Journal of Elementology ISSN 1644-2296 Przygocka-Cyna K., Grzebisz W., Łukowiak R. 2018. Effect of bio-fertilizer amendment on agrochemical properties of soil cropped with vegetables. J. Elem., 23(1): 163-177. DOI: 10.5601/jelem.2017.22.3.1476 ORIGINAL PAPER EFFECT OF BIO-FERTILIZER AMENDMENT ON AGROCHEMICAL PROPERTIES OF SOIL CROPPED WITH VEGETABLES* 1Katarzyna Przygocka-Cyna, Witold Grzebisz, Remigiusz Łukowiak Chair of Agricultural Chemistry and Environmental Biogeochemistry Poznan University of Life Sciences, Poland AbstrAct Reuse of agricultural waste as bio-fertilizer is a key to closing the nutrient cycle and saving mineral fertilizers. To analyze the effect of two bio-fertilizers in an interaction with three crops grown in a consecutive crop rotation, a micro-plot experiment on poor loamy sand was carried out. Three series of a two-factorial experiment included two bio-fertilizers as the first factor with contrasting ratios of biomass ash and solids of BD (BAD): FE1 (2.2:1) and FE2, (1:2.2). The second factor was a dose of the applied BAD: 0.0, 20, 40, 80, 160, and 320 g m-2. Two crops were grown in a fixed cropping sequence: radish à green bean àradish. The applied BAD ferti- lizers resulted in very rapid changes in the soil pH, the content of plant available nutrients as well as heavy metals. The pre-plant patterns of the content of nutrients and heavy metals were only slightly modified by the grown crops. The key reason of the observed stability in soil agro- chemical properties was the variability in the plant available P, which significantly affected the content of the other elements, including heavy metals. The effect of phosphorus depended on the amount of BD introduced into the soil. Its higher doses resulted in a simultaneous increase in both the soil pH and the P and Ca content. An insufficiently low dose of BAD led to nutrient exhaustion, which was most severe with respect to K, Mg and Ca. The content of available heavy metals under the effect of the soil available P can be taken controlled, provided the con- tent of available K and Zn is increased. Keywords: BD, biomass ash, soil, nutrient availability, availability of heavy metals, cropping sequence. Katarzyna Przygocka-Cyna PhD, Chair of Agricultural Chemistry and Environmental Biogeo- chemistry, Poznan University of Life Sciences, Wojska Polskiego 71F street, 60-625 Poznań, Poland, e-mail: [email protected] ; [email protected] * This study is part of the research conducted within the EKO-BIONOM project of the NCBiR titled: Pro-ecological manufacturing organic-mineral fertilizers based on: by products of burning and bio-gasification of biomass. Agreement No. GEKON1/05/214543/38/2015. 164 INTRODUCTION A standard way to improve soil fertility is to apply lime, originating from local resources of lime rocks, and mineral fertilizers, derived from non-rene- wable resources. In the Polish agriculture, K and P are production limiting factors (Tujaka, Terelak 2012). An alternative pathway for replenishing these two nutrients, in order to keep the soil fertility high, is to use bio-fertilizers produced from agricultural waste (Nkoa 2014). The latest studies focus mainly on ashes from combustion of different kinds of agriculture biomass or forest debris. The key attribute of bio-ash is its high alkalinity. Hence, it can be used for soil pH amelioration. It is well documented that bio-ash is a good source of P, K, and Ca, but the content of any specific nutrient depends on the incinera- ted plant species. Therefore, bio-ash is considered a good soil amendment, increasing soil fertility. One of the key disadvantages of bio-ash is the content of heavy metals, mainly lead and cadmium, which is a potential threat to any food cycle (CiesielCzuk et al. 2011, james et al. 2012, PiekarCzyk 2013). The biogas production in the last decade has become one of the most pro- gressing branches in agriculture waste utilization. The EU policy for renewable energy assumes a 25% contribution of biogas in the coming future. The main sources of raw material are farmyard manure and maize silage (Holm-NielseN et al. 2009, Łagocka et al. 2016). The real challenge for both biogas producers and scientists is to develop an efficient strategy of biogas slurry management. Biogas digestate (BD) contains 2-5% of dry