From field soil sampling to watershed model: Upscaling by integrating information entropy and interpolation method

Area of Science:

• Environmental Science
• Hydrology
• Soil Science

Background:

• Accurate soil property data is vital for watershed hydrology and non-point source (H/NPS) modeling.
• Current methods for soil sampling and data integration present challenges in improving model accuracy and cost-effectiveness.

Purpose of the Study:

• To optimize soil sampling points and properties for enhanced H/NPS modeling.
• To evaluate the impact of different parameterization schemes on H/NPS model performance.
• To develop a cost-effective framework for upscaling soil sampling information.

Main Methods:

• Information entropy and spatial interpolation were used to optimize soil sampling.
• Soil properties including bulk density, saturated hydraulic conductivity, and available water capacity were parameterized.
• The Soil and Water Assessment Tool (SWAT) was employed to test different parameterization schemes.

Main Results:

• Optimized sampling required more points for SOL_BD, SOL_K, and SOL_AWC.
• The new scheme improved Nash-Sutcliffe Efficiency (NSE) by 22.8% and R² by 10.5% compared to traditional databases.
• Entropy-based optimization reduced sampling points by 13.2%, offering a more cost-effective solution.
• Sampled properties significantly impacted NPS modeling, particularly nitrogen, more than hydrological simulations.

Conclusions:

• The proposed entropy-based framework effectively optimizes soil sampling for improved H/NPS modeling.
• This method provides a more accurate and cost-effective approach to watershed modeling.
• The framework is generalizable for upscaling field soil data to watershed scales, enhancing H/NPS simulations.