Objective Straws contain large amount of nutrients, yet they are rarely considered in nutrient management practices. We studied the potential of replacing chemical fertilizers with straw nutrients and the annual variation patterns, aims to achieve more scientific and precise fertilizer management and sustainable land use.

Methods From 2022 to 2024, a three-year field trial was conducted at Yanjia Gang Farm in Harbin, Heilongjiang Province, where is typical black soil farmland aregion, and the test crop was corn. The NPK fertilizer application rates under full straw return, recommended using the Nutrient Expert system, was used as the control treatment (NE). Based on the rates, straw nutrient substitution treatments were established by reducing N/K fertilizer amounts equivalent to 25%, 50%, 75%, and 100% of the total nutrient through straw incorporation: NE-50%SN, NE-75%SN, NE-100%SN, NE-25%SK, NE-50%SK, NE-75%SK, and NE-100%SK. Additionally, treatments without N or K fertilizer respectively on the basis of NE (NE-N, NE-K) were established for calculating relevant indicators, and local farmer’s practice (FP) serving as the no straw return control at the same time. Corn yield, plant nutrient content, and soil fertility indicators at the 0−60 cm depth were measured annually at harvest.

Results Compared with the NE treatment, the FP treatment applied 36.1% extra chemical fertilizers (N+P₂O₅+K₂O), however, reduced yield by 5.0%, leading to accumulated deficits of N 10.1 kg/hm² and K 117.8 kg/hm² in the soil, and 30.8% higher nitrogen accumulation in the soil profile below 40 cm, indicating a risk of soil nitrogen leaching. Compared to NE, straw nitrogen substitution treatments (NE-50%SN, NE-75%SN, and NE-100%SN) did not significantly effect on maize yield, NE-75%SN and NE-100%SN improved nitrogen fertilizer utilization rate and partial factor productivity, but resulted accumulated N deficits of 18.8 and 29.3 kg/hm² over three years, while NE treatments achieved a cumulative N surplus of up to 22.6 kg/hm² over the three years. The NE-100%SN treatment significantly reduced mineral nitrogen content by 58.4% compared to NE, with no significant differences among other treatments. All substitution ratio treatments reduced mineral nitrogen leaching into deeper soil layers. Compared to NE, corn yield showed no significant changes in straw potassium substitution treatments (NE-25%SK, NE-50%SK, NE-75%SK, and NE-100%SK). The NE-75%SK and NE-100%SK treatments enhanced partial factor productivity of potassium, while accumulating deficits of 12.4 and 36.8 kg K/hm² over three years. The NE treatment achieved the highest cumulative K surplus of 65.6 kg/hm² over three years. Compared to NE soil, all substitution ratio treatments significantly reduced available potassium content by 21.9%, 15.6%, 14.5%, and 13.8%, respectively, with no significant differences among substitution ratios. Treatment with substitution ratios of 50% or higher (NE-100%SK, NE-75%SK, and NE-50%SK) reduced slow-release potassium content by 31.2%, 26.4%, and 20.9% compared to NE soil. Differences among treatments significantly decreased with increasing soil depth.

Conclusions During the three-year experimental period, even subtracting the nitrogen and potassium amounts equivalent to those introduced by complete straw return from the chemical fertilizer application rates recommended by the Nutrient Expert system (NE) did not lead to a significant reduction in maize yield. At the same time, it substantially increased the nitrogen and potassium fertilizer utilization efficiency and partial factor productivity, while reducing the leaching of nitrogen and potassium into deeper soil layers. However, from the perspective of soil nutrient apparent balance, when the reduction in nitrogen and potassium fertilizer application exceeded half of the nitrogen and potassium supplied by straw, the soil nitrogen and potassium levels showed a deficit state. Therefore, it is not recommended to replace chemical fertilizers with straw nitrogen/potassium on a large scale over the long term. The appropriate substitution scale requires further research.