Fertility Impact of Separate and Combined Treatments with Biochar, Sewage Sludge Compost and Bacterial Inocula on Acidic Sandy Soil

The short-term effects of processed waste materials: sewage sludge compost (up to 0.5%), biochar made of paper sludge and grain husk (BC) (up to 2%) combined with plant growth-promoting rhizobacterial (PGPR) inoculum, on the fertility of acidic sandy soil at 65% of field capacity were tested in a pot experiment in separate and combined treatments. The soil pH, organic matter content, total and plant-available nutrients, substrate-induced respiration, arbuscular mycorrhizal fungal (AMF) root colonisation parameters and maize (Zea mays L.) biomass were investigated in experiments lasting two months. The positive priming (21% organic matter loss) induced by BC alone was not observed after combined application. The combination of compost and PGPR with 1.5% BC resulted in 35% higher P and K availability due to greater microbial activity compared to BC alone. Only compost applied alone at 0.5% gave a 2.7 times increase in maize biomass. The highest microbial activity and lowest AMF colonisation were found in combined treatments. In the short term the combined application of BC, compost and PGPR did not result in higher fertility on the investigated soil. Further research is needed with a wider range of combined treatments on acidic sandy soil for better understanding of the process.


Introduction
On sandy soils crop production is limited by several factors, the most important of which are low water retention capacity and nutrient content [1]. Due to the texture of these soils, organic matter (OM) is mineralised at a higher rate, leading to reduced fertility [2]. Fertilisation is less effective on these soils, as the nutrients added with mineral fertilisers have low colloid content and are easily leached [3]. On such soils irrigation and/or soil amelioration are prerequisites for safe cultivation, especially in the case of less drought-resistant crops like maize.
One possible way of improving such degraded or inherently unfavourable soils is to incorporate organic materials. Composts could be useful amendments for sandy soil, as they may increase the

Experiment Setup
Two experimental layouts were used to test the materials in soil: a completely randomised block for testing the separate effects of the additives, and a Box-Wilson method for testing their combined effects. The essence of this last method is that by changing the amendment doses in a specific order the number of treatment combinations can be significantly reduced [27]. In both experiments the pots contained 1.5 kg soil (pot volume: 1l). The highest dose of additives used in separate applications was determined on the basis of the quantities used in practice. Based on earlier experiments [28] the applied maximum dose was 45 t/ha = 1.5%wieght/weight (%w/w) for BC, while the manufacturer recommended a maximum dose of 15 t/ha = 0.5%w/w for compost. The doses of PGPR were determined by preliminary experiments performed on maize by the manufacturer. The lower doses were tested because measurable effects may already occur at these application rates on this low fertility soil. The treatment combinations in the experiments are shown in Table 2. The highest dose of BC and PGPR was the same in the separate and combined treatments, but the highest dose of compost in the combined treatments was only 0.33%w/w, to prevent the compost from suppressing the expected effects of the other amendments. Each treatment was performed in three or 10 replicates according to the experimental layout ( Table 2). The soil and compost were air-dried, then sieved through a 2 mm mesh. The treatments were set up in 1l pots, 13 cm in height, 13 cm wide at the top and 9 cm at the base. The bottoms of the pots were sealed so that no leaching occurred during the experiment. The component(s) of a given treatment were mixed thoroughly with 1.5 kg soil, then placed in the pots and wetted to 65% of maximum field capacity. Each pot was then weighed and kept in a dark room for two weeks at a temperature of 20 • C for incubation. During incubation and plant growth the water loss was monitored by the gravimetric method twice a week and the missing moisture was replaced by irrigation. According to Kang et al. [29] 65% of field capacity can be considered as a satisfactory moisture content for maize growth. After incubation, the pots were placed in a growth chamber with a 12/12 h photoperiod and a temperature setting of 26/16 • C, representing day (600 µmol/m 2 /s photon flux density) and night phases. The test plant was maize (Zea mays L., Mv 277), for which the tested soil was relatively unfavourable, so more pronounced treatment effects could be expected [30]. Two dressed seeds of maize were sown in each pot, the less developed of which was removed after germination. At the end of the two-month growth period the above-ground biomass of the plants was harvested and weighed, after which soil and plant samples were prepared for analysis.

