Performance of two winter cover crops and their impacts on soil properties and two subsequent rice crops in Dongting Lake Plain, Hunan, China

Winter cover cropping is a common practice in paddy soils in southern China. This two year study (2008–2010) was undertaken to determine dry matter productivity and N uptake of two winter cover crops: ryegrass (Lolium multiflorum Lam.) and Chinese milkvetch (Astragalus sinicus L.) and their effects on soil properties as well as two subsequent rice (Oryza sativa L.) crops in Dongting Lake Plain in Hunan, China. Cover crops were broadcast sown in October, approximately 10 d before the harvest of late rice. Ryegrass was first cut in March and finally harvested in April coincident with the plough of milkvetch. Two consecutive rice crops were grown following the cover crops without application of chemical N. Averaged over two years, ryegrass produced more aboveground dry matter and accumulated higher amounts of N than milkvetch (4.7 Mg ha−1 vs. 2.4 Mg ha−1, 128 kg ha−1 vs. 71 kg ha−1). Compared to CK (bare fallow without weed), both ryegrass (RG) and milkvetch (CMV) treated plots had lower soil bulk density at 0–15 cm and 15–30 cm depths. At the harvest of cover crops, CMV increased soil NH4+-N concentration at 0–10 cm, 10–20 cm and 20–30 cm depths. In contrast, RG decreased soil NH4+-N and NO3−-N concentrations in the three depths. Averaged over four sampling dates in the double rice seasons in 2010, soil microbial biomass C was slightly different between RG (257 mg kg−1) and CMV (249 mg kg−1), and these were significantly higher than CK (218 mg kg−1). Soil microbial biomass N was highest for CMV (15.4 mg kg−1), followed by RG (11.3 mg kg−1) and CK (6.1 mg kg−1). Grain yield and total N yield of early rice were 0.6 Mg ha−1 and 11 kg ha−1 higher for CMV, respectively, and 1.0 Mg ha−1 and 20 kg ha−1 lower for RG as compared with CK averaged over years. Grain yield and total N yield of late rice were not significantly affected by cover crops.