Physico-empirical methods for estimating soil water characteristic curve under different particle size

4th International Conference on Water Resource and Environment

地球科学;生态环境科学

Zhao, Y.Q.^1,2,3 ; Wang, W.K.^1,2 ; Wang, Z.F.^1,2 ; Li, L Chen^1,2

School Environment Science and Engineering, Chang'An University, Xian

710054, China^1

Key Laboratory of Subsurface Hydrology and Ecological Effect, Xian

710054, China^2

Groundwater Monitoring Station, Xianyang

712000, China^3

关键词: Coefficient of determination;    Different particle sizes;    Prediction precision;    Root mean square errors;    Soil particle size;    Soil-water characteristic curve;    Spherical particle;    Water characteristics;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/191/1/012018/pdf
DOI  :  10.1088/1755-1315/191/1/012018

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Soil water characteristic curve (SWCC) is an important property of soil, but directly measuring methods are costly and time-consuming. Many attempts have been made to predict the water characteristic curve indirectly from soil particle size distribution (PSD) and basic physical properties. In this study, the Arya-Paris (A-P) model and Non-similar midia concept (NSMC) model were applied to predict the SWCC indirectly with six soil samples in arid regions. The objective of this study was to compare the two physico-empirical methods and assess the applicability of them under different particle size of soil. The results showed that the A-P model was superior to NSMC model on prediction precision, and more suited to predict silty sand and silt, those soils with medium texture. While using Arya-Paris method, the scaling parameter, α (parameter used to scale pore lengths based on spherical particles to natural soil pore lengths), was estimated by non-linear fitting, linear fitting and as a constant. It was found that the A-P model with non-linear α that used improved logistic growth equation was the best method to estimate SWCC, which had a maximum coefficient of determination (R2) and minimum root mean square error (RMSE).