Discussion
Our results demonstrated that soil N- hydrolyzing enzyme activities (including NAG and LAP), and the specific enzyme activities varied with land use change and season in the Danjiangkou Reservoir area of central China (Table 2; Figs. 1-3). Notably, whether conversion of uncultivated land to the afforested lands (woodland and shrubland) or cropland enhanced NAG and LAP enzyme activities in this region. With land use changing, multiple factors including availability of nutrients, soil microbial attributes and soil properties could cause variation in soil enzyme activities and EES (Luo et al., 2020; Zhao et al., 2018).
Firstly, greater SOC and SON contents in afforested soils and cropland soils compared to the uncultivated land (Table 1; Fig. 5) made larger contribution to increases in NAG and LAP enzyme activities, because SOC and SON could provide more substrates and energy for enzymatic reactions (Feng et al., 2018). The significant positive relationships of soil N-hydrolyzing enzyme activities with SON, SOC (Table 1; Table S1) further confirmed that the SON and SOC were importance source for microbes producing hydrolytic enzymes (Wallenstein & Weintraub, 2008; Yu et al., 2017). Increased N-hydrolyzing enzyme activities in the afforested soils could also be attributed to larger LN contents as labile organic matter were preferentially used by microorganisms (Knops et al., 2002). The tightly positive relationship between LN and hydrolyzing enzyme activities were found in our Pearson’s correlation analysis (Table 1). Secondly, the higher soil C: N ratios (i.e., the C: N ratio of organic soils, C: N ratio of labile pool, C: N ratio of recalcitrant pool) could possibly lead to higher N-hydrolyzing enzyme activities as microbial activities was more susceptible to N restriction (Mendham et al., 2004). This speculation was supported by the significant positive correlation between soil N-hydrolyzing enzyme activities and three kinds of C: N ratios across all land types (Table 1). Thirdly, our results showed that two N- hydrolyzing enzyme activities was negatively correlated with NH4+-N and NO3--N (Table 1). This result also agrees with the results of other researchers, which showed that increased inorganic N availability could inhibit the synthesis of N-hydrolyzing enzyme activities (Allison et al., 2007; Stursova et al., 2006). Additionally, our results showed that change in soil pH under land use change negatively affected soil N-hydrolyzing enzyme activities (Table 1 and Table S1), which highlighted the evidence that soil pH was a strong control on enzyme activities with soil pH values of approximately 8 (Cenini et al., 2016).
As expect, part of our finding was consistent with our hypothesis 1 that afforestation would increase soil N-hydrolyzing enzyme activities compared with uncultivated soils. Whereas, cultivation also increased soil N-hydrolyzing enzyme activities compared with those in the uncultivated land soil, which was contrary to part of the hypothesis 1 (Figs. 1-2). This could attribute to the N fertilization could not meet the nutritional needs of crops and microorganisms in cropland. Generally, N- hydrolyzing enzyme activities decreased with the soil depth (Figs. 1-2), primarily due to the decrease in C and N availability with increasing soil depth (Bowles et al., 2014). Meanwhile, the seasonal pattern of both NAG and LAP enzyme activities with higher levels in summer than other seasons could be attributed to higher microbial biomass in summer (Fig. 2). Globally, temperature has positive influence on the activities of N-hydrolyzing enzymes (Sinsabaugh et al., 2008). However, there was no significant relationship between soil temperature and soil N-hydrolyzing enzyme activities in our study region (Table 1). Instead, we found the positive relationship between soil total PLFA biomass and soil NAG and LAP enzyme activities (Table 1), due to more favorable temperature in summer (Zhang et al., 2016; Zhang et al., 2019a; Table 1)
Specific enzyme activities increased with both cultivation and afforestation (Fig. 3), confirming that increased N turnover rate and enhanced N mineralization via N-hydrolyzing enzyme activities (Wang et al., 2012). It has been suggested that specific enzyme activities can decouple the soil enzyme activities from changes in SON contents, which can serve as an improved indicator of the co-metabolism of microbial communities (Raiesi and Beheshti 2014; Sinsabaugh et al., 2009). We also found specific enzyme activities were significantly lower in cropland compared to afforested soils (Fig. 3), being consistent with previous results that N fertilization in cropland reduced the activity of NAG and LAP per unit of soil SON (Cenini, 2016). The addition of inorganic forms of N to soils could provide N forms available for microorganism to be assimilated, which could explain the lower N-hydrolyzing enzyme activities and lower specific enzyme activities in cropland compared to the afforested lands.
Previous studies indicated that soil ecoenzymatic C: N, C: P and N: P ratios reflected the metabolic requirements and nutrient availability of microbes in the environments (Cui et al., 2018; Waring et al., 2014). The soil ecoenzymatic C: N and C: P ratios were averaged 0.72 and 0.63 in our dataset, respectively (Fig. 4), which was higher than the global values (0.62 and 0.13) (Sinsabaugh et al., 2008, 2009). This finding implied that soil microbial growth and C uptake in our study area could be limited by N and P availability (Xu et al., 2017). Peng and Wang (2016) also reported higher soil ecoenzymatic N: P ratio in temperate grasslands than in tropical soils, indicating higher N demand compared to P demand in temperate grassland. Meanwhile, we found that soil ecoenzymatic C: N and C: P ratios were higher in the afforested lands and croplands than in the uncultivated land, indicating that greater microbial demand for C relative to N and P with afforestation and cultivation. Higher soil ecoenzymatic C: N and C: P ratios in cropland land and afforested land could be related to the different nutrient status of soil microorganism (He et al., 2020). According to resource allocation strategy, microbes produce additional N-hydrolyzing enzymes (indicated by low soil ecoenzymatic C: N ratios) in nutrient-limited environments (Xu et al., 2017). However, the soil ecoenzymatic C: N ratios increased with increasing soil C: N ratios in this present study (Table 1). We inferred that the enzymatic acquisition did not match the resource stoichiometry because SOM with labile and recalcitrant fractions, possibly only a small labile fraction could be used by microorganisms (Kamble and Bååth, 2014). This speculation was supported by the higher RIC and RIN index (i.e. larger recalcitrant fraction) in afforested soils (Table 1).
In addition, we found that soil ecoenzymatic N: P ratio was significantly higher in cropland than other land uses (Fig. 4), due to P fertilizer in cropland. During afforestation, plant debris input in woodland and shrubland, possibly increased soil N content (Cheng et al., 2013; Zhang et al., 2016), while soil P was mainly derived from mineral decomposition and fertilizer. The addition of P fertilizer to farmland could weaken the P limitation. All ecoenzymatic ratios (ecoenzymatic C: N ratios, ecoenzymatic C: P ratios and ecoenzymatic N: P ratios) decreased with depth in afforested soils, suggesting that microbes could invest more N- and P-hydrolyzing enzymes than C-hydrolyzing enzymes, especially in deep soils. Notably, soil enzymatic C: N: P hydrolyzing enzyme activities averaged from 1: 1.06: 1.12 to 1: 2.20: 2.50 in all land use types (Table 3). The ratios of soil C-, N-, and P-hydrolyzing enzyme activities converged on 1:1:1 at the global scale, that defined the boundaries of microbial responses to changes in resource availability and resource allocation patterns depending on nutrient demands (Sinsabaugh et al., 2008). Thus, our results significantly deviated from 1:1:1. It indicated that soil microbes were co-limited by C and P with the imbalance between resource supply and production N and P limitation in our study region.