Introduction

Biochars are by-products of the pyrolysis process of burning biomass from either plant or animal origin heated (>250 °C) in a low or nil oxygen environment1. The carbon (C), ash and nutrient contents of biochars vary depending on the biomass source used as feedstock and the production temperature employed2. Biochars contain highly stable forms of C which have the potential to remain in the soil for hundreds of years3,4,5 but the ash only remains for a short time. Biochars are attracting attention as a stable soil amendment, as a C sink in agricultural soils, and as a stimulant of soil biological fertility6 and agricultural productivity7,8,9. Some biochars applied at high rates have been shown to increase soil water holding capacity either directly due to their high surface areas10 or indirectly in association with increases in soil organic carbon11. Biochars contain traces to high concentrations of nutrients including phosphorus (P)8,12. Plant responses to application of biochars to soil range from increased to the decreased colonisation of roots by arbuscular mycorrhizal (AM) fungi depending on P and nitrogen (N) concentrations in the biochars13,14,15,16,17. Biochar may also contain unwanted compounds such as crystalline silica, dioxin, phenolic compounds, volatile compounds and heavy metals based on sources18,19,20 which could influence mycorrhizal colonisation. Variability in the quality of biochars may lead to diverse effects on mycorrhizal colonisation, with the potential to enhance P uptake and growth of plants to different degrees, including positive responses under drought stress21.

There is an ongoing debate about the agronomic benefit of biochar on crop growth, soil fertility and AM symbioses. Where biochars have been shown to increase plant growth and yield22,23 and the involvement of AM fungi has been proposed, a range of possible mechanisms has been suggested14,24. Some studies reported an increase in mycorrhizal colonisation of roots in response to application of biochars without identifying the involvement of specific mechanisms15,25,26,27. It has been claimed that biochar can increase microbial activity, including that of mycorrhizal fungi, by providing a favourable microhabitat24,25,27. Extraradical hyphae of AM fungi can extend into biochar fragments buried in soil, sporulation can occur inside biochar pores28, and increased plant P uptake can result13. However, despite demonstrated influences of biochars on colonisation of roots by AM fungi, comparative effects of a wide range of biochar sources on AM colonisation have not been investigated.

Nevertheless, it is expected that the nature and extent of response to soil amendment with biochar will depend on the feedstock used to make the biochar and that this will be associated with variation in biochar physical characteristics such as pore sizes24,29, ash content and chemical characteristics30. Potential interactions between biochar and colonisation of roots by AM fungi could occur directly by altering the growth of hyphae in the soil prior to the colonisation of roots17, and indirectly by altering root growth and subsequent colonisation15,17. Biochar can also alter colonisation of roots by stimulating the growth of hyphae in the soil before the establishment of mycorrhizal symbiosis under water-limiting conditions21.

While biochar can influence colonisation in wheat roots by AM fungi both in the field and under glasshouse conditions15,26, a plant growth benefit gained from amending the soil with biochar can vary according to the properties of the biochar used31. Therefore, our study extended previous investigations by comparing 34 different biochars sourced from 17 biochar feedstocks. The aim was to determine the effects of biochar on plant growth and some aspects of soil fertility which could depend on interactions between biochar and indigenous AM fungi present in the agricultural soil used. It compared biochar effects on mycorrhizal colonisation, plant growth, and P nutrition of subterranean clover and wheat when applied to P-deficient agricultural soil. We tested the hypothesis that the higher P concentration in biochar will reduce mycorrhizal colonisation in roots of both subterranean clover and wheat by a naturally occurring community of AM fungi in the agricultural soil.

Results

Characteristics of biochars

The range of characteristics of the 34 different biochars (Table 1) demonstrated considerable variability, including their pH and water holding capacity (WHC). The pH of the biochars ranged from 3.7 to 12.3; most biochars were mild to strongly alkaline, and only two were acidic (B56 produced from green waste: pH 4.7 and B71 produced from woods and stored for long unknown periods: pH 3.7). The WHC of the biochars varied from 47% (for B70 made from oil mallee wood) to about 460% (for B43 made of from rice husk). The total C content of biochars varied from 19.29% (for B41 made from biosolids) to 85.43% (for B33 made from non-activated sawdust). The cases of low C are likely to be due to the presence of inorganic solids, including ash in the biochar. The total N concentration of the biochars varied from 0.06% (for B30 made from sawdust) to 2.56% (for B41 made from biosolids). Total P concentrations in the biochars ranged from <0.01% (for B32 made from sawdust) to 1.89% (for B54 made from poultry litter) and 2.83% (for B37 which was a biochar-mineral complex artificially formulated).

