This article has been peer reviewed and accepted for publication. It is in production and has not been edited, so may differ from the final published form.
Traditional manual tillage significantly affects soil redistribution and CO2 emission in agricultural plots on the Loess Plateau
Traditional manual tillage using hand tools is widely adopted by local farmers in hilly and mountainous regions in China and many Southeast Asian countries. Manual tillage could result in severe soil erosion redistributing slopes from the upslope areas (erosion) to the lower slopes (deposition). This soil redistribution process may potentially affect the soil carbon cycle, but few studies have quantified soil CO2 emission under different manual tillage practices. Here we evaluated the soil redistribution and its impacts on in-situ CO2 emission as affected by manual tillage of different intensities on three short slopes representing typical cultivated landscapes on the Loess Plateau. Soils were removed at depths of 2 cm, 6 cm and 10 cm by three types of hand tools, hoe, mattock and spade, respectively, at the upslope, and accumulated at the downslope subsequently, to simulate soil erosion and deposition processes by traditional manual tillage. Across the tilled hillslopes, soil CO2 emission was reduced at sites of erosion while enhanced at sites of deposition. Tillage with greater intensity resulted in greater change in CO2 emission. This change in soil CO2 emission was largely associated with the depletion of soil organic carbon (SOC) stocks at erosion sites and the increments of SOC available to decomposition at deposition sites. Moreover, with increasing tillage intensity, soil redistribution by manual tillage shifted the hillslope from a C sink to C neutral or even a C source. Furthermore, manual tillage resulted in substantial changes in soil CO2 emission and redistributed soil in amounts that dwarf animal-powered tillage. Our results implied that manual tillage-induced soil redistribution could have a large impact on the C balance across the local landscape and therefore may have considerable implications for the estimates of regional and global C budgets.
SR16157 Accepted 21 August 2017
© CSIRO 2017