the non-contributing area over landscape for different events. The results of MESH-PRIMA can help in updating the non-contributing area map and in identifying pluvial/nival flooding hazard, which is useful in flooding contexts.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    48
    References
    0
    Citations
    NaN
    KQI
    []
    Baidu
    map