The experiment was performed at a farm of the Rice Research Institute of Sichuan Agricultural University Chengdu, China (30°35′N, 103°45′E), during 2015–2017. The soil had a sandy loam texture with a total N of 1.96 g/kg, organic matter of 26.00 g/kg, available N of 29.13 mg/kg, available P of 81.60 mg/kg, and available K of 85.98 mg/kg at the time of experiment initiation. The region is classified as humid subtropical with a monsoon climate, and the meteorological data for the experimental years, which were measured at a weather station near the experimental site, are shown in Figure 1 .

The experiment comprised three treatments in a single-factor randomized block design with three replicates during 2015–2017. The treatments were designed as follows: wheat–rice rotation with wheat straw mulching during the rice season (WRS), oilseed rape–rice rotation with oilseed rape straw mulching in the rice season (ORS), and fallow–rice rotation without straw mulching (FRN), which served as the control ( Table 1 ). The plot size was 4.8 m × 3.3 m, and plots were separated by a 0.4 m wide alley with a plastic film inserted into the soil to form a barrier. Wide–narrow row spacing cultivation was used for rice plantation in the present study. Wheat and oilseed rape straws were cut into 5–10 cm long pieces and mulched on a wide row immediately after rice transplantation. Alternate dry/wet irrigation was applied, and the seedlings were transplanted in shallow water (1–2 cm). The field was submerged in a 2 cm water layer for 5–7 days after transplanting to ensure that the seedlings turned green and survived. Thereafter, the water was drained from the field until booting, and the soil water content accounted for 70–80% of the saturated water content. The field was dried during the ineffective tillering stage. The field was again submerged in a 1–3 cm water layer at the booting stage. Irrigation with 3 mm water was carried out in the bolting stage of wheat and oilseed rape, and rain-fed irrigation was applied in other growth periods. The total N applied in the rice and upland crop seasons was 135 and 48 kg/ha, respectively, which was much lower than the 180 and 120 kg/ha applied in conventional cultivation reported by Peng et al. (2002). Details of other cultivation measures and fertilizer applications are shown in Tables 1 , 2 .

The ¹⁵N mini-plot experiment was performed along with a field experiment in 2017. Four metal frames without bottoms (80 cm long × 70 cm wide × 50 cm high) were installed 30 cm deep in the soil and 20 cm above the soil surface around 10 adjacent rice plants in each field plot. The ¹⁵N-labeled urea and straw were applied as described in Table 3 . The ¹⁵N abundance and total N content of labeled wheat and oilseed rape straw were 0.763 atom% and 0.749% and 0.634 atom% and 0.620%, respectively. The application times of fertilizers and other management practices were the same as those in the field experiment.

The data were statistically analyzed to test the level of significance using a single-factor randomized block design. Analysis of variance was performed using SPSS Version 12.0 and Sigma Plot 12.0 to test the effects of treatments and interactions.

N accumulation increased with the growth and development of rice plants, but the amount of N uptake at 30 DAT-HS was the highest, followed by HS-MS ( Figure 2 ). Straw mulching promoted the absorption of N by rice plants, especially after 30 days of transplanting, and the effect of ORS was stronger than that of WRS. In 2016, the rice N uptake of WRS at 20–30 DAT and 30 DAT-HS was significantly decreased by 27.56% and increased by 14.91%, respectively, compared to that of FRN, whereas the rice N uptake of ORS was 3.35–11.64% higher than that of FRN during the entire rice growth period. The rice N accumulation of ORS was 4.21% lower at 30 DAT-HS but 8.49, 54.12, and 9.32% higher at 0–20 DAT, 20–30 DAT, and HS-RS, respectively, compared to that of WRS. In 2017, except for 0–20 DAT, rice N uptake of WRS and ORS was 10.10–95.22% higher than that of FRN. Rice N uptake of ORS was 0.92 and 9.70% lower at 20–30 DAT and HS-RS, respectively, but 14.86 and 13.12% higher at 0–20 DAT and 30 DAT-HS, respectively, compared to that of WRS. Although the amount of N accumulation differed among the three treatments at different stages, the total amount of N accumulated at the mature stage was the highest for ORS, followed by that for WRS, which was 10.09–29.67% and 2.99–22.19% higher than that of FRN.

Total N accumulation in rice plants increased in WRS mainly because of the increase in the uptake of soil N, whereas in ORS it was because of the synchronous increase in the uptake of soil N and fertilizer N ( Figure 3 ). Straw mulching significantly promoted the uptake of N from the soil by rice plants from 30 DAT to MS. The amount of soil N uptake by rice plants with straw mulching was 11.68–12.01% lower than that of FRN at 20 DAT, but 14.79–29.43%, 40.11–40.64%, and 21.75–26.82% higher than that of FRN at 30 DAT, HS, and MS, respectively. The effects of oilseed rape and wheat straw mulching on N fertilizer uptake by rice plants were significantly different. The amount of fertilizer N uptake by WRS rice plants from 20 DAT to HS was lower than that of FRN, whereas ORS promoted fertilizer uptake by rice plants throughout the growth stages. As a result, the total amount of fertilizer N uptake by rice plants of WRS and ORS was 46.33 and 61.67 kg/ha, which was 9.02 and 45.10% higher than that of FRN, respectively. Although straw mulching promoted N absorption by rice plants, the amount of straw N absorbed by rice plants was 1.06–1.20 kg/ha, only accounting for 0.62–0.66% of the total N uptake.

