Optimization of carbon stock models to local conditions using farmers' soil tests: A case study with AMGv2 for a cereal plain in central France

Abstract

AbstractSoil organic carbon (SOC) is associated with environmental benefits and crop productivity; therefore, monitoring of SOC stocks is important when assessing the sustainability of agricultural systems. This study applied a hybrid method and the results of on‐farm soil tests to calibrate the AMGv2 model and adapt it to local conditions. The method was applied to soils in the Limagne plain (France), which shifted to cereal production since the 1960s. Based on analysing 988 soil test data, covering the period from 1954 to 2019, it was found that SOC stocks showed a significant decrease in the three main soil types of the study area. An optimization procedure estimated that the initial ratios of stable to total carbon fall in the ranges 0.42 and 0.46 for vertisols, 0.48 and 0.52 for calcisols, and 0.56 and 0.60 for fluvisols. Simulations using these values estimated that SOC stocks declined between 1960 and 2018 by between −31 and −17%, depending on soil type. The optimized model was used to forecast the evolution of SOC stocks up to 2050. These simulations showed a further decline in SOC stocks with continuation of current practices, even assuming a 15% increase in crop yields. They indicated that stopping straw exports would stabilize stocks, while a systematic introduction of cover crops would increase stocks about 3.8% over the period considered. It is concluded that this hybrid procedure can improve the adaptation of predictive models to local conditions.