Band application of acidified slurry as an alternative to slurry injection in a Mediterranean double cropping system: Agronomic effect and gaseous emissions

Agriculture, Ecosystems and Environment 267 (2018) 87–99

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee

Band application of acidified slurry as an alternative to slurry injection in a Mediterranean double cropping system: Agronomic effect and gaseous emissions

T

⁎

David Fangueiroa, , José L.S. Pereirab,c, Irene Fragab, Sónia Surgya, Ernesto Vasconcelosa, João Coutinhod a

LEAF, Instituto Superior de Agronomia, Universidade de Lisboa, Tapada da Ajuda, 1349-017 Lisboa, Portugal CITAB, University of Trás-os-Montes and Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal CI&DETS, Escola Superior Agrária de Viseu, Instituto Politécnico de Viseu, Quinta da Alagoa, 3500-606 Viseu, Portugal d Centro de Química, University of Trás-os-Montes and Alto Douro, Quinta de Prados, 5000-801 Vila Real, Portugal b c

A R T I C LE I N FO

A B S T R A C T

Keywords: Acidification Gaseous emissions Organic fertiliser Oat Maize Injection Cattle slurry Nutrient recovery

Injection is the recommended technique for slurry application to soil in most European countries but its utilization at the farm scale is quite limited, namely in countries from southern Europe, due to the strong investment needed in machinery and problematic utilization in stony and/or heavy soils. Acidification of animal slurry has proved to be efficient at minimising NH3 emissions but little is known about its impact on other greenhouse gas (GHG) emissions or agronomic effect, particularly in Mediterranean conditions. In the present study, we evaluate the potential of band application of acidified slurry as an alternative to raw slurry injection, in terms of agronomic effects and NH3 and GHG emissions, for two different Mediterranean soils (a sandy and a sandy-loam soil) where a double-cropping system (oat during winter and maize during spring/summer) was run over 3-years. Five treatments were tested in 1 m2 field plots: 1) control (non amended soil); 2) injected slurry (IS); 3) band application of raw slurry followed by soil incorporation (SS); 4) band application of acidified slurry followed by soil incorporation (AS); 5) band application of acidified slurry with no soil incorporation (ASS). An amount of slurry equivalent to ∼90 and 170 kg N ha−1 was applied before oat and maize sowing, respectively. The dry matter yields obtained with the AS treatment, in both the maize and oat crops, were mostly similar to or higher than those of IS, while ASS led - on some occasions - to small decreases in dry matter yield relative to IS, namely in the sandy soil. Treatment AS led also to apparent N and P recovery values (ANR and APR, respectively) similar to or higher than those of IS, except in the sandy soil during oat growth. After six consecutive slurry applications, a significant decrease of pH and an increase of the extractable S content were observed in soil receiving acidified slurry, relative to soil amended with non-acidified slurry. Significant NH3 emissions were observed only in SS treatment during all the experiment. Of the total N applied, the amount lost as N2O did not differ significantly among the amendment treatments during the oat growth. However, the cumulative N2O emissions from IS were significantly higher, relative to SS, AS and ASS, during maize growth. Higher cumulative CH4 emissions were observed during maize growth relative to oat growth, namely from IS compared to all other treatments. Band application of acidified slurry without soil incorporation reduced the N2O and CH4 emissions by 65% and 40%, respectively, relative to IS. The soil characteristics had no significant effect on the gaseous emissions for the acidified slurry treatments. It can be concluded that band application of acidified slurry followed by soil incorporation is an efficient solution to provide nutrients to plants while minimising NH3 and GHG emissions and can thus be proposed as an alternative to injection. Nevertheless, the impact of acidified slurry application on soil properties needs to be monitored in the long term.

⁎

Corresponding author. E-mail address: [email protected] (D. Fangueiro).

https://doi.org/10.1016/j.agee.2018.08.011 Received 18 December 2017; Received in revised form 9 August 2018; Accepted 12 August 2018 0167-8809/ © 2018 Elsevier B.V. All rights reserved.

