Flexible vector-based spatial configurations in land models

Summary

This paper presents an alternative vector-based approach to configuring land models that directly incorporates unique combinations of soil types, land cover, and hydrologically significant geographical features (elevation, slope, aspect), rather than using predetermined grid sizes. Using the Variable Infiltration Capacity (VIC) model, the authors demonstrate that flexible spatial configurations can achieve comparable accuracy to finer-resolution grids whilst capturing important differences in internal model states and fluxes across spatial scales. The findings suggest that adoption of flexible spatial configurations by the land modelling community could improve representation of water and energy fluxes at scales relevant to specific research objectives.

UK applicability

This methodological approach is applicable to UK hydrological modelling and catchment-scale water resource assessment, particularly where terrain heterogeneity and variable soil-vegetation combinations are important. The framework could enhance UK land-surface and hydrological models used in flood forecasting, drought prediction, and climate impact assessments.

Key measures

Basin-average streamflow; snow water equivalent (SWE); evapotranspiration (ET); model calibration accuracy; spatial variability of internal processes

Outcomes reported

The study implemented a vector-based spatial configuration approach in the Variable Infiltration Capacity (VIC) land model and demonstrated how different spatial configurations affect simulations of streamflow, snow water equivalent (SWE), and evapotranspiration (ET). Results showed that lower-complexity model configurations, once calibrated, can achieve similar accuracy to more complex configurations but may produce contradictory internal states and fluxes.

Topic tags