Biochar amendment decreases soil microbial biomass and increases bacterial diversity in Moso bamboo (Phyllostachys edulis) plantations under simulated nitrogen deposition

Biochar was produced by pyrolysis of Moso bamboo chips at 600 °C (Yaoshi charcoal industry Company, Hangzhou, China). The main characteristics of the biochar were: pH (H₂O) = 9.67; bulk density = 0.53 g cm⁻³; cation exchange capacity (CEC) = 14.9 cmol kg⁻¹; carbon content = 81.73%; N content = 0.57%; and C/N ratio = 143.4.

The study site was in Qingshan Town, Lin'an City (30°14'N, 119°42'E), Zhejiang Province, China, which has a monsoonal subtropical climate. The mean annual precipitation is 1 420 mm, and the mean annual temperature is 15.6 °C, with maximum and minimum temperatures of 41.7 °C and −13.3 °C, respectively. The area receives an average of approximately 1 847 h of sunshine per year and an average of 230 frost-free days per year.

The Moso bamboo plantations in the study site are intensively managed, and are subjected to annual fertilization (67.5 kg N, 11.8 kg P, and 74.7 kg K ha⁻¹), artificial ploughing, and weeding. There are 11 understory plant species, dominated by Viola prionantha, with a mean height of 0.1 m, forest floor coverage of 5%, and total herbal biomass of 14.6 kg ha⁻¹. The soils are classified as Ferrisols derived from granite (Song et al 2015).

In November 2012, 12 plots of 20 × 20 m were established in the Moso bamboo plantations. In accordance with the local background N deposition of 30–37 kg N ha⁻¹ yr⁻¹ (Xie et al 2008, Cui et al 2014) and the widely used method to double and triple the local deposition rate, to simulate additional N deposition (Song et al 2015), these treatments included (1) low N, 30 kg N ha⁻¹ yr⁻¹ (N30); (2) medium N, 60 kg N ha⁻¹ yr⁻¹ (N60); (3) high N, 90 kg N ha⁻¹ yr⁻¹ (N90); and (4) N-free condition (control). Three replicate plots per treatment were established. The original surface soil characteristics of the study plots are shown in table 1. Beginning in January 2013 and each month thereafter, quantitative NH₄NO₃ was dissolved into 10 L of water and then sprayed evenly onto the ground of each plot for a total of 12 equal applications over the entire year. A more detailed description of the N treatment is given in Song et al (2015). In September 2014, (at 21 months of N addition treatment), two subplots of 10 × 10 m were established in each N deposition plot and the control plot. Each subplot was isolated using a plastic panel with a depth of 1 m, to avoid disturbing the nearby area. The two subplots were amended once with either 20 t ha⁻¹ biochar (BC20) or 40 t ha⁻¹ biochar (BC40). Biochar was applied via top soil incorporation (Verheijen et al 2009): biochar was homogenously spread across the plot and then ploughed in, to mix well with the top 30 cm of soil. All plots still received the quantitative N addition monthly as described above.

In July 2015, ten soil cores (0–20 cm depth) were collected randomly using a soil auger (3.5 cm) from each subplot, mixed until homogenized, and transported to the laboratory in an incubator. The samples were sieved through a 2 mm sieve, and subsamples were stored at -80 °C until further molecular analysis. The remaining subsamples were used to determine the MBC and the physicochemical properties of the soil.

Soil pH was measured using a pH meter (FE20, Mettler Toledo, Switzerland) after preparing a soil:water (1:2.5 w v⁻¹) suspension by shaking for 30 min (Bao 2008). The soil organic carbon (SOC) and total nitrogen (TN) in the extracts was determined using an elemental analyzer (ElementarVario EL III, Germany). Available phosphorus (AP) was analyzed using the molybdenum blue method (Watanabe and Olsen 1965). NO₃⁻ and NH₄⁺ were extracted using 2 M KCl, steam distillation, and titration (Mulvaney 1996). MBC was estimated using the chloroform fumigation-extraction method (Vance et al 1987).

High-throughput sequencing of DNA from the soil samples was used to identify the species present and the functional diversity of microbial populations; this method provides a more detailed description of microbial communities than traditional methods, such as cloning libraries (Roesch et al 2007).

Total DNA was extracted from 0.25 g of soil using the Ezup Column Soil DNA Purification Kit (Sangon Biotech, Shanghai, China) according to the manufacturer's instructions. DNA concentrations were then determined using a Nano Drop ND-1000 UV-Vis spectrophotometer (Thermo Scientific, Rockwood, TN, USA).

The V4 hypervariable regions of bacterial 16S rRNA genes were selected for amplification using the primer pair 515 f/806r (Zhou et al 2015). PCR conditions and products were as described previously (Li et al 2016). The 16S rRNA amplicons were sequenced using the Illumina Miseq platform at Novogene (Beijing, China).