Table 1 Properties of biochars made from different feedstocks used in this experiment.
Full size table

Biochar effects on colonisation of roots by AM fungi

Addition of the biochars to soil had variable effects, ranging from positive to negative, on the extent of colonisation of roots by AM fungi for both subterranean clover and wheat (p < 0.01; Fig. 1a,b). The highest level of mycorrhizal colonisation (%) of roots was observed for biochars B64 and B65 (both produced from oil mallee) in subterranean clover (Fig. 1a) and for biochars B69 (oil malle), B60 (wood waste) and B30 (sawdust) in wheat (Fig. 1b). The lowest mycorrhizal colonisation (%) was observed both in subterranean clover and wheat when poultry litter biochar B54 (biochar produced from poultry litter contains total P 1.89%, see Table 1) was applied. Overall, mycorrhizal colonisation (%) was generally higher in roots of subterranean clover than wheat across all biochars, and effects of biochars on root colonisation in these two test plant species were correlated (r = 0.40, p < 0.05).

Figure 1
figure 1

The effect of biochars on arbuscular mycorrhizal colonisation (%) in roots of (a) subterranean clover (lsd = 8.0, p < 0.05) and (b) wheat (lsd = 2.8, p < 0.05) after eight weeks of growth. Bar bigger than control (B0) with an asterisk (*) indicate increase but bar smaller than the control with an asterisk (*) indicate decrease. Bars do not have an asterisk do not differ.

Full size image

Biochar effects on plant growth and P nutrition

Shoot and root growth of subterranean clover and wheat responded to biochar application as either increases or decreases depending on the sources of biochar (p < 0.05; Figs 2 and 3). In subterranean clover, eight biochars significantly increased shoot DW, and it was decreased by one biochar (Fig. 2) whereas root DW was increased by five biochars and decreased by two biochars. Similarly, P uptake in subterranean clover was increased by 16 biochars and decreased by only one biochar (p < 0.05; Fig. 2). In wheat, 12 biochars significantly increased shoot DW, and it was decreased by six biochars (Fig. 3) whereas root DW was increased by eight biochars and decreased by three biochars. For P uptake in wheat, increases were observed for 29 biochars and a decrease for only one biochar (p < 0.05; Fig. 3).

Figure 2
figure 2

The effect of biochars on (a) shoot DW, (b) root DW, and (c) P uptake of subterranean clover after eight weeks of growth. Bar bigger than control (B0) with an asterisk (*) indicate increase but bar smaller than the control with an asterisk (*) indicate decrease. Bars do not have an asterisk do not differ.

Full size image
Figure 3
figure 3

The effect of biochars on (a) shoot DW, (b) root DW, and (c) P uptake of wheat after eight weeks of growth. Bar bigger than control (B0) with an asterisk (*) indicate increase but bar smaller than the control with an asterisk (*) indicate decrease. Bars do not have an asterisk do not differ.

Full size image

Biochar effects on soil P availability at the end of plant growth cycle

Soil P fractions after harvest were significantly influenced by biochar source (Fig. 4). The highest level of residual soil P at the time of plants was harvested observed for soil amended with biochars B37 (biochar-mineral complex), B44 (cow manure) and B54 (poultry litter). The second highest residual soil P was recorded for soil amended with biochars B58 (green waste enhanced with N) and B59 (green waste). Similar amounts of available P remained in soil amended with most of the other biochars at the end of plant growth cycle. Microbial biomass P was lowest in soils amended with biochars B27 (jarrah wood), B54 (poultry litter), B55 (green garden organics) and B68 (oil mallee) following eight weeks of plant growth. Soil organic P concentration was highest in soil amended with biochar B41 (biosolid) and lowest in soil amended with biochar B33 (sawdust) following eight weeks of plant growth.

Figure 4
figure 4

The effect of biochars sources on (a) inorganic Colwell P (lsd = 3.1, p < 0.05), (b) microbial biomass P (lsd = 0.5, p < 0.05) and (c) organic P (lsd = 9.6, p < 0.05) in soil after 8 weeks of growth. Bar bigger than control (B0) with an asterisk (*) indicate increase but bar smaller than the control with an asterisk (*) indicate decrease. Bars do not have an asterisk do not differ.

