Plant Genus (Acacia and Eucalyptus) Alters Soil Microbial Community Structure and Abundance Within Revegetated Shelterbelts

Abstract

The composition of native woody vegetation communities in restored agricultural landscapes can alter plant-soil interactions and thus ecological function. This study investigated the effects of primary species groups (Acacia, Eucalyptus) used in shelterbelts in south-eastern Australia on soil microbial community structure in comparison with those under adjacent pastures. The quantity and structure of bacterial and fungal communities was analysed using quantitative PCR (qPCR) and terminal restriction fragment length polymorphism (TRFLP). The fungal:bacterial ratio was highest under Eucalyptus trees), but did not differ between Acacia and pasture. Both fungal and bacterial communities varied at different depths in the soil profile and across sampling periods. The dominant tree genus was a significant factor in observed differences in bulk soil bacterial and fungal communities compared to the surrounding pasture. The effect of tree type on fungal communities was more pronounced than on bacterial communities. Fungal communities not only differed between shelterbelts and pastures, but also between Acacia and Eucalyptus dominated shelterbelts and these patterns were consistently observed at different soil depths. Shelterbelt systems which contain a mixture of different plant genera have the potential to exhibit greater microbial community heterogeneity than single species plantations. Maintaining a variety of tree species during afforestation may therefore contribute to conserving important bacterial and fungal diversity in agricultural landscapes.

Key words: Acacia, Eucalyptus, shelterbelts,soil microbial communities, T-RFLP, agricultural restoration

1. Introduction

In south-eastern Australia, strips of planted native trees and shrubs (shelterbelts) are frequently established to restore ecosystem services altered by agriculture (Cleugh et al., 2002; Hobbs, 1993) or to reduce land degradation (e.g. soil erosion) (Bird, 1998; Bird et al., 1993). The establishment of shelterbelts contributes to the conservation of biodiversity, in particular for mammals, birds and amphibians (Lindenmayer and Hobbs, 2004). Along with biodiversity and other benefits, conversion from agriculture to forest systems affects soil physical and chemical properties. This is particularly well studied for soil carbon storage (Hoogmoed et al., 2012; Paul et al., 2018; Paul et al., 2002; Paul et al., 2015). Much less attention has been given to how soil bacterial and fungal community structure and function are impacted by the identity of the tree species used, or the restoration process itself (Cavagnaro et al., 2016). However, understanding the key drivers that influence microbial community composition is an important element of managing the effects of land use change.

Interactions between above ground and below ground components of ecosystems has received increasing attention as awareness grows of their importance in driving ecosystem processes (Bardgett and Wardle, 2010; Bardgett et al., 1998; Wardle et al., 2004). One area of particular interest is the effect of plant community structure on soil microbial community structure and function (Prescott and Grayston, 2013). Tree species that occupy a site may alter the rate of fundamental soil processes, such as nutrient cycling and carbon dynamics, through differences in plant microbial associations, amount and quality of leaf and root litter, and root exudates (Bauhus et al., 1998; Hobbie, 1992)

These differences can control the composition and function of soil communities (Bardgett and Wardle, 2010) and more specifically microbial communities (Zak et al., 2003). For example, variation in microbial communities in bulk forest floor samples can be related to different associated tree species. These differences are largely due to physico-chemical properties of plant litter from different tree types (Ndaw et al., 2009; Schutter and Dick, 2001) as is clearly illustrated for comparisons between conifers and broadleaved species (Bauhus et al., 1998; Gartzia-Bengoetxea et al., 2016; Hagen-Thorn et al., 2004; Thoms et al., 2010).

The total amount of organic matter input is highest in the topsoil and this is the soil stratum in which the largest observed effects of tree species on soil chemical properties are found (Augusto et al., 2003; Hagen-Thorn et al., 2004; Thoms et al., 2010). Lejon et al. (2005) found distinct bacterial and fungal community profiles in mineral soil (0–5, 10–20 cm depth) in pure plots of spruce, Douglas fir, oak and beech in France. This is supported by (Jiang et al., 2010) who found increased catabolic diversity (CLPP, community level physiological profiles) and distinct bacterial and fungal communities in soil beneath broadleaf and mixed species plantations compared with soil beneath conifers in the top 0-10 cm.

Much of the previous work examining the effects of plant species on microbial communities has focused on comparisons between conifer and broadleaf species. In Australia, two of the most widespread and ecologically important genera are Acacia and Eucalyptus, which form dominant components of many ecosystems. Importantly, these groups are also widely used in land reclamation and restoration activities, including in the establishment of shelter belts in agro-ecological landscapes. The quality of Acacia and Eucalyptus litter varies greatly (Hoogmoed et al., 2014b). Eucalyptus litter tends to have low nitrogen and high phenolic content, while Acacia tends to have higher N concentrations (Hoogmoed et al., 2014a). These differences, along with associated dominant symbiotic relationships, i.e., Acacia sp. with rhizobial N-fixing bacteria (Burdon et al., 1999; Thrall et al., 2000) and Eucalyptus spp. with mycorrhizal fungi (Warcup, 1980) may foster different microbial communities within the bulk soil and thus impact on ecosystem function.

Disentangling the effect of different plant species on soil microbial communities is further complicated by microbial interactions with soil edaphic properties. Recent studies of the biogeography of soil microorganisms have demonstrated strong relationships between the soil microbial community diversity and structure and factors such as pH, texture, organic matter content and C:N ratio of the soil (Bissett et al., 2010; Brockett et al., 2012; Fierer and Jackson, 2006; Fierer et al., 2009; Rousk et al., 2010; Siciliano et al., 2014).

In order to better understand plant-soil interactions within restored agricultural landscapes we investigated the effects of dominant plant type on soil microbial communities using shelterbelts containing Acacia and Eucalyptus trees adjacent to grazed pastures. We hypothesised that:

1) soil microbial communities would differ under different tree types, with an increase in the fungal:bacterial ratio under Eucalyptus relative to Acacia due to dominant symbiotic relationships;

2) soil microbial communities and soil edaphic properties at different depths under each tree genus would also differ, and that;

3) changes in microbial community composition would correspond to changes in soil chemical properties driven by differences in plant type.

