Ammonia volatilization with swine slurry injection and use of nitrification inhibitor

Submitted on June 23 , 2015 and accepted on April 18 , 2017. 1 Part of doctorate’s thesis of the first author. 2 Universidade do Estado de Santa Catarina, Centro de Ciências Agroveterinárias, Lages, Santa Catarina, Brazil. sr_roiber@yahoo.com.br; andreiapatricia74@yahoo.com.br; waltersbjr@hotmail.com; alvaro.mafra@udesc.brariandreola@yahoo.com.br; lgatiboni@gmail.com *Corresponding author: sr_roiber@yahoo.com.br Ammonia volatilization with swine slurry injection and use of nitrification inhibitor


INTRODUCTION
Many areas dedicated to swine farming in southern Brazil may be impacted by excessive and successive applications of swine slurry (SS), resulting in environmental constraints.The application of this organic material to the soil surface may promote ammonia (NH 3 ) volatilization and cause losses of up to 50% of the applied nitrogen (N), reducing its fertilizing potential (Cameron et al., 2013).
Technologies used to mitigate the polluting effects and to enhance agronomic use of N present in animal wastes have been evaluated.The injection of SS is an recommended alternative and it is already adopted in countries of temperate climate, proving to be efficient in reducing volatilization of NH 3 (Pote & Meisinger, 2014).This reduction by soil injection is justified by the lower exposure to air and the increase in the adsorption of ammonium (NH 4 + ) in the soil, due to the greater contact with soil particles (Webb et al., 2014).
Another alternative is the use of nitrification inhibitors, which have been investigated in many countries in order to reduce nitrate (NO 3 -) leaching (Zaman & Blennerhassett, 2010;Zhang et al., 2015;Gonzatto et al., 2016).Dicyandiamide (DCD) is one of the nitrification inhibitors used in other countries.This compound has bacteriostatic properties involved in the oxidation of NH 4 + to nitrite (NO 2 ), reducing the action of bacteria Nitrosomonas (Singh & Verma, 2008), by temporarily blocking the ammonium monooxygenase enzyme, which prolongs permanence of NH 4 + in the soil.Thus, it is necessary to evaluate whether the application of this inhibitor interferes with the levels of NH 4 + of the soil and affects the losses of N by volatilization (Vander Zaag et al. 2011).
These strategies used to reduce the environmental impact of SS, such as the injection and the use of the nitrification inhibitor, may present variable results on the volatilization of NH 3 (Kim et al., 2012), depending on the soil conditions and the form in which the organic waste was applied.
Thus, the hypothesis of this work is that the combined use of the nitrification inhibitor and SS injection reduces N losses by NH 3 volatilization, which would increase the efficiency of the use of this organic material as fertilizer.
The objective of this work was to evaluate the effect of SS and urea in two modes of application in the soil (injected and superface), and the use of nitrification inhibitor on NH 3 volatilization in a controlled environment, upon varying conditions of texture and pH of the soil.