Full size image

Correlations between soil or biochar properties and plant parameters

Significant Pearson’s correlation coefficients (r) were observed between extractable soil P and all measured parameters of plant growth and % mycorrhizal colonisation (Table 2). Soil NaHCO3-extractable P (Colwell P) was correlated with wheat shoot biomass, uptake of P in wheat plants (Table 2), with biochar total P (r = 0.74*, P < 0.05), and with biochar total N (r = 0.39*, P < 0.05).

Table 2 Pearson’s correlation coefficients (r) between the soil or biochar properties, measured parameters of plants and % mycorrhizal colonisation after 8 weeks of growth.
Full size table

Significant Pearson’s correlation coefficients (r) were also observed between the soil pH and concentration of P in plant shoot tissues (Table 2). The availability of P in soil and % mycorrhizal colonisation were negatively correlated (Table 2). Soil microbial P was not significantly correlated with % mycorrhizal colonisation, plant growth or P uptake (P > 0.05), but organic P concentration in soil after plant growth was significantly and negatively correlated with % mycorrhizal colonisation in both plants (P < 0.05). Soil organic P concentration was positively correlated with plant growth and P uptake for wheat but negatively correlated for subterranean clover (P < 0.05). Biochar P concentration was negatively correlated with % mycorrhizal colonisation but positively correlated with plant growth and P uptake in wheat but not for subterranean clover (P < 0.05). Biochar N and K concentrations were both negatively correlated with % mycorrhizal colonisation (P < 0.05). Across all biochars, there was a significant correlation between % mycorrhizal colonisation and plant growth (p < 0.05) and with P uptake (p < 0.05) for subterranean clover. In contrast, there was no significant correlation between % mycorrhizal colonisation and plant growth for wheat (p > 0.05). There was a no significant correlation between shoot biomass of subterranean clover and wheat (p > 0.05) or between root biomass of subterranean clover and wheat (p > 0.05). However, there was a positive correlation between % mycorrhizal colonisation of subterranean clover and wheat (p < 0.05; Fig. 5).

Figure 5
figure 5

Correlation between subterranean clover mycorrhizal colonisation (%) vs wheat mycorrhizal colonisation (%) shown in scatter plot.

Full size image

Principal component analysis (PCA) revealed patterns in plant parameters across all samples (Fig. 6a,b). For subterranean clover, PCA axis 1 was mostly influenced by soil organic and inorganic P and soil pH, whereas PCA axis 2 was most influenced by biochar pH and biochar WHC (Fig. 6a). However, the reverse occurred for wheat (Fig. 6b). The scatterplot showed slightly overlapping groups of samples corresponding to particular biochar types. The sample differentiation pattern concerning plant parameters was pronounced regarding biochar sources. The most symptomatic differences were observed for subterranean clover among biochar types. Increased % mycorrhizal colonisation in subterranean clover corresponded to enhanced P concentration with both shoots and roots, whereas wheat was characterised by lower % mycorrhizal colonisation and increased microbial biomass P.

Figure 6
figure 6

Principal component analysis ordination diagram Axis 1 (PC 1) vs. Axis 2 (PC 2) of (a) subterranean clover and (b) wheat parameters (mycorrhizal colonisation, shoot and root mass, P concentration in the shoots) for samples of the 17 sources of feedstocks plus a control. The percentage of total variance as explained by each axis is shown.

Full size image

Discussion

Biochars prepared from different feedstock sources can differ in their influence14 including their effects on colonisation of roots by AM fungi and P availability, but the mechanisms involved are poorly understood. It was expected that biochars of different feedstock sources would differ in the extent to which they influenced root growth and that this would be associated with differences in nutrient concentrations in the biochar and the plant species used. It has been observed previously that P and N, in particular, may become more available to plants after soil is amended with some forms of biochar8,32 and this could increase root growth and % mycorrhizal colonisation. Increased root growth could also reduce the proportion of roots colonised by AM fungi, but increase the volume of roots colonised21.

