Response of soil carbon dioxide emissions to no-tillage and moldboard plow systems on Andosols in a humid, subtropical climate, Japan

Abstract

Characterization of soil carbon dioxide (CO2) emissions can help us to understand the effects of agricultural management on climate change and the ability of soil to sequester atmospheric carbon. However, there is a lack of studies on the effects of tillage and cover crops on soil CO2 emissions on Andosols in the humid, subtropical climate. The present study aimed to quantify soil CO2 emissions and identify the factors accounting for CO2 emissions among no-tillage (NT) and moldboard plow (MP) systems under different cover crop treatments at an experimental site established in 2002 in Kanto, Japan. A split-plot field experiment was conducted from May 2017 to November 2018 comprising two tillage systems (NT and MP) and three cover crops: fallow, hairy vetch (HV), and rye (RY). Higher soil CO2 daily emissions at 80% of the measurements and significantly greater CO2 seasonal and annual emissions were observed in the NT system as compared to the MP system. The relationship between CO2 daily flux and soil temperature was well-described by exponential functions (p < 0.001). Soil volumetric water content (θv) was significantly positively associated with soil CO2 daily emissions when soil temperature ranged from 10 °C to 20 °C (p < 0.05) and 20 °C to 30 °C (p < 0.01), and the response of soil CO2 flux to soil temperature and θv was well-described by a multiplicative model (RMSE: 1.2920, R2: 0.7777). Although there was no difference in soil temperature between NT and MP, significantly higher soil θv was found in NT than in MP. In addition, significant exponential correlations were observed between soil organic carbon (SOC) stocks and annual CO2 emissions in four different soil layers (0–2.5 cm, 0–7.5 cm, 0–15 cm, and 0–30 cm). The HV and RY cover crops had a positive effect on SOC stocks. NT increased SOC stocks significantly compared to MP at 0–2.5 cm (p < 0.001) and 0–7.5 cm (p < 0.05) depth while no significant SOC stock difference at 0–15 cm and 0–30 cm depth between NT and MP were observed. These results suggest that surface SOC enrichment and higher soil θv result in higher soil CO2 emissions in 15-year NT systems on Andosols in the humid, subtropical climate of Kanto, Japan. The carbon sources and their proportions of soil CO2 emissions require further clarification to properly assess the benefits of NT in soil carbon sequestration.