Sequence analysis was carried out using the software QIIME (Quantitative Insights into Microbial Ecology) as described by Fadrosh et al (2014). Briefly, sequence reads were extracted, primers were removed, and the sequences were trimmed to remove low quality sequences. The trimmed and unique sequences were used to define the number of operational taxonomic units (OTUs) at the 97% similarity level. Shannon and Simpson diversity indices and Chao1 and ACE richness estimators were calculated using the open-source software mothur (Schloss et al 2009) to estimate bacterial diversity and richness.

Differences in the concentrations of MBC and in the Shannon index among treatments were tested using one-way analysis of variance (ANOVA) and post-hoc multiple comparisons were conducted using the Bonferroni method. The combined effect of N deposition and biochar amendment on MBC and diversity was examined using two-way ANOVA and the Bonferroni test. The relationship between soil properties and α-diversity was investigated using Pearson correlation coefficients. Data analyses were conducted using SPSS v 18.0 for Windows (SPSS Inc., Chicago, Illinois).

Principal coordinate analysis and the unweighted pair group method with arithmetic mean (UPGMA) clustering were conducted on the weighted UniFrac based on a previously published protocol (Kuczynski et al 2012). In addition, multiple correlations between environmental variables (SOC, TN, C/N rate, AP, NO₃⁻, NH₄⁺, and pH) and community composition were analyzed using a redundancy analysis in Rv.2.8.1 with the vegan package.

Zoom In Zoom Out Reset image size

Compared with the control, soil MBC was 22.3% higher in the BC20 treatment and 17.5% lower in the BC40 treatment (P < 0.05) (figure 1). In plots without biochar amendment, MBC was significantly higher in the N addition (N30, N60, and N90) treatments than in the control. Soil MBC was significantly lower in the combined N addition (N30 and N60) and biochar amendment (BC20 and BC40) treatments than in the N addition treatments alone (P < 0.05). MBC did not significantly change in the combined N90 and BC20 or BC40 treatments compared with that in N90 treatment alone (P > 0.05). Soil MBC was significantly higher in the combined BC20 and N30 or N60 treatments than in the combined BC40 and N30 or N60 treatment (P < 0.01). A two-way ANOVA showed biochar amendment and N addition significantly affected soil MBC, both independently and in combination (P < 0.001) (table S1 available at stacks.iop.org/ERL/13/044029/mmedia).

More than 38 000 valid reads were obtained for each replicate treatment. The median sequence length of each read was 253 bp. A total of 59 352 OTUs was detected using 97% identity as the cutoff, representing 45 phyla and 470 genera. At the phylum level, the dominant phyla were Proteobacteria (38.6%), Acidobacteria (23.4%), Actinobacteria (8.3%), and Verrucomicrobia (7.6%) (figure 2). Twenty-six phyla occurred with relatively low abundance and accounted for less than 1% of the total. At the genus level, the dominant genera were Acidibacter (3.6%), Candidatus Nitrosotalea (2.6%), Acidothermus (2.6%), Rhizomicrobium (2.3%), and Sphingomonas (2.2%).

Zoom In Zoom Out Reset image size

In the N-free conditions, the relative abundance of Acidobacteria and Chloroflexi was lower in the BC20 and BC40 treatments than in the control, whereas Gemmatimonadetes and Thaumarchaeota exhibited the opposite trends. The relative abundance of Proteobacteria and Actinobacteria was higher in the BC20 treatment than in the control, but was lower in the BC40 treatment. In the no-biochar-amendment condition, N addition increased the relative abundance of Crenarchaeota and Actinobacteria but decreased the relative abundance of Proteobacteria.

The relative abundance of Proteobacteria was higher in the combined N addition (N30 and N60) and biochar amendment (BC20 and BC40) treatments than in the N addition treatments alone, whereas Actinobacteria and Gemmatimonadetes exhibited the opposite trends. The relative abundance of Proteobacteria and Verrucomicrobia was higher in the combined N90 and BC20 than in the N90 treatment alone, whereas it was lower in the combined N90 and BC40 treatment. The relative abundance of Actinobacteria was higher in the combined N90 and BC20 or BC40 treatments than in the N90 treatment alone, whereas Gemmatimonadetes exhibited the opposite trend. The relative abundance of Acidobacteria in combined N addition (N30, N60, and N90) and biochar amendment (BC20 and BC40) treatments was different from that in the N addition treatments alone.

The relative abundance of Proteobacteria was higher in the combined BC20 and N-free or N30 treatments than in the combined BC40 and N-free or N30 treatments, whereas Acidobacteria abundance showed the opposite trend. The relative abundance of Acidobacteria was higher in the combined BC20 and N60 or N90 treatments than in the combined BC40 and N60 or N90 treatments, whereas Proteobacteria abundance showed the opposite trend. The relative abundance of Actinobacteria was lower in the combined BC20 and N addition treatments than in the combined BC40 and N addition treatments.