Effect of biochars on mycorrhizal colonisation, plant growth and P nutrition have been reported previously for a small number of biochars using both wheat and subterranean clover15,33. In the present experiment, where we compared a much wider range of biochar sources, the higher the nutrient (P and N) concentrations in biochar, the lower was mycorrhizal colonisation and vice versa. Mycorrhizal colonisation (assessed as % of root length colonised) decreased with excess available P derived from biochars and soil32,34,35. Our experiment provides further support for the role of the mycorrhizal symbiosis in soil fertility and plant P nutrition when some sources of biochar are applied36,37. This needs further evaluation in other soil types and environmental conditions. Previous studies showed biochars had relatively small amounts of nutrients available to plants, which favoured colonisation of roots by microbes (including AM fungi)23,38,39. P-rich biochars could function as slow-release fertiliser40, thus available P is a factor that needs to be considered when selecting biochars for use as a soil amendment; an increase in available P in soil was observed in our experiment for many biochars.

Soil amendment with biochars can change soil physicochemical properties which may lead to increases in soil pH and nutrient availability and subsequent alterations in root colonisation by mycorrhizal fungi13,23. In our experiment, correlations between mycorrhizal colonisation and plant growth as well as between mycorrhizal colonisation and P uptake were observed in subterranean clover but not in wheat. This is likely to be because wheat is less responsive to mycorrhizas than subterranean clover41. Plant P uptake has been shown to increase with increasing level of biochar application in some agricultural ecosystems7,33. It is possible that biochar applications up to a certain level may stimulate mycorrhizal colonisation, leading to increased P uptake, but when applied at higher levels that enrich soil P beyond that required to overcome a deficiency for plant growth, the response may disappear32.

The physical and chemical properties of biochars (see Table 1) used in this experiment varied widely because the biochars were made from various feedstock sources and pyrolysis conditions42. Some biochar has the potential to provide a habitat for soil microbes, but this capability depends on the physical properties of biochar such as porosity and surface area24. Soil solution, air and water diffuse through the biochar pores facilitating soil microbes to colonise the biochar24,43,44. However, these claimed mycorrhizal interactions with biochars14,27 have only been investigated for a relatively small number of biochars24. For example, woody biochar from Pinus radiata increased fungal and bacterial abundance and supported a higher abundance of P solubilising bacteria45. Fungi, especially saprotrophic fungi, were shown to colonised biochar particles in association with decomposing fibrous organic matter46. Furthermore, any stimulation of colonisation of roots by AM fungi may also depend on soil characteristics29 and soil water availability21.

Although most biochar amendments used in our experiment increased mycorrhizal colonisation, plant growth and P uptake under glasshouse conditions in plastic lined confined pots, the scarcity of published data restricts evaluations of the potential of biochars produced from various sources for use as amendments under field conditions. Furthermore, the influence of biochars on mycorrhizal colonisation and effects on plant growth and P uptake under mixed cropping systems varied with biochar source and the reasons for this may be complex. This is the first report that demonstrates, for a large number of biochars with a wide range of characteristics, biochar effects on colonisation of roots by AM fungi. However, where biochars are produced as a soil amendment, appropriate application levels32 and the mechanisms underlying their effectiveness need to be investigated before field application is widely recommended47. Further studies on the effect of biochars which have a high potential as a habitat for mycorrhizal fungi are also necessary to assess bio-physicochemical interactions between biochar particles, soil, microbes and roots48.

As most of the biochars used here were prepared from diverse feedstocks that contain varying amounts of C and ash (which is unknown) they are likely to be extremely heterogeneous and impure when mixed with soil49. The ash that arises during the anaerobic burning process (pyrolysis) contains calcium carbonates (hence pH effects), struvite and calcium phosphate, and sometimes silicates, for example, in rice husks biochar (hence fertiliser effects). Conclusions about biochar efficacy are often based on investigations of only one to a few sources of biochar. However, the wide range of responses observed in our study of biochars from 17 different feedstocks (including multiple sources from similar feedstocks), highlights the need for caution when generalising about the effects of biochar. As such, we note conspicuous differences in biochar properties (e.g. the low pH in B32 and the differences between B55 and B56), suggesting that biochar’s final properties cannot be easily predicted from the information provided about the specific charring process. We demonstrated substantial differences in the efficacy of different sources of biochar in relation to the colonisation of roots by AM fungi. The possibility of sourcing large numbers of biochars produced under equivalent conditions is very unlikely and extends likely differences in biochar responses, even from the same feedstock. The conditions under which biochars are produced further compounds variability in efficacy. Therefore, in order to evaluate the potential benefits of biochars for use in the field, investigations need to focus on the responses of forms of biochars that are available locally.