2. Methods

2.1 Study sites

The study sites were located near Murrumbateman in New South Wales, Australia (34°57′0″S, 149°01′0″E). The region is predominantly agricultural land dominated by sheep grazing. The climate is temperate, with warm to hot summers and cool winters. The average annual rainfall is 627 mm (Australian Bureau of Meterology, 2014). Rainfall is highest in spring (October and November) and lowest in winter (June and July).

2.2 Soil Sampling

Two shelterbelts (strips of planted native trees and shrubs sown in rows) were established in 1995 by property owners, one dominated by Acacia species, the other by Eucalyptus species. The shelterbelts were approximately 100 metres apart. The shelterbelts and adjacent pastures were sampled for soil microbial communities in October 2011 and April 2012. The Eucalyptus shelterbelt was comprised mostly of Eucalyptus macarthurii, Eucalyptus blakelyi, Eucalyptus viminalis and Eucalyptus melliodora. The Acacia shelterbelt included Acacia mearnsii, Acacia baileyana, Acacia decurrens, Acacia cardiophylla and Acacia cultriformis. The two shelterbelts ran parallel to each other (approximately 100m apart). Within each of the shelterbelts, the soil under six trees was sampled at 10 metre intervals along transects (Acacia spp. N=6 trees, Eucalyptus spp. N=6).

Directly under each sampled tree, at 4 compass points equidistant from the trunk (50 cm out), soil cores (30cm deep x 5cm wide) were taken in the mineral soil. Soil for each individual tree was then bulked by soil depth, 0-10 cm, 10-20 cm and 20-30 cm (Fig. S1). In 2012 all 6 trees were sampled at 10-20 cm and 20-30 cm. However in 2011, only 3 trees of each genus were sampled at 10-20 cm and 20-30 cm due to resource constraints. The four samples collected from each depth interval per tree were bulked together. In 2011, we also sampled between tree rows. Two inter-row samples were collected per soil depth interval in the open space next to each sampled tree (Fig. S1). The two samples per depth interval were bulked. Within the pastures, transects were laid out, parallel to but approximately 15 metres away from each shelterbelt. Soil cores were taken in the mineral soil at at the same depths as in shelterbelts. A total of three pasture composite samples were taken per depth interval from each of the pasture transects (n= 6 pasture samples per sampling period).

2.3 Soil analysis

Soil moisture content was determined gravimetrically by oven drying 10 g subsamples at 105°C until weight became stable. After the removal of a subsample for molecular analysis the remainder of the bulked soil samples were air-dried and pH and electrical conductivity (EC) were analysed in a 1:5 water suspension which was shaken end over end for 1 hr, centrifuged down and read immediately with a Thermo Scientific Orion 3-Star portable meter. Total carbon (C) and total nitrogen (N) were determined on an oven dried basis on finely ground soil using the dry combustion method on an elemental analyser (vario MAX CNS). The remaining parameters [particle size (sand, silt and clay), organic C, nitrate N (NO₃-N), ammonium N (NH₄⁺-N), sulphur (S), phosphorus (P), potassium (K),  Exchangeable cations (Ca, Mg, Na, K, Al) and trace elements (Cu, Fe, Mn Zn)] were determined by the CSBP Soil and Plant Analysis Laboratory (Perth Australia). Organic C was approximated by the Walkley Black method (Walkley and Black, 1934) P and K measured according to Colwell (1963), S was extracted using the Blair/Lefroy method (Blair et al., 1991), and NO₃-N and NH₄⁺-N were extracted with 1M potassium chloride and measured on a Lachat Flow Injection Analyser. DPTA trace elements were measured by atomic absorption spectroscopy (Rayment and Higginson, 1992).

2.4 DNA extraction

The 108 individual soil composite samples were frozen at -20° C until analysis. Total soil DNA was extracted from 0.25 g subsamples using the Power soil DNA extraction Kit (MoBio Laboratories, Carlsbad, California, USA), with the following modification to manufacturer’s instructions; samples were shaken in a Bio101 Savant bead beater on the maximum speed for 2 min instead of vortexing (Bissett et al., 2011). DNA quality was assessed on an agarose gel and DNA quantity was determined using PicoGreen dsDNA Quantitation Reagent (Invitrogen). This DNA was used as a template for T-RFLP and qPCR analyses of bacterial and fungal communities.

2.5 Real time qPCR for Fungi and Bacteria Abundance

The abundances of bacterial 16S and fungal ITS rRNA genes were determined using quantitative real time PCR (qPCR) as described in Fierer et al. (2005). Primers used for bacterial ITS were Eub338 (Lane, 1991) and Eub518 (Muyzer et al., 1993) and for fungal ITS primers were 5.8s and ITS1F. The qPCR assays were performed in triplicate 10µl reactions containing 5 µl SsoAdvancedTM SYBR® green 246 Supermix (Bio-Rad, Hercules, CA), 0.4 µl of forward and reverse primers (10 µM), 2.2 µl H₂O and 2 µl template DNA. Quantification of bacterial 16S rRNA genes was carried out under the following conditions, 95^oC for 2 min followed by 95^oC for 10s, 53^oC for 30s and 72^oC for 20s. Fungal quantification was as follows: 95^oC for 1 min followed by 95^oC for 5s, 53^oC for 20s and 72^oC for 20s.  40 cycles were run for bacteria and fungi, followed by a melting curve to confirm the amplified products were of the appropriate size. For each plate a standard curve was run with a 10 fold dilution series from 10⁹ copies to 10⁴ copies. Standards were prepared from PCR product generated using the primers above (from Pseudomonas fluorescens strain 5 and Fusarium oxysporum vasinfectum),and purified using the Agencourt PCR purification system. Standards were quantified using Picogreen (Invitrogen, Grand Island, NY, USA). Amplification efficiencies were > 90% and R² values were > 0.99 for bacterial and fungal calibration curves.