MATERIAL AND METHODS
The experiment was carried out in a greenhouse under a completely randomized design, with three replications, in a 4 x 2 x 2 x 2 factorial scheme, which were, as follows: four fertilizers (urea, swine slurry (SS), SS+ nitrification inhibitor dicyandiamide-DCD) and control); two pH conditions (natural and corrected); two modes of fertilizer application (injected and surface); and two soils (clayey and sandy).
The experimental units consisted of polyethylene pots with a capacity of 700 mL, whose dimensions were 7.5 cm in diameter, containing 250 g of dry soil, maintained with moisture of 70% of field capacity (FC) for Rhodic Kandiudox (clayey soil) and 60% of FC for the Typic Hapludult (sandy soil).Moisture was previously tested, allowing aerobic conditions for biological activity and good physical condition, not causing deformation of the soil aggregates during the handling and setting up of the experiment.
Swine slurry was collected from anaerobic manure storage tanks from a production system of swine finishing.SS was characterized according to a methodology described by Tedesco et al. (1995), with dry matter of 156 g kg -1 ; total-N: 8.2 kg m -3 ; ammoniacal-N: 4.2 kg m -3 ; nitric-N: 0.1 kg m -3 ; pH: 6.6.The rate of SS applied was 21 m 3 ha - 1 , based on the recommendation of 140 kg ha -1 of N to reach an yield of 8 Mg ha -1 of mayze (CQFS-RS/SC, 2016).For the mineral fertilizer, the conventional urea (45% N ) was applied at the same N rate of the SS.
The nitrification inhibitor used was dicyandiamide (DCD).This product is presented as a white, synthetic powder, consisting of 81% of DCD and 6.5% of N-(nbutyl) thiophosphoric triamide (NBPT) in its formulation.It is commercially used in the United States and tested under experimental conditions in Brazil.The inhibitor was mixed to SS at the time of soil application at the rate of 10 kg ha -1 of active ingredient.
Fertilizer injection was performed on a tray by evenly mixing the sources of N (SS and urea (diluted in water)) with 250 g of soil.Moisture was standardized using water in the same volume of the SS, for the treatments with urea and in the control.For surface application, the sources of N were distributed with the aid of a pipette over soil surface in the pots.
The capture of soil volatilized NH 3 was based on a work carried out by Tasca et al., (2011), performed in falcon tubes with a capacity of 15 mL, containing 10 mL of H 3 PO 4 0.5 N with glycerin (1%) and two tapes of filter paper (1 x 8 cm) soaked in this solution to increase NH 3 contact surface with H 3 PO 4 .The falcon tubes were inserted by 2 cm into the soil, in the center of each experimental unit, where the pots were closed with a lid, which had six 2-mm holes to allow air circulation.Evaluations of NH 3 were performed daily from the first to the eighth day and from the 11 th to the 14 th day after fertilizer application, with monitored temperatures in the greenhouse (Figure 1).
The amount of volatilized NH 3 was determined daily by steam trapping in Kjeldahl semi-micro apparatus with distillation of a 10 mL aliquot, adding 10 mL of NaOH 10 M in each sample (Tedesco et al., 1995).
By the end of the evaluations, the daily and accumulated ammonia emissions were calculated, by discounting the value of the control for each fertilizer.The total accumulated from the fertilizers was expressed as percentage of applied N (equivalent to 140 kg ha -1 ).Emission of accumulated NH 3 was adjusted by Mitscherlich equation, Eq.2 (Clay et al., 2012): (1) Where: A and b -constants of the model, where A is the maximum theoretical value of accumulated ammonia and b is the adjustment coefficient; Y and x -dependent and independent variables, respectively.
Analysis of variance was performed by the F test and means of the experiments were compared by Tukey's test (P < 0.05).For the emissions of accumulated ammonia and percentage of lost N, the fertilizers and their modes of application were compared within the same pH; and each fertilizer with the same mode of application at different pH, both evaluating the soils separately.The statistical package used was SAS (2007).

Treatments main effect on daily volatization of ammonia
The effects on the daily emissions of NH 3 observed in clayey soil occurred for the type of fertilizer (Figure 2A) on the 3 rd , 6 th and 7 th day of evaluation.For the application mode and pH, an effect was observed on the 3 rd and 6 th day after fertilizer application (Figure 2C; Figure 2E).In the sandy soil, daily emissions of NH 3 displayed responses to fertilizers on the 1 st , 2 nd and 8 th day (Figure 2B), while for the application mode, an effect was found on the 1 st , 2 nd , 5 th and 8 th day (Figure 2D).In the pH factor, the effects occurred on the 1 st , 3 rd , 4 th , 7 th , 11 th and 14 th (Figure 2) days.
The highest volatilizations were observed on the third day after fertilizer application (Figures 2A and Figure 2B), in both soils.At the peak of volatilization, the addition of SS differed from the other treatments in the clayey soil (Figure 2A), which can be attributed to the high initial concentration of ammoniacal N added into the soil, what is in agreement with a work carried out with SS (Misselbrook et al., 2002).
By adding urea, volatilization starts after enzymatic hydrolysis and ammoniacal N release.In the sandy soil, no difference was found between the fertilizers on the third day of evaluation, probably because of the soil structure, which may have favored nitrate leaching.
The use of DCD together with the SS in the clayey soil (Figure 2A) reduced the NH 3 emission when compared to the SS without DCD on the third day, while in the sandy soil, DCD had no effect.The higher presence of NH 4 + in the soil with the use of DCD did not increase NH 3 emission because soils with a greater clay content have a lower tendency to lose N in the form of NH 3 due to their higher buffering capacity and the higher capacity of retaining ammonium.
For the mode of application of the N sources to the soil, the injection showed lower emission of NH 3 in relation to the surface application at the peak of volatilization in the clayey soil (Figure 2C).This reduction is due to the lower exposure of the manure to the air and to the higher N-ammoniacal retention in the soil particles (Dell et al., 2012).In the sandy soil, no difference in the volatilization of NH 3 was found between modes of application at the peak of emission (Figure 2D).The effect of temperature on volatilization of NH 3 may also have favored this loss.Tasca et al. (2011) found that the emissions occurring with the addition of urea in a Cambisol were 30% higher at 35 o C than at 18 o C. In this experiment, the peak of volatilization coincided with maximum temperature higher than 35 ºC (Figure 1), mainly when the fertilizers were applied on the surface.
Correcting soil pH (Figure 1E, Figure 2F) influenced the volatilization of NH 3 on both soils on the third evaluation day, when limed soil had higher emission.The absence of acid sites (H + ) prevents ammonia gas from returning to the mineral N (NH 4 + ) form, which contributes to the intensification of emissions.