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

1. Introduction

slurry acidification has also some impact on P availability to plants (Roboredo et al., 2012). Information about the impact of acidified slurry application to soil on NH3 and greenhouse gas emissions in Mediterranean conditions is still scarce or non-existent (Fangueiro et al., 2015c, 2016). On the other hand, by minimising NH4+ losses, slurry acidification might lead to higher N2O emissions after soil application since more substrate will be available for nitrification and denitrification processes. Nevertheless, recent studies indicated that the nitrification process is inhibited or almost delayed in soil amended with acidified slurry relative to non acidified slurry (Fangueiro et al., 2016). Therefore, lower N2O emissions could be expected from soil amended with acidified slurry relative to non acidified slurry. Our main hypothesis was that band application of acidified slurry is almost as efficient as slurry injection with regard to minimising NH3 emissions without increasing N2O emissions. Hence, the main objective of the present work was to evaluate the potential of band application of acidified slurry as an alternative to raw slurry injection, in terms of agronomic effects and NH3 and GHG emissions, for two different Mediterranean soils where a double-cropping system (oat followed by maize) was run over 3-years. Band application of raw slurry followed by incorporation was also considered, as the traditional method, and the impact of acidified slurry incorporation into soil following band application was also tested. The sub-objectives were: 1) to compare the fertilizing value of raw and acidified slurry; 2) to determine the impact of soil incorporation of acidified slurry on plant yield and nutrient removal; 3) to compare the effects of raw slurry injection and band application of acidified slurry on the agronomic value; 4) to assess the effects of the slurry management strategies tested here on general soil properties after three years of application.

Dairy cattle production in some European countries has been forced to undergo an important concentration and industrialization, resulting in a significant production of manure, namely slurry (liquid manure) that is now close to 55 million tonnes per year in Europe (Foged et al., 2011). Dairy cattle slurry is traditionally applied to agricultural soil as a source of nutrients and, over the last decade, the use of dairy slurry has been promoted as a substitute for or complement to mineral fertilizer in order to decrease production costs and increase nutrients recycling at the farm scale. The Portuguese dairy cattle production is concentrated in the North West region, where a double cropping system is traditionally used (Fangueiro et al., 2008). A spring/summer crop (mainly maize silage) is grown from April to the end of August and a second crop (oat or ryegrass) is grown from October until March. Before each crop, cattle slurry is applied as a basal fertilization that enables dairy farmers to apply most of the slurry produced during the growing period. This double cropping system is used in other European regions, such as Galicia (Spain) and the Po Valley (Italy) (Ovejero et al., 2016). The use of cattle slurry for partial or complete replacement of mineral fertilizer has been studied in different pedo-climatic conditions (Cavalli et al., 2016; Schröder et al., 2013) and the results show that this practice led to yields similar to those obtained with mineral fertilizers (Cavalli et al., 2016; Webb et al., 2013). However, slurry management, namely during and after soil application, leads to emissions of carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) and ammonia (NH3), all with important impacts on climate change, acid rain and ozone formation in the troposphere (Sommer et al., 2013; Bittman et al., 2014). Among them, the greatest concerns are related to NH3 emissions (Sommer et al., 2013) - which, besides decreasing the plant N use efficiency of animal slurry, have a strong impact on ecosystems (Bittman et al., 2014). Furthermore, the N:P ratio of the slurry does not match the N and P demands of the crops, leading to over-application of P when the application is based on N. Strict conditions for slurry application to soil have arisen in some European countries to minimize the associated environmental impact, namely gaseous emissions. The common solution to minimize NH3 losses in many countries, such as Portugal and Spain, is based on slurry incorporation into the soil after its surface broadcasting application, but this incorporation has to be performed as soon as the application is done. However, this solution is not as efficient as slurry injection at minimising NH3 emissions and cannot be applied in permanent grassland (Webb et al., 2014). Furthermore, surface broadcasting by a splash plate applicator, has been banned in some countries of Northern Europe due to its strong impact on NH3 emissions (Webb et al., 2010). For this reason, slurry injection into the 7–10 cm soil layer is today compulsory in several countries, since it reduces NH3 emissions strongly compared to surface broadcasting (Carozzi et al., 2013; ten Hoeve et al., 2016). The impact of slurry injection on other gases, such as N2O and CH4, is still not clear and previous studies led to contradictory results (Rodhe et al., 2006; Langevin et al., 2015). Indeed, Bhandral et al. (2009) concluded, from a field experiment, that slurry injection is efficient to minimize NH3 emissions without increasing N2O emissions but recently, Duncan et al. (2017) reported that slurry injection lead to a significant increase of N2O emissions (84–1152%) relative to surface broadcast. Furthermore, slurry injection implies a strong investment in specialised machinery that requires more energy consumption and may not be applicable in stony soils and/or small plots (Webb et al., 2010). More recently, slurry acidification before application to soil has been proposed as a solution to minimize NH3 emissions during and after soil application by surface banding (Kai et al., 2008; Bittman et al., 2014; Fangueiro et al., 2015a, 2015b; Cocolo et al., 2016; GómezMuñoz et al., 2016). Furthermore, laboratory studies indicated that

2. Material and methods 2.1. Study site and soils characteristics The present study was performed in lysimeters located at the experimental facilities of the Instituto Superior de Agronomia-LisbonPortugal (N 38.708098; W 9.185001). Thirty lysimeters (1 m × 1 m × 1 m) were used in this experiment, half of them containing a Haplic Arenosol with a sandy texture (70% coarse sand, 7% fine sand, