3.1 Straw decomposition and nitrogen release
The decomposition of field-incorporated straw was primarily concentrated in the crop growth period during the first year of straw incorporation (April 15 to October 15, 2018), showing a dynamic change that was initially fast and then slow. The straw decomposition speed was the fastest in the first two months of the first year of decomposition (April 15 to June 15, 2018) and then gradually slowed (Fig. 2). The proportions of decomposed straw were significantly different among the treatments. Specifically, the proportion of decomposed straw was lowest in the AS treatment, followed by the SM treatment, while the proportions of decomposed straw were not significantly different among the B15, B30, and B45 treatments. On June 15, 2018 (two months of decomposition), the proportions of decomposed straw in the AS, SM, B15, B30, and B45 treatments were 19.5%, 20.4%, 42.8%, 43.6%, and 40.3%, respectively. On October 15, 2018 (the first crop growth period during the decomposition period), the proportions of decomposed straw in the AS, SM, B15, B30, and B45 treatments were 41.7%, 56.2%, 67.2%, 63.6%, and 64.2%, respectively. From April 15, 2019, to October 15, 2019 (the second crop growth period during the decomposition period), the decomposition speed of the SM treatment was faster than those of the other treatments. By October 15, 2019, the proportions of decomposed straw in the AS, SM, B15, B30, and B45 treatments were 54.4%, 75.2%, 78.9%, 77.8%, and 77.1%, respectively.
During decomposition, the net nitrogen release of the straw gradually increased over time, showing different patterns of variation in different treatments. On October 15, 2018 (the first crop growth period during the decomposition period), the net nitrogen release in the AS, SM, B15, B30, and B45 treatments accounted for 35.9%, 38.9%, 42.8%, 38.6%, and 39.5%, respectively, of the initial nitrogen content in the straw. On October 15, 2019 (the second crop growth period during the decomposition period), the net nitrogen release in the AS, SM, B15, B30, and B45 treatments accounted for 55.4%, 65.8%, 71.7%, 68.2%, and 64.0%, respectively, of the initial nitrogen content in the straw (Fig. 3a). The 15N abundance of straw in each treatment gradually decreased over time during decomposition, and the reduction in 15N abundance during the first crop growth period was not significantly different among the treatments. Through October 15, 2018, the 15N abundance decreased by 4.60% on average. The abundance of 15N in different treatments began to show differences during the second crop growth period. The abundance of 15N in the AS, SM, and B15 treatments started to show a significant decrease on June 15, August 15, and October 15, 2019, respectively. In contrast, the abundance of 15N in the B30 and B45 treatments only showed a slight reduction. Through October 15, 2019, the abundance of 15N in the AS, SM, B15, B30, and B45 treatments decreased by 13.2%, 11.4%, 9.16%, 6.78%, and 6.70%, respectively (Fig. 3b). The initial 15N abundance (0.632%) of field-incorporated straw was higher than the natural abundance (0.365%) of 15N in the environment, which indicates that there was nitrogen exchange between the straw and the environment during straw decomposition. The amount of nitrogen released from the straw in each treatment gradually increased over time during decomposition, and the differences in the amount of nitrogen released from the straw between the treatments were not significant. From April 15 to August 15, 2018, the nitrogen release speed of the straw was the fastest, and the mean amount of nitrogen released from each treatment accounted for 43.1% of the initial nitrogen content in the straw. The mean amount of nitrogen released from each treatment accounted for 48.4% of the initial nitrogen content in the straw on October 15. Through October 15, 2019 (over two crop growth periods during the straw decomposition period), the mean nitrogen release amount reached 82.3% of the initial nitrogen content of the straw (Fig. 3c). The straw nitrogen uptake increased slowly over time in the first crop growth period during the straw decomposition period, and the differences in straw nitrogen uptake between treatments were not significant; the mean nitrogen uptake accounted for 9.29% of the initial nitrogen content of the straw. The straw nitrogen uptake began to show differences between treatments in the second growth season during the straw decomposition period. The nitrogen uptake amounts in the AS and SM treatments significantly increased, while the nitrogen uptake amounts in the B15, B30, and B45 treatments did not change appreciably. Through October 15, 2019, the nitrogen uptake amounts in the AS, SM, B15, B30, and B45 treatments accounted for 26.9%, 20.8%, 15.6%, 11.5%, and 11.7%, respectively, of the initial nitrogen content in the straw (Fig. 3d).
Figure 4 shows the monthly mean nitrogen release and uptake over two consecutive sampling intervals during the decomposition process of field-incorporated straw. The amount of nitrogen released per month was relatively high during the first two months of the decomposition of the field-incorporated straw and then decreased with increasing decomposition time. In the first crop growth period during straw decomposition, the monthly nitrogen release from the AS and SM treatments from April 15 to June 15, 2018, was significantly lower than that from the other treatments. During the second crop growth period, the monthly nitrogen release from the AS treatment from April 15 to June 15, 2019, was significantly higher than that from the other treatments (Fig. 4a). During the first two months of the first crop growth period, the monthly nitrogen uptake of field-incorporated straw was also relatively high and then gradually decreased with increasing decomposition time. During the first crop growth period, there was no significant difference in the monthly nitrogen uptake of straw from April 15 to August 15, 2018, and the difference between the treatments became evident after August 15, 2018. During the second crop growth period, as noted in April 2019, from September 15 to June 15, the monthly nitrogen uptake in the AS treatment was significantly higher than that in the other treatments, and the variation trend was the same as that of the straw nitrogen release speed, while the other periods did not show the same pattern (Fig. 4b).
