Response of Sugarcane Parameters to Different Fertilization
The response of sugarcane parameters to different fertilizations: biochar (BC), organic matter (OM), filter mud (FM) and control (CK) were assessed. Compared with CK treatment, BC, OM and FM treatments did not improve the sucrose content, stem diameter and stalk height (Figure 1A,B,D). On the other hand, BC and FM significantly increased (p < 0.05) sugarcane stalk weight and ratoon weight compared to CK and OM (Figure 1C,E), while chlorophyll content peaked significantly (p < 0.05) under BC, FM and OM treatments compared with CK treatment (Figure 1F).
Soil Properties and Enzyme Activities
Compared to CK treatment, BC and OM treatments significantly improved (p < 0.05) NH4+-N in 0-20 soil layer. However, FM treatment significantly decreased (p < 0.05) NH4+-N in 0-20 soil depth compared to CK treatment (Figure 2A). Compared to CK treatment, soil NO3--N significantly increased (p < 0.05) in all the treatments in 0-20 cm soil profile (Figure 2B). In 0-20 cm soil depth, soil OM significantly increased (p < 0.05) under BC and FM treatments compared to CK treatment. However, compared to CK treatment, soil OM was not influenced under OM amended soil across the entire soil layers (Figure 2C). It was also observed that soil TC was enhanced significantly (p < 0.05) under all the amended soils compared to CK treatment in soil profile 0-20 cm (Figure 2,D). Moreover, Compared to CK treatment, soil TN and TC/TN were not significantly impacted under BC, FM and OM treatments, especially in upper the soil profile (0-20 cm) (Figure 2E,F). Soil AK was significantly higher (p < 0.05) 0-20 and 20-40 cm soil depths compared to CK treatment, but significantly decreased (p < 0.05) in soil depth 0-20 cm under FM and OM treatments compared to CK treatment. We also observed that soil AK was significantly more (p < 0.05) under FM and OM treatments in 20-40 and 40-60 cm soil profiles compared to CK treatment (Figure 2,G). Soil AP revealed no significant change in the entire soil layers in BC treatment compared to CK, however, FM treatment significantly increased (p < 0.05) AP in 0-20 and 20-40 cm soil depths compared to CK treatment, while in OM treatment, AP significantly increased (p < 0.05) in 0-20 cm soil layer, but was not significantly impacted in 20-40 and 40-60 cm soil depths compared to CK (Figure 2H). Compared to CK treatment, soil pH was not affected under BC treatment across the entire soil profiles. However, soil pH significantly reduced (p < 0.05) under FM treatment in soil 20-40 and 40-60 cm soil profiles. Whereas in OM treatment, the soil pH in 0-20 and 20-40 cm soil layers significantly reduced (p < 0.05), but was remarkably high (p < 0.05 in) in 40-60 cm soil depth compared to CK (Figure 2I). We also observed that soil EC increased significantly (p < 0.05) under BC and FM treatments in soil depths 0-20 cm and 0-20 and 20-40 cm compared to CK treatment, respectively. While OM treatment significantly diminished soil EC in soil depth 0-20 cm compared to CK treatment (Figure 2,J). While soil SWC significantly increased under OM treatment compared with the other treatments specifically in 20-40 cm (Figure 3K).
Meanwhile, β-glucosidase was significantly low (p < 0.05) in BC, and OM treatments, and enhanced under OM treatment in 0-20 cm soil depths compared with CK treatment, (Figure 3L). Under both FM and OM treatments, phosphatase was considerably improved (p < 0.05) in 0-20 cm soil profile, while in BC amended soil revealed no difference was observed compared to CK (Figure 3M). Moreover, urease activity under BC, FM and OM treatments was significantly higher (p < 0.05) in 20-40 cm soil profile compared to CK (Figure 3N). The analysis also showed that cellulose activity decrease with soil profile under BC and FM treatments compared to CK treatment. However, cellulose activity significantly reduced (p < 0.05) in 0-20 cm soil depth in OM treatment compared to CK treatment (Figure 3O).
nifH Gene Copies and Alpha Diversity
Compared with CK treatment, BC and OM amended soils significantly diminished the nifH gene in 0-20 cm soil profile. On the other hand, FM treatment significantly increased the nifH gene in 20-40 and 40-60 cm soil layers compared with CK treatment. Regarding different soil depths, we observed that the nifH gene was stable in all the three soil depths in BC amended soil, but higher in 0-20 cm soil layer compared with 20-40 and 40-60 cm soil profile in CK treatment. Furthermore, our analysis indicated that the nifH gene was significantly high (p < 0.05) in 20-40 cm soil layer compared with 0-20 cm soil depth under FM treatment, and also decreased with soil depth in OM treatment (Figure S1A). Diazotrophic community diversity was analyzed accordingly in every sample along with various soil depths using diversity estimator (Shannon and Simpson) and richness (Ace and Chao1). The observed diversity and richness under various soil amendments exhibited little or no significant change in the entire soil profiles compared with CK (Table S2).