2.6 T-RFLP

Bacterial and fungal soil communities were analysed using terminal-restriction fragment length polymorphism (T-RFLP) (Thies, 2007). Amplification of bacterial 16S rRNA genes was performed using the 6-carboxyfluorescein (FAM)-labelled primer FAM27f (5′-AGTTTGATCMTGGCTCAG-3′)and primer BAC519r (5′-GWATTACCGCGGCKGCTG-3′). Fungal internal transcribed spacer (ITS) region was amplified from soil DNA using fungal primers: 6-carboxyfluorescein (FAM)-labelled primer FAMITS1-F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) (Gardes and Bruns, 1993) and ITS4 (5-TCCTCCGCTTATTGATATGC-3′) (White et al., 1990). 50 μL PCR reactions contained 2 ng of extracted DNA, 1 U of MyTaq DNA Polymerase (bioline) and 1x supplied buffer, and 0.5 µM of each of the oligonucleotide primers. Fungi PCR mix was as above with the addition of 0.5 µg BSA (Promega). Bacterial PCRs were run at 94°C for 5 min, 40 cycles of 94°C for 30s, 58°C for 30s and 72°C for 45s followed by a final extension step at 72°C for 7 min.

We confirmed amplification in a 1% agarose gel, the PCR product was then purified using the Agencourt AMPure XP system, according to the manufacturer’s instructions. Approximately 25ng of the FAM-labeled PCR products were digested with the restriction endonucleases HinfI, in a volume of 25µLthat included 5 U of HinfI and the supplied buffer at 1X. Restriction digests were incubated at 37 °C for 4 h. Digested products were isopropanol precipitated, resuspended in 10µL of formamide containing LIZ600 size standard (Applied Biosystems), then denatured at 95°C for 2 min. T-RFLP profiles were generated by separating the fragments on an ABI 3130xl Genetic Analyser (Applied Biosystems, UK).

Terminal-restriction fragments (T-RFs) generated by the sequencer were exported and initially analysed for absolute peak areas using GeneMapper™ 4.0 (Applied Biosystems, UK). First, T-RF profiles were checked for stable baselines, voltage and calibrations and profiles were trimmed to between 50 and 600bp to eliminate T-RFs caused by primer-dimers while retaining fragments within the linear range of the internal size standard. Data were then uploaded to the online TRFLP analysis software package T-REX (http://trex.biohpc.org/; (Culman et al., 2009)). Noise filtering was carried out by removing peaks whose area was less than twice the standard deviation computed over all peaks (Abdo et al., 2006). T-RFs were aligned where peaks with values less than the threshold (1 bp) were identified and grouped into a T-RF (Smith et al., 2005). The relative abundance of a detected T-RF within a given T-RFLP profile was calculated as the respective signal area of each peak divided by the total peak area of all the peaks of the T-RFLP profile.

2.7 Statistical analysis

An analysis comparing soil properties and microbial communities under trees compared to inter-row areas was conducted on the 2011 data using two way ANOVA and PERMANOVA. Initial analyses showed that inter-row soil samples were not significantly different to those taken from under trees, thus inter-row and under-tree data were combined for the 10-20cm and 20-30cm depths for 2011. This resulted in equal numbers of replicates between sampling years (n=6 samples per depth interval per genus). Statistical analysis followed a factorial design [site type (Acacia, Eucalyptus, pasture)*soil depth*year] for all analyses where possible. ANOVA was used to test for significant differences in soil edaphic properties using the design above, followed by Tukey HSD post hoc tests for differences between groups. When necessary, data was log transformed to achieve homogeneity of variances and normality prior to analysis. All statistical tests were performed at P=0.05.

Multivariate analysis was used to assess differences in microbial community composition. Relative abundances of TRFs identified were used to produce Bray Curtis similarity matrices based on samples without the need for any transformations or normalisations. Non-diversity data were normalised and Euclidean distances used to calculate distance matrices. Transformation was necessary for some environmental variables as indicated from a draftsman plot (in these cases data were log transformed). ANOSIM and PERMANOVA (Anderson, 2001) were used to test the significance of site type, depth and sampling year on differences in community structure. General differences between soil microbial communities were visualised using non metric multidimensional scaling (nMDS) and canonical analysis of principle coordinates (CAP) (Anderson and Willis, 2003) computed from the similarity matrices.

Distance Based Linear Models (DistLM) (Legendre and Anderson, 1999) were used to test for differences in community composition and how these related to predictor environmental variables. For DistLM, co-varying edaphic variables (r²>0.75) were found and only one such variable was run in the model. Results were interpreted such that this single variable represented the behaviour of all variables with which it was found to co-vary. The stepwise model selection procedure (Anderson et al., 2008) was used to identify variables that explained significant amounts of variation in bacterial and fungal communities. Corrected Akaike information criterion (AICc) was used to select the predictor variables. Marginal tests were conducted to assess the statistical significance and percentage contribution of each soil edaphic variable alone (Anderson and Cribble, 1998; Legendre and Anderson, 1999; McArdle and Anderson, 2001). Distance-based redundancy analysis (dbRDA) was used for graphical visualization of the DistLM results. Statistical analysis was conducted using GenStat software 16^th edition package (VSN International, 2011) and multivariate tests were conducted using PRIMER 6 (Clarke and Gorley, 2006; Clarke and Warwick, 2001) and PERMANOVA + (Anderson et al., 2008).

3. Results

3.1 Soil edaphic properties

Site type (Acacia, Eucalyptus, pasture) was significant for all soil properties except C:N ratio and Ca (Table S1). Soil depth alone was also significant for all soil properties. Soil chemical properties such as total C, N, P declined and C:N ratio and pH increased with increasing soil depth across all site types. Interaction effects of site type*depth and treatment*year showed more variability in relation to interactions with measured soil properties (Table S1). Pairwise comparisons (Table 1 & 2) showed a significant difference between Acacia and Eucalyptus soil within the top 0-10cm for carbon density, P, K, Cu, Fe, Zn, Al, Mg, Na, NO₃ and S. No significant difference was observed between Acacia and Eucalyptus soils at 0-10cm for total N and C (P>0.05), however the C:N ratio was significantly lower under N-fixing Acacia (12:2) compared to Eucalyptus (13:1). The pasture soils had a significantly lower C:N ratio (11:2) than either Acacia or Eucalyptus. Below 10cm, treatment effects were reduced for most major soil edaphic properties (Fig. 1). PCA analysis showed a shift primarily along PC1 in pH and bulk density, which separated the top 0-10cm soils from samples deeper in the soil profile (Fig. 1).

