Introduction

It is likely that climate change-induced hydrological variations will threaten water resources for both rainfed and irrigated agriculture (FAO 2008). Agricultural drought may impair food security and economic prosperity in number of countries in the world (Schewea et al. 2014). All types of drought (meteorological, hydrological and agricultural) are interrelated, but agronomic drought is reported to be the most frequent (Lal 2009). It is affected mainly by the available water capacity, which depends on the soil properties, especially organic carbon contents and aggregation (Reich and Eswaran 2004; Bot and Benites 2005; Lal 2009), and approaches to alleviate water scarcity in agricultural production usually include soil organic matter (SOM) increases (Lal 2008).

Biochar (BC) has recently been proposed as an option for improving soil fertility, for carbon sequestration and greenhouse gas emission reductions (e.g. Woolf et al. 2010; Lehmann 2007a; Jeffery et al. 2011). BC increases soil organic carbon stocks, i.e. the stable organic matter fraction, and thus may have the potential to alleviate climate change problems (Lehmann 2006, 2007b; Laird 2008; Sohi et al. 2010; Atkinson et al. 2010). It may significantly improve nutrient availability, either by nutrient delivery from the BC itself, or by changes in nutrient retention and cycling, and thus the growth of plants (Glaser et al. 2002; Chan et al. 2007; Renner 2007; Lehmann and Joseph 2009). It has been shown to reduce nutrient leaching (Ventura et al. 2013) and greenhouse gas emissions (Kammann et al. 2012; Cayuela et al. 2013), and it may stimulate microbial activity (Singh et al. 2010). However, positive effects are not always guaranteed (Jeffery et al. 2011). With regard to water supply, it was reported that biochar improved the structure and water holding capacity (WHC) (Brodowski et al. 2006; Clough et al. 2013; Laird et al. 2010; Kammann et al. 2011) and soil hydraulic conductivity of the soil (Steiner et al. 2007; Karhu et al. 2011). Novak et al. (2009) reported an increase in WHC from 6.7 % in control to 15.9 % due to the addition of switchgrass biochar in loamy sand. An increase in the water supply to plants grown on sandy soils amended with BC has been reported by Buss et al. (2012) and Kammann et al. (2011), while others reported improvements in soil characteristics pertinent to soil-plant water relations, including the structure and development of micro-pores (Cheng et al. 2006; Bornemann et al. 2007; Major et al. 2009; Liu et al. 2012).

So-called “humic acid” can be bought as commercial products. They are complex organic molecules mostly generated from Leonardite, a brown coal precursor, as sodium or potassium salts. Such products have been shown to lead to changes in the surface chemistry of soil solids and to improve soil fertility (Amirbahman and Olson 1995). Humic compounds in general can have multiple beneficial effects on soil functions including biological activity, nutrient availability, cation exchange capacity, pH buffering, carbon sequestration, and soil-water relations (Drozd et al. 1997; Piccolo et al. 1996; Schnitzer 2000). They are reported to improve stress tolerance of plants by exerting hormone-like effects and stimulating the activity of microorganisms including those that produce growth promoting hormones (Zhang et al. 2005), e.g. root growth promotion (Vaughan and Malcom 1985; Trevisan et al. 2010). Therefore, theoretically, humic acid products could add to the beneficial effects of fresh BC additions on soil fertility and productivity, particularly under water limited conditions.

Drought stress can severely influence the plant metabolism such as physiological, biochemical, and molecular components of photosynthesis. It primarily causes stomatal closure at the whole-plant level to minimize further water loss (Cornic 1994; Lawlor 1995), ultimately reduces inflow of CO2 into mesophyll tissue (Flexas et al. 2006) and therefore decreases photosynthesis (Mwanamwenge et al. 1999; Yordanov et al. 2000). Water stress also inhibits the photosynthetic electron transport rate through photosystem II ‘PSII’ (Chakir and Jensen 1999), reducing the photosynthetic efficiency of plants while increasing non-photochemical quenching (heat dissipation). Thus it can be expected that plant-physiological reactions may detect the effects of BC or HAP on plant-soil water relations before the effects are detectable in the whole-plant biomass yield.

The main aim of the study was to evaluate the complementary or synergistic advantages of biochar and a commercial humic acid product on growth, water relations and photosynthesis of maize at limited and frequent (H2O limit , H2O frequ ) water supply. We hypothesised that (1) BC amendments to sandy soil will improve the water retention capacity, and thereby increase photosynthesis and plant-soil water relations, resulting in higher biomass yields in a “the more BC the better” manner; that (2) BC loaded with HAP will further improve soil and plant water relations compared to BC alone, and that (3) any beneficial BC and/or HAP effects will, in relative terms, be more pronounced when the water supply is limiting and plants experience drought stress than with frequent water supply.