3.2 Composition and succession of bacterial communities in straw during straw decomposition
In the experiment, a total of 64 straw samples were collected on four sampling dates (April 15, 2018, June 15, 2018, June 15, 2019, and June 15, 2020) to determine the straw bacterial communities. A total of 3,663,154 high-quality sequences were obtained after excluding chimeras and short sequences, and these sequences were clustered into 10,676 OTUs at 97% similarity. During straw decomposition, the Shannon and Chao1 indices varied at different incorporation depths and first increased and then decreased over time. On June 15, 2018, there was no significant difference in the Shannon and Chao1 indices between the treatments. On June 15, 2019, the Shannon and Chao1 indices of the SM and B15 treatments were significantly lower than those of the other treatments. On June 15, 2020, the Shannon and Chao1 indices of the SM, B15, B30, and B45 treatments were significantly higher than those of the AS treatment, which indicates that the abundance and diversity of bacterial communities in the SM, B15, B30, and B45 treatments were significantly higher than those in the AS treatment (Fig. 5). The NMDS analysis showed that the bacterial communities in the straw changed significantly after decomposition. Specifically, the bacterial communities in the straw on June 15, 2018 (after two months of decomposition), were significantly different from those on June 15, 2019 (after 14 months of decomposition). and June 15, 2020 (after 26 months of decomposition), while the difference between the latter two dates was relatively small. The incorporation depth also affected the bacterial community composition. Specifically, the straw decomposed aboveground (AS and SM) and the straw decomposed underground (B15, B15, B30, and B45) had different bacterial community compositions (Fig. S1). According to the differences between the treatments, all the treatments were divided into five subgroups: no decomposition (CK), two months of aboveground decomposition (P_Te), two months of underground decomposition (M_Te), two, 14 and 26 months of aboveground decomposition (P_Tf), and 14 and 26 months of underground decomposition (M_Tf).
Before straw incorporation, the main bacterial phylum was Cyanobacteria (40.01%), and the main bacterial order was Chloroplast (39.95%). Straw decomposition reduced the relative abundances of Cyanobacteria and Chloroplasts while increasing the relative abundances of Rhizobiales and Burkholderiales. (Fig. 6). The dominant bacterial phyla after straw decomposition were Proteobacteria, Actinobacteriota, and Bacteroidota (Fig. S2). The main bacterial orders were Burkholderiales, Sphingomonadales, and Rhizobiales (Fig. 6). The main bacterial genera were Massilia, Sphingomonas, and uncultured (Fig. S3). STAMP analysis showed that Burkholderiales, Flavobacteriales, Micrococcales, Pseudomonadales, Rhizobiales, Sphingomonadales, Sphingobacteriales, and Xanthomonadales were the main differentially abundant bacteria after straw decomposition (Fig. S4). The main bacterial communities in the aboveground and underground straw decomposition treatments were different, and the bacterial communities changed over time during straw decomposition (Fig. 6). During the entire decomposition process, the relative abundance of Burkholderiales in the P_Te treatment was significantly higher than that in the other treatments (Fig. 7a). The relative abundances of Micrococcales and Pseudomonadales in the M_Te treatment were significantly higher than those in the other treatments (Fig. 7c, Fig. 7d). The relative abundances of Flavobacteriales and Sphingobacteriales were the highest after two months of decomposition (Fig. 7b, Fig. 7g), while that of Rhizobiales was the lowest after two months of decomposition (Fig. 7e). The relative abundance of Sphingomonadales in the P_Tf treatment was significantly higher than that in the other treatments (Fig. 7f). The relative abundance of Xanthomonadales in the M_Te and M_Tf treatments was significantly higher than that in the other treatments (Fig. 7h).
3.4 Relationship between nitrogen release from straw and microbes during decomposition
The network analysis showed that the bacteria that had a strong relationship with straw nitrogen release and that uptake changed over time during decomposition (Fig. 9). On June 15, 2018, seven bacterial orders, including Burkholderiales, had a strong association with nitrogen release (Fig. 9a). On June 15, 2019, 10 bacterial orders, including Sphingomonadales, had a strong association with nitrogen uptake, and among them, five bacterial orders had a strong association with both nitrogen release and uptake (Fig. 9b). On June 15, 2020, 17 bacterial orders, including Rhizobiales, had a strong association with nitrogen uptake, and among them, six bacterial orders had a strong association with both nitrogen release and uptake (Fig. 9c). Spearman correlation analysis showed that the relative abundances of Burkholderiales, Flavobacteriales, Micrococcales, Pseudomonadales, and Sphingobacteriales were significantly positively correlated with the decomposition speed, nitrogen release speed, and 15N abundance of the straw. The relative abundance of Rhizobiales was significantly negatively correlated with the decomposition speed, the nitrogen release speed, and the 15N abundance of the straw. The relative abundance of Sphingomonadales was significantly negatively correlated with the 15N abundance of the straw. The relative abundances of Sphingobacteriales, Flavobacteriales, and Pseudomonadales were significantly negatively correlated with the nitrogen uptake speed of the straw (Fig. 10).