3.2 Fungal:bacteria ratio

F:B ratios were primarily influenced by site type (P=0.002) and year (P<0.001) (Fig. 2).  The F:B ratio was highest under Eucalyptus trees, but did not differ between Acacia and pasture (Fig. 2). Across years, the F:B ratios were overall higher in 2011 than in 2012.

3.3 Bacterial and fungal TRFLP community profiles

The composition of fungal and bacterial communities as determined by T-RFLP differed significantly between site types (P=0.001), soil depth (fungi: P=0.002 fungi; bacteria: P=0.001) and year (P=0.001) (Table 3, Fig. 3). To investigate treatment effects further, bacterial and fungal community profiles were analysed separately for each sampling year (Table 4) as bacterial and fungal community data indicated treatment*year effects (Table 3). Fungal communities under all treatments differed from each other across both sampling years (2011 and 2012: P=0.001) (Table 4, Fig. 4). Differences between fungal communities were also influenced by depth*site type interactions where within each depth profile fungal community composition between site type differed, however there was no significant difference in composition between soil depths for each site type (Table 4, Fig. 4).

There were no significant differences in the composition of soil bacterial communities under Acacia versus Eucalyptus within either sampling year (Table 4). However in 2012 bacterial community composition under Acacia and Eucalyptus trees differed to those in the pasture (Table 4). There was a significant interaction between depth and year for bacterial communities (Fig. 5). Bacterial community composition between soil depths varied within each year, as well as within the same soil depth across years.

3.4 Soil edaphic properties and bacterial and fungal communities

The relationships between soil bacterial and fungal communities were assessed using DistLM and visualised with dbRDA. Co-varying variables (r>0.75) removed included total N, P, NO₃-N, NH₄₊-N, Fe, Mn, Zn, Al and EC. Marginal tests indicated that Cu (8.03%, p=0.001), Na (7.62%, p=0.001), Mg (6.23%, p=0.001) most strongly correlated to fungal community composition on their own (Table 5). In contrast, total C (21.45%, p=0.001), K (14.11%, p=0.001), pH(14.06 %, p=0.001), Mg (7.86%, p=0.001) and C:N ratio(7.51%, p=0.001)most strongly correlated to bacterial community composition (Table 6). Using AICc and a stepwise procedure for fungi, six soil physiochemical variables, Cu, silt, K total C, pH and Mg accounted for 26% of the variation in fungal community composition. Fungal communities primarily separated out by Cu and silt – these variables also differentiated pasture communities from those under Acacia and Eucalyptus (Fig. 6). For bacteria, four soil physiochemical variables, total C, pH, moisture, and Cu accounted for 30% of the variation in community composition (Fig. 6). Bacterial community composition primarily separated out by total C and pH which differentiated communities in the 0-10 cm layer from those at depths >10cm.

4. Discussion

Overall, our results demonstrate that microbial community composition can be significantly influenced by the presence of different dominant tree species in revegetated shelterbelt systems, and that these communities in turn differ from those occurring in adjacent pastures. However the effect of tree type on fungal communities was more pronounced than on bacterial communities. Our original hypothesis that F:B ratios would be higher under Eucalyptus than under Acacia due to dominant symbiotic relationships was supported. Plant typeeffects within shelterbelts were most dominant within the first 0-10 cm of the soil profile, after which microbial communities were influenced to a greater extent by edaphic properties. This is consistent with other studies showing that organic inputs and plant effects are greatest in the upper soil stratum (Hoogmoed et al., 2012; Li et al., 2018).

With regard to revegetation using woody plants, soil development has been shown to be largely influenced by the dominant tree species which includes plant inputs such as the quantity and quality of the litter (Šnajdr et al., 2013). For example, a meta-analysis by Paul et al. (2002) quantifying changes in soil C under large-scale afforestation or reforestation found that the most important factors affecting change in soil C were previous land use, climate and the type of forest established. Most soil C was lost when softwoods, particularly Pinus radiata plantations, were established on previously improved pastoral land in temperate regions. Accumulation of soil C was greatest when deciduous hardwoods, or N₂-fixing species (either as an understorey or as a plantation), were established on previously cropped land in tropical or subtropical regions. With respect to changes in soil C following afforestation, on average, the <10 cm (or <30 cm) layers generally decreased by 3.46% per year (or 0.63% per year) relative to the initial soil C content during the first 5 years, followed by a decrease in the rate of decline and eventually recovery to C levels found in agricultural soils at about age 30.

In our study, we observed no significant differences in total N, K or C between pasture and Eucalyptus despite 16 years since establishment. This may indicate that the decline in properties may have reached its peak and nutrients are starting to recover. Mendham et al. (2003) and O’connell et al. (2003) also found soil organic C and N were not significantly influenced by afforestation 7–11 years after plantation establishment in south-western Australia.

We were also unable to detect significant differences in total N and C levels between Acacia and Eucalyptus stands. There was  however a trend of increasing N, NO₃-N and NH₄⁺-N under Acacia and a slight increase in C, resulting in a significantly lower C:N ratio in the top 0-10cm under Acacia as compared to Eucalyptus. This may be indicative of higher N concentrations in Acacia litter and increased N in root exudates from legumes (Warembourg et al., 2003) resulting in increased available soil N (Hoogmoed et al., 2014a). It is also possible that a lower C:N ratio under N-fixers may favour bacterial over fungal decomposers (Fierer et al., 2009). Such a shift  towards bacterial dominance, in turn may slow the rate of decomposition of organic matter and increase the rate of soil C sequestration (Hoogmoed et al., 2014a).