Material and methods

Experimental setup and growth conditions

The sandy soil used in this study was obtained from the plough layer of the agricultural experimental station of the Institute for Plant Breeding and Agronomy I, Justus Liebig University Giessen at Gross-Gerau, Germany. The site is located (49°45′N and 8°29′E, 90–145 m above sea level) in the upper Rhine Valley with the river Main to the North, River Rhine to the west and Odenwald mountains to the east. The soil was formed from Rhine sand deposits and the agricultural area is frequently irrigated during hot spring or summer dry spells. The soil was silty sand that consisted of 85.2 % sand, 9.6 % silt, and 5.2 % clay. It contains a low amount of organic carbon (0.592 %) and total N (0.057 %), CAL-P 92.2 mg kg−1, CAL-K 124.5 mg kg−1, Mg 35.5 mg kg−1 and a pH (0.01 M CaCl2) of 6.31. Before use, the soil was air-dried, thoroughly mixed, and sieved (≤5 mm). Prior to the start of the experiment, the water holding capacity (WHC) was determined for soil or soil-BC mixtures as described by Kammann et al. (2011): the entire Mitscherlich pot was submersed in distilled water for 24 h and then allowed to drain for another 24 h with the soil surface covered. Pre- and post-pot weights were then compared to calculate the WHC in g H2O per g of dry soil. For seed germination and early plant growth, the WHC of the respective soil/soil-BC mixtures was adjusted to 60 % by daily watering. Growing plant weight in pots was accounted for by harvesting additional replicates that were grown for this purpose.

Biochar was produced from wood-chip sievings at 550–600 °C (Pyreg GmbH, Dörth, Germany). Collectively, the feedstock was comprised of wood chip sievings (needles, bark, twig pieces and small wood chips) of Picea abies (70%) and deciduous wood sievings of Fagus sylvatica (30%); needles roughly contributed 30% to the total feedstock. It contained 74.4 % C, <1 % H, 10.6 % O, 0.56 % N, 0.163 % P, 0.607 % K, 0.33 % Na, 1.907 % Ca, 0.209 % Mg, 0.259 % Fe, 0.002 % Cu and 0.017 % Zn. The particle size fractions were as follows: >6.3 mm 0.5 %, 3.15–6.3 mm 24.2 %, 2–3.15 mm 7.1 %, 1.6–2 mm 20.9 %, 1–1.6 mm 4.9 %, 0.63–1 mm 17 %, 0.1–0.63 mm 25.3 % and <0.1 mm 0.1 %. The BC, was sieved (≤2 mm) to get a 100 % mixture of particle size between <0.1 and 2 mm. It was oven dried at (105 °C) before use. The humic acid product (HAP; granulated potassium salt, 100 % water soluble) is a commercial product of Humintech GmbH, Germany, marketed as POWHUMUS® WSG 85.

In this three-factorial completely randomized greenhouse study, each of the 36 Mitscherlich pots (0.30 m in diameter and 0.175 m in height; 3 replicates per treatment) was filled with 6 kg of soil, or soil-BC mixture according to the following factors (1) “biochar” including i) 0 BC, ii) 1.5 % BC or 34.26 Mg ha−1, iii) 3 % BC or 68.53 Mg ha−1, (2) “humic acid product”, including iv) HAP, v) 1.5 % BC + HAP, and vi) 3 % BC + HAP. In all cases HAP was added at a rate equivalent to 8 kg ha−1 as recommended by the manufacturers. (The third factor, “water”, applied frequently or limited, is explained below.) HAP was applied in solution either directly to the soil (HAP-control) or after loading onto the required amount of biochar. To provide similar conditons, BC was moistened to 40 % of its WHC with the HAP solution to deliver an amount equivalent to 8 kg ha−1 when the respective amount of BC was added. The HAP-loaded BC was dried and applied to the soil during mixing and pot-filling as described above.

Soil in each pot was fertilized with 13 g of a compound fertilizer (Nitrophoska special blue) that contained 12 % nitrogen as NH4NO3, 5.2 % phosphorous as Ca(H2PO4)2.2H2O), 14.1 % potassium as K2SO4 and KCl, 1.2 % magnesium as MgSO4 × 7 H2O, 6 % sulphur, 0.02 % boron as H3BO3 and 0.01 % zinc as ZnSO4 × 7 H2O. In addition, 20 ml of micronutrient solution was added to each pot and thoroughly mixed. One litre of this solution contained 6.4 g copper (CuSO4 × 5 H2O), 14.3 g zinc (ZnSO4 × 7 H2O), 8.2 g manganese (MnSO4 × H2O), 0.86 g boron (H3BO3), and 0.06 g molybdenum (ammonium molybdate).