matter, and numerous nutrients, easily available to crop plants (möller and müller 2012, kaTaki et al. 2017). However, there are some disadvantages of slurry management. First of all, pi- pes need to be installed for long distance transportation directly into fields. Secondly, the EU law prohibits slurry application to arable soils during winter months (Łagocka et al. 2016). In a whole-year operation of a biorafinery it is necessary to account for the cost of tanks for temporary storage of slurry in investment calculations. The disadvantages of separation of solids from the slurry are energy demand and loss of ammonia (møller et al. 2000). The simultaneous utilization of both groups of waste (ash and BD) as soil amendments can be considered as a way to close cycles of numerous nutrients (arTHursoN 2009). The solid fraction of BD, due to its high content of easily degradable organic matter and bio-ash as a mineral alkaline compo- nent, can be used to formulate a new type of bio-fertilizer (kolář et al. 2008). It should be assumed that any type of bio-fertilizer based on bio-ash requires a validation study oriented on its effect on both soil fertility and its heavy metal content (sHarma et al. 2017). The main objective of this study has been to evaluate the effect of two contrary types of bio-fertilizers, based on bio-ash and on the solid fraction of biogas digestate, on soil pH and the content of available nutrients and heavy metals. Another objective was to define optimum doses of each tested fertilizer required to maintain or even increase the soil fertility level in a nutrient exhausting, vegetable cropping sequence. 165 MATERIAL AND METHODS The effect of bio-fertilizers based on biomass ash and biogas digestate (BAD) on soil agrochemical characteristics was evaluated in three series of artificially prepared micro-plots with consecutively grown vegetables. Soil used in the study was topsoil characterized by a loamy sand texture, classi- fied as light soil. The agrochemical status of the soil was evaluated based on the Mehlich 3 method (meHliCH 1984). The soil pH was measured in a 1:2.5 soil 0.01 M CaCl suspension. Detailed description of the chemical properties 2 of the soil under study clearly indicates slightly acid pH and very low con- tent of available magnesium (Mg) and calcium (Ca), but not of phosphorus (P) and potassium (K) – Table 1. A two-factorial experiment was arranged as follows: 1. T he first factor: two types of BAD fertilizers composed of biomass ash (BA) + a solid residue of biogas digestate (BD) + phosphoric rock (PR) + elemental sulfur (S0). The tested BAD were composed based on con- trary shares of the first two components: a. FE1: BA-55% + BDs-25% + PR-15% + S0-5%; b. FE2: BA-25% + BDs-55% + PR-15% + S0-5%; Table 1 Soil geochemical status after two-week pre-plant incubation, mg kg-1 soil Level of pH P K Mg Ca Mn Fe Cu Zn Pb Cd Ni Factor factor T0# 5.40 158 144 69 569 50.0 140 1.11 3.10 4.20 0.13 0.50 BAD FE1 5.90 164a 166 80 612a 49.3 145 1.11 3.32 4.22 0.13 0.44 Fertilizer FE2 5.92 178b 167 79 738b 51.5 153 1.17 3.71 4.45 0.15 0.42 F test 0.01 6.3* 0.0 0.0 7.5* 2.8 2.1 1.6 3.0 2.8 3.5 0.4 0 5.35 147a 160ab 74ab 576a 55.9b 163 1.09ab 3.35ab 4.27 0.08a 0.22a 20 5.89 167ab 139a 67a 538a 48.3a 149 1.11ab 3.26a 4.36 0.14b 0.46b Dose, (D) 40 5.80 162ab 130a 68ab 543a 46.6a 143 1.02a 3.18a 4.28 0.14b 0.48b (g m-2) 80 6.03 166ab 147ab 74ab 648ab 50.7ab 146 1.17ab 3.28a 4.40 0.15b 0.45b 160 6.32 182bc 185b 86b 844bc 49.3ab 146 1.13ab 3.50ab 4.28 0.15b 0.47b 320 6.08 204c 238c 106c 900c 51.7ab 146 1.32b 4.54b 4.43 0.17b 0.49b F test 0.8 8.6*** 18.0***13.2*** 7.9*** 4.2* 1.1 2.7* 3.4* 0.2 13.6*** 9.3*** F test for the interaction BAD x D 0.21 1.32 1.60 0.34 2.68* 0.23 0.94 0.11 0.60 0.41 0.70 0.20 # Soil agrochemical status before the experiment; a numbers marked with the same letter are not significantly different; ***, **, * indicate significant differences at p < 0.001, p < 0.01, and p < 0.05, respectively. 