Chemical and Biological Analysis
The pH was measured according to ISO 10390:2005 in a 1:2.5 soil:water suspension 12 h after mixing [31]. The OM content was determined using the modified Walkley-Black method [32]. The organic carbon content of BC was measured by incineration [33]. The CaCO 3 content was measured using the Scheibler gas-volumetric method [34]. The carbonates present in the sample were converted into CO 2 by the addition of hydrochloric acid. The plant-available phosphorus (P) and potassium (K) contents in the soil were determined in ammonium-acetate lactate extract (AL-P 2 O 5 , AL-K 2 O) using the Egner-Riehm-Domingo method [35]. The total nitrogen (N) content of soil and plants was determined with the Kjeldahl method [36], digesting the organic matter so that both total organic and inorganic N content could be measured. The NH 4 -N and NO 3 -N contents were measured in KCl extracts using the titrimetric method [35]. The pseudo total element contents were determined with the ICP-AES method (Jobin-Yvon Ultima 2) after microwave teflon bomb digestion with aqua regia [37] using Merck calibration standards and following the manufacturer's instructions. In each ICP measurement session the extract of a standard soil sample was also analysed as a control. The calibration curves were determined after every 12th sample.
Substrate-induced respiration (SIR) [38] was measured according to Szili-Kovács et al. [39]. Samples with a 2:1 water to soil ratio were incubated after the addition of glucose. The intensity of AMF colonization (M%) and the arbusculum richness (A%) in the roots were calculated using a five-class system [40] after observing 30 randomly selected root segments, each 1 cm in length. Root samples from the pots were cleared in KOH solution (15%w/w) and stained with aniline blue [41].

Statistical Analysis
The separate treatment effects were analysed using one-way ANOVA. Significant differences between the treatment means were calculated using the least significant difference (LSD) test at the p < 0.01, p < 0.05 and p < 0.1 levels. The results of the combined applications were evaluated using analysis of variance and regression analysis [28]. As the coefficient of determination (R 2 ) shows whether the model fits the data, only variables for which the Box-Wilson model gave R 2 values higher than 60% are discussed here, since this indicates that changes in these variables could be explained to at least a moderate extent by the model equation. The variance of these variables was determined using the F-test. Variability between the samples was determined by means of principal component analysis (PCA). Statistica v.9 (StatSoft Inc., Tulsa, OK, USA) software was used for the statistical evaluation.

Soil pH and OM Content
The application of 1.5%w/w BC alone significantly increased the soil pH to 5.9, while it rose to 5.5 in the 0.5%w/w compost treatment (Tables 3-5). Table 6 presents the R 2 and p values for parameters exhibiting R 2 values greater than 60% in the combined treatments, while Table 7 contains the mean values and LSD 5% values of these parameters for each treatment combination. In combined treatments compost and PGPR had an additive effect, but BC was decisive for pH change due to its application rate (Tables 6 and 7). Table 3. Significant effect of individual biochar (BC) treatment levels (%w/w) on the properties of acidic sandy soil and maize. Data are mean ± sd of the replicates.    The OM% of BC, measured with the Walkley-Black method, was almost three times higher than that of compost and can be considered as a labile fraction that can be mineralised in the soil [42]. The maximum dose of BC alone caused a 19% increase in OM%. There was no significant OM% increase in response to compost alone, because the standard deviation of the OM% values (0.06%) exceeded the OM increase that could be expected in compost treatments (0.01-0.04%) ( Table 3). The measured OM% increment in treatments with BC alone were 70% lower on average than the expected value based on the amount of OM added (Figure 1). This means that, on average, the soil OM content in BC-treated soils after harvest was 21% lower than expected. In the combined treatments (1-9, 11, 13 and 15) when compost and PGPR were applied with BC, around 100% of the added OM could be found in the soil after harvesting the maize biomass. The exception was treatment No. 9, in which the highest BC dose (2%) was applied ( Figure 1). In the case of treatments 4 and 7 the high OM values measured may have been caused by undetectable plant residues in the soil sample.

Properties
Agronomy 2020, 10, x FOR PEER REVIEW 7 of 16 the OM increase that could be expected in compost treatments (0.01-0.04%) ( Table 3). The measured OM% increment in treatments with BC alone were 70% lower on average than the expected value based on the amount of OM added (Figure 1). This means that, on average, the soil OM content in BC-treated soils after harvest was 21% lower than expected. In the combined treatments (1-9, 11, 13 and 15) when compost and PGPR were applied with BC, around 100% of the added OM could be found in the soil after harvesting the maize biomass. The exception was treatment No. 9, in which the highest BC dose (2%) was applied ( Figure 1). In the case of treatments 4 and 7 the high OM values measured may have been caused by undetectable plant residues in the soil sample.    Table 2. Legend: light grey: original soil OM content; dark grey: OM added in individual treatments; medium grey: OM content measured in the soil after the experiment.  Table 2. Legend: light grey: original soil OM content; dark grey: OM added in individual treatments; medium grey: OM content measured in the soil after the experiment.