The Shannon index in the BC20 and BC40 treatment was significantly higher than in the control (P < 0.05). In the no-biochar-amendment condition, N addition significantly decreased the Shannon index (P < 0.05), particularly in the N60 treatment. The Shannon index was significantly higher in the combined N addition (N30, N60, and N90) and biochar amendments (BC20 and BC40) treatments than in the N addition treatments alone (figure 3).

Zoom In Zoom Out Reset image size

A cluster analysis of the similarity of the soil microbial communities showed two groups of bacterial communities (figure 4). N60-BC20, N0-BC40, N60-BC0, N30-BC0 and N0-BC0 treatments clustered into group I, and the other N addition treatments clustered into group II, irrespective of biochar amendment. This clustering was further confirmed by a principal coordinate analysis (figure S2). The combination of N addition and biochar amendment accounted for 46.73% of the variation in microbial community structure; N addition accounted for 28.56% of the variation, and biochar amendment accounted for 18.17%. Two-way ANOVA showed biochar amendment and N addition significantly affected soil bacterial diversity independently and in combination (P < 0.05) (table S1).

Zoom In Zoom Out Reset image size

Soil properties are provided in figure S1. The relationships between the α-diversity of the soil microbes and soil properties are shown in table 2. The Shannon index was significantly, positively correlated with pH, and significantly and negatively correlated NH₄⁺ concentrations (P < 0.05). The Shannon and Chao1 indices were significantly and negatively correlated with SOC, C/N ratio and AP concentrations (P < 0.05). The soil MBC was significantly and positively correlated with C/N ratio and AP concentrations (P < 0.05).

A redundancy analysis was used to analyze the correlation of the main soil properties and the microbial community (figure 5). Of the evaluated attributes, AP, pH, and NO₃⁻concentrations were significant conditional effects, while NH₄⁺, C/N ratio, TN, and SOC concentrations had indicative effects on the community.

Zoom In Zoom Out Reset image size

Our results demonstrated that biochar amendment significantly increased microbial diversity, moderate biochar amendment (20 kg ha⁻¹) significantly increased the soil MBC, but higher levels of biochar amendment (40 kg ha⁻¹) significantly decreased MBC. These results partially support our first hypothesis that biochar amendment increases soil MBC and bacterial diversity. Some studies have reported that soil microbial diversity is closely correlated to soil pH (Nicol et al 2008, Lauber et al 2009). Our results here also suggested that pH was significantly and positively correlated with soil bacterial diversity, which was higher after biochar amendment relative to the control (figure S1(g)). This conclusion is supported by previous studies that Anthrosols typically have higher pH and microbial diversity relative to adjacent soils in the Brazilian Amazon (O'Neill et al 2009). Some studies found that SOC was significantly and positively correlated with soil MBC (Chen et al 2011, Zhou et al 2014). However, a Pearson correlation analysis showed that MBC was not significantly or positively correlated with SOC (P > 0.05) but was significantly and positively correlated with C/N ratio. We found that SOC and C/N ratio significantly decreased after biochar amendment relative to the control (P < 0.05) (figures S1(a) and (c)), which supported the interpretation that BC40 significantly decreased MBC. It has also been reported that biochar amendment can decrease soil MBC (Lehmann et al 2011, Dempster et al 2012). For example, Dempster et al (2012) found that soil MBC decreased with increasing rate of addition of jarrah (Eucalyptus marginata) wood biochar, possibly as biochar input decreased the mineralization rate of the SOC. However, our results did not explain why BC20 increased soil MBC. We speculate that addition of soil MBC is related to other physiochemical properties of soil and unique properties of biochar. For example, the nutrients in the biochar components (abundant K) may have provided an energy source for microorganisms (Pietikäinen et al 2000, Watzinger et al 2014). Previous studies also indicated that the large surface area and porous structure of biochar create a favorable habitat for microbial colonization (Pietikäinen et al 2000, Ezawa et al 2002, Saito and Marumoto 2002, Luo et al 2013). Kolb et al (2009) found that charcoal addition improved the soil surface area, promoted soil microbial growth, and thus significantly increased soil microbial biomass and activity. Zavalloni et al (2011) also observed that biochar absorbed toxic and harmful substances in the soil, indirectly increasing soil MBC. In addition, biochar also improved the water holding capacity and CEC (Warnock et al 2007, Wang et al 2012, Maestrini et al 2014, Perry et al 2015). These alterations to soil biochemical reaction conditions stimulated soil enzyme activities, produced more chemicals (Lee et al 2010), and promoted the growth and reproduction of soil microorganisms, thereby improving soil MBC (Wu et al 2015). This suggests that the effects of biochar amendment on soil microbes are controlled by multiple environmental factors and that the direction and magnitude of the effects depend on the biochar amendment rate.

