Revisiting the pseudo continuous pedotransfer function concept: Impact of data quality and data mining method

• Published in: Geoderma 2014-08, Vol.226-227, p.31-38
• Main Authors: Haghverdi, A., Öztürk, H.S., Cornelis, W.M.
• Format: Article
• Language: English
• Subjects: Agronomy. Soil science and plant productions; Biological and medical sciences; Density; Earth sciences; Earth, ocean, space; Exact sciences and technology; Fluid dynamics; Fluid flow; Fundamental and applied biological sciences. Psychology; Mathematical analysis; Neural network; Neural networks; Pseudo continuous PTF; Soil (material); Soils; Support vector machine; Support vector machines; Surficial geology; Training; Water retention curve
• ISSN: 0016-7061; 1872-6259
• DOI: 10.1016/j.geoderma.2014.02.026

The pedotransfer function (PTF) concept has been widely used in recent years as an indirect way to predict soil hydraulic properties, particularly the water retention curve (WRC). The pseudo continuous (PC) approach allows us to predict water content at any predefined matric head, resulting in an almost continuous WRC. When combined with powerful pattern recognition approaches, a PC-PTF can be trained to learn the shape of WRC from a discrete set of measured points, unlike traditional parametric PTFs which follow a predefined WRC shape dictated by the selected soil hydraulic equations. The purpose of this study was to investigate the impact of two elements on the performance of a PC-PTF: (i) data mining method (neural network, NN, versus support vector machine, SVM) and (ii) distribution and density of the provided water retention data in the training phase. Two datasets from Turkey and Belgium, consisting of mainly fine and coarse-textured soils, respectively, were employed. Multiple scenarios containing different combinations of measured water retention points in the training phase were defined. The lower root mean square error (RMSE) on average (0.044cm3cm−3) attained with the NN-based PTF shows that it is a better option than SVM (RMSE of 0.052cm3cm−3) for deriving PC-PTFs. The accuracy of PC-PTF was firmly dependent on the presence of measured water retention points in the entire range of WRC. Applying different scenarios revealed that a well distributed set of measured water retention points in the training phase could result in up to 0.03cm3cm−3 reduction in RMSE values. •The pseudo continuous PTF (PC-PTF) was utilized to predict water retention curve.•Support vector machine (SVM) and neural network (NN) were examined.•The impact of distribution and density of water retention data was investigated.•The NN worked better than SVM.•The PC-PTF was sensitive to the quality of the data in training phase.•The NN-PTF performed satisfactory with limited but well-distributed data.