N release from straw increased gradually with the growth of rice plants ( Figure 4 ). Except for the release rate at MS in 2016, the cumulative release amount and rate of WRS were higher than those of ORS at all stages. The total amount of N released from WRS was 24.76 and 34.97 kg/ha, which was 10.1 and 40.87% higher than that from ORS in 2016 and 2017, respectively. The N release amount and rate of WRS and ORS first decreased and then increased, and the release rate was the highest at 20 DAT, accounting for 21.20–52.77% and 19.68–39.41% of the total release in WRS and ORS, respectively. For different stages, the amount of straw N released from WRS at 0–20 DAT and 20–30 DAT was 69.70 and 19.54% higher than that from ORS in 2016, respectively, which significantly increased by 34.62–149.78% during the entire rice growth period in 2017. Only 3.04–4.82% was absorbed and utilized by rice plants out of the total 20.31–34.97 kg/ha of N released by straw mulching, and most of it remained in the soil ( Figure 5 ). The amount of N released by WRS was higher than that released by ORS; however, the utilization rate was 1.78% lower than that of ORS.

N use efficiency increased with straw mulching; however, the absorption and utilization rates of N applied to rice at different growth stages were significantly different ( Figure 6 ). N uptake of base fertilizer in WRS decreased by 19.03–68.48% at different growth stages, whereas that of tillering and panicle fertilizer in WRS first decreased and then increased, reaching 30.03 and 10.66% at MS, respectively, compared to that of FRN. N uptake of base, tillering, and panicle fertilizer in ORS increased at different growth stages (except for tillering fertilizer at 20 DAT), reaching 12.01–41.69%, 5.53–39.44%, and 22.94–50.39%, respectively, compared to that in FRN. Therefore, the N utilization rate of the base fertilizer of WRS and ORS decreased by 3.52% and increased by 5.91%, respectively, whereas that of tillering and panicles in WRS and ORS increased by 5.85 and 5.35%, and 7.68 and 25.30%, respectively, compared to that in FRN, which led to the increase in total N fertilizer utilization by 2.84 and 14.20%, respectively ( Figure 5 ).

The present study showed that the total N uptake by rice plants of WRS and ORS was higher than that by plants of FRN, but N uptake decreased at 0–20 DAT ( Figure 1 ). In paddy systems, rapid straw decomposition may also lead to a high accumulation of allelochemicals during early straw incorporation and cause rapid changes in soil properties, inhibit root growth, and decrease N uptake (Liu et al., 2007; Yan et al., 2019). Except for the allelochemicals, the decrease in N uptake by rice plants in the early growth stages may also be attributed to the following: i) Some amount of the N released by the base-tillering fertilizer was absorbed by the straw. Chen et al. (2022) found that the N content of mulched straw increases after fertilization, thus reducing the ammonia volatilization loss of fertilizer N. The results of the mini-plot experiment with ¹⁵N-labeled urea showed that fertilizer N in the undecayed straw was 1.55–2.97 kg/ha at the mature stage of rice, which indicated that a large amount of fertilizer N was absorbed by the straw in the early stage, and part of fertilizer N was not released into the soil until rice maturity. ii) After returning to the field, straw decomposes rapidly in the early stage, which rapidly increases the soil C content, leading to an unbalanced soil C/N ratio and resulting in competition for N between microorganisms and rice plants (Zhang et al., 2010; Kuzyakov and Xu, 2013). However, no-tillage with straw mulching increases soil bacterial community diversity (Luo et al., 2020) and the activities of soil enzymes, including invertase, acid phosphatase, and urease (Iqbal et al., 2021); improves soil N retention capacity, and reduces N loss risk (Yang et al., 2022b), all of which lead to more efficient N recycling in cropping systems (Yang et al., 2023). Therefore, WRS and ORS promoted N uptake by rice plants in the later growth stages (from HS to MS) and significantly increased the total N accumulation in rice plants under no-tillage with straw mulching conditions. This reveals that the straw mulching of dryland crops has the potential to reduce N input and increase rice yield under paddy-upland rotation with no-tillage.

The results of this study showed that total N accumulation in rice plants increased in the WRS and ORS, which was mainly attributed to an increase in the uptake of soil N ( Figure 3 ). Soil is the main N source for crop uptake, and approximately 50–80% of the N absorbed by rice during its entire growth period is sourced from the soil (Zhu, 2008). Additional N sources promote the absorption of soil N by plants, which was defined as the “priming effect” or “added N interaction” by Hamid and Ahmad (1993). Under the paddy-upland rotation, N application during the rice season increases the absorption of soil N from previous crops by rice plants (Zhou et al., 2020). In this study, the soil N uptake increased to 97.89, 124.15, and 119.18 kg/ha for FRN, WRS, and ORS, respectively. Straw mulching promotes the establishment of soil microbial colonies and increases fungal and enzyme activities (Bolinder et al., 2020; Dhaliwal et al., 2020), which significantly improves the mineralization and release of soil N, increases the transformation of ammonium and nitrate, and promotes the absorption of soil N by rice plants (Quan et al., 2018). In addition, although the straw released 20.31–34.97 kg/ha N in the rice season, there was little absorption by rice plants ( Figure 5 ). Li et al. (2009) reported that more than 50% of the N in crop straw is refractory organic matter, which can only be absorbed by plants after transformation by microorganisms. Therefore, N released by straw is mainly retained in the soil during the short period of crop growth. This could explain why straw mulching not only increases N uptake by plants but also improves soil N content (Liu et al., 2021). The transformation and utilization of straw N require further study to reveal the mechanism of long-term straw return to improve soil fertility and crop yield.