Estimating soil water characteristic curve using landscape features and soil thermal properties

Abstract

Soil water characteristic curve (SWCC) is a very important soil function with several environmental applications. Using pedotransfer functions as a quick and easy way to estimate the SWCC has been proposed to avoid its difficult and time-consuming measurements. Although numerous studies have been conducted on pedotransfer functions, the use of landscape features and soil thermal properties as predictors to estimate SWCC has received little attention. Therefore, in this study, the landscape data and soil thermal properties were used as predictors to estimate SWCC. We used 157 soil samples collected from an area within the central part of the U.S. Great Plains by Mohanty et al. (2002). By developing pedotransfer functions, SWCC was estimated using the landscape features and soil thermal properties in 21 pedotransfer functions (PTFs). Splitting the data, 100 and 57 soil samples were used for training and testing, respectively. The results showed that the landscape and geographic data did not provide an accurate estimate of the SWCC, while the soil thermal properties were very effective in the estimation of SWCC. Such a result may be due to the effect of soil textural and structural properties on both soil moisture and thermal properties. Also, comparing the results of the PTFs of this study and those of previously developed PTFs confirmed the superiority of the PTFs developed in this study.