3.4. Relationships among the SOC fractions, enzyme activities, physical and chemical characteristics
A correlation analysis (Table 4) demonstrated that the MBC content displayed an extremely significant positive correlation with the POC and catalase, and displayed an extremely significant negative correlation with EOC (with a correlation coefficient of 0.911). The POC was significantly correlated with the SOC, urease, sucrase, total N and P, however, no significant correlations were observed with amylase, catalase, total porosity, and bulk density. The SOC was significantly correlated with urease, sucrase, total N, total P, total porosity, and bulk density. Soil sucrase activity displayed an extremely significant correlation with amylase and urease, with respective correlation coefficients of 0.597 and 0.848. Physical and chemical characteristics of the soil (i.e., total N, total P, total porosity, and bulk density) displayed strong positive correlation with urease and surase.
4. Discussion
4.1. Soil carbon fraction ofdifferent type vegetations
Vegetation is one of the most important components of an ecosystem, and its community succession has a significant effect on the SOC content (Deng et al., 2018; Solomon et al., 2007). This study demonstrated that SOC content in a forest was significantly higher than in shrublands and grasslands (Fig. 2D). Root exudates and litter from forest vegetation both strongly affected the organic carbon content in the soil and promoted the effectiveness of forest nutrients (Qiao, Miao, Silva, & Horwath, 2014). At the same time, forest vegetation can also alter the forest environment, reducing solar radiation and temperature differences, increasing soil moisture (Özkan & Gökbulak, 2017), and creating a stable environment for litter decomposition. All of this causes the SOC content of forest to be higher than that of shrublands and grasslands. Moreover, due to the higher coverage of herbaceous vegetation and abundant species density (Table 1), more surface litter increases the sources of organic carbon (Zhang et al., 2019), making the SOC content of desert grassland vegetation higher than that of both HR and CK vegetation. Meanwhile, the SOC content of the four vegetation types was not only due to organic carbon inputs, but was also affected by enzyme activities and soil physical-chemical characteristics. A correlation analysis between SOC contents and soil physical-chemical characteristics and enzyme activities further confirmed these results (Table 4).
The MBC content in the soil of HR was significantly higher than in the soil of GL (Fig. 2A). On the one hand, HR vegetation has a wide horizontal root structure and can quickly grow new shoots (Letchamo et al., 2018). These new shoots increase soil porosity (Table 2) and oxygen content during the growth process, and increase soil aerobic microbial activity. On the other hand, the root nodules of HR can fix atmospheric nitrogen and improve soil fertility (their annual average nitrogen accumulation is 17,475 kg/hm2) (Ruan & Li, 2002). Studies have shown that increasing soil N can promote microbial activity and increase the decomposition rate of soil organic matter (Nottingham et al., 2012; Sistla, Asao, & Schimel, 2012), thereby reducing SOC content. The partial shading effect of XS vegetation reduces the soil temperature and the activity of soil microorganisms (Jiménez, Tejedor, & Rodríguez, 2007). Therefore, the soil MBC content is highest in HR vegetation.
The changes in soil POC and SOC are consistent (Fig. 2C) across different types of vegetation, while the changes in EOC and SOC differ (Fig. 2B). Since the soil in this study was obtained from different types of vegetation, different physical and chemical properties (Table 2) regulate the decomposition rates in the soil (Xu et al., 2016). Various surface litter can significantly change the input of soil organic matter (Thorburn, Meier, Collins, & Robertson, 2012), which affects the EOC content in the surface soil (DuPont, Culman, Ferris, Buckley, & Glover, 2010). At the same time, the higher soil temperature and the lower soil water content, which may potentially create more beneficial conditions to enhance labile SOC fractions (Chen et al., 2016). However, the decomposition of plant litter is the most complex ecological process in the biosphere (Méndez, Martinez, Araujo, & Austin, 2019). Therefore, due to the differences in tree species composition, litter quantity and quality, soil microbial group composition, soil moisture, temperature, and nutrient input, different vegetation types have significant differences in soil active organic carbon components (Yang et al., 2018; Soucémarianadin et al., 2018).
The content of activated carbon in the soil under the four vegetation types was greater in the upper layer than in the lower layer. This was mainly because the soil active organic carbon largely depends on the total organic carbon content of the soil. Total SOC decreased as soil depth increased (Fig. 2D), however, the litter on the upper layer not only provides a significant amount of organic carbon for the soil, but also provides the surface soil with a high concentration of nutrients (Table 2), providing stable conditions for growing fine roots in the topsoil layer. Litter and root exudates have become an important source of soil active organic carbon after they are decomposed by microorganisms (Weintraub, Scott-Denton, Schmidt, & Monson, 2007).