N, P and K Contents in Soil and Plants
BC2 and BC3 treatments significantly increased the plant-available NO 3 -N and the total and plant-available P and K contents in the soil, but not that of total N (Table 3). PGPR application caused a decrease in plant-available K in the soil, while compost increased the total P, K and plant-available P contents. In response to BC alone the AL-P 2 O 5 and AL-K 2 O contents rose to a greater extent than the total P and K contents in soil, since about 80% of both the P and K content was in plant-available form (Tables 1 and 3). In these treatments the P and K contents also increased in maize (Tables 3-5). Although compost raised the K (treatment C3) and P (treatments C1-3) contents of the soil, there was no significant change in the plant P and K contents. However, the K and P uptake of maize increased almost three and two times, respectively, in the 0.5%w/w compost treatment compared to the control (data not shown). The inoculum itself was capable of improving the nutrient-supplying ability of the soil: as a result of PGPR treatment the plant P and K concentrations rose significantly ( Table 5).
The combined treatments indicated that BC was decisive for changes in the ratio of plant-available K to total K content, as it increased plant-available K to a greater extent (Tables 6 and 7). Both BC and compost led to a rise in the ratio of plant-available P to total P content, indicating increased P availability (Table 6).
Compost and PGPR had different effects on P and K availability when applied alone or in combined treatments. Compost alone did not affect the ratio of available and total K in the soil compared to the control (ratio: 2.4%), while the inoculum decreased it to a value of 1.6%. However, when combined with 1.5%w/w BC (combined treatments 1, 3, 5 and 7) both compost and PGPR improved P and K availability. In these treatments the ratio of available to total K was 35% higher on average (10.7%) compared to 1.5%w/w BC alone (7.9%). A 35% difference was also found in these treatments for P availability.
In the combined treatments the P and K contents of the maize biomass were primarily dependent on the BC treatment, but the K content was also influenced by PGPR (Table 6).
Regarding soil N, there was no significant change in the NH 4 -N content (14 mg/kg in the control soil), but the NO 3 -N content rose significantly in response to BC2 and BC3 treatments and decreased significantly after treatment with compost alone (treatments C1-3) (Tables 3 and 4). The combined treatments revealed that the change in soil NO 3 -N content was mainly influenced by BC ( Table 6). The C1-3 treatments decreased the maize N content, but the N uptake tripled in the 0.5%w/w (treatment C3) treatment compared to the control.

Maize Biomass, SIR and AMF Colonisation
In the separate treatments only compost (treatments C1-3) significantly increased the maize biomass. In the combined treatments there was no significant difference between the dry weights of the plants (Tables S1 and S2). The average biomass in the combined treatments was 2.9 g dry matter/pot, which was statistically equal to the value of the control treatment ( Table 3). The 1.5%w/w dose of BC alone resulted in a 1.5 times increase in SIR, while the 0.5%w/w dose of compost alone doubled it (Tables 3 and 4). SIR was not influenced by PGPR addition. The combined application of the materials led to an increase in microbial biomass compared to the untreated soil (data not shown due to the low coefficient of determination). The lowest SIR value (1.18 ± 0.03 µg CO 2 -C/g soil/hour) was recorded in treatment 2, given a 0.5%w/w BC dose combined with 0.25%w/w compost and PGPR, and the highest (1.86 ± 0.22 µg CO 2 -C/g soil/hour) in treatment 7, given a 1.5%w/w BC dose with 0.08%w/w compost and PGPR ( Table 2).
The application of BC alone caused an increase in the colonization intensity in treatment BC3 (M%) and arbuscular richness in treatments BC1-3 (A%) of indigenous AMF, while the arbuscular richness declined considerably in soils treated with compost (treatments C1-3) (Tables 3 and 4). In combined applications the mycorrhizal parameters indicated that the infectivity of the indigenous AMF community had been inhibited (data not shown due to the low coefficient of determination).
Even the highest value of M% (17.84) was considerably lower than in any of the treatments where BC, compost and PGPR were applied alone.

