Chong Chen Emmanuel Arthur Markus Tuller Hu Zhou Xiang Wang Jianying Shang Kelin Hu Tusheng Ren
The adsorbed water content is an attractive predictor for estimation of soil specific surface area (SSA) as its measurement is less laborious and more cost effective than standard laboratory techniques. We analysed the effects of total specific surface area (SSAtot), external specific surface area (SSAex) and internal specific surface area (SSAin) on water vapour sorption on 21 soil samples, and proposed models for estimating SSAex, SSAtot and SSAin from organic carbon content and the slope of soil water vapour sorption isotherm (SL0.5) and water content (WC0.5) at a water activity of 0.5. The results indicated that the variation of correlation coefficients between water content change at 0.05 water activity interval (WCC0.05) and SSAtot with water activity was mainly due to the differences in water vapour adsorption in interlayer spaces at different water activity levels. Furthermore, in the soil water content range from ~0.001 to ~0.03 g g−1 (within the water activity range from ~0.45 to ~0.75), water vapour adsorption was related closely to SSAex. Cross-validation results on estimated SSAex, SSAtot and SSAin of the 21 samples produced root mean square error (RMSE) values less than 9.39 m2 g−1 and Nash-Sutcliffe model efficiency coefficients (E) greater than 0.93, suggesting that the proposed models provided reasonable estimates of SSA components. Thus, soil water vapour adsorption data can be applied to simultaneously estimate SSAex, SSAtot and SSAin. Variation of correlation between WCC0.05 and SSAtot with water activity depended on SSAin. Soil water vapour sorption is related closely to SSAex at water activities from ~0.45 to ~0.75. The proposed models yield reasonable estimates of SSAtot, SSAex and SSAin. Adsorbed water content can be used as a predictor for SSA estimation. Highlights: Variation of correlation between WCC0.05 and SSAtot with water activity depended on SSAin. Soil water vapor sorption relates closely to SSAex at water activities from ~0.45 to ~0.75. The proposed models yield reasonable estimates of SSAtot, SSAex, and SSAin. Adsorbed water content can be used as a predictor for SSA estimation.
Keywords: external surface area, internal surface area, model development and evaluation, soil watervapour adsorption, total soil specific surface area