Observed and theoretical accumulated ammonia volatilization
By analyzing NH 3 theoretical maximum loss (Table 1), it can be observed that with the addition of SS in the clayey soil with corrected pH on the surface application, the emission would reach 21.7 mg kg -1 , indicating that on day 14, 92% of maximum volatilization would be reached.However, the greatest difference between the observed value and the theoretical maximum occurred in the treatment where SS + DCD was applied in the clayey soil at natural pH and with surface application, which represented an additional loss of 29.8%, indicating that on the 14 th day, 70% of theoretical maximum volatilization would be reached.
In the sandy soil, accumulated theoretical losses that had the largest difference between the observed accumulated value and the theoretical (Table 1), occurred due to the application of SS (17.8 mg kg -1 ) and SS + DCD (17.8 mg kg -1 ) on the surface with adjusted pH, followed by SS injection with adjusted pH (15.4 mg kg -1 ), where both indicate that 80%, 81% and 78% of the maximum volatilization would be reached on the 14 th day, respectively.

Effect of interaction of treatments on accumulated ammonia volatilization
It is observed in the accumulated emissions of ammonia (Table 2) that in clayey soil at natural pH (4.8), the highest emission occurred in the SS + DCD on the surface (8.48 mg kg -1 ), which was similar to that where SS (6.32 mg kg - 1 ) and urea (6.69 mg kg -1 ) were added.The fertilizers injected in the clayey soil at natural pH presented the lowest accumulated NH 3 volatilization, which is equivalent to the addition of urea and SS on the surface.For the emissions that had occurred in the clayey soil under adjusted pH (6.0) the addition of SS to the surface showed the highest accumulated NH 3 emission (19.9 mg kg -1 ), differing from the other treatments.
The highest emissions due to the addition of SS to the surface in the clayey soil at natural and corrected pH may be attributed to the high ammoniacal N content of the SS (4.2 kg m -3 ), which promotes NH 3 volatilization, and to the surface application, which facilitates gas exchange.In addition, the acidity correction results in a lower amount of H + ions and there would be less transformation of NH 3 into NH 4 + .The accumulated emission of NH 3 with addition of urea on surface of the clayey soil (Table 2) may be related to the ammonification of urea, which increases the pH of the soil in micro sites, due to the consumption of protons (H + ) and, consequently, increasing NH 3 emissions (Chen et al., 2013).Webb et al. (2014), injected animal manure into a clayey and sandy soil, and found a reduction in NH 3 emission in relation to surface application.Effects of DCD on ammonia volatilization can be variable.Zaman & Blennerhassett (2010), mixed DCD with animal urine and observed an increase in NH 3 emissions; on the other hand, Pujol (2012), added DCD to SS and concluded that there was no increase in NH 3 emission in a sandy soil.Ni et al. (2014), applied DCD with urea on the surface of a sandy soil, and did not observed any increase in accumulated ammonia emissions.
By comparing the two pH conditions (Table 2) in the clayey soil, the pH correction resulted in the highest ammonia emissions in relation to the natural pH occurred, except for the application of SS + DCD on the surface (5.85 mg kg -1 ).
In the emissions verified in the sandy soil (Table 2) under natural pH (4.2), no effect of DCD was found.The highest emissions occurred in the treatments with SS + DCD (12.72 mg kg -1 ) and SS (9.90 mg kg -1 ), both applied on the surface.Aita et al. (2014) studied the injection of SS + DCD in sandy soil and verified that the emissions of NH 3 were smaller, in relation to the superficial application.For the limed sandy soil (pH 6.8) all treatments had higher NH 3 emissions, compared to the natural pH, except for the injected treatments, where the lowest emission was observed in the SS + DCD treatment (8.53 mg kg -1 ).Tasca et al. (2011), evaluating the volatilization of NH 3 with the addition of urea in Cambisols with corrected (6.0) and natural pH (5.5), concluded that volatilization of NH 3 increased in the limed soil.
The highest accumulated emissions observed with fertilizers applied on the surface are in accordance with Gonzatto et al. (2013), who by adding 60 m 3 ha -1 SS on the surface of an sandy soil, observed an increase in the volatilization of NH 3 .