Relative Abundance of nifH Genes and Dominant Diazotrophic Phyla and Genera
Compared to control, the different amendments showed little or no significant difference in the various soil depths and treatments (Figure S1A). The dominant diazotrophic relative abundance distribution pattern was examined in different soil depths (0-20, 20-40 and 40-60 cm) at the phyla and genera levels. In soil depth 0-20 cm, we detected Proteobacteria (71.1-80.2%) and Cyanobacteria (8.6-15.3%) as the most dominant diazotrophic phyla. Moreover, in 20-40 cm soil profile, Proteobacteria (88.6-94.4%) and Cyanobacteria (0.0-2.8) were the dominant diazotrophic phyla. Whereas in 40-60 cm soil layer, Proteobacteria (82.9-88.4%) was the dominant diazotrophic phyla (Figure 3A). However, we observed that FM, OM and BC amended soils revealed little impact on diazotrophic phyla compared with the CK in the entire soil profiles (Figure S1B-I). At diazotrophic genera level, Geobacter (89.8-94.3%), Anaeromyxobacter (3.2-5.1%), Burkholderia (0.8-2.2%), Azotobacter (0.1-1.7%), Desulfovibrio (0.3-1.5%), Anabaena (0.4-1.0%) and Enterobacter (0.1-0.5%) were the dominant bacterial genera in 0-20 cm soil depth. Furthermore, Geobacter (90.6-94.0) and Anaeromyxobacter (4.7-6.6%) were the dominant diazotrophic genera in soil depth 20-40 cm. In 40-60 cm soil profile, Geobacter (83.7-89.5%) and Anaeromyxobacter (10.0-16.1%) were highly abundant (Figure 3B). Further analysis showed that a vast majority of diazotrophic genera were significantly altered in the different soil profiles under the different soil amendments (Figure S1J-S). Noticeably, Anabaena was significantly (p < 0.05) higher in 0-20 cm in BC amended soil than the other treatments (Figure S1J). Burkholderia, Desulfovibrio and Enterobacter in FM and BC amended soils at 0-20 and 20-40 cm soil layers improved significantly (p < 0.05), respectively, followed by Methylomonas in 20-40 cm under BC treatment (p < 0.05) relative to that under CK, OM, and FM (Figure S1M-O,Q). Venn diagram analysis further revealed the unique and overlap enriched genera among various treatments and soil profiles. It was observed that 1 genus was enriched in both CK and BC, 3 in FM and none in OM (Figure 3C). Meanwhile, 8 genera were enriched in 0-20 cm and 1 in both 20-40 and 40-60 cm soil depths (Figure 3D).
Multivariate ANOVA Analysis for the Impact of Soil Depth and Fertilizations on Diazotrophic Parameters and Edaphic Factors
Multivariate ANOVA analysis was leveraged to test the effects of soil depth gradient and fertilization and their association with different soil parameters relating to N2-fixers, namely, OTUs, Shannon, chao1, coverage, nifH gene copies and edaphic factors, such as urease, cellulase, glucosidase and phosphatase (Table 1). Multivariate ANOVA analysis revealed that soil depth significantly (p < 0.05) impacted diazotrophic species richness indice (Chao1) and N2-fixers alpha diversity index (Shannon), followed by diazotrophic coverage. However, soil depth had no impact on nifH gene copy number. The analysis also revealed that both soil depth and treatment had a significant impact on bacteria OTUs, while treatment had little impact on coverage. Moreover, we observed that soil enzyme activities, namely, urease, glucosidase and acid phosphatase, followed by cellulase were affected to a greater extent by the different soil depths compared with fertilization (Table 1). Regarding soil biochemical properties, namely, soil pH, available phosphorus (AP), available potassium (AK), total carbon (TC), total nitrogen (TN) and organic matter (OM), ammonium (NH4+N) and nitrate (NO3-N). It was observed that soil depth was a potential environment gradient affecting the edaphic factors than fertilization. Both treatment and soil depth had little impact on soil TC/TN. However, both treatment and soil depth revealed no impact on soil OM content (Table 2).