Discussion
Each of the materials applied had several positive effects on soil properties, but only compost resulted in an improvement in soil fertility, defined here as the plant biomass produced on it. Differences in early growth stages may determine the final biomass and yield of maize [43]. Thus, in contrast with expectations, the combined application of BC, compost and PGPR had no positive synergistic effect on soil fertility on this acidic sandy soil in the short term, when the water supply was satisfactory for maize [4,21,22,30]. The failure of BC to influence soil fertility can probably be attributed to the short experiment time and the laboratory conditions [44].
Though BC is basically used to amend physical soil properties, in the short term it may significantly influence chemical soil properties, though this depends on the type of BC [4,12,22]. In a similar short-term pot experiment, reported by Wang et al. [14], the application of BC without fertiliser resulted in plant yield depression. In the present experiment this adverse effect was not observed for BC alone, probably due to its significant labile fraction, which provided nutrients, thus avoiding the need for fertiliser addition.
BC and compost influenced the availability of nutrients in three ways: through their own nutrient and OM content, their pH-enhancing effect and the changes induced in the activity of the microbes present in the soil. PGPR exerted its effect by influencing the soil microbial community.
The most important of the possible longer term effects of the amendments is their effect on soil OM. Although BC is a recalcitrant material, it may induce the decomposition of soil OM due to its labile fraction. The application of BC alone probably triggered a positive priming effect in the soil, causing the native OM to be mineralised and resulting in lower OM% than in combined treatments with compost and PGPR. The 21% discrepancy between measured OM% and the expected value fits into the range described by Whitman et al. [45]. The OM% of the BC was relatively high (22.5%), and this labile fraction can be mineralised in the soil within a short time [46,47]. The N content of the BC was also comparatively high (0.96%), which may also have facilitated positive priming in the soil [6,48]. In the combined treatments the priming could have been inhibited by PGPR, which had an antibacterial impact due to the presence of Paenibacillus peoriae, and may thus have had a negative effect on the soil microbe community [49]. Compost may also have alleviated priming by decreasing the C/N ratio of the added organic materials in the soil [50].
BC and compost had similar CaCO 3 % contents, so both materials increased the soil pH due to the liming effect. This may be the main beneficial effect on soil fertility on this acidic sandy soil, as also reported in the study of van Zwieten et al. [11]. As a result, the liming effect directly influenced the N cycle in the soil. Both BC and compost influenced the NO 3 -N content, but in contrasting ways as a function of their liming effect. Ammonification is a less pH-dependent process than nitrification, so the higher pH in the BC treatments could have facilitated an increase in NO 3 -N content, while NH 4 -N remained unchanged [19]. In the case of compost the slighter pH increment could have resulted in less intensive nitrification. In response to compost the NO 3 -N content in the soil declined, which could be attributed to plant uptake, since a considerable quantity of plant biomass was formed in this treatment, resulting in greater N uptake [51]. The ability of BC to promote nitrification and raise pH was also manifested in the combined treatments, resulting in higher NO 3 -N contents than when BC was applied alone, due to the joint N content of compost and BC [52]. In the combined treatments the high C/N ratio of BC could have led to the immobilisation of N, resulting in limited uptake by maize [52]. Despite the satisfactory P and K content, this may have inhibited biomass growth.
The increase in soil pH also had a certain effect on P and K availability. In combined treatments changes in the P and K concentrations in plant biomass and soil were mostly related to the nutrient content of BC, the pH-enhancing effect of which resulted in the optimum pH range for nutrient availability and plant uptake [44,53]. These treatments not only supplied rapidly utilisable carbon sources from the compost, but also provided protection to microorganisms due to the pore volume of a high dose of BC [54]. The consequent higher microbial activity and mineralisation could explain the 35% higher P and K availability values in these treatments compared to BC alone.
The exogenous OM provided by compost was more favourable for microbial decomposition, having a lower C/N ratio than BC. In addition, the compost itself contained a substantial number of microorganisms, resulting in higher SIR values in compost treatments than in BC treatments [5,55]. As also reported by Hussain [4], the microbial activity was higher in treatments with BC, compost and PGPR than when these materials were applied alone, but no significant differences were observed between the combined treatments. There was also a significant increase in plant biomass in the compost treatment, probably associated with the greater quantity of roots and root exudates, which may also have increased the microbial biomass in this treatment [56].
The ability of compost to influence P content and availability could be related to its high P content, but could also be attributed to the fact that the enhanced microbial activity induced by the compost solubilised organically bound P [57]. Separate PGPR application was also able to mobilise the soil P content through the activity of mineral phosphate-solubilising strains (Bacillus aryabhattai) [58], though this effect was only manifested in the maize P content. Although the inoculum did not contain K-solubilising bacteria, it promoted K uptake by maize and might have created more favourable conditions for mycorrhizal symbiosis (Table 5), thus enhancing both P and K mobilisation and plant uptake. In combined treatments this strengthened the effect of BC on the AL-K 2 O content (Table 6) [59,60], but for the other soil properties the effect of inoculum was suppressed by that of compost and BC, as also observed by Ohsowski et al. [22].