166 2. T he second factor: control treatment and five doses of BAD: 0; 20; 40; 80; 160; 320 g m-2. A detailed description of the chemical composition and amounts of parti- cular elements incorporated into the soil is presented in Table 2. Two crops Table 2 Chemical composition and the amount of elements incorporated into the soil with BAD fertilizers Fertilizer Content of Fertilizer dose (g m-2) type the element 20 40 80 160 320 FE1 Element (g kg-1) amount of the element (g m-2) P 36.8 0.75 1.50 3.00 6.01 12.0 K 31.1 0.66 1.33 2.66 5.31 10.6 Ca 57.4 1.28 2.56 5.11 10.2 20.5 Mg 9.1 0.19 0.38 0.77 1.53 3.06 S 48.0 0.96 1.92 3.84 7.68 15.4 (mg kg-1) amount of the element (mg m-2) Fe 13586 271.7 543.4 1086.7 2173.8 4347.5 Mn 1392.4 27.9 55.7 113.9 222.8 445.6 Zn 205.1 4.10 8.20 16.40 32.8 65.6 Cu 58.3 1.17 2.33 4.67 9.33 18.7 Pb 21.8 0.44 0.87 1.75 3.49 6.98 Cd 3.6 0.07 0.14 0.29 0.58 1.15 Ni 14.9 0.30 0.60 1.19 2.38 4.77 FE2 (g kg-1) amount of the element (g m-2) P 36.4 0.73 1.46 2.91 5.82 11.65 K 19.7 0.39 0.79 1.58 3.15 6.31 Ca 40.4 0.81 1.61 3.23 6.46 12.91 Mg 5.2 0.10 0.21 0.42 0.84 1.68 S 48.0 0.96 1.92 3.84 7.68 15.36 (mg kg-1) amount of the element (mg m-2) Fe 6907.6 138.2 276.3 552.6 1105.2 2210.4 Mn 682.7 13.7 27.3 54.6 109.2 218.5 Zn 181.0 3.62 7.24 14.48 28.96 57.91 Cu 49.0 0.98 1.96 3.92 7.83 15.66 Pb 14.1 0.28 0.56 1.13 2.26 4.52 Cd 2.1 0.04 0.08 0.17 0.33 0.67 Ni 16.0 0.32 0.64 1.28 2.56 5.11 167 grown consecutively in a cropping sequence of radish à snap bean à radish were tested. The BAD fertilizers were applied in one dose at the beginning of the experiment and mixed into the soil layer of 7 cm. Mineral nitrogen at 4 g N m-2 was applied to all plots with radish, including the BAD control. P and K fertilizers were not applied there. The content and quantities of nutrients and heavy metals added in the soil with the BAD fertilizers are presented in Table 2. They are expressed in kg and g per ha in order to demonstrate the applied quantities. The water moisture regime during the whole experiment, which was 5.5 months, was kept at the field capacity. The incubation period, preceding crop cultivation, lasted 14 days. Composite soil samples (0-7 cm) were collected four times: at the beginning and at the end of each plant’s growing season (where the termination of one plant’s growing process was the onset of the growth of the next one). The soil sam- ples were then air-dried and crushed to pass a 2-mm mesh size sieve. The extractable nutrients and heavy metals were determined based on the Mehlich 3 method (meHliCH 1984). The content of available P in the extract was determined colorimetrically, while the content of K, Mg and Ca, Fe, Mn, Zn, Cu, Pb, Cd, and Ni was determined using FAAS. The experimental data were subjected to a conventional analysis of variance using the computer program Statistica 12®. Differences between treatments were evaluated with the Tukey’s test. In tables and figures, equations of the F test results (***, **, * indicate significance at the P < 0.1%, 1%, and 5%, respectively) are given. RESULTS AND DISCUSSION The pre-plant evaluation of soil agrochemical properties Application of BAD resulted in a soil pH change, showing a constant tendency, i.e. pH increasing progressively with the increasing fertilizer doses (Table 1). The initial content of soil available phosphorus (P) was high. The FE2 showed a significantly higher impact on the P content compared to FE1. However, the amounts of P incorporated into the soil with both BAD fertilizers were the same. Therefore, the observed increase in the content of plant available P can only be explained by the amount of degradable organic substances introduced into soil via BAD fertilizer. The FE2 was 2.2-fold richer in organic substances than FE1. The C:N ratio in the solid fraction of BD tends to be very narrow, ranging from 3.0 to 8.5 (møller, müller 2012). Hence, the FE2 composition could create much more favourable conditions for efficient hydrolysis of phosphoric rock (PR), which was the primary form of BAD fertilizers. This hypothesis is supported by the study of loria and sawyer (2005), who recorded a higher content of available P in response to an application of BD. This effect was probably enforced by the intensive release of protons from decomposing digestate (De la FueNTe et al. 2013). 