4.1 Microbial groups along soil depth
The decline in the abundance of microbial PLFAs with soil depth (Table 1, P <0.05) could arise from the reduction in energy (carbon) and nutrient availability along the profile. A reduction in the ratio of fungi to bacteria with soil depth (Figure 1a) is also an indication that the fungal population declined much faster than that of bacteria along soil profiles. Similarly, the decreasing trends of the fungi to G−bacteria and G+ bacteria to actinomycetes ratios, as well as the increasing G+ to G−bacteria ratio with soil depth could mean that microbial dependence on nutrients adhered to the following order: fungi > G− bacteria> G+ bacteria > actinomycetes. Fungal abundance and activity are considered low in paddy soil because of prolonged anaerobic conditions. However, in paddy soils, rice plants could release oxygen from the root through aerenchyma (Frenzel et al., 1992) and created an aerobic environment in the rhizosphere. Previous studies have demonstrated that fungi have the highest capacity for assimilating rhizodeposits in paddy soils (Ge et al., 2016). Such translocation to fungi can be rapidly detected in PLFAs due to the direct connection of fungal hyphae and mycorrhizal fungi with rice roots (Yuan et al., 2016). Therefore, oxygen limitation caused by root biomass reduction with soil depth could largely suppress fungal abundance. G− bacteria have been reported to preferentially use labile compounds such as rhizodeposits and exudates from plant biomass, and G+ bacteria are able to use recalcitrant compounds from SOM (Kramer and Gleixner, 2008; Fanin et al., 2019). Zhu et al. (2017)also found that G+ bacteria increased after rhizo- and micro-C addition. Labile substrates decrease with increasing soil depth, thereby causing G− bacteria to decline faster than G+ bacteria. As actinomycetes are well known for their key role in degrading complex compounds (Acosta-Martínez et al., 2008), they are relatively the least sensitive to labile substrate reduction with soil depth. Thus, the sensitivity trends of the PLFAs with soil depth were not altered by fertilization type.
The decrease in MBC and MBN with soil depth corresponded with the decrease in SOC and available N content along the soil profile (Table 2). This indicates that the growth of microorganisms is related to the availability of C and N nutrients (Loeppmann et al., 2016; van Leeuwen et al., 2017). SOC levels are fundamentally determined by the balance between organic matter inputs and their losses through decomposition (Six, Frey, Thiet, & Batten, 2006). In our current study, rhizodeposition and fertilization were the two main sources of microbial C nutrient availability. Zhu et al. ( 2017) found that up to 45% of rice rhizo-C was stabilized within SOC. In paddy soils, more than 50% of the total root biomass is allocated to the first 5 cm of surface soil; therefore, the upper soil layer can receive more C input than the soils below (Li et al., 2004; Li & Yagi, 2004). The decrease in the δ13C isotopic signature of soil and PLFA from C4 plants with soil depth after C3−C4 vegetation change in a previous study also demonstrated the decreasing influence of plants on deeper soil (Kramer & Gleixner, 2008). Higher MBC and MBN contents were observed at 0–20 cm in the control, indicating that soil layers with fresh C input can increase microbial biomass. RDA results also showed an obvious separation of the 0–20 and 20–40 cm soil layers, even in the control without fertilizer, confirming the rhizodeposition effect on the 0–20 cm soil layer (Figure 3a).