Effect of the interaction between treatments on the percentage of N lost by volatilization
The percentage of N lost by NH 3 volatilization (Table 3) in the clayey soil showed no difference between treatments with natural pH.A difference was found between the two pH conditions in the treatment with SS + DCD applied to the surface, where the largest loss of N (22.4%)occurred in the corrected pH.
In the sandy soil (Table 3) in natural pH, the highest percentage of lost N occurred in the SS + DCD applied on the surface (6.8%), equivalent to the SS with no DCD.In this same soil with corrected pH, the lowest N loss was observed in the injected SS + DCD, which is similar to the treatment with injected urea.Treatments under acid soil conditions showed lower N loss, except the SS + DCD surface treatment which was similar to the condition of corrected soil (Table 3).

Figure 1 :
Figure 1: Minimum, maximum and average temperature during experiment period, under controlled conditions.

Figure 2 :
Figure 2: Main effect for daily ammonia volatilization from application of fertilizers (Urea, SS, SS+DCD, and Control), with two application modes (Injected and Surface), two pH conditions (Natural and Corrected) in a Rhodic Kandiudox (A, C, E) and Typic Hapludult (B, D, F).Vertical bars: represent the minimum significant difference by the Tukey's test (p < 0.05); ns: not significant.SS: swine slurry.DCD: Dicyandiamide.

Table 1 :
Observed and theoretical accumulated ammonia (mg kg -1 ), with addition of fertilizers (SS, SS + DCD, urea) applied in two conditions of pH (natural and corrected) and two modes of application (injected and surface), in a Rhodic Kandiudox and a Typic Hapludult soil SS: swine slurry; DCD: Dicyandiamide; 1a Cumulative loss values discounted from the emissions of the control; 1b Equation of accumulated ammonia loss adjusted by Mitscherlich's model.

Table 2 :
Volatilization of accumulated ammonia, with addition of fertilizers (Urea, SS, SS + DCD,) applied in two pH conditions (natural and corrected) and two modes of application (injected and surface), in a Rhodic Kandiudox and a Typic Hapludult soil

mg kg -1 ) (mg kg -1 )
LSD: least significant difference; CV: coefficient of variation; Upper case letters compare fertilizers applied in the same soil in different pH; Lower case letters compare fertilizers with mode of application in the same soil at equal pH.Tukey (p < 0.05).

Table 3 :
Percentage of nitrogen loss in relation to the applied nitrogen from fertilizers (Urea, SS, SS + DCD) applied in two pH conditions (natural and corrected) and two modes of application (injected and surface), in a Rhodic Kandiudox and a Typic Hapludult soil : least significant difference; Upper case letters compare fertilizers applied in the same soil in different pH; Lower case letters compare fertilizers with mode of application in the same soil at equal pH.Tukey (p < 0.05). LSD