A simplified modelling approach for quantifying tillage effects on soil carbon stocks

Abstract

SummarySoil tillage has been shown to affect long‐term changes in soil organic carbon (SOC) content in a number of field experiments. This paper presents a simplified approach for including effects of tillage in models of soil C turnover in the tilled‐soil layer. We used an existing soil organic matter (SOM) model (CN‐SIM) with standard SOC data for a homogeneous tilled layer from four long‐term field experiments with conventionally tilled (CT) and no‐till (NT) treatments. The SOM model was tested on data from long‐term (>10 years) field trials differing in climatic conditions, soil properties, residue management and crop rotations in Australia, Brazil, the USA and Switzerland. The C input for the treatments was estimated using data on crop rotation and residue management. The SOM model was applied for both CT and NT trials without recalibration, but incorporated a ‘tillage factor’ (TF) to scale all decomposition and maintenance parameters in the model. An initial value of TF = 0.57 (parameter uncertainty, PU = 0.15) for NT (with TF set to 1.0 for CT) was used on the basis of a previous study with observations of soil CO2 respiration. The simulated and observed changes in SOC were then compared using slopes of linear regressions of SOC changes over time. Results showed that the SOM model captured observed changes in SOC content from differences in rotations, N application and crop residue management for conventional tillage. On the basis of SOC change data a mean TF of 0.48 (standard deviation, SD = 0.12) was estimated for NT. The results indicate that (i) the estimated uncertainty of tillage effects on SOC turnover may be smaller than previously thought and (ii) simple scaling of SOM model parameters may be sufficient to capture the effects of soil tillage on SOM turnover in the tilled layer. Scenario analyses showed that the average extra C input needed to compensate for soil tillage was 762 (SD = 351) kg C ha−1 year−1. Climatic conditions (temperature and precipitation) also affected how much extra C was needed, with substantially larger inputs being required for wetter and warmer climates.