Objectives In general, crop residue return is an important measure to improve the soil fertility, nutrient content, and productivity of cropland. We analyzed the annual dynamics of soil organic carbon (SOC), nitrogen (N), phosphorus (P), potassium (K) content, and maize yield under different residue returning methods in the black soil region of Northeast China, and studied the availability of soil nutrients based on soil stoichiometric characteristics.

Methods A long-term maize residue return experiment was established in the black soil region in 2012 with three treatments: residue removed (CK), residue incorporated into 0–20 cm soil (RI), and residue covered on the soil surface (RC). The content of SOC, total nitrogen (TN), total phosphorus (TP), total potassium (TK) in different soil layers and maize yield were measured after the treatments were imposed for 1-year (2013), 3-year (2015), 5-year (2017), and 8-year (2020). The soil stoichiometric characteristics (C∶N, C∶P, N∶P) were calculated, and the correlations between the indexes were analyzed.

Results With the increase of experimental years, SOC content in 0–20 cm soil depth decreased and TN content was not influenced under CK, whereas the content of SOC and TN under RI and RC both increased, with the highest contents in RC treatment. The SOC and TN under RI treatment were uniformly distributed in the plough layer; while the SOC and TN under RC treatment was mainly increased in top 0–5 cm layer, especially from the 5th to 8th year of the experiment. The TP and TK contents in all the three treatments increased slowly with elongation of experiment, and no significant difference was found among treatments. Soil C∶N ranged from 9.87 to 12.56, which was significantly negatively correlated with soil TN content (P<0.05), but not affected by residue returning method. The soil C∶P of all treatments ranged from 31.15 to 39.87 and was not affected by residue returning method. Soil N∶P of all treatments ranged from 2.81 to 3.72, N∶P did not differ significantly among treatments in topsoil. As the experiment progressed, C∶N, C∶P and N∶P in CK treatment showed a remarkable decrease (P<0.05), C∶N in RI and RC treatments also decreased significantly (P<0.05), whereas N∶P in RC treatment increased significantly only in the third year of the experiment (P<0.05). The average annual maize yield of each treatment ranged between 11025 kg/hm² and 11127 kg/hm², and there was no significant difference among treatments.

Conclusions Under the consecutive 8-year of maize cropping, the decomposition rate of soil organic matter accelerated significantly, regardless of residue return. Long-term residue removal markedly increased the availability of P while intensifying the limitation of N in black soil. Although residue return increased the accumulation of SOC, N, P, and K, but did not significantly change the availability of P and the limitation of N, and the maize yield was not significantly increased. The contents of SOC, N, P, and K in 0–20 cm plough layers were relatively uniform when residue was incorporated into soils, while the contents were the highest when residue covered on the soil surface.