Abstract

Determining soil water retention curve by field methods is expensive, laborious and time consuming. Several indirect methods have been developed to circumvent these setbacks. However, tropical soils have hardly benefited from these innovations. In this work, pedo-transfer functions (PTFs) and fractal analysis were used to predict the soil water characteristic curve of a Rhodic Ferralsol, collected from the three western provinces of Cuba. Experimental and predicted soil water content at different suction heads were correlated for each model. The accuracy of the predictions was evaluated by the mean difference (MD) and the root mean squared difference (RMSD). With the exception of the Tomasella et al.'s [Soil Sci. Soc. Am. J. 64 (2000) 327] PTF, the results revealed poor agreement between predicted and measured values for the other functions. The one-way ANOVA, as well as Duncan's multiple range tests, conducted on the average mean difference (〈MD〉) and the average root mean squared difference (〈RMSD〉) showed significant differences intermodels. Variations intermodels accounted for 91% of the total variation in the case of 〈MD〉, while for 〈RMSD〉, the contribution was 85% to the total variation.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call