Modelling long-term impacts of fertilization and liming on soil acidification at Rothamsted experimental station

Abstract

Liming is widely used to reduce the impacts of soil acidification and optimize soil pH for agricultural production. Whether models can simulate the effect of liming on soil pH, and base saturation (BS), and thereby guide lime application, is still largely unknown. Long-term experimental data from a grassland (Park Grass, 1965–2012) and arable land (Sawyers Field, 1962–1972) at Rothamsted Research, UK, were thus used to assess the ability of the VSD+ model to simulate the effects of long-term fertilization and liming on soil acidification. The VSD+ model was capable of simulating observed soil pH and BS changes over time in the long-term liming experiments, except for a treatment in which sulphur (S) was added. Normalized Mean Absolute Errors (NMAE) and Normalized Root Mean Square Errors (NRMSE) of simulated and observed pH values, averaged over the observation periods varied between 0.02 and 0.08 (NMAE) and 0.01–0.05 (NRMSE). The acidity budget results for Park Grass suggest that nitrogen (N) transformations contributed most to acidity production, causing predominantly aluminium (Al) exchange in the topsoil (0–23 cm) followed by base cation (BC) release, but in the treatment with S addition, BC uptake had a nearly similar effect on acidity production. However, in Sawyers Field, the acidity budget suggested that BC uptake was the dominant cause of soil acidification, while the impacts of N transformations were limited. Liming was found to sufficiently replenish BC and decrease Al exchange in the topsoil layer. Overall, the VSD+ model can adequately reconstruct the impacts of fertilizer and liming applications on acid neutralizing processes and related soil pH and BC changes at the soil exchange complex.