Objectives The paddy soils in coastal saline areas are high in salt content, which may potentially slow down the decomposition rate of straw, thereby affecting rice yield and nitrogen fertilizer use efficiency. We investigated the appropriate amounts of straw return, nitrogen application rates, and the effects of using straw decomposition agents, aiming to provide reasonable technical approaches for soil fertility improvement, high rice yield, and efficient nitrogen fertilizer utilization in coastal saline paddy fields.

Methods Field experiments were conducted in 2022 and 2023 in Caofeidian District, Tangshan City, Hebei Province. A split-split plot design was employed. Under the conditions of no application decomposition agent HT20 (D0) and an application rate of 45 kg/hm² (D1), treatments with half amount of straw return (HS, 5250 kg/hm²) and full straw return (S, 10500 kg/hm²) were set up respectively. For each straw return amount, three nitrogen application levels were established: a 40% reduction in nitrogen fertilizer (N1), a 20% reduction (N2), and the conventional nitrogen rate (375 kg/hm², N3). Additionally, two control treatments were included: a blank control (CK), and a control with only the conventional nitrogen rate applied (F). Soil samples (0−20 cm) and aboveground plant samples were collected during the rice tillering, full heading, grain filling, and maturity stages. Soil physicochemical properties, plant biomass, and nitrogen content were analyzed. Yield and its component factors were measured at maturity, and N accumulation, translocation characteristics, and nitrogen fertilizer use efficiency were calculated.

Results At rice maturity stage, compared to the F control, the D0HSN3, D1HSN3, and D1SN3 treatments increased the average soil total N over two years by 2.11%−7.59%; all the 12 combined treatments increased soil alkali-hydrolyzable N by 11.76%−28.71%; D1N2 and D1N3 treatments increased NH₄⁺-N by 26.21%−42.68%; D0N3, D1N2, and D1N3 treatments increased NO₃⁻-N by 32.94%−74.76%; all the 12 treatments increased soil organic carbon by 1.83%−15.95%; D0N2, D0N3, D1SN2, and D1SN3 treatments increased soil EC by 4.20%−7.29%; all the 12 treatments increased CEC by 4.43%−12.23%. There were no significant differences in pH values among different treatments during various rice growth stages. Compared to the F treatment, eight N2 and N3 combined treatments increased nitrogen accumulation in the aboveground parts of rice from the heading to maturity stages by 0.93%−2.78%; four N2 treatments improved the apparent nitrogen recovery rate by 11.16%−12.95% and the nitrogen agronomic efficiency by 23.50%−27.11%, with the D1HSN2 treatment showing the highest values. Random forest analysis indicated that N accumulation, thousand-grain weight, soil NH₄⁺-N, and soil organic carbon (SOC) content were important factors affecting rice yield. Removing these four variables respectively led to decreases in model prediction rates of 41.39%, 23.43%, 22.94%, and 11.83%. N accumulation, soil NH₄⁺-N, and CEC were important factors affecting the apparent nitrogen recovery rate. Removing these three variables respectively led to decreases in model prediction rates of 62.83%, 28.06%, and 13.86%. TOPSIS comprehensive evaluation showed that the D1HSN2 combination was the optimal treatment, followed by the D0HSN2 combination.

Conclusions Reducing the conventional nitrogen application rate by 20% while returning straw to the field significantly increased the contents of organic carbon, available N, EC, and CEC in coastal saline paddy soils. Applying a straw decomposition agent further improved soil physicochemical properties. Under a 20% reduction in the conventional nitrogen application rate, returning half the amount of straw was more beneficial for increasing rice yield and achieving a higher apparent nitrogen recovery rate compared to full straw return. Applying a straw decomposition agent further enhanced the apparent nitrogen recovery rate. Based on the comprehensive evaluation results, a nitrogen application rate of 300 kg/hm² combined with 5250 kg/hm² of straw return, along with either applying 45 kg/hm² of the decomposition agent HT20 or not applying any decomposition agent, can be recommended as the fertilization mode for current coastal saline paddy fields.