The increase in the F:B ratio observed under Eucalyptus shelterbelts compared to pasture soil is similar to observed higher F:B ratios in forest soils generally compared to agricultural or grassland soils (Bailey et al., 2002; Bossuyt et al., 2001; Högberg et al., 2007; Treseder, 2004). Forest systems receive litter which is relatively more recalcitrant than the plant litter in pastures and grassland soils, leading to an increased abundance of fungi which can more readily break down low quality substrates (Six et al., 2006). Six et al. (2006) showed that fungal-dominated soils have slower C turnover rates because fungi incorporate more C into biomass than bacteria. Fungi also have more recalcitrant cell walls and facilitate C stabilisation and protection by increasing soil aggregation.

In general, it is considered that bacteria require more N per unit biomass C accumulation than fungi (De Deyn et al., 2008). The higher available NO₃-N and NH₄⁺-N under Acacia may thus support higher bacterial abundance contributing to lower F:B ratios relative to Eucalyptus. Fungal to bacterial gene ratios have been shown to be significantly correlated to soil C:N ratio and pH at the field scale (Fierer et al., 2009; Rousk et al., 2010). However at finer spatial scales (i.e. plot based shelterbelts) we were unable to link variation in the F:B gene ratio with any soil edaphic variables, including the C:N ratio.

Plant species that vary in ecophysiological traits (e.g. biochemistry, metabolism, gas exchange, leaf structure and function, nutrient and biomass allocation), will exert an influence on soil biota through differential rhizo-deposition and litter quality (Bardgett et al., 1999). For example, legumes are known to produce root exudates rich in N (Warembourg et al., 2003) , increasing available N in the soil. Nitrogen rich systems have been shown to decrease fungal abundance (de Vries et al., 2012). Several recent studies have compared the influence of N-fixing legumes vs. other functional plant groups on soil microbial communities (Bini et al., 2013; Hoogmoed et al., 2014a; Milcu et al., 2013; Strecker et al., 2015). Our data supports the idea that under functionally different dominant tree genera within a revegetated shelterbelt system, there are changes in the bulk soil fungal and bacterial communities compared to the surrounding pasture, although in our study Acacia and Eucalyptus were more similar to one another than to pastures.

The responses to different plant types in shelterbelts relative to the pasture system were more pronounced in the fungal community. This may be because bacteria are generally less adapted to decompose recalcitrant litter (e.g. from Acacia and Eucalyptus) than fungi (Henriksen and Breland, 1999; Six et al., 2006; Van Der Heijden et al., 2008). As a consequence the effect of plant type on soil fungal communities may be greater, particularly within the organic and first few centimetres of the mineral soil. A higher C:N ratio under Eucalyptus may also promote fungal species capable of degrading low quality substrates such as phenolic compounds (Six et al., 2006; Swift, 1982). The less pronounced changes in bacterial communities we observed may also indicate resistance to land use change within the bacterial community due to their rapid life cycles (Vries et al., 2012).

Like Thoms et al. (2010) our sampling scheme to a depth of 30cm offered a deeper insight into the soil layers that may be connected to the main rooting depth. Resource availability, which is mainly characterized by decreasing C concentration and a reduction in C quality with increasing soil depth (Richter and Markewitz, 1995), is the principal factor governing composition of the microbial community across soil profiles (Fierer et al. 2003).  The decrease in C with depth and the increase in C: N ratio within our study across site types is consistent with this. In turn we observed a significant effect of site type and depth on fungi (Table 4, Figure 4). These findings are consistent with the study of Lejon et al. (2005) which showed that fungal communities up to a depth of 30 cm could be separated out by plant genus. Our results indicate that discrimination of bacteria communities under each plant genus between soil depths was not as robust beyond a depth of 10 cm, where deeper samples were more similar to each other than to communities in the 0-10 cm samples.

Temporal shifts in microbial community composition may arise due to variation in soil temperature and moisture (Cao et al., 2010; Moore-Kucera and Dick, 2008; Myers et al., 2001). Not surprisingly, soil moisture was lower under shelterbelt trees than in pastures. This may be partly because the presence of Acacia and Eucalyptus species have been related to high water repellence in soils (Crockford et al., 1991; Doerr et al., 1998). Stand density may also influence soil moisture in these systems where high above ground biomass exerts a large demand for water (Bréda et al., 1995); as such we observed reduced soil moisture with increasing depth compared to within pasture systems. The soil surface is likely to experience wider variation (daily and seasonally) in soil temperatures and soil moisture than soils found at greater depth (Brady and Weil, 2002).

However, we found that between sampling years soil moisture availability did not differ with soil depth. Soils <10cm deep experienced no greater variation in soil moisture than those deeper down. This is also reflected in the fact that we found no statistical differences in fungal communities between sampling years within each soil depth. Contrary to this, bacterial community composition varied considerably between sampling periods within soil depths, indicating that factors other than environmental conditions may influence bacterial communities in these shelterbelt systems.

The effect of plant genus is further complicated as several other studies suggested that soil characteristics (Fierer et al., 2009; Girvan et al., 2003; Singh et al., 2008) are the most important factors in shaping microbial community structure. Unlike other studies which have shown that pH plays a significant role in explaining variation in soil microbial communities at various spatial scales from distances <1m to field and continental scales (Fierer and Jackson, 2006; Lauber et al., 2009; Rousk et al., 2010), pH did not show the highest correlation with community composition (accounting for 3.6 and 14% of the variation in fungal and bacterial community composition respectively). There was however a range of soil physiochemical factors important for shaping microbial community structure under Acacia, Eucalyptus and pastures (Figure 6).

Among the soil physiochemical factors we investigated, multivariate analysis of data suggested that variation in bacterial communities was associated with total C (including co-varying variables such as total N, NO₃-N and NH₄⁺-N , P, and carbon density). These variables collectively accounted for 21.5% of the variability in bacterial community composition. The fungal community responded to different physiochemical factors with copper explaining the greatest precent of the variation (8.03%). Soil moisture had little influence on  fungal communities, while soil properties had a comparatively stronger influence in shaping fungal community structure (Singh et al., 2009).

5. Conclusion

The results of this study comparing revegetated shelterbelt and pasture sites under the same environmental conditions indicate that plant genus can strongly influence fungal and to a lesser extent bacterial community composition. Changes in fungal soil communities were seen at different depths under Acacia and Eucalyptus. Within these shelterbelt systems, the effect of tree genus on soil microbial communities was stronger than that of individual soil properties. Overall, this effect of plant genus on soil communities highlights the importance of mixed species plantings as well as which species are included in revegetation mixtures. These can lead to increased heterogeneity within shelterbelt systems compared to pastures which may be vital for ecosystem functions such as decomposition and nutrient cycling.