168 Fig. 1. The effect of BAD fertilizers on the available calcium content following the pre-plant incubation. a Numbers marked with the same letter are not significantly different In the second group of nutrients, composed of potassium (K), magnesium (Mg) and calcium (Ca), only the latter element responded significantly to the type of BAD fertilizer. As shown in Figure 1, the net increase in the content of available Ca was influenced by the type of BAD fertilizer. For FE1, it star- ted from the dose of 1.6 t ha-1, whereas for FE2 the effect was seen since the dose of 0.8 t ha-1. In addition, the latter one raised the Ca content in the soil treated with 3.2 t ha-1 to over 1000 mg kg-1, i.e. improving the fertility class of the treated soil. This result cannot be explained by the amount of calcium applied to the soil with this dose of BAD. It was much higher in FE1. The observed phenomenon was probably due to the activity of micro- organisms, which received a higher amount of organic matter from digestate (De la FueNTe et al. 2013). The presented data clearly indicate that the low doses of BAD were not high enough to increase the content of available cations. The observed trend was even more conspicuous for K and Mg, as there was a noticeable correlation between their content. The relationship between K and Mg is presented below: Mg = 0.321K + 25.8 for R2 = 0.76, and n = 36. The third group of nutrients was represented by micronutrients. In ge- neral, they did not respond to the BAD type of fertilizer. Based on the micro- nutrient rank order proposed by zBíral (2016), the content of manganese (Mn), iron (Fe), and zinc (Zn), which was averaged over the experimental treatments, can be classified as falling within the medium class, whereas copper (Cu) content belonged to the low class. The impact of a BAD dose was nutrient specific. The content of available Mn and Fe decreased significantly in response to the increasing doses of BAD. Iron requires much more atten- tion. As shown in Figure 2, the content of available Fe decreased, following the pH value of 5.2. The content of Cu and Zn showed a progressing, yet unstable response to the increasing doses of BAD. 169 Fig. 2. The content of plant available iron following the pre-plant incubation as a function of soil pH variability The fourth group of elements comprises heavy metals, represented by lead (Pb), cadmium (Cd) and nickel (Ni). It ws only the content of Pb that did not respond to either of the experimental factors. It is interesting to notice that the Pb content in the soil showed a positive correlation with the P con- tent and a negative correlation with the Mg content. In general, the stepwise regression corroborates the above conclusion: Pb = 1.43 + 0.008P*** – 0.011Mg*** + 0.674Cu** + 0.011Fe*** for R2 = 0.84 and n = 36. The post-harvest evaluation of soil agrochemical properties The post-harvest evaluation of soil agrochemical characteristics was con- ducted on soil samples taken up directly after the harvest of each of the particular crop cultivated in the cropping sequence composed of radish-green bean-radish (Table 3). Soil pH The key soil property, i.e. soil pH, is in general, considered as the main cause of soil agrochemical variability (oHNo and eriCH 1990). The general pattern of soil pH response to increasing doses of both fertilizers was esta- blished during the growth of the first crop, i.e. radish. Its cultivation resulted in a significant drop in soil pH in the plot fertilized with 0.2 t ha-1 of BAD as compared to the N control. The further pH trend was significantly affected by an interaction of the fertilizer type, its doses and the crop (Figure 3). The appli- cation of FE2 resulted in an exponential increase in soil pH, excee- ding 6.5 in response to the BAD dose higher than 0.8 t ha-1. The effect of FE1 was much weaker, reaching only 5.5 in the plot with 0.8 t ha-1 The initial . 170 Table 3 Soil geochemical status after final harvest (mg kg-1 soil) Level Factor of pH P K Mg Ca Mn Fe Cu Zn Pb Cd Ni factor Re# 5.66a 142a 179c 86c 634c 55b 161b 1.14 3.60 4.08a 0.08a 0.16a Crop (C) Be 5.55a 178b 122a 75a 496a 52a 153a 1.15 3.76 4.66b 0.19b 0.57c Re^ 6.22b 142a 138b 82b 563b 53a 169c 1.11 3.68 4.10a 0,09a 0.27b F test 35.6*** 195.7*** 45.1*** 33.2*** 15.0*** 8.5*** 30.3*** 1.9 1.0 88.3*** 486.7*** 281.7*** FE1 5.55a 149a 140a 79a 492a 53 163b 1.12 3.53 4.30 0.11a 0.34 BAD fertilizer FE2 6.08b 158b 153b 83b 637b 53 159a 1.15 3.83 4.26 0.12b 0,33 F 58.8*** 27.8*** 5.7* 12.4*** 49.9*** 0.1 7.1** 2.5 10.2** 0.9 7.7** 0.1 0 5.81c 146a 141ab 77a 667c 54ab 157 1.10a 3.42 4.26 0.12ab 0.31 20 5.44a 152ab 