EVALUATION AND IMPROVEMENT OF BUFFER-pH LIME REQUIREMENT METHODS

Publisher Website
Google Scholar
Abstract
In this experiment, we sought to determine the accuracy and useful range of six buffer-pH lime requirement (LR) tests, improve their accurate range, and select the best methods to determine the LR of acid mineral soils by routine analysis. Four single-buffer methods (Woodruff, Shoemaker et al. (SMP-SB), Adams and Evans, and Mehlich) and two double-buffer methods (Yuan and McLean et al. (SMP-DB)) were verified by comparing the estimated to the incubation LR, using correlation and regression techniques. Seventy acid mineral soils requiring from 1.4 to 40 milliequivalents CaCO3/100 grams to achieve pH 6.5 were used in this experiment.The proportion of soil acidity reacting with the LR buffers decreased as the LR of soils increased. Because this proportion is not constant (linear) as is generally assumed, but tends to be slightly exponential (curvilinear), the calibrations of these buffers were found to be more accurate for a particular range of soils. Accordingly, when used as proposed, the Woodruff, SMP-SB, and SMP-DB buffer methods predicted on the average 96, 130, and 102 percent of the actual requirements of soils requiring less than 10 meq CaCO3/100 g to attain pH 6.5, but predicted 69, 98, and 90 percent of the actual requirements for soils requiring between 10 and 40 meq CaCO3/100 g, respectively. The accuracy of these methods can be improved substantially by using two linear or a single curvilinear regression equation to correct for the exponential relationship between the proportion of soil acidity determined by such a buffer and the incubation LR. As these regression equations make the Woodruff and Shoemaker et al. single-buffer methods as accurate as the more involved Yuan and McLean et al. double-buffer methods, these single-buffer methods are recommended for routine LR determination. In this experiment, we sought to determine the accuracy and useful range of six buffer-pH lime requirement (LR) tests, improve their accurate range, and select the best methods to determine the LR of acid mineral soils by routine analysis. Four single-buffer methods (Woodruff, Shoemaker et al. (SMP-SB), Adams and Evans, and Mehlich) and two double-buffer methods (Yuan and McLean et al. (SMP-DB)) were verified by comparing the estimated to the incubation LR, using correlation and regression techniques. Seventy acid mineral soils requiring from 1.4 to 40 milliequivalents CaCO3/100 grams to achieve pH 6.5 were used in this experiment. The proportion of soil acidity reacting with the LR buffers decreased as the LR of soils increased. Because this proportion is not constant (linear) as is generally assumed, but tends to be slightly exponential (curvilinear), the calibrations of these buffers were found to be more accurate for a particular range of soils. Accordingly, when used as proposed, the Woodruff, SMP-SB, and SMP-DB buffer methods predicted on the average 96, 130, and 102 percent of the actual requirements of soils requiring less than 10 meq CaCO3/100 g to attain pH 6.5, but predicted 69, 98, and 90 percent of the actual requirements for soils requiring between 10 and 40 meq CaCO3/100 g, respectively. The accuracy of these methods can be improved substantially by using two linear or a single curvilinear regression equation to correct for the exponential relationship between the proportion of soil acidity determined by such a buffer and the incubation LR. As these regression equations make the Woodruff and Shoemaker et al. single-buffer methods as accurate as the more involved Yuan and McLean et al. double-buffer methods, these single-buffer methods are recommended for routine LR determination. © Williams & Wilkins 1981. All Rights Reserved.