AbstractA physico‐empirical model proposed by L.M. Arya and J.F. Paris in 1981 for predicting the soil moisture characteristic curve using particle size and bulk density data was compared with a commonly used laboratory method using undisturbed core samples and a pressure plate extractor, for precision in estimating in situ water retention measured with paired neutron probe and tensiometer installations. An empirical factor for the Arya and Paris (A&P) model was found to average close to the value (α = 1.38) proposed by its authors. Values for this experiment were different between texture classes. Loam and silt loam values were constant, while values for soils of sand, loamy sand, and sandy loam texture were nonconstant. The empirical factor was calibrated for (i) loam and coarse loam; (ii) silt loam; and (iii) sand, loamy sand, and sandy loam texture groups. The sandy soil values required linear calibration as a function of soil water potential. Results for the calibrated A&P model compared well, overall, with laboratory determinations on soils ranging in texture from sand to loam and silt loam. Laboratory estimates were slightly better for A and B horizon materials, while A&P estimates were better for C horizons. Error for estimation of field volumetric water content using A&P and lab methods was approximately twice the error for estimation of θ using a second field neutron‐tensiometer installation within one meter of the first field apparatus.
Read full abstract