Five seeds of maize (cv. DKC-3399) were sown into each pot on May 29, 2012. After emergence the two relatively weaker seedlings were removed to maintain three healthy plants per pot. For the first 4 weeks of the experiment, soil water was maintained at 60 % WHC (as optimum watering) of soil or soil-BC mixtures by daily watering. On the 29th day after sowing (DAS), 3 replicates from each treatment were picked at random for frequent (H2O frequ ) or limited (H2O limit ) watering regimes, respectively. For H2O frequ , soil moisture was maintained at 60 % WHC by daily adjustment on a balance to the desired target weight. However, in the H2O limit treatment, the supply of water was reduced to 25–30 % WHC (e.g. not a sudden decrease from 60 to 25–30 % WHC was imposed but it was done be gradual decrease in terms of 2–3 days with the plants’ water consumption) until wilting symptoms became visible; wilting symptoms first occurred in the controls without biochar/HAP amendment. For the H2O limit treatment the pure control treatment (no BC/HAP) was the benchmark: The same amount of water that was daily provided to the control (with the first wilting symptoms visible) was applied to the BC, HAP and BC-HAP treatments, no matter if BC, HAP or BC-HAP treatments may have needed more water than the benchmark control treatment to reach the WHC of 25–30 %. In this way a moderate drought stress was imposed, with equal rainfall/water supply for all H2O limit treatment pots.

Chlorophyll content, photosynthesis, transpiration, and relative water content

Relative chlorophyll contents were measured with the SPAD-502 device (Minolta, USA) on the first fully developed leaf at the leaf base, middle and tip on 29th and 66th DAS (day after sowing) for all three plants in a pot; values were averaged per pot. Measurements of leaf transpiration and/or chlorophyll fluorescence were carried out on the same day for all treatments after achieving visible symptoms of water stress in the H2O limit treatments. Chlorophyll a fluorescence imaging techniques were used to monitor photosynthetic performance of plants (Schurr et al. 2006; Baker 2008). These techniques allow the estimation of the relative quantum efficiency of the electron transport through the photosynthesis apparatus, photosystem II (PSII) which reacts to environmental stresses (Ort and Baker 2002). The operating efficiency of PSII is characterized by two factors: a) the efficiency by which excitation energy is transferred to photo-synthetically active (open) PSII reaction centres, which can be estimated by the rate of heat dissipation in PSII antennae (non-photochemical quenching); and b) the electron transport efficiency of PSII to acceptors (photochemical quenching); the latter depends on the availability of CO2 or suitable electron sinks in the chloroplasts (Baker et al. 2001). A Junior-PAM, i.e. a miniaturised Pulse-Amplitude-Modulated photosynthesis yield analyser (Company Walz, Effeltrich, Germany), was used to image chlorophyll fluorescence kinetics parameters. Measurements were performed according to Schreiber et al. (1986) at 62th DAS on the adaxial side of same leaf on which transpiration measurements were made. The plants were dark-adopted for a minimum of 30 min prior to the measurements and the value of minimum fluorescence (Fo) was obtained by applying a modulated light (<0.1 μmol photon m−2 s−1) and that of maximum florescence (Fm) after imposing a saturating pulse of 10,000 photons (μmol m−2s−1) for 0.6 s (Pfundel 2007). The photochemical utilization, Y(II) or effective photochemical quantum yield, was calculated as:

$$ Y(II)=\left(F'm\hbox{--} F\right)/F'm $$
(1)

where F is steady-state fluorescence in the light and F’m is maximum fluorescence in the light when saturating light imposed (Genty et al. 1989). The non-regulated heat dissipation Y(NO) and non-photochemical heat dissipation Y(NPQ) were calculated according to Kramer et al. (2004). We also calculated apparent photosynthetic electron transport rate (ETR) by using Y(II) and photosynthetic active radiation (PAR, μmol photons m−2·s−1). The ETR calculation was made according to Schreiber et al. (1994) as:

$$ ETR=0.5\times Y(II)\times PAR\times 0.84\mu mol{m}^{-2}{s}^{-1} $$
(2)

where 0.5 is the fraction of excitation energy distributed to PSII and 0.84 is a standard factor representing the fraction of incident light absorbed by a leaf.

Stomatal resistance (S. cm−1) and transpiration (mmol m−2 s−1) were measured between 9 a.m. and 12 p.m. in the last week prior to harvesting, i.e. 60th DAS, on the first fully developed leaf using a steady-state porometer LI-1600 (LI-COR, Inc. LTD., Lincoln, USA).

Before harvesting the plants on 66th DAS, the relative water content (RWC) of the first fully developed leaf was determined by taking leaf discs of 0.013 m diameter (3 leaves of 3 plants per pot). After noting the fresh weight (FW), leaf discs were floated overnight on well watered filter paper in glass petri plates for rehydration at 4 °C. Turgid weight (TW) was then taken after gently blotting water from the surface of the leaf discs using tissue paper. Leaf samples were oven-dried at 70 °C for 48 h to obtain the dry weight (DW) and RWC computed by using the equation:

$$ RWC\left(\%\right)=\left[\left(FW\hbox{--} DW\right)\right]/\left[\left(TW-DW\right)\right]\times 100 $$
(3)

Osmotic potential (Ψπ)

Leaf samples (first fully developed leaf) were frozen at −80 °C just after excision from intact plants. For measurements, frozen leaf samples were brought to room temperature, cut into small pieces, put in Eppendorf tubes, and incubated at 100 °C in a water bath for 15 min. Leaf sap was collected for Ψπ (osmotic potential) determination (in MPa). The 50 μl of leaf sap was taken in eppendorf tubes and Ψπ was measured by using the freeze-point depression method with a cryo-osmometer (type, 030 Gonotec, Germany).

