3.5 | Carbon pool management index
After ten years, the SOC lability (L) and SOC lability index (LI) of the 0-10 cm and 35-50 cm depths were higher under CF treatment than those under BF or LF treatment, whereas the C pool index (CPI) of the topsoil (0-10 cm) was lower under CF than that under BF and LF and CMI of the 0-10 cm and 10-20 cm depths were greater under BF than those under LF and CF (Table 2). For tillage effect, the L and LI at the 0-10 cm and 20-35 cm depths under rotational tillage systems (NS, SP, and PN) were higher than those under PP tillage (p < 0.05). And compared with PP, rotational tillage systems significantly increased CPI at the 0-10 cm and 10-20 cm depths (p < 0.05). Meanwhile, CMI was higher under rotational tillage systems (NS, SP and PN) than that under PP tillage at the 0-10 cm, and 20-35 cm depths. Moreover, NS and SP also strongly increased CMI in the 35-50 cm soil layer (p < 0.05).

3.6 | Crop yields

The wheat and maize yields from 2008 to 2016 were strongly affected by fertilization treatments and tillage systems (Table S5 and Figure 6). For fertilization effect, the yield of wheat and maize were both followed by BF > CF > LF. Compared with LF, the yields with BF were significantly increased by 11.70% for wheat on average and 8.31% for maize in average (p < 0.05). For tillage effect, rotational tillage systems (NS, SP and PN) produced higher wheat and maize yields than the PP tillage. More specifically, compared with PP, wheat yields on average were increased by 12.71%, 8.89%, and 12.83% in NS, SP, and PN, maize yields in average were increased by 14.05%, 8.83%, and 12.59% in NS, SP, and PN, respectively. On the whole, the minimum average yield of wheat and maize were found in LF+PP treatment, the maximum average yield of wheat and maize were found in BF+NS treatment.
Regression analysis showed the positive relationships were found between crop (wheat and maize) yields and SOC and it’s labile fractions (Figure 7). The correlation coefficients of crop yields (wheat yield, maize yield) and SOC were smaller than the correlation coefficients of crop yields and soil labile C fractions (ROC, MBC, DOC and POC), indicating that the positive influences of labile C fractions on crop yields were more significant than those of SOC. Wheat and maize yields were positively correlated with CMI.

4 | DISCUSSION

4.1 | Impacts of long-term tillage rotation and fertilization on SOC content and SOC stocks

In this study, fertilization and tillage significantly affected SOC content and stock, whereas their interaction was slight. The SOC stock under BF was markedly greater than that under CF and LF at the 0-50 cm depth. This was primarily since balanced fertilization was beneficial to crop growth and increased the amount of returned crop residues. Compared to CF, BF and LF reduced the application amount of nitrogen and phosphate fertilizer. Lu et al. (2011) indicated that excessive application of N fertilizer might reduce the ratio of C: N in soil, accelerated decomposition of crop straw by soil microorganisms, result in the lower carbon sequestration. In other words, the increase of SOC is more related to the stabilization rate of input C, which also was confirmed by that the linear correlation coefficient between SOC storage and the stabilization rate (0.96) was higher than that between SOC storage and input C (0.55) (Figure 5). In our study, BF and LF increased in SOC contents at the 0-10 cm and 10-20 cm depth due to the higher stabilization rate of input C. In addition, N and P nutrients enriched by long-term high rates of fertilizers could accelerate the decomposition of SOC (Luo et al., 2019), which may be one reason to explain the SOC content in CF lower than in BF and LF.
At the end of the ten-year experiment, the higher levels of SOC stocks were found at the 0-50 cm depth under rotational tillage systems (NS, SP, and PN) than under PP. That can be related to the higher plant biomass C input under rotational tillage systems (Figure 1) and the lower mineralization of SOC. Previous studies reported that conservational tillage, such as no tillage, could reduce soil disturbance, decreases the mineralization rate of SOC and promote the accumulation and humification of crop residues returned to the soil (Mazzoncini et al., 2013). Based on the observed SOC changes in different soil layers, the rotational tillage systems (NS, SP, and PN) increased the SOC content and SOC stock accumulation of the 0-10 cm depth compared to PP. The reason for this difference is that rotational tillage causes less soil disturbance, which allows the returned crop residues to cover most of the soil surface (He et al., 2019). The rotational tillage systems also increased SOC content in > 10 cm layers, and the effect of NS at 10-20 cm and 35-50 cm depths was obvious. This can be attributed to two factors: (1) The rotational tillage systems increased the incorporation of crop residues by input at depth where there may be a greater chance of protecting SOC by the combination with the mineral matrix (Hou et al., 2012); (2) subsoil tillage can break the bottom layer of the plow pan and promote crop root growth, which increases the production of root residues and secretions (Leonard et al., 2012). Poeplau and Don (2013) suggested that subsoil tillage can increase the production of crop root exudates and root litter and therefore increase the amount of SOM in the subsoil.