132a 77a 451ab 53a 165 1.12ab 3.56 4.30 0.11a 0.34 40 5.47ab 147a 135ab 79a 422a 52a 163 1.10a 3.63 4.25 0.12a 0.33 Dose (D) (g m-2) 80 5.86cd 148ab 139ab 82ab 537b 52a 159 1.10a 3.48 4.25 0.11a 0.34 160 6.13cd 157b 158b 82ab 644c 54ab 160 1.19b 3.63 4.29 0.12ab 0.35 320 6.18d 173c 174c 87c 665c 56b 161 1.19b 4.36 4.33 0.14b 0.34 F test 13.6*** 24.1*** 6.9*** 7.4*** 19.1*** 3.1* 1.8 5.3*** 2.6* 0.4 4.6*** 0.7 F test for the interaction C x BAD 5.88** 1.49 1.88 3.22* 2.10 0.32 0.74 0.89 0.66 1.39 2.29 1.15 C x D 2.14* 0.89 1.09 1.54 1.64 0.45 0.32 1.35 0.86 0.82 1.05 1.48 BAD x D 7.29*** 1.58 3.35** 3.78** 6.61*** 2.21 1.84 1.28 2.64* 0.60 0.39 0.22 C x BAD x D 0.37 0.61 0.41 0.30 0.89 0.54 0.70 0.47 0.67 0.89 0.49 0.61 a Numbers marked with the same letter are not significantly different; # Re – radish; ^Be – green bean; ***, **, * indicate significant differences at p < 0.001, p < 0.01, and p <0.05, respectively. pattern of the pH response to BAD was enforced during green bean cultiva- tion. Its main feature was a significant drop in soil pH, irrespective of the fertilizer type. Radish, following green bean, caused a reverse effect of BAD on soil pH, although maintaining the primary pattern. The sudden drop in soil pH in response to the low doses of BAD is the most intriguing aspect of its impact on soil agrochemical characteristics. It indicates a low soil buffer capacity of soil in the study in response to the introduction of easily decom- posable organic matter. This phenomenon is frequently observed in soil amended with organic matter with a narrow C:N ratio (arTHursoN 2009). BAD fertilizers represent the same case. 171 Fig. 3. The response of soil pH to impact of the type and dose of BAD fertilizers in consecutive cropping sequences Macronutrients The second group of soil properties studied is represented by the content of available macronutrients (Table 3). All of them showed a significant res- ponse to the cropping sequence. The content of available P was significantly lower following radish as compared to green bean. At the same time, it was much higher in the soil treated with FE2. The effect of BAD doses was highly specific. A significant increase as compared to the N control was first recor- ded in the treatment of 1.6 t ha-1 of the applied fertilizer. The lack of interac- tion between the experimental factors indicates a constant effect of FE2 on P availability. The highest post-harvest P content in the soil cropped with green bean can be explained by the strong acidifying potential of this crop. As a result, this process leads to the increasing dissolution of P unavailable resources in both soil and applied fertilizers (arCaND, sCHNeiDer 2006). The other three macronutrients, such as K, Mg and Ca, show an opposite response to crops grown in the studied sequence. A significantly lower con- tent of all the three nutrients was recorded in the soil cropped with green bean. The strongest difference was found for potassium. The key reason was a high requirement of green bean for this nutrient (Przygocka-cyna et al. 2018). The advantage of FE2 over FE1 was mostly revealed in treatments with low doses of BAD and the P content never dropped below that in the N control (Figure 4). The K content showed a very similar pattern to the soil pH variability in response to BAD doses. The K content remained at the N control level, provided 1.6 t ha-1 of BAD was applied. This pattern was even more pronounced for Mg and Ca (Figures 5 and 6). The stabilizing effect of FE2 on the content of available Mg appeared first at a BAD dose of 0.4 t ha-1, and at a dose of whereas 3.2 t ha-1 of FE1. These two figures clearly point to digestate as the key factor affecting K nd Mg release from the soil exchange complex (chung and zasoski 1994). The observed phenomenon can be expla- 172 Fig. 4. The effect of BAD fertilizers on the post-harvest content of plant available potassium. a numbers marked with the same letter are not significantly different Fig. 5. The effect of BAD fertilizers on the post-harvest content of plant available magnesium. a numbers marked with the same letter are not significantly different Fig. 6. The effect of BAD fertilizers on the post-harvest content of plant available calcium. a numbers marked with the same letter are not significantly different