References

Abdo, Z., Schüette, U.M.E., Bent, S.J., Williams, C.J., Forney, L.J., Joyce, P., 2006. Statistical methods for characterizing diversity of microbial communities by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes. Environmental Microbiology 8, 929-938.

Anderson, M.J., 2001. Permutation tests for univariate or multivariate analysis of variance and regression. Canadian Journal of Fisheries and Aquatic Sciences 58, 626-639.

Anderson, M.J., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA + for PRIMER: Guide to Software and Statistical Methods. PRIMER-E,, Plymouth.

Anderson, M.J., Willis, T.J., 2003. Canonical analysis of principal coordinates: a useful method of constrained ordination for ecology. Ecology 84, 511-525.

Augusto, L., Dupouey, J.-L., Ranger, J., 2003. Effects of tree species on understory vegetation and environmental conditions in temperate forests. Annals of Forest Science 60, 823-831.

Australian Bureau of Meterology, 2014. Climate statistics for Australian locations.

Bardgett, R.D., Wardle, D.A., 2010. Aboveground-belowground linkages: biotic interactions, ecosystem processes, and global change. Oxford University Press Oxford.

Bardgett, R.D., Wardle, D.A., Yeates, G.W., 1998. Linking above-ground and below-ground interactions: how plant responses to foliar herbivory influence soil organisms. Soil Biology and Biochemistry 30, 1867-1878.

Bauhus, J., Paré, D., Côté, L., 1998. Effects of tree species, stand age and soil type on soil microbial biomass and its activity in a southern boreal forest. Soil Biology and Biochemistry 30, 1077-1089.

Bini, D., Santos, C.A.d., Bouillet, J.-P., Gonçalves, J.L.d.M., Cardoso, E.J.B.N., 2013. Eucalyptus grandis and Acacia mangium in monoculture and intercropped plantations: Evolution of soil and litter microbial and chemical attributes during early stages of plant development. Applied Soil Ecology 63, 57-66.

Bird, P., 1998. Tree windbreaks and shelter benefits to pasture in temperate grazing systems. Agroforestry systems 41, 35-54.

Bird, P., Bicknell, D., Bulman, P., Burke, S., Leys, J., Parker, J., Van der Sommen, F., Voller, P., 1993. The role of shelter in Australia for protecting soils, plants and livestock, The Role of Trees in Sustainable Agriculture. Springer, pp. 59-86.

Bissett, A., Richardson, A., Baker, G., Wakelin, S., Thrall, P., 2010. Life history determines biogeographical patterns of soil bacterial communities over multiple spatial scales. Molecular ecology 19, 4315-4327.

Bissett, A., Richardson, A.E., Baker, G., Thrall, P.H., 2011. Long-term land use effects on soil microbial community structure and function. Applied Soil Ecology 51, 66-78.

Blair, G.J., Chinoim, N., Lefroy, R.D.E., Anderson, G.C., Crocker, G.J., 1991. A soil sulfur test for pastures and crops. Australian Journal of Soil Research 29, 619-626.

Brady, N., Weil, R., 2002. The nature and proprieties of soils. Prentice Hall, New Jersey, USA.

Bréda, N., Granier, A., Aussenac, G., 1995. Effects of thinning on soil and tree water relations, transpiration and growth in an oak forest (Quercus petraea (Matt.) Liebl.). Tree physiology 15, 295-306.

Brockett, B.F.T., Prescott, C.E., Grayston, S.J., 2012. Soil moisture is the major factor influencing microbial community structure and enzyme activities across seven biogeoclimatic zones in western Canada. Soil Biology and Biochemistry 44, 9-20.

Burdon, J., Gibson, A., Searle, S.D., Woods, M., Brockwell, J., 1999. Variation in the effectiveness of symbiotic associations between native rhizobia and temperate Australian Acacia: within‐species interactions. Journal of Applied Ecology 36, 398-408.

Cao, Y., Fu, S., Zou, X., Cao, H., Shao, Y., Zhou, L., 2010. Soil microbial community composition under Eucalyptus plantations of different age in subtropical China. European Journal of Soil Biology 46, 128-135.

Cavagnaro, T.R., Cunningham, S.C., Fitzpatrick, S., 2016. Pastures to woodlands: changes in soil microbial communities and carbon following reforestation. Applied Soil Ecology 107, 24-32.

Clarke, K., Gorley, R., 2006. PRIMER v6: user manual/tutorial (Plymouth routines in multivariate ecological research). Plymouth: Primer-E Ltd.

Clarke, K., Warwick, R., 2001. Change in Marine Communities: An approach to statistical analysis and interpretation. PRIMER-E: Plymouth, UK.

Cleugh, H.A., Prinsley, R., Bird, P.R., Brooks, S.J., Carberry, P.S., Crawford, M.C., Jackson, T.T., Meinke, H., Mylius, S.J., Nuberg, I.K., Sudmeyer, R.A., Wright, A.J., 2002. The Australian National Windbreaks Program: overview and summary of results. Australian Journal of Experimental Agriculture 42, 649-664.

Colwell, J., 1963. The estimation of the phosphorus fertilizer requirements of wheat in southern New South Wales by soil analysis. Australian Journal of Experimental Agriculture 3, 190-197.

Crockford, H., Topalidis, S., Richardson, D., 1991. Water repellency in a dry sclerophyll eucalypt forest—measurements and processes. Hydrological Processes 5, 405-420.

Culman, S.W., Bukowski, R., Gauch, H.G., Cadillo-Quiroz, H., Buckley, D.H., 2009. T-REX: software for the processing and analysis of T-RFLP data. BMC Bioinformatics 10, 171.

de Vries, F.T., Manning, P., Tallowin, J.R., Mortimer, S.R., Pilgrim, E.S., Harrison, K.A., Hobbs, P.J., Quirk, H., Shipley, B., Cornelissen, J.H., 2012. Abiotic drivers and plant traits explain landscape‐scale patterns in soil microbial communities. Ecology letters 15, 1230-1239.

