The Mineral Fertilizer-Dependent Chemical Parameters of Soil Acidiﬁcation under Field Conditions

: Soil acidiﬁcation in agroecosystems is a natural process that could be accelerated, mainly by the inappropriate application of mineral fertilizers, or prevented, by sustainable management practices. On the basis of a three-year ﬁeld study in a grassland agroecosystem, the impact of di ﬀ erent rates of fertilization with nitrogen (N), phosphorus (P), and potassium (K) on soil chemical parameters related to soil acidity was evaluated. It was found that high-rate fertilization with ammonium nitrate accelerated the soil acidiﬁcation process, which was additionally intensiﬁed by the application of superphosphate and potassium salt. The sum of exchangeable base cations, the values of base saturation and hydrolytic acidity in the soil reﬂected the interactions between the applied NPK-fertilizer levels. Considering chemical parameters related to soil acidity studied in this experiment, it seems that the best strategies for mitigating soil acidiﬁcation in grasslands are reducing nitrate leaching, changing fertilizer types and increasing the input of base cations.


Introduction
Soil acidification, one of the primary constraints for the global productivity and sustainability of agriculture, affects almost 40% of the world's farmlands. It threatens agroecosystem functions and services and leads mainly to: (i) a decrease in cation exchange capacity and a reduction in the potential of soils for the retention of macro-and microelements in cation forms, which are subsequently susceptible to losses; (ii) a limitation of biological processes in soils that have remarkable cascading impacts on biodiversity and nutrient cycling, mainly nitrogen and carbon, including the formation of greenhouse gases; (iii) less availability/deficiencies of nutrients; (iv) the mobilization of potentially toxic elements and their subsequent transfer in the food chain; (v) the depression of crop growth and yields [1][2][3][4][5][6][7].
Although soil acidification is a naturally existing process, it has been intensified recently as a result of the overuse of nitrogen fertilizers [5,6,[8][9][10]. Today approximately fifty percent of the world's population is provided with food obtained with the use of them. The anticipated population increase to almost ten billion by 2050 will elevate agricultural production by another 50-98%. Increased food Sustainability 2020, 12, 7165 2 of 11 demand and dietary changes are likely to contribute to the further intensified use of fertilizers and thus soil acidification [11][12][13].
The main mechanisms of soil acidification induced by nitrogen fertilizers are connected with: (i) the displacement of acid ions and basic cations from the sorption complex, which leads to an increase in active acidity and the leaching of base cations with accompanying anions, respectively; (ii) the absorption by plants and/or microorganisms of the NH 4 + and the secretion of H + (physiological acidity), cation has the potential to contribute to intensifying acidification [1,2,4,5,9,14,15]. Gundersen et al. [16] estimated that annually one kilomol of hydrogen ions per hectare can be liberated as the result of the leaching of 14 kg NO 3 -N ha −1 . It is generally assumed that soil acidity rises with increasing doses of N fertilizers and their acidification potential decreases as follows: urea > ammonium nitrate = urea-ammonium nitrate > calcium-ammonium nitrate > calcium nitrate. Additionally, agroecosystems intensively fertilized with N may act as a source for acidifying gaseous pollutants (NO x and NH 3 ) [12]. According to a meta-analysis made by Tian and Niu [14], soil pH dropped linearly with the rising doses of nitrogen applied, which was to a greater extent caused by ammonium nitrate and urea than by NH 4 -form fertilizers. The dynamics of the soil acidification in response to nitrogen application depend on abiotic as well as biotic factors (e.g., climate, land cover, and soil properties). Generally, high precipitation, especially when it outweighs evaporation, low temperatures, and plant species, found for example in the grasslands, uptaking high amounts of base cations and/or NH 4 + -N or secreting organic acids, accelerate acidification, while large contents of organic and mineral colloids, particularly with a predominance of permanent, pH-independent charges on their surfaces (e.g., vermiculite and montmorillonite) and high base cation pool mitigate this process [3,10,14]. Until now, the majority of studies concerning soil acidification has focused on nonagricultural areas [10] or took into account only nitrogen fertilizers in arable lands, while very little research has attempted to assess their increasing levels and interactions with phosphorus and potassium fertilizers under grassland agroecosystems.
The aim of this study was to evaluate the impact of different rates of fertilization with nitrogen (N), phosphorus (P), and potassium (K) and their interactions on chemical parameters related to soil acidity (pH, hydrolytic acidity, exchangeable base cations, and cation exchange capacity).

