Modelling soil attribute depth functions with equal-area quadratic smoothing splines

The objective of this paper is to test the ability of equal-area quadratic splines to predict soil depth functions based on bulk horizon data. In addi tion, the possibility of improving the prediction quality by the use of add itional samples from the top and/or bottom of soil profiles along with hori zon data is examined. The predictive performance of the splines is compared with that of exponential decay functions, and 1st and 2nd degree polynomia ls. In addition, the predictive quality of the conventional horizon data is examined. The measure of predictive performance used is the root mean squa re error values calculated from differences between the 'true' depth functi on and the fitted depth function. The 'true' depth functions were derived f rom the intensive sampling and laboratory analysis of soil profiles. Three soil profiles were sampled; a Red Podzolic Soil (Red Kurosol), Podzol. (Aer ic Podosol) and Krasnozem (Red Ferrosol). The soil attributes that were mea sured included; pH, electrical conductivity (EC), clay %, sand %, organic c arbon %, gravimetric water content at - 33 kPa and air dry. The results cle arly indicated the superiority of equal-area quadratric splines in predicti ng depth functions. Such splines depend on a parameter, lambda that control s goodness-of-fit vs, roughness. Their quality of fit varied with the lambd a value used and it was found that a lambda value of 0.1 was the best overa ll predictor of the depth functions. The results also showed that using add itional samples from the top and/or bottom of the soil profiles improved th e prediction quality of the spline functions. (C) 1999 Elsevier Science B.V . All rights reserved.