Quantification of sugars

Water soluble sugar contents were determined by the Ludwig and Goldberg (1956) method after drying and grinding of first fully developed leaf. A 0.5 g of dry, ground leaf material was taken in 20 ml screw cap glass tubes. Subsequently, 10 ml of deionised water was added and final weight of the glass tubes was recorded. The tubes were incubated in a water bath at 100 °C for 1 h and deionised water was added where needed. Extract was filtered (Rotilabo-activated carbon filter papers round, Ø 185 mm) and stored in a refrigerator at 4 °C. One ml of hot water extract (diluted as necessary) was pipetted in another screw capped tube and 2 ml of anthrone reagent was added. The mixture was again incubated for 11 min in a water bath at boiling temperature. Afterwards the reaction was terminated by rapidly cooling the glass tubes in an ice bath. The observations were taken at 630 nm by Beckman photometer ‘(Beckman Coulter inc., Fullerton, USA) using deionised water as blank and final sugar concentration was calculated on the basis of a glucose standard curve (12.5–100 mg L−1).

Final harvest

At 66th DAS, the plants were clipped at the soil surface and data on plant height and fresh weight of leaves and stems were recorded. Dry mass was recorded after drying at 70 °C for 48 h. Root biomass was collected by sieving the soil from each pot through a 2 mm mesh sieve and gently but thoroughly washing the sieved roots with tap water. Roots were blotted and dried at 70 °C.

Soil respiration

Soil respiration was measured within 30 min after removing the plant tops from pots using a LI-8100 soil efflux chamber system (LICOR, Nebraska, USA). The large survey chamber (0.2 m diameter) fitted exactly to the brim of the Mitscherlich pots that were used in the experiment. The offset (height between soil surface and pot brim) of each pot was entered into the LI-8100-driving software for calculation of the correct system volume and thus of the soil CO2 efflux. Measurement time and observation delay were set to 60 and 20 s, respectively, to provide sufficient time for chamber-volume mixing and CO2 concentration increase. The increase in CO2-concentration always showed a linear slope with R 2 > 0.99. The flux was calculated automatically by the LI-8100 software that used the ideal gas law and linear regression. The respiration rate is given as CO2 flux in μmol m−2 s−1.

Carbon and nitrogen content of biomass, NUE and WUE

The leaf and stem biomass of the three plants from each pot was milled using a Retsch mill type SM300 (Hahn, Germany) with a 0.5 mm sieve. An aliquot of the plant material (~200 mg) was combusted in a CN analyzer (Vario MAX, Elementar Analysensysteme Gmbh, Hanau, Germany) for the determination of the N concentration. Nitrogen use efficiency (NUE) was calculated as above ground biomass dry matter produced per unit of fertilizer-N applied. Water use efficiency of productivity (WUEP) was calculated on the basis of dry matter produced (g) per unit of water consumed.

Soil mineral nitrogen and moisture contents

Gravimetric soil moisture content was measured after completely removing roots from the soil. Soil mineral nitrogen (NO3 −and NH4 +) was quantified using the methods of Keeney and Nelson (1982). A 20-g portion of soil was mixed with 80 ml 2M KCl, shaken for 1 h at 100 rpm and filtered (Round filter ø 70 mm S and S type 595). Concentrations of NH4 +-N and NO3 −-N were determined colorimetrically using an auto-analyzer (Seal, Germany).

Statistical analyses

The effects of all three factors (BC, HAP and water regime) were determined using three way analysis of variance (ANOVA) unless stated otherwise. Means were separated at the P ≤ 0.05 level with the Tukey HSD test. Data were occasionally log-transformed to ensure normal distribution (Komolgorov-Smirnov test) or homogenous variances (Levene median test). Linear regression analyses were also performed to describe the relationship among different parameters. All statistical tests were performed using Sigma Plot 11.0 (Systat, Inc., Richmond, USA).

Results

Plant growth and yield response

Plant growth and productivity was significantly enhanced (6.5 to 7.9 %) by addition of BC which was largely the result of greater stem heights and weights. Water limitation reduced the plant biomass by 35 % while the root:shoot ratio was increased (Fig. 1a, Table 1, Table S1). No difference was found between 1.5 and 3 % BC addition compared to the respective control in any parameter. Addition of HAP had no significant effect on yield parameters (Fig. 1a, Table 1, Table S1) with the exception of a significant negative effect (p ≤ 0.033) of the BC x HAP interaction on the root mass.