Doerr, S.H., Shakesby, R.A., Walsh, R.P., 1998. Spatial variability of soil hydrophobicity in fire-prone eucalyptus and pine forests, Portugal. Soil Science 163, 313-324.

Fierer, N., Jackson, J.A., Vilgalys, R., Jackson, R.B., 2005. Assessment of soil microbial community structure by use of taxon-specific quantitative PCR assays. Applied and environmental microbiology 71, 4117-4120.

Fierer, N., Jackson, R.B., 2006. The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Sciences of the United States of America 103, 626-631.

Fierer, N., Strickland, M.S., Liptzin, D., Bradford, M.A., Cleveland, C.C., 2009. Global patterns in belowground communities. Ecology Letters 12, 1238-1249.

Gardes, M., Bruns, T.D., 1993. ITS primers with enhanced specificity for basidiomycetes‐application to the identification of mycorrhizae and rusts. Molecular ecology 2, 113-118.

Gartzia-Bengoetxea, N., Kandeler, E., Martínez de Arano, I., Arias-González, A., 2016. Soil microbial functional activity is governed by a combination of tree species composition and soil properties in temperate forests. Applied Soil Ecology 100, 57-64.

Girvan, M.S., Bullimore, J., Pretty, J.N., Osborn, A.M., Ball, A.S., 2003. Soil Type Is the Primary Determinant of the Composition of the Total and Active Bacterial Communities in Arable Soils. Applied and Environmental Microbiology 69, 1800-1809.

Hagen-Thorn, A., Callesen, I., Armolaitis, K., Nihlgård, B., 2004. The impact of six European tree species on the chemistry of mineral topsoil in forest plantations on former agricultural land. Forest Ecology and Management 195, 373-384.

Henriksen, T.M., Breland, T.A., 1999. Decomposition of crop residues in the field: evaluation of a simulation model developed from microcosm studies. Soil Biology and Biochemistry 31, 1423-1434.

Hobbie, S.E., 1992. Effects of plant species on nutrient cycling. Trends in Ecology & Evolution 7, 336-339.

Hobbs, R.J., 1993. Can revegetation assist in the conservation of biodiversity in agricultural areas? Pacific Conservation Biology 1, 29.

Hoogmoed, M., Cunningham, S., Thomson, J., Baker, P., Beringer, J., Cavagnaro, T., 2012. Does afforestation of pastures increase sequestration of soil carbon in Mediterranean climates? Agriculture, Ecosystems & Environment 159, 176-183.

Hoogmoed, M., Cunningham, S.C., Baker, P., Beringer, J., Cavagnaro, T.R., 2014a. N-fixing trees in restoration plantings: Effects on nitrogen supply and soil microbial communities. Soil Biology and Biochemistry 77, 203-212.

Hoogmoed, M., Cunningham, S.C., Baker, P.J., Beringer, J., Cavagnaro, T.R., 2014b. Is there more soil carbon under nitrogen-fixing trees than under non-nitrogen-fixing trees in mixed-species restoration plantings? Agriculture, Ecosystems & Environment 188, 80-84.

Jiang, Y.M., Chen, C.R., Liu, Y.Q., Xu, Z.H., 2010. Soil soluble organic carbon and nitrogen pools under mono-and mixed species forest ecosystems in subtropical China. Journal of Soils and Sediments 10, 1071-1081.

Lane, D., 1991. 16S/23S rRNA sequencing. Nucleic acid techniques in bacterial systematics, 125-175.

Lauber, C.L., Hamady, M., Knight, R., Fierer, N., 2009. Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Applied and environmental microbiology 75, 5111-5120.

Legendre, P., Anderson, M.J., 1999. Distance-based redundancy analysis: testing multispecies responses in multifactorial ecological experiments. Ecological Monographs 69, 1-24.

Lejon, D.P., Chaussod, R., Ranger, J., Ranjard, L., 2005. Microbial community structure and density under different tree species in an acid forest soil (Morvan, France). Microbial Ecology 50, 614-625.

Li, J., Tong, X., Awasthi, M.K., Wu, F., Ha, S., Ma, J., Sun, X., He, C., 2018. Dynamics of soil microbial biomass and enzyme activities along a chronosequence of desertified land revegetation. Ecological Engineering 111, 22-30.

Lindenmayer, D., Hobbs, R., 2004. Fauna conservation in Australian plantation forests–a review. Biological Conservation 119, 151-168.

Mendham, D.S., O’Connell, A.M., Grove, T.S., 2003. Change in soil carbon after land clearing or afforestation in highly weathered lateritic and sandy soils of south-western Australia. Agriculture, Ecosystems & Environment 95, 143-156.

Milcu, A., Allan, E., Roscher, C., Jenkins, T., Meyer, S.T., Flynn, D., Bessler, H., Buscot, F., Engels, C., Gubsch, M., 2013. Functionally and phylogenetically diverse plant communities key to soil biota. Ecology 94, 1878-1885.

Moore-Kucera, J., Dick, R.P., 2008. PLFA profiling of microbial community structure and seasonal shifts in soils of a Douglas-fir chronosequence. Microbial Ecology 55, 500-511.

Muyzer, G., De Waal, E.C., Uitterlinden, A.G., 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59, 695-700.

Myers, R.T., Zak, D.R., White, D.C., Peacock, A., 2001. Landscape-level patterns of microbial community composition and substrate use in upland forest ecosystems. Soil Science Society of America Journal 65, 359-367.

Ndaw, S., Gama-Rodrigues, A., Gama-Rodrigues, E., Sales, K., Rosado, A., 2009. Relationships between bacterial diversity, microbial biomass, and litter quality in soils under different plant covers in northern Rio de Janeiro State, Brazil. Canadian journal of microbiology 55, 1089-1095.

O’connell, A., Grove, T., Mendham, D., Rance, S., 2003. Changes in soil N status and N supply rates in agricultural land afforested with eucalypts in south-western Australia. Soil Biology and Biochemistry 35, 1527-1536.

