Comparison of laboratory methods for predicting soil lime requirements of Luvisol in Dera district, the Northwestern highlands of Ethiopia

Abstract

This study addresses the growing concern over acidification of agricultural soils in the Northwest Ethiopia highlands. It aims to identify the most effective lime requirement (LR) assessment methods and calibrate them for estimating LR of specific acid soils in Ethiopia. The methods compared were the Shoemaker-McLean-Pratt buffer method (SMP), Woodruff method, modified Adams-Evans buffer method (MAEB), buffering capacity (BC), and exchangeable acidity (EA). Soil samples were taken from ten acid-prone soils of the Dera district at a depth of 0 - 30 cm. Treatments ranged from 0 - 200% of the expected lime (CaCO3) rate for each approach; a randomised complete block design with three replications was used.
Results indicated that LR calculated by MAEB, Woodruff, and SMP was too high, raising soil pH above the target of 6.0. Conversely, the LR determined by BC was too low, followed by EA, to achieve the expected pH of 6. Average LR for the soils was 2.12, 2.34, 8.01, 9.38, and 13.27 t/ha for EA, BC, MAEB, SMP, and Woodruff methods, respectively. The BC technique was found to better predict the LR for a target pH of 6.
Principal component analysis revealed that three predictors - initial pH, organic matter (OM), and exchangeable aluminum - accounted for 97.6% of the LR variation. Initial pH, OM, and exchangeable acidity were strong predictors of LR, while clay content was not a significant predictor and was removed from the model. Regular updates on LR determination techniques are recommended.