Calibrating the data logger…
Ecohydrological processes in mountainous watersheds are complex due to steep elevation gradients, diverse soil types, and climatic influences on snowmelt and vegetation growth. This study uses a SIMTOP model to simulate runoff and hydrological partitioning within the NCAR Community Land Model CLM 2.0 to showcase an ecohydrological zonation in mountain regions. SIMTOP modifies classical TOPMODEL by (1) using an exponential function to estimate fractional saturation in steep slopes' topographic index distribution, and (2) modeling subsurface runoff as a water table depth function using minimal parameters to allow global use. Validation on both watershed (Sleepers River, USA) and global (UNH-GRDC) scales showed greater runoff simulation than baseline models. In order to capture ecohydrological behavior, 3-meter time-series Planet Scope NDVI images from the East River Watershed were analyzed using unsupervised machine learning, which can classify images without prior knowledge of patterns. Clustering identified regions characterized by distinct interactions between snowmelt and vegetation, which were greatly impacted by microtopographic features. In low-slope, high TWI zones, there was enhanced plant productivity followed by quick senescence at low moisture, especially in forb-dominated regions. In contrast, high slope, low TWI zones with sagebrush had diminished productivity. Furthermore, flood-prone basins within the Central Sudetes, Poland, were modeled using the TOPMODEL approach coupled with Monte Carlo calibration. Notably, the model exhibited considerable skill (NSE = 0.78) in about half of the catchments, highlighting its responsiveness to the duration of input data and catchment features. These findings in combination illustrate that ecohydrological zonation based on TOPMODEL and high-resolution remote sensing significantly enhances the understanding of vegetation–hydrology–topography interactions and helps to pinpoint ecologically sensitive regions in mountainous terrains.
Items connected by shared entities, co-authorship, citations, or semantic similarity.