The solution of many field‐scale flow and transport problems requires estimates of unsaturated soil hydraulic properties. The objective of this study was to calibrate neural network models for prediction of water retention parameters and saturated hydraulic conductivity, Ks, from basic soil properties. Twelve neural network models were developed to predict water retention parameters using a data set of 1209 samples containing sand, silt, and clay contents, bulk density, porosity, gravel content, and soil horizon as well as water retention data. A subset of 620 samples was used to develop 19 neural network models to predict Ks. Prediction of water retention parameters and Ks generally improved if more input data were used. In a more detailed investigation, four models with the following levels of input data were selected: (i) soil textural class, (ii) sand, silt, and clay contents, (iii) sand, silt, and clay contents and bulk density, and (iv) the previous variables and water content at a pressure head of 33 kPa. For water retention, the root mean square residuals decreased from 0.107 for the first to 0.060 m3 m‐3 for the fourth model while the root mean square residual Ks decreased from 0.627 to 0.451 log(cm d‐1). The neural network models performed better on our data set than four published pedotransfer functions for water retention (by ≈0.01–0.05 m3 m‐3) and better than six published functions for Ks (by ≈0.1–0.9 order of magnitude). Use of the developed hierarchical neural network models is attractive because of improved accuracy and because it permits a considerable degree of flexibility toward available input data.
In many vadose zone hydrological studies, it is imperative that the al., 1998). Far fewer alternatives exist for unsaturated soil's unsaturated hydraulic conductivity is known. Frequently, the Mualem-van Genuchten model (MVG) is used for this purpose be-hydraulic conductivity. Although some pedotransfer cause it allows prediction of unsaturated hydraulic conductivity from functions are available (Saxton et al., 1986; Schuh and water retention parameters. For this and similar equations, it is often Bauder, 1986; Vereecken et al., 1990), pore-size distriassumed that a measured saturated hydraulic conductivity (K s) can bution models by Burdine (1953) and Mualem (1976), be used as a matching point (K o) while a factor S L e is used to account among others, are more popular. for pore connectivity and tortuosity (where S e is the relative saturation Generally speaking, the Burdine and Mualem models and L ϭ 0.5). We used a data set of 235 soil samples with retention infer the pore-size distribution of a soil from its water and unsaturated hydraulic conductivity data to test and improve preretention characteristic. By making assumptions about dictions with the MVG equation. The standard practice of using K o ϭ continuity and connectivity of pores, integral expres-K s and L ϭ 0.5 resulted in a root mean square error for log(K) sions can be derived that describe unsaturated conduc-(RMSE K) of 1.31. Optimization of the matching point (K o) and L to the hydraulic conductivity data yielded a RMSE K of 0.41. The fitted tivity in terms of water content or pressure head. A K o were, on average, about one order of magnitude smaller than general expression can be given as (after Hoffmannmeasured K s. Furthermore, L was predominantly negative, casting Riem et al., 1999): doubt that the MVG can be interpreted in a physical way. Spearman rank correlations showed that both K o and L were related to van
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.