Site Description and Experimental Design
A three-year field study was conducted at the Experimental Farm in Elizówka (51 • 17 20.739" N 22 • 34 32.766" E) on Luvisol (loamy sand) with an acidic reaction (pH = 4.9 ± 0.05), a low content of organic carbon according to the European Database on Soils (10.4 ± 0.1 g SOC kg −1 ), a very high availability of phosphorus (77.2 ± 0.78 mg P kg −1 ), a high availability of potassium (187.0 ± 1.23 mg K kg −1 ), and a low availability of magnesium (44.0 ± 0.49 mg Mg kg −1 ). The test plant was timothy grass (Phleum pratense L.), variety Skrzeszowicka. The annual precipitation was 428.4.0 mm, 442.4 mm and 454.0 mm in the first, second, and third year of the experiment, respectively. Average monthly temperatures in January and July were: −3.2 and 20.4, −5.0 and 18.5, −3.7 and 20.2, respectively. The field experiment was set up in a split-split-plot design with four replicates. The experiment scheme included three factors: (i) three levels of nitrogen rates (120 kg N ha −1 (N1), 240 kg N ha −1 (N2), 360 kg N ha −1 (N3)); (ii) three levels of phosphorus rates (39.4 kg P ha −1 (P1), 69.8 kg P ha −1 (P2), 104.7 kg P ha −1 (P3)); and (iii) three levels of potassium rates (83 kg K ha −1 (K1), 166 kg K ha −1 (K2) and 249 kg K ha −1 (K3)). The lowest NPK rate (N1, P1, K1) was applied according to fertilizer Nitrogen in the form of ammonium nitrate (AN) (34% N) was used in three split doses at the beginning of the growing season and after the first and second cut (in May and July, respectively). The whole dose of P in the form of granulated triple superphosphate (SP) (20.1% P) and a 1 2 dose of K in the form of potassium salt (PS) (47.3%) were applied prior to the sowing of timothy seeds. In the second and the third year of the field study, PK fertilization was used after the 3rd cut (in September). The second rate of K was incorporated in the soil after the first cut (in May) each year.

Soil Sampling and Analysis
Soil samples were collected after each cut at the depth 0-20 cm in accordance with ISO 10381-1:2002 [17], air dried and sieved (<2 mm).
In the prepared soil samples the chemical properties were determined as follows: pH in 1 mol KCl dm −3 (potentiometric method) [18], hydrolytic acidity (HA) and exchangeable base cations (BC) with Kappen's method [19]. The Cation Exchange Capacity of the soil (CEC) was calculated according to the following formula: CEC = HA + BC. The data were presented in the paper in the form of the means of the three years of studies.

Statistical Analysis
The statistical analysis of the results was performed using Statistica PL software (Tulsa, OK, 78 USA). A three-way analysis of variance (ANOVA) and Tukey's mean separation were used to determine the statistical significance at p < 0.05. The Pearson's linear correlation coefficient was calculated with a significance level of p < 0.05 for all studied parameters.