Fig 1
figure 1

Impact of biochar application (BC 0, 1.5 and 3 %) with or without humic acid product (HAP) addition under two water regimes (frequent or limited supply) on a) aboveground dry matter yield (bars show means of stem (lower bar part) plus leaves (upper bar part); error bars give the standard deviation of the aboveground biomass; n = 3), b) water use efficiency of productivity, (error bars = stdev. of means, n = 3); means with similar letters are not significantly different. Lower-case letters show differences due to the BC treatment within “Frequent H2O” while upper case letters show differences within “Limited H2O” when the water treatment effect was significant, respectively; the factor HAP was not significant, see statistical results, Table 1

Full size image
Table 1 Results of three way ANOVA’s with factors biochar (BC 0, 1.5 and 3 %) and humic acid product (HAP) addition under frequent and limited water supply on Leaf DW = leaf dry weight (g), Stem DW = stem dry weight (g), Total DM = total aboveground dry biomass (g) and WUEp = Water use efficiency of productivity (g kg−1)
Full size table

Water use efficiency of productivity and plant soil N dynamics

Biochar addition at 1.5 % was best to improve WUEP in the H2O frequ treatment while in the H2O limit treatment, BC addition did not significantly improve WUEP (Fig. 1b; Table 1). This is reflected in a significant BC x H2O interaction (Table 1). Frequent watering generally reduced the WUEP significantly. HAP addition had no effect on WUEp.

Tissue N concentrations decreased significantly by 23.4 to 22.9 % due to BC addition (Fig. 2a). Since the biomass increased, the NUE was significantly improved by 6.5 to 7.8 % (Fig. S1; Table S1); the 1.5 and 3 % addition results did not differ from each other (Fig. S1). Limited water supply resulted in significantly higher (53 %) tissue N concentrations as compared to frequent water supply (Fig. 2a; Table S1). This reduced in turn the NUE by 35.5 % in the limited compared to frequent water supply (Fig. S1; Table S1). Both, tissue N concentration and NUE were not significantly influenced by HAP addition (Table S1).

Fig 2
figure 2

Impact of biochar application (BC 0, 1.5 and 3 %) with or without humic acid product (HAP) addition under two water regimes (frequent or limited supply) on a aboveground tissue N concentration, b soil NO3 −-N left at harvesting; (bars show means + stdev., n = 3); means with similar letters are not significantly different. Letters as described in Fig. 1

Full size image

After the plant harvest, the NO3 −-N concentration in the control soils was zero, either in the H2O frequ or H2O limit treatment (Fig. 2b, Table S1). However, with BC, significant NO3 −-N concentrations were detected, and significantly more NO3 −-N was retained in the H2O limit than H2O frequ treatment. NH4 +-N amounts were negligible, with no significant effect of the treatments (not shown). HAP addition did not affect mineral N concentration.

Plant water relations and photosynthetic response

Limited soil water always significantly impacted the plant physiological parameters including the osmotic potential. However, BC addition supported the plants at both water regimes. Biochar addition significantly increased the relative water content (RWC) and the osmotic potential of the leaves (Ψπ) and generally enhanced transpiration due to significant decreases in stomatal resistance (Tables 2 and 3). The BC-induced improvements (most probably plant water availability) increased Ψπ while decreasing sugar concentrations compared to the control (Tables 2 and 3). This is further indicated by a significant negative correlation between the Ψπ and sugar concentration (Fig. 3a). HAP significantly decreased RWC, Ψπ and stomatal resistance but had no effect on transpiration (Tables 2 and 3). Biochar addition in the H2O limit treatments significantly decreased the chlorophyll content (Tables 2 and 3).

Table 2 Influence of biochar (BC 0, 1.5 and 3 %) and humic acid product (HAP) addition under frequent and limited water supply on photosynthesis and water/osmotic relations of maize; means ± stdev., n = 3; Chlorophyll (SPAD values) ETR = electron transport rate (μmol m−2 s−1), Y(II) = effective photochemical quantum yield, Y(NO) = regulated heat dissipation, Y(NPQ) = non-regulated heat dissipation, Y(II)/Y(NPQ) (ratio), Ψπ = leaf osmotic potential (MPa), RWC = leaf relative water content (%), Rs = stomatal resistance (S. cm−1) and Transpiration (mmol m−2 s−1)
Full size table
Table 3 Results of three and two way ANOVA’s with factors biochar (BC 0, 1.5 and 3 %) and humic acid product (HAP) addition under frequent and limited water supply on photosynthesis and water/osmotic relations of maize; deviation = F-value, p = p-value; Chlorophyll (SPAD values) ETR = electron transport rate (μmol m−2 s−1), Y(II) = effective photochemical quantum yield, Y(NO) = regulated heat dissipation, Y(NPQ) = non-regulated heat dissipation, Ψπ = leaf osmotic potential, RWC = leaf relative water content (%), Rs = stomatal resistance (S. cm−1) and Transpiration (mmol m−2 s−1)
Full size table
Fig 3
figure 3