Methods

Experimental design

Wheat (Triticum aestivum L. var. ‘Wyalkatchem’) and subterranean clover (Trifolium subterraneum L. var. ‘Seaton Park’) were grown together for 8 weeks under glasshouse conditions in an agricultural soil collected from Mingenew, Western Australia following its amendment with one of 34 biochars plus one control where no biochar was applied to the soil. Biochar codes (B27 and so on in Table 1) were used according to the Australian National Project Biochar Database (funded by the Department of Agriculture, Australian Government). There were three replicates of each treatment.

Each pot contained 1.4 kg air-dry soil which had been mixed thoroughly with one of the 35 biochar treatments at the rate of 1% (w/w). The 1.3 L plastic pots were sown with seeds of wheat and subterranean clover and thinned to 2 plants of each species per pot after germination. Each pot was watered to 70% of field capacity based on the daily addition of water to the prescribed weight.

Soil and biochar characteristics

Soil (0–10 cm) was collected from a subterranean clover/wheat rotation near Mingenew, Western Australia (latitude 29°19′, longitude 115°44′). Mingenew has a Mediterranean climate and a mean annual rainfall of 400 mm (with 80% falling in the May to October growing season). The soil contained 85% sand, 3% silt and 12% clay and was classified as a Tenosol (sand over gravel)50 and Humic Dystroxerepts51; the 0–10 cm layer was analysed for basic properties. The pH of the soil was 4.8 measured in 0.01 M CaCl2 at 1:5 (w/v) ratio. Organic matter was 10.3 g/kg soil measured by dry combustion using an elementar (vario MACRO CNS; Elementar, Germany). The soil contained 0.6 g/kg total N, 7 and 5 mg/kg NO3-N and NH4-N respectively. This soil was chosen because its loamy sand texture gave it a low capacity to retain P and the low available P (7.5 mg/kg) content was suitable for a mycorrhizal response in both subterranean clover and wheat. Soil P and K were measured using the 0.5 M NaHCO3 extraction method52. Soil available P, organic P and microbial biomass P were extracted after harvest of plants52,53,54. The P concentration in the extracts was measured in a spectrophotometer55.

The 35 biochars including one control (no biochar amended) used in this evaluation were produced from 17 sources of feedstocks (see Table 1) manufactured under different temperature conditions and sieved to 2 mm. The pH of the biochars was measured in 0.01 M CaCl2 at 1:5 (w/v) ratios. Water holding capacity of the biochars was measured using a gravimetric method56. A subsample of biochar was finely ground before total C and N contents were determined by dry combustion using an elementar (vario MACRO CNS; Elementar, Germany). Total P in biochars was measured after digested in 3:1 HNO3-HClO4 and P measured in solution by the molybdenum-blue method55.

Assessment of colonisation by AM fungi

At harvest after washing with tap water, sub-samples of roots (0.5 g) were cut into approximately 1 cm pieces and cleared in 10% KOH, acidified and stained with Trypan blue (0.05%) in lactoglycerol (1: 1: 1.2/ lactic acid: glycerol: water)57. Mycorrhizal colonisation (%) of roots by AM fungi was assessed using the gridline root intercept method under a microscope at 100 × magnification as % root colonised57.

Plant analyses

At harvest, shoots were cut from each plant and roots were washed free of soil and organic matter. Shoots and roots (after a defined weight of roots was removed for assessment of AM colonisation) were dried at 60 °C for at least 72 h to determine the shoot and root dry weights (DW). Oven-dried shoots were ground and digested in 3:1 HNO3-HClO4 mixture and the P concentration in the digest measured by the molybdenum-blue method55. Shoot P uptake was calculated by multiplying shoot P concentration by shoot weight.

Statistical Analyses

Statistical analyses were carried out using Genstat (v.18). One-way analysis of variance was used to detect significant effects of biochars on all soil and plant parameters measured. The least significant difference (LSD) was applied to test significance between means. The significant Pearson’s correlations between the measured soil and plant parameters after eight weeks of wheat and subterranean clover growth were tested. Plant parameters (shoot and root mass, P concentration in shoots) and mycorrhizal colonisation were also explored with principal component analysis (PCA) to identify the association between these traits and to recognise the grouping of samples, associated with the soil properties and 17 different category based on biochar sources, with their similar source and characteristics. The analysis was based on the correlation matrix and multivariate analyses performed for each plant species separately.