Soil organic carbon dynamics within density and particle-size fractions of Aquic Cambisols under different land use in northern China

A long-term crop residue incorporation and tillage experiment was initiated in 1999 on cropland that was converted from natural grassland in 1974. Three management factors included in the experiment were land use, tillage systems (tillage/no-till) and methods of maize (Zea mays L.) straw incorporation (whole straw, and pulverized straw and no straw). Soil samples were obtained in 2010, and 13C natural abundance analysis was used to estimate the maize-derived C in density fractions (light fraction, LF; heavy fraction, HF), particle-size fractions (> 50 μm, 50–5 μm, 5–2 μm, 2–0.2 μm and < 0.2 μm), and original SOC decomposition. There occurred a 14–23% decrease in the SOC concentration or a loss of 0.166–0.225 Mg C ha− 1 yr− 1 upon conversion of natural grassland to cropland. Organic carbon (OC) in the LF was a sensitive indicator reflecting this reduction in SOC concentration. Conversion to no-till did not significantly increase the SOC concentration, compared with tillage systems. The data on OC concentration in particle size fractions and those on maize-derived OC in clay and silt fractions indicate that the most stable OC was associated with the silt fraction, while OC bound to clay was important in the short to medium term. Incorporation of pulverized straw over 11 years contributed 2 to 2.4 Mg of maize-derived C ha− 1 into the SOC stock, significantly higher than that by incorporation of whole straw. However, because the whole straw incorporation retained more of the antecedent SOC, it effectively increased the SOC stock. These data suggest that incorporation of whole straw combined with tillage could be an effective practice for improving the SOC stock in croplands of the North China Plain.