Impact of biochar application (BC 0, 1.5 and 3 %) with or without humic acid product (HAP) addition under two water regimes (frequent or limited supply) on a correlation between osmotic potential and leaf sugar concentrations b correlation between stomatal resistance and electron transport rate, ETR (means ± stdev., n = 3). Dots within symbols indicate HAP treatment. (Treatment means and statistical results see Tables 2 and 3)

Full size image

As expected, water limitation negatively influenced photosynthesis: it decreased the photosynthetic electron transport rate (ETR) and the effective quantum yield (Y(II)), at the costs of increased heat dissipation Y(NO) and non-photochemical quenching Y(NPQ) in PSII (Tables 2 and 3). This resulted in a significant decrease of Y(II)/Y(NPQ), a ratio between the effective photochemical quantum yield and non-photochemical quenching. Biochar addition did not increase ETR and Y(II), but it increased Y(NO), reduced Y(NPQ), and thus increased the Y(II)/Y(NPQ) ratio (Tables 2 and 3). Also, HAP increased the Y(II)/Y(NPQ) ratio (Tables 2 and 3). The significant positive influence of BC on photosynthesis was coupled with better water supply by the BC amended treatments, as indicated by a significant negative correlation between ETR and stomatal resistance (Fig. 3b).

The HAP addition, in the absence of BC, appeared to have a positive influence on photosynthetic parameters (Table 2). Therefore, the effect of HAP was further investigated by two way ANOVAs with the factor water regime, excluding the data sets with the factor BC (Table 3, bottom). This confirmed that pure HAP addition indeed had a significantly positive effect on photosynthetic parameters which was masked by the effects of BC when the factor BC was included in the three way ANOVAs (Table 3).

Soil moisture contents and respiration (CO2 efflux) at harvesting

Gravimetric soil moisture content measured after the harvest was significantly higher in BC amended soil and higher in the H2O frequ than H2O limit treatment, respectively (Fig. S2b). In the H2O frequ treatment addition of 1.5 and 3 % BC increased the soil moisture by 48 and 129 %, respectively (Fig. S2b; Table S1) while HAP had no impact. Soil respiration (CO2 efflux) measured directly after cutting the plant tops (with the roots still in the soil) was significantly increased with BC addition in the H2O frequ treatment. Water shortage reduced soil respiration on average by 39 % (Fig. S2a; Table S1). HAP addition significantly (p ≤ 0.039) decreased the CO2 efflux relative to the treatment with no HAP (Fig. S2a, Table S1).

Discussion

Hypotheses revisited: expected and unexpected effects

Biochar addition clearly improved plant-soil water relations and plant eco-physiological traits, resulting in significantly increased maize biomass as observed earlier (Kammann et al. 2011; Yamato et al. 2006; Sukartono et al. 2011; Uzoma et al. 2011a, 2011b). However, increasing the BC amendment rate did not have linearly positive effects; responses at 1.5 and 3 % addition were mostly identical. We also hypothesized that HAP loading would improve the performance of BC, and that the beneficial effects will be more pronounced at limited compared to frequent water supply which was not the case.

Biochar effects on plant water relations and dry matter yield

Many studies report that biochar addition can considerably promote the water holding capacity (disturbed soil samples/mixtures), or field capacity (undisturbed soil cores) of sandy soils in particular (Abel et al. 2013; Artiola et al. 2012; Belyaeva and Haynes 2012; Case et al. 2012; Kammann et al. 2011, 2012; Kinney et al. 2012; Liu et al. 2012; Novak et al. 2012 and Rajkovich et al. 2012), but also in other soil types (Chan et al. 2007; Glaser et al. 2002 and van Zwieten et al. 2010c). Indeed significant increases of 12.5 and 24.7 % in the WHC were observed with 1.5 and 3 % BC addition to the sandy, SOC-poor soil, respectively. Although the available water capacity (AWC) (soil moisture between field capacity and permanent wilting point) was not determined, it was very likely enhanced. In other studies, the permanent wilting point was increased slightly with biochar addition (Abel et al. 2013; Cornelissen et al. 2013; Utomo 2013 or Brecht 2012) and interestingly, the amount of water held at field capacity increased to a larger extent than that held at the permanent wilting point, i.e. increase of AWC. Therefore, in the current study, the significantly increased WHC is taken as indication for an overall increase in the plant-available water that the BC-amended soil is able to deliver.

Improved biomass yields with biochar addition in greenhouse (Buss et al. 2012; Kammann et al. 2011; Mulcahy et al. 2013) as well as in field studies (Liu et al. 2012; Major et al. 2010; Vaccari et al. 2011; Baronti et al. 2014); were often attributed to an improved soil water supply. The two water regimes applied here were chosen to differentiate between growth-promoting effects caused by higher water availability (the H2O frequ treatment: WHC increase with biochar provided by daily adjustment to 60 % WHC), and positive effects “beyond more water supply” (the H2O limit treatment at the verge of drought stress, with equal reduced daily water supply to all treatments).