Results and Discussion
The obtained results demonstrated that the application of NPK fertilizers accelerated the acidification process of the studied soil. The significant decrease in soil pH, by 0.5 and 0.2 units on average over three years, occurred under the influence of the triple (360 kg N ha −1 ) and double (240 kg N ha −1 ) doses of nitrogen, respectively (Table 1). Other authors also reported that fertilization with ammonium nitrate reduces the soil pH in timothy agroecosystems. The soil pH decreased from 5.9 to 5.3 following the applications of NH 4 NO 3 in a seven-year grassland. Annual treatment with 270 kg N ha -1 as AN induced the pH to drop by almost 1.3 units after 25-year-cropping of Phleum pratense L. [20,21]. Interestingly, a meta-analysis of 106 studies [14] showed that, due to ecosystem adaptation, a significant impact of nitrogen use on soil acidification was found in studies with a duration of less than 20 years. Table 1. Impact of NPK fertilizers on soil pH.
pH declined significantly when the fertilizer level exceeded 200 kg N ha −1 , while below that threshold only slight changes in its value were found. Extensive inputs of ammonium based fertilizers in grasslands accelerate soil acidification predominantly by enhancing the nitrification of ammonium nitrogen to nitrate nitrogen, followed by nitrate leaching accompanied with base cations, promoting their uptake in excess of anions, and BC removal in the repeatedly harvested plant biomass [5,6,20,21,23]. Moreover, 16% of N presented in NH 4 NO 3 as ammonium cation could induce physiological acidity and displace base cations binding to the soil surface, increasing their susceptibility to leaching loss and reducing acid-buffering capacity [14].
Sustainability 2020, 12, x FOR PEER REVIEW 4 of 11 The significant differences in soil pH values observed in the experiment were undoubtedly influenced by the nitrogen rates applied (Table 1, Figure 1). Some researchers [22] stated that soil pH declined significantly when the fertilizer level exceeded 200 kg N ha −1 , while below that threshold only slight changes in its value were found. Extensive inputs of ammonium based fertilizers in grasslands accelerate soil acidification predominantly by enhancing the nitrification of ammonium nitrogen to nitrate nitrogen, followed by nitrate leaching accompanied with base cations, promoting their uptake in excess of anions, and BC removal in the repeatedly harvested plant biomass [5,6,20,21,23]. Moreover, 16% of N presented in NH4NO3 as ammonium cation could induce physiological acidity and displace base cations binding to the soil surface, increasing their susceptibility to leaching loss and reducing acid-buffering capacity [14].  environment the grass yield might be reduced, nitrogen use efficiency decreased, and the excess of reactive nitrogen that appears could be subjected to losses generating H + ions [10,24]. This mechanism was not confirmed statistically in the conducted experiment, probably due to the high resistance of Phleum pratense L. to acidification (9 on a ten-point scale) [25].
In the present study, the acidifying effect of ammonium nitrate was intensified by phosphorus and potassium fertilization; and significant interactions between NP, NK, PK, and NPK were observed (Table 1, Figure 1). The lowest pH values (4.1) were noted in N 3 P 2 K 2 , N 3 P 2 K 3 , N 3 P 3 K 1 , N 3 P 3 K 2 and N 3 P 3 K 3 treatments.
Water-soluble phosphorus fertilizers-chemically acid, but physiologically alkaline, contributed to soil acidification to a lesser extent than the nitrogen one, by approximately 0.4 unit at the highest dose ( Table 1). The amount of lime required to neutralize their effect equals 15 kg CaCO 3 per 100 kg of SP, while for AN-61 kg CaCO 3 [3,6].
Additionally, the dissolution of superphosphates in soils leads to the formation of H + ions. The influence of fertilization with potassium chloride on lowering the soil pH is mainly related to the activation of exchangeable and physiological acidity and the leaching of the accompanying anion (Cl). The application of potassium fertilizers also results in the displacement of Ca 2+ ions from the sorption complex, yield increasing, and BC removal [1,3,26]. Although the low initial pH value could weaken the acidifying effect of applied NPK fertilizers in our experiment [10], their double and triple doses induced a change in acidification class of the soil from acidic to very acidic (Table 1, Figure 1).
Hydrolytic acidity as a quantitative indicator of soil acidification was closely related to the pH values (r = −0.950). Hence, the application of elevating doses of nitrogen, phosphorus, and potassium fertilizers contributed to a significant increase in HA, by 45.8%, 31.2%, and 12.9% respectively (Table 2), which was in line with the previous studies [27]. Blaser et al. [28] stated that the largest contents of exchangeable hydrogen ions were determined for pH < 5.2. Similarly, as in the case of the acidification qualitative indicator, i.e., pH, synergism was observed in the impact of the analyzed fertilizers on this parameter (Table 2, Figure 2). Under the combined treatments of the highest doses of NP, NK, and PK, the largest values of HA, 5.15 cm(+) kg −1 , 4.62 cm(+) kg −1 and 4.56 cm(+) kg −1 , respectively, were found ( Figure 2). The greatest acid saturation percentage was found in the plots with the double and triple doses of N, P, and K, which resulted in an increase in the degree of soil degradation level from very weak (25-40%) to poor (40-55%) and medium (55-70%) [3].   With increasing hydrolytic acidity, decreasing amounts of base cations remained in the exchangeable form (r = −0.788). Liberated hydrogen ions undoubtedly replaced the exchange sites of base cations in the colloids, and displaced the BC into the soil solution, liable to leaching losses, which implied a reduction in the acid buffering capacity of the soil [10,26]. According to Bolan [2], for each kmol of hydrogen ion remaining in the soil following nitrate leaching 50 kg of CaCO 3 is required to neutralize the acidifying effect that arises. The obtained linear relationship between BC and soil pH (r = 0.799) in the experiment demonstrated that soil pH was predominantly influenced by the exchange of base cations with H + ions in the range of pH from 4 to 7 [14,29]. The magnitude of leaching loss of BC cations is usually as follows: Ca 2+ > Mg 2+ > K + > Na + , i.e., in the order they are present in soils. In the conducted experiment, the exchangeable base cations capacity was also determined mainly by the contents of calcium (r = 0.989) and magnesium (r = 0.804) ions.