The availability of P has a direct effect on the AMF colonization of the roots. The compost treatment significantly decreased the value of AMF-A%, indicating that an improvement in nutrient supplies could substantially reduce the dependence of plants on symbiotic organisms [61]. BC may stimulate symbiosis on such low fertility soil, but high soil P content may inhibit AMF infectivity and colonisation [62]. When BC was applied alone symbiosis was stimulated, but in combined treatments the enhanced nutrient availability limited fungal growth.
PCA was performed on all the treatments to obtain a better understanding of the interactions taking place in the combined treatments, and the data showed that two principal factors explained 68.61% of the variance (Figure 2). The exogenous OM provided by compost was more favourable for microbial decomposition, having a lower C/N ratio than BC. In addition, the compost itself contained a substantial number of microorganisms, resulting in higher SIR values in compost treatments than in BC treatments [5,55]. As also reported by Hussain [4], the microbial activity was higher in treatments with BC, compost and PGPR than when these materials were applied alone, but no significant differences were observed between the combined treatments. There was also a significant increase in plant biomass in the compost treatment, probably associated with the greater quantity of roots and root exudates, which may also have increased the microbial biomass in this treatment [56].
The ability of compost to influence P content and availability could be related to its high P content, but could also be attributed to the fact that the enhanced microbial activity induced by the compost solubilised organically bound P [57]. Separate PGPR application was also able to mobilise the soil P content through the activity of mineral phosphate-solubilising strains (Bacillus aryabhattai) [58], though this effect was only manifested in the maize P content. Although the inoculum did not contain K-solubilising bacteria, it promoted K uptake by maize and might have created more favourable conditions for mycorrhizal symbiosis (Table 5), thus enhancing both P and K mobilisation and plant uptake. In combined treatments this strengthened the effect of BC on the AL-K2O content (Table 6) [59,60], but for the other soil properties the effect of inoculum was suppressed by that of compost and BC, as also observed by Ohsowski et al. [22].
The availability of P has a direct effect on the AMF colonization of the roots. The compost treatment significantly decreased the value of AMF-A%, indicating that an improvement in nutrient supplies could substantially reduce the dependence of plants on symbiotic organisms [61]. BC may stimulate symbiosis on such low fertility soil, but high soil P content may inhibit AMF infectivity and colonisation [62]. When BC was applied alone symbiosis was stimulated, but in combined treatments the enhanced nutrient availability limited fungal growth.
PCA was performed on all the treatments to obtain a better understanding of the interactions taking place in the combined treatments, and the data showed that two principal factors explained 68.61% of the variance (Figure 2).  While Factor1 correlated primarily with BC, Factor2 correlated with compost treatment. Except in the case of AL-K 2 O, significant changes that could be attributed to PGPR were only observed when it was applied alone. These results show that BC had a significant effect on macronutrient availability, which is decisive for plant production. The compost had the greatest influence on the soil biota, which may be related to the easier mineralisation of the organic compounds in this material due to its lower C/N ratio [16].

Conclusions
The main benefit of the combined application of BC, compost and PGPR on this acidic sandy soil was that it prevented the positive priming effect observed when BC was applied alone. When combined with BC, compost and PGPR increased the ratio of available to total P and K concentrations via the intensification of soil microbial activity, while compost and PGPR alone did not increase P and K availability. The microbial activity in the soil was mainly stimulated by the OM content of BC and compost and by the pH changes they caused, while the negative influence of the high P content of compost on the AMF parameters was mitigated by BC. In a sandy soil of this type, if the water supplies are adequate, an increase in biomass can only be achieved in the short term in response to the easily accessible nutrient content of the compost, while BC and PGPR are ineffective in this respect at the applied doses. When applied in combination with BC, the ability of compost to increase plant biomass may be counteracted by N immobilisation by BC, so on BC-amended soils with adequate water supplies it may be necessary to use more than the recommended doses of compost or other organic fertilisers in order to increase yields. Further research in connection with the use of biochar on low fertility soil will need to include a wider range of organic fertiliser and inoculum doses. Experiments carried out under field conditions over a number of growing periods can be expected to give a better picture of the possible synergistic effects of the tested materials on soil fertility, which could not be detected in the present study.

Conflicts of Interest:
The authors declare that there is no conflict of interest.