The biomass production results clearly show that biochar caused not only an improvement effect at higher WHC (H2O frequ treatment), but also when this surplus water supply was not allowed (H2O limit treatment). However, it was unexpected that in both water treatments the biomass increases due to BC amendment had the same relative magnitude; and that the positive biomass response was not linearly increasing with increasing biochar additions. The relationship followed a saturation curve with no difference between 1.5 and 3 % BC additions for most of the measured parameters. In the H2O frequ treatment, the water use efficiency of productivity, WUEP, was significantly increased only with 1.5 % but not 3 % BC amendment which was surprising. Therefore other, competing mechanisms may have ameliorated linear water-related effects of BC addition, such as phyto-hormonal signalling (Graber et al. 2010; Jaiswal et al. 2014), or nitrate capture (Ventura et al. 2013). We argue that biochar may have immobilised/adsorbed mineral-N which was therefore be unavailable for plant uptake because (i) significantly larger nitrate amounts were extracted from the biochar but not control treatments at the end of the study, and because (ii) a reduced N uptake into the plant biomass was observed. However, other reasons for the lack of a direct correlation to biochar addition cannot be ruled out and deserves further study.

Biochar effects on plant physiology

Generally, water limitation impairs photosynthesis by increasing stomatal resistance or through metabolic limitations (Cornic 2000; Lawlor 2002). However, plants have evolved not only osmotic and stomatal regulation mechanisms (Jones and Sutherland 1991) to cope with water shortages, but also defence strategies (xanthophyll cycle, photorespiration etc.) to alleviate the harmful effects of excessive energy under such stress conditions (Ort and Baker 2002). Biochar amendments improved the leaf osmotic potential Ψπ of Chenopodium quinoa plants which grew significantly better with addition of peanut hull biochar, either at sufficient water supply or drought (Kammann et al. 2011); the same was observed here with maize, and a woody biochar. The accumulation of sugars or other osmotically active substances lowers Ψπ under drought stress to maintain turgor, stomatal opening, photosynthesis and growth to a certain extent (Bolaños and Edmeades 1991; Kakani et al. 2011). In our study BC improved the osmotic potential which closely correlated to lower accumulations of soluble sugars. This corresponded to reduced stomatal resistance, larger transpiration rates, and higher relative water contents of the leaves at the harvest with BC additions.

The PAM chlorophyll fluorometer permits the assessment of excitation energy fluxes at PSII in three different pathways, termed Y(II), Y(NO) and Y(NPQ), which adds up to an unity. Any one or two of these can increase or decrease at the rate of the remaining one(s) in PSII (Kramer et al. 2004). Moreover, Genty et al. (1989, 1990) reported that Y(II) is directly related to the rate of CO2 assimilation in the leaf. In this study, BC amendment (without HAP) caused a relative increase in the electron transport rate (ETR) and Y(II) in PSII. Thus the ratio of the effective photochemical yield to the non-photochemical quenching Y(II)/Y(NPQ) significantly increased with BC addition so that more excitation energy was directed into the photosynthetic yield instead of energy loss. The more efficient photosynthetic energy gain finally resulted in higher biomass with BC amendment.

Stresses generally reduce photosynthetic efficiency and CO2 fixation. For example Qu et al. (2013) reported that combined salt and potassium stress significantly decreased Y(II) and increased Y(NPQ) or Y(NO) in maize. Other researchers have also reported lower photosynthetic CO2 gain due to declined Y(II) under severe drought stress e.g. in cucumber (Li et al. 2008). In the only other study where BC was applied to herbaceous plant species under salt stress (Abutilon theophrasti Medik. and Prunella vulgaris L.), and where photosynthetic performance was measured, Thomas et al. (2013) found no significant influence of BC amendments on photosynthetic carbon gain (Amax), chlorophyll fluorescence (Fv/Fm) or on water use efficiency. The authors amended BC at rates of 5 and 50 Mg ha−1, the higher rate of which is in between the BC application we used in this study (1.5 and 3 % correspond to 34.26 and 68.53 Mg ha−1, respectively). Their findings are in contrast to this study where significant improvements were observed with biochar addition, which is the first report of its kind to our knowledge, in both water treatments.

Taken together, the results indicate that the yield improvements were not only caused by an improved water supply (as evidenced by the results of the H2O limit treatment), but rather by subtle improvements of the plant water status and stomatal conductance, and thus changes in the performance of the photosynthetic apparatus (PSII photochemistry). Thus, biochar amendment increased the overall potential for photosynthetic carbon gain. The results clearly demonstrate that PSII photochemistry was positively impacted by BC soil amendment, even despite reductions observed in the relative chlorophyll content (see below). Biochar therefore acted dominantly along the ‘water-effect route’ of plant physiology.

