Models for Predicting Composition and Temporal Structure of Chemical Parameters for Leachates from Amended Acid Soils

Abstract

The composition of drainage water from an Ultisol amended with limestone, sugar foam waste, dolomite, and gypsum was studied by using the lysimetric technique. The typical rainfall regime for the area from which the soil samples were obtained was simulated in the laboratory for seven weeks. The behavior of the samples was found to vary markedly with the type of amendment applied to the soil. The structure of the time series for four of the 13 chemical parameters studied in the tests was analyzed and discriminated by using various statistic tests. Multiple linear regressions models of the autoregressive and exponential smoothing types were constructed for pH, EC, , and , and those providing the best fits selected. Finally, concentration plots for each element were constructed as a function of the concentration factor for the individual parameter that proved the best marker and was most strongly affected by the concentration–dilution effect of the electrolyte.

Keywords:

• Soil
• Acidity
• Liming
• Sugar foam waste
• Leachates
• Cross-correlation function
• Exponential smoothing
• Autoregressive model
• Concentration plots

Acknowledgment

This work was funded by the Junta Castilla y León (project no. LE 04/01).