Increasing mineral nitrogen additions resulted in a significant decrease, by 25.9%, in the contents of BC (Table 3). The application of phosphorus and potassium fertilizers intensified this process and statistically significant interactions between NP, NK, PK, and NPK were found (Table 3, Figure 3). The lowest value of BC (2.91 cm (+) kg −1 ) and base saturation were observed in the N 3 P 2 K 2 treatment. Long-term applications of ammonium nitrate greatly reduced the levels of exchangeable calcium and magnesium in other experiments [30]. Under NP and NPK treatment, BC contents decreased sometimes by over 80% [9,10,31]. The meta-analysis made by Lucas et al. [16] suggested that nitrogen fertilization in grasslands caused a decrease in exchangeable forms of calcium, magnesium, and potassium by about 24% only in experiments which lasted less than 5 years. With increasing hydrolytic acidity, decreasing amounts of base cations remained in the exchangeable form (r = −0.788). Liberated hydrogen ions undoubtedly replaced the exchange sites of base cations in the colloids, and displaced the BC into the soil solution, liable to leaching losses, which implied a reduction in the acid buffering capacity of the soil [10,26]. According to Bolan [2], for each kmol of hydrogen ion remaining in the soil following nitrate leaching 50 kg of CaCO3 is required to neutralize the acidifying effect that arises. The obtained linear relationship between BC and soil pH (r = 0.799) in the experiment demonstrated that soil pH was predominantly influenced by the exchange of base cations with H + ions in the range of pH from 4 to 7 [14,29]. The magnitude of leaching loss of BC cations is usually as follows: Ca 2+ > Mg 2+ > K + > Na + , i.e., in the order they are present in soils. In the conducted experiment, the exchangeable base cations capacity was also determined mainly by the contents of calcium (r = 0.989) and magnesium (r = 0.804) ions.
Increasing mineral nitrogen additions resulted in a significant decrease, by 25.9%, in the contents of BC ( Table 3). The application of phosphorus and potassium fertilizers intensified this process and statistically significant interactions between NP, NK, PK, and NPK were found (Table 3, Figure 3). The lowest value of BC (2.91 cm (+) kg −1 ) and base saturation were observed in the N3P2K2 treatment. Long-term applications of ammonium nitrate greatly reduced the levels of exchangeable calcium and magnesium in other experiments [30]. Under NP and NPK treatment, BC contents decreased sometimes by over 80% [9,10,31]. The meta-analysis made by Lucas et al. [16] suggested that nitrogen fertilization in grasslands caused a decrease in exchangeable forms of calcium, magnesium, and potassium by about 24% only in experiments which lasted less than 5 years.   In the present study, changes in the contents of BC were closely related to cation exchange capacity (r = 0.570). The influence of mineral fertilizers on this parameter, with the average value of 8.76 cm (+) kg −1 , showed less differentiation than in the case of the other properties discussed (Table 4, Figure 4). In a previous study, cation exchange capacity was not influenced by nitrogen doses [32] or NPK fertilization [33]. According to Šimanský et al. [34], statistically significant changes in CEC values occur only at certain doses of nitrogen fertilization.  In the present study, changes in the contents of BC were closely related to cation exchange capacity (r = 0.570). The influence of mineral fertilizers on this parameter, with the average value of 8.76 cm (+) kg −1 , showed less differentiation than in the case of the other properties discussed (Table 4, Figure 4). In a previous study, cation exchange capacity was not influenced by nitrogen doses [32] or NPK fertilization [33]. According to Šimanský et al. [34], statistically significant changes in CEC values occur only at certain doses of nitrogen fertilization.

Conclusions
The study revealed that high-rate fertilization with ammonium nitrate accelerated the soil acidification process, which was additionally intensified by the application of superphosphate and potassium salt. The lowest pH values noted in N 3 P 2 K 2 , N 3 P 2 K 3 , N 3 P 3 K 1 , N 3 P 3 K 2 , and N 3 P 3 K 3 treatments were predominantly influenced by the exchange of base cations with hydrogen ions. The values of exchangeable base cations, their saturation levels and hydrolytic acidity in the soil reflected the interactions between the applied NPK-fertilizer rates.
Taking into account chemical parameters related to soil acidity as studied in this experiment, it seems that the best strategies for mitigating soil acidification in grasslands, recommended for further research, are: reducing nitrate leaching mainly by balancing fertilization and improving nitrogen use efficiency; changing fertilizer types to those that induce lesser acidification, e.g., phosphate rock; and increasing the input of base cations via liming or organic fertilization.