The relative abundance of Proteobacteria and Actinobacteria was higher in the BC20 treatment than in the control and BC40, whereas Acidobacteria had a lower abundance. This demonstrated that biochar amendment altered soil microbial composition in Moso bamboo forests, which concurs with Khodadad et al (2011) and Perry et al (2015). Jaiswal et al (2017) observed a significant decrease in Acidobacteria abundance in the biochar-enriched soil rhizosphere. Acidobacteria generally prefer oligotrophic environments and lower pH soils; therefore, it is possible that the reduction in Acidobacteria was a response to the biochar-related shift to a more neutral and copiotrophic environment (Jaiswal et al 2017). Xu et al (2014) found that biochar amendment in an acidic soil (pH 4.48) changed the relative abundance of some microbes that are related to C and N cycling. Sun et al (2016) showed that the relative abundance of Actinobacteria in biochar pellets was higher than that in adjacent control soils. In general, Actinobacteria (especially Actinomycetales) can depolymerize polyphenols in leaf litter, converting them into small soluble molecules; thus, they play an important role in the decomposition of organic matter (Zhao et al 2015, Dai et al 2017). The effects of biochar amendment on soil microbial community structures are complex and variable, and are closely related to time, and types and rates of biochar amendment and soil properties (Wu et al 2015, Dai et al 2017).

N addition alone (30, 60, or 90 kg N ha⁻¹ yr⁻¹) significantly increased soil MBC but significantly decreased soil bacterial diversity (figures 1 and 3). After BC20 and BC40 amendment, MBC decreased by 6.9%–25.7% and 5.1%–45.8%, respectively, compared with N addition treatments alone. Moreover, the significant positive effects of N addition on MBC disappeared after biochar amendment of N90 plots. This could be explained by a decreased C/N ratio and SOC in the combined N addition (N30 and N60) and biochar amendment (BC20 and BC40) treatments relative to the N addition treatments alone (figure S1(a) and (c)). In addition, biochar amendment reversed the reduction in soil microbial diversity caused by N addition (figure 3). One possible explanation may be that biochar created a favorable habitat for microbial colonization due to the large surface area and porous structure of biochar and its physicochemical properties, such as alkalinity, which might improve pH, water holding capacity and CEC (Jaafar et al 2015). These results support our second hypothesis by suggesting that biochar amendment could offset the positive effects of N deposition on MBC and the negative effects on bacterial diversity. Biochar amendment of N addition plots did not significantly change the dominant microbial phyla. The reason for this may be that N addition had a greater effect on soil microbial composition than biochar amendment, which had been ongoing at the study site for 21 months prior to the biochar amendment experiment, whereas the biochar amendment was conducted only for 10 months. The cumulative effects of biochar amendment on soil microbes may not have been evident in the short duration of the experiment.

MBC was significantly higher after BC20 amendment of N30 and N60 soils compared to the BC40 amendment of these soils, indicating that the higher biochar amendment rate resulted in a stronger mitigating effect. This finding supports our third hypothesis. One possible reason for this effect is that biochar amendment changes the soil environment in a dose-dependent manner (Dempster et al 2012). Although the microbial components varied under the combined biochar and N treatments, external characteristics were less variable compared with MBC.

Previous studies observed that soil microbial community structure and microbial diversity were closely correlated to soil pH (Nicol et al 2008, Lauber et al 2009, Kuang et al 2013). Our results here also demonstrated that the bacterial diversity and community composition were correlated with soil pH. The bacterial community composition was mainly correlated with AP and NO₃⁻ concentrations. Zhao et al (2014) also observed that Betaproteobacteria and Chloroflexi are positively correlated with soil AP content. Belén et al (2014) reported that the bacterial community structure is closely correlated with the NO₃⁻ concentration, whereas Zhou et al (2015) found that nitrate concentration might shape microbial communities in cultivated black soils.

Changes in microbial community structure might influence microbial diversity (Fang and Mo 2006). In this study, the C/N ratio affected microbial community structure and, thus, were significantly correlated with bacterial diversity. A similar result was reported by Zhou et al (2015). Chu et al (2010) observed that the C/N ratio is significantly correlated with the relative abundance of dominant phyla in Arctic tundra soils, consistent with the abundance and composition of soil bacterial communities in the Changbai Mountains (Shen et al 2013). Our results indicate that C/N ratio and AP concentration are the key controlling factors for MBC, while pH, C/N ratio, and AP concentration are the key controlling factors for diversity and composition of microbial communities in Moso bamboo forest.