4.2 | Impacts of long-term tillage rotation and fertilization on soil labile C fractions and CMI

In our report, the ROC, DOC, and POC contents under the BF were higher than those under the LF or CF at the 0-10 cm depth. This was primarily because of the higher C input in the BF. Simultaneously, BF also increased DOC at the 10-20 cm and 20-35 cm depth compared to LF and CF, due to the migration of DOC with soil moisture (Kaiser and Kalbitz, 2012). Increasing the application of nitrogen and phosphorus fertilizers can stimulate soil microbial activity and therefore increase MBC concentrations (Ghosh et al., 2018), and we came to a similar conclusion that the content of MBC in the all layers (0-50 cm) under BF and CF was higher than that under LF. Compared with PP, the rotational tillage systems (NS, SP, and PN) increased the content of soil labile C fractions (ROC, DOC, MBC and POC) in soil surface layer (0-10 cm) by creating an environment in which more crop residues covered on the soil surface. Crop residues provide substrates for soil microorganisms and promote the accumulation of soil labile C (Jharna et al., 2018). In contrast, PP tillage weakened the physical protection of SOC, exposed protected SOC to microbial decomposition, accelerated mineralization of active organic matter in newly turned topsoil, thus increasing the loss of soil labile C (Chen et al., 2009). The rotational tillage systems also increased the content of soil labile C fractions in deep soil (20-35 cm and 35-50 cm), which may be as a result of the increased amount of root debris and exudate returning to the soil (Hou et al., 2012).
In our study, fertilizer application had a significant impact on CMI at the 0-10 cm and 10-20 cm depths and tillage had a significant influence on CMI in all layers (0-50 cm) (Table 2). The positive linear correlation between CMI and carbon input was found in the present study, indicates that higher crop residue inputs caused higher CMI. Similar observations were reported by Chatterjee et al. (2018). The CMI at the 0-10 cm depth was lager with BF than that with CF; rotational tillage systems (NS, SP, and PN) increased CMI compared to PP tillage. These results suggested that soil fertility can be healthily developed by balanced fertilization or rotational tillage. The value of CMI depends on SOC content and activity (Blair et al., 1995), that is increasing organic carbon input or reducing SOC mineralization in BF and rotational tillage could cause higher SOM content and labile organic C, thereby resulting in higher CMI.

4.3 | Impacts of long-term tillage rotation and fertilization on crop yields

In this study, BF with appropriate reductions in N and P fertilizer rates and supplemental K fertilizer improved wheat and maize yields, which was confirmed by Yang et al., (2006). This suggested that optimal yields cannot be obtained in the Loess Plateau by the fertilization methods of applying high N and P fertilizer and no K fertilizer. For tillage, the rotational tillage systems (NS, SP, and PN) increased wheat and maize yields compared with PP tillage. This is because the tillage rotation systems can prevent the decline in SOC caused by long-term intensive tillage and enhance soil fertility (Bhattacharyya et al., 2012). Water shortage is another important factor restricting crop growth in the Loess Plateau. Tillage rotation can reduce soil water evaporation by covering a large of crop residue on soil surface (Yu et al., 2020). In addition, the higher SOC content produced by tillage rotation promotes water storage and water absorption by crop, thereby increasing crop yields (Manns and Berg, 2014).
In our study, crop yields were significantly positively correlated with SOC and soil labile C fractions (ROC, MBC, DOC, and POC) (Figure 7), indicating that the increased content and activity of SOC had a positive impact on crop production (Li et al., 2016). The correlation coefficients of crop yield and soil labile organic C fractions were greater than those of crop yield and SOC, suggesting that soil labile organic C contributes more significantly to increased crop yields. Therefore, agricultural practices that increase the active components of SOC are crucial for maintaining soil fertility and increasing crop yields.

5| CONCLUSIONS

Our results demonstrate that fertilization and tillage practices affect the SOC pool in China’s Loess Plateau. Balanced fertilization (BF) and rotational tillage (NS, SP, and PN) significantly increased SOC stocks, and NS rotation combined with BF produced the highest SOC stock among all treatments. SOC stock accumulation was positively correlated with plant biomass C input and with the stabilization rate (SR) of returned plant biomass C, indicate that BF and rotational tillage had positive effects on SOC sequestration by the increases of input-C and it’s stabilization rate. The SOC, ROC, DOC, and POC contents were greater under BF than those under CF at the 0-10 cm depth. Meanwhile, rotational tillage systems increased the soil labile C contents at the 0-10 cm, 20-35 cm and 35-50 cm depths. BF and rotational tillage also significantly improved the CMI and soil quality due to changes in the content and activity of SOC. In addition, fertilization and tillage practices affected SOC content, soil labile C content and CMI, which in turn affect crop yields. BF and rotational tillage were effective in increasing the yields of wheat and maize. The highest average yields of wheat (increased by 30.93%) and maize (increased by 20.39%) were found in BF+NS treatment. We found that NS tillage with a balanced application of nitrogen phosphorus and potassium fertilizers could increase SOC sequestration, improve soil quality, and increase maize and wheat yields in China’s Loess Plateau.

ACKNOWLEDGEMENTS

This work was funded by the Special Fund for Agro-scientific Research in the Public Interest of the Ministry of Agriculture, China (Grant no. 201503116), and the National Natural Science Foundation of China (51879224).