Paul, K.I., England, J.R., Baker, T.G., Cunningham, S.C., Perring, M.P., Polglase, P.J., Wilson, B., Cavagnaro, T.R., Lewis, T., Read, Z., Madhavan, D.B., Herrmann, T., 2018. Using measured stocks of biomass and litter carbon to constrain modelled estimates of sequestration of soil organic carbon under contrasting mixed-species environmental plantings. Science of The Total Environment 615, 348-359.

Paul, K.I., Polglase, P.J., Nyakuengama, J.G., Khanna, P.K., 2002. Change in soil carbon following afforestation. Forest Ecology and Management 168, 241-257.

Paul, K.I., Roxburgh, S.H., England, J.R., de Ligt, R., Larmour, J.S., Brooksbank, K., Murphy, S., Ritson, P., Hobbs, T., Lewis, T., Preece, N.D., Cunningham, S.C., Read, Z., Clifford, D., John Raison, R., 2015. Improved models for estimating temporal changes in carbon sequestration in above-ground biomass of mixed-species environmental plantings. Forest Ecology and Management 338, 208-218.

Prescott, C.E., Grayston, S.J., 2013. Tree species influence on microbial communities in litter and soil: Current knowledge and research needs. Forest Ecology and Management 309, 19-27.

Rayment, G., Higginson, F.R., 1992. Australian laboratory handbook of soil and water chemical methods. Inkata Press Pty Ltd.

Rousk, J., Bååth, E., Brookes, P.C., Lauber, C.L., Lozupone, C., Caporaso, J.G., Knight, R., Fierer, N., 2010. Soil bacterial and fungal communities across a pH gradient in an arable soil. The ISME journal 4, 1340-1351.

Schutter, M., Dick, R., 2001. Shifts in substrate utilization potential and structure of soil microbial communities in response to carbon substrates. Soil Biology and Biochemistry 33, 1481-1491.

Siciliano, S.D., Palmer, A.S., Winsley, T., Lamb, E., Bissett, A., Brown, M.V., van Dorst, J., Ji, M., Ferrari, B.C., Grogan, P., Chu, H., Snape, I., 2014. Soil fertility is associated with fungal and bacterial richness, whereas pH is associated with community composition in polar soil microbial communities. Soil Biology and Biochemistry 78, 10-20.

Singh, B.K., Dawson, L.A., Macdonald, C.A., Buckland, S.M., 2009. Impact of biotic and abiotic interaction on soil microbial communities and functions: A field study. Applied Soil Ecology 41, 239-248.

Singh, B.K., Nunan, N., Ridgway, K.P., McNicol, J., Young, J.P.W., Daniell, T.J., Prosser, J.I., Millard, P., 2008. Relationship between assemblages of mycorrhizal fungi and bacteria on grass roots. Environmental Microbiology 10, 534-541.

Six, J., Frey, S., Thiet, R., Batten, K., 2006. Bacterial and fungal contributions to carbon sequestration in agroecosystems. Soil Science Society of America Journal 70, 555-569.

Smith, C.J., Danilowicz, B.S., Clear, A.K., Costello, F.J., Wilson, B., Meijer, W.G., 2005. T‐Align, a web‐based tool for comparison of multiple terminal restriction fragment length polymorphism profiles. FEMS Microbiology Ecology 54, 375-380.

Šnajdr, J., Dobiášová, P., Urbanová, M., Petránková, M., Cajthaml, T., Frouz, J., Baldrian, P., 2013. Dominant trees affect microbial community composition and activity in post-mining afforested soils. Soil Biology and Biochemistry 56, 105-115.

Strecker, T., Barnard, R.L., Niklaus, P.A., Scherer-Lorenzen, M., Weigelt, A., Scheu, S., Eisenhauer, N., 2015. Effects of Plant Diversity, Functional Group Composition, and Fertilization on Soil Microbial Properties in Experimental Grassland. PLoS ONE 10, e0125678.

Swift, M., 1982. Basidiomycetes as components of forest ecosystems, Symposium series-British Mycological Society.

Thies, J.E., 2007. Soil microbial community analysis using terminal restriction fragment length polymorphisms. Soil Science Society of America Journal 71, 579-591.

Thoms, C., Gattinger, A., Jacob, M., Thomas, F.M., Gleixner, G., 2010. Direct and indirect effects of tree diversity drive soil microbial diversity in temperate deciduous forest. Soil Biology and Biochemistry 42, 1558-1565.

Thrall, P.H., Burdon, J., Woods, M.J., 2000. Variation in the effectiveness of symbiotic associations between native rhizobia and temperate Australian legumes: interactions within and between genera. Journal of Applied Ecology 37, 52-65.

Van Der Heijden, M.G., Bardgett, R.D., Van Straalen, N.M., 2008. The unseen majority: soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecology letters 11, 296-310.

Vries, F.T., Manning, P., Tallowin, J.R., Mortimer, S.R., Pilgrim, E.S., Harrison, K.A., Hobbs, P.J., Quirk, H., Shipley, B., Cornelissen, J.H., 2012. Abiotic drivers and plant traits explain landscape‐scale patterns in soil microbial communities. Ecology letters 15, 1230-1239.

VSN International, 2011. GenStat for Windows 14th Edition. VSN International, Hemel Hempstead.

Walkley, A., Black, I.A., 1934. An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil science 37, 29-38.

Warcup, J., 1980. Ectomycorrhizal associations of Australian indigenous plants. New Phytologist 85, 531-535.

Wardle, D.A., Bardgett, R.D., Klironomos, J.N., Setälä, H., Van Der Putten, W.H., Wall, D.H., 2004. Ecological linkages between aboveground and belowground biota. Science 304, 1629-1633.

Warembourg, F., Roumet, C., Lafont, F., 2003. Differences in rhizosphere carbon-partitioning among plant species of different families. Plant and Soil 256, 347-357.

White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. PCR protocols: a guide to methods and applications 18, 315-322.

Zak, D.R., Holmes, W.E., White, D.C., Peacock, A.D., Tilman, D., 2003. Plant diversity, soil microbial communities, and ecosystem function: are there any links? Ecology 84, 2042-2050.