Table 1
Multivariate ANOVA for the effects of soil depth, and different fertilizers on nifH OTUs, diversity, species richness, coverage, gene abundances and soil enzyme activities
Factor
|
|
OTUs
|
Shannon
|
Chao1
|
Coverage
|
nifH
|
Urease
|
Cellulase
|
Glucosidase
|
Acid phosphatase
|
Treatment
|
|
**
|
NS
|
NS
|
*
|
***
|
NS
|
NS
|
NS
|
NS
|
Depth
|
|
**
|
***
|
***
|
**
|
NS
|
***
|
***
|
***
|
***
|
T x D
|
|
***
|
***
|
***
|
***
|
***
|
***
|
***
|
***
|
***
|
Depth stands for soil depth with 0-20 cm, 20-40 cm and 40-60 cm soil layers, treatment stands for fertilizer different fertilization with CK, BC, FM and OM.
Table 2
Multivariate ANOVA for the effects of soil depth, and different fertilizers on soil biochemical properties
Factor
|
pH
|
EC
|
SWC
|
TN
|
TC
|
TC/TN
|
AP
|
OM
|
AK
|
NO3--N
|
NH4+-N
|
Treatment
|
NS
|
NS
|
NS
|
NS
|
NS
|
*
|
NS
|
NS
|
NS
|
NS
|
NS
|
Depth
|
***
|
***
|
***
|
***
|
***
|
*
|
***
|
NS
|
***
|
***
|
***
|
T x D
|
***
|
***
|
***
|
***
|
***
|
*
|
***
|
**
|
***
|
***
|
***
|
Depth stands for soil depth with 0-20 cm, 20-40 cm and 40-60 cm soil layers, treatment stands for fertilizer different fertilization with CK, BC, FM and OM.
Diazotrophic Community Composition under Contrasting Fertilization in Different Soil Profiles
Principal coordinates analysis (PCoA) was adopted to assess N2-fixers community composition in different soil profiles (0-20, 20-40 and 40-60 cm) and the different soil amendments. The analysis demonstrated that N2-fixers community composition in the entire soil depth revealed distinct patterns compared with the different treatments (Figure 4A,B). Redundancy analysis (RDA) was then employed separately in two soil depths (0-20 and 20-60 cm) to assess the impact of soil biochemical properties on diazotrophic community composition at the phyla level. The analysis showed that soil AP (R2 = 1.1860, p < 0.05), EC (R2 = 1.0933, p < 0.05), NH4+-N (R2 = 1.0915, p < 0.05), TN (R2 = 1.9840, p < 0.05), OM (R2 = 1.8575, p < 0.05), followed by pH (R2 = 1.5793, p < 0.01) and AK (R2 = 1.5232, p < 0.01) were the major impact factors shifting diazotrophic community composition, while TC (R2 = 1.5702, p < 0.05) was the manor driver influencing diazotrophic community composition in 0-20 cm soil profile (Figure 4C). Whereas in 20-60 cm soil layer, soil AP (R2 = 0.4968, p < 0.001), AK (R2 = 0.4273, p < 0.001), NO3--N (R2 = 0.7832, p < 0.001) were the major impact factors shifting diazotrophic community composition, while TC (R2 = 0.2532, p < 0.01) and EC (R2 = 0.2184, p < 0.01) were the manor drivers altering diazotrophic community composition in 20-60 cm soil profile (Figure 4D).
Correlation between Edaphic Characteristics and Diazotrophic Community Composition
Network correlation analysis was used to measure the possible interaction between environmental variables and specific diazotrophic community composition. The patterns in network structure demonstrated diazotrophic genera related to Proteobacteria demonstrated a significant and positive association with significant a vast majority of edaphic factors. Noticeably, Azoarcus, Dechloromonas had a positive relationship with soil NO3--N, AP, AK, EC, NH4+-N, TC and soil TN, but showed a negative relationship with soil pH. However, Anaeromyxobacter showed a negative correlation with soil with important soil variables, namely, NO3--N, AP, AK, EC, NH4+-N. Moreover, Zoogloea was negatively associated with soil NH4+-N, EC, TC, AK, TC, pH, NO3--N, but revealed a positive correlation with soil AP, TN, pH. Whereas genus classified as Cyanobacteria, such as Microcoleus exhibited a positive relationship with soil EC, AP, NO3--N, TN, TC, AK and NH4+-N (Figure 5A). Additionally, the taxonomic composition of nifH OTUs showed a significant and positive correlation with a vast majority of the edaphic factors, including NO3--N, TC, EC, NH4+-N, AK, TN and AP (Figure 5B). Later, Pearson’s correlation coefficients were employed separately in the three soil profiles (0-20, 20-40 and 40-60 cm) to have a better understanding of how soil biochemical properties affected diazotrophic phyla and genera community composition. The analysis showed that soil biochemical properties were significantly associated with a vast majority of diazotrophic genera than the phyla, especially in 0-20 cm soil layer (Figure 5C,D).