Biochar effects on nitrogen dynamics

Lehmann et al. (2003) observed lower N uptake by cowpea in an Anthrosol due to charcoal addition. Similarly, in this study, BC addition decreased maize N uptake and decreased the leaf chlorophyll content. The reduced N uptake was likely not the result of N losses, as NO3 −-N was still present in BC treated soils even after the harvest.

If the amount of N removed with the above-ground plant material and the amount of mineral N left in the (BC-amended) soil is summed up, no differenc exist between treatments. Soil N retention was also observed by van Zwieten et al. (2010a); Taghizadeh-Toosi et al. (2011); Rajkovich et al. (2012) or Prendergast-Miller et al. (2011). However, there are studies where the N uptake was increased by increasing rates of BC addition, depending on plant species, soil bio-chemical properties and type of biochar (van Zwieten et al. 2010b; Chan et al. 2007). Lehmann et al. (2003) reported that plant productivity was increased even by 50 % less foliar N uptake; in our study the NUE was also increased at lower foliar N contents. It is unclear if (and if so, how) the remaining mineral N, mostly NO3 −-N, was bound to the biochar particles. Furthermore, it is unclear if the plants were unable to retrieve the mineral N that was extractable with KCl at the end of the study; or if there was no need for the plants to take up the remaining soil mineral N. Thus the question remains if the increase in NUE was a genuine physiological response of the maize plants, or if the plants were not able to take up the N. In the latter case the improved NUE with biochar would rather be a demonstration of their physiological plasticity. In line with Clough et al. (2013), our results suggest that the nature of the mineral N retention in the biochar-amended soil is more complex than we know so far and deserves further investigations.

Biochar and water treatment effects on soil respiration

The argument that water effects dominated the measured responses is backed up by the soil respiration measurements taken directly after the harvest. Effect of BC addition on soil respiration and CO2 efflux can vary considerably depending on biochar feedstock, soil type and moisture conditions, long-term land use and other factors that impact the soil microbial community (Bamminger et al. 2013; Kammann et al. 2012; Kolb et al. 2009; Spokas and Reicosky 2009; Ulyett et al. 2014; van Zwieten et al. 2010c; Zimmerman et al. 2011 and Hilscher et al. 2009). When plant roots were included as done here, soil respiration increased with BC addition, concomitantly with the root mass (Major et al. 2010). Here, the root mass was unchanged, but in the H2O frequ treatment soil respiration significantly increased with BC addition, whereas with limited water supply soil moisture and soil respiration were unchanged which is in line with the results of Zhang et al. (2012) or Kammann et al. (2011). The soil CO2 efflux was largely predicted by soil moisture with an exponential rise function (R 2 = 0.83; p ≤ 0.0001) but not by root mass (not shown). Therefore, this study showed that the most important effect of biochar was on the improvement of the water supply.

Effect of the added humic acid product

Humic acid products have been found to improve respiration and photosynthetic performance of plants before, by modifications in mitochondria functioning and chloroplasts (Orlov and Sadovnikova 2005); HAP amendment is often discussed for growth improvement (Trevisan et al. 2010). In our study, however, beneficial HAP effects were restricted only to small improvements in PSII photochemistry, increased stomatal conductance and Ψπ. This became only visible when the factor HAP was tested alone, omitting BC amendments. Either the stronger BC effects masked the smaller HAP effects; or, alternatively, HAP as complex organic molecules were adsorbed onto the BC surfaces and thus unavailable for interaction with the plant roots, since BC is a known strong adsorber of a variety of organic compounds such as PAHs (Smernik 2009; Schimmelpfennig and Glaser 2012; Hilber et al. 2012; Quilliam et al. 2013).

Conclusions

Global climate change has strong impacts on precipitation patterns and thus soil water resources. Therefore effective and farmer-friendly countermeasures are urgently needed. We observed that biochar can improve soil-plant water relations. The beneficial biochar effect enrolled via positive (i.e. self-reinforcing) feedback loops within the plants’ eco-physiological response capabilities: BC addition (1) increased Ψπ, RWC and transpiration while decreasing soluble sugars and stomatal resistance; (2) decreased chlorophyll contents with higher N leftover in soil, but still improved NUE and PSII photochemistry efficiency, as indicated by increased Y(II)/Y(NPQ) ratios; (3) improved WUEp even with daily adjustment to optimal WHC, and finally (4) these improvements resulted in a higher plant biomass yield. Biochar and HAP amendments both had positive effects on plant water and photosynthetic parameters (with BC > HAP). However, when used in combination, BC overruled the smaller positive effect of HAP, presumably due to sorption of HAP. Thus, HAP loading on BC, or their combined use, did not provide an improvement. Nitrogen retention in BC-amended soils deserves further investigations because the results suggested restrictions for plant N uptake, which may have been the cause for the U-shaped or saturation-type responses in WUEp or biomass production to increasing amounts of biochar, respectively. For future research the use of well-designed watering regimes may be helpful to identify and develop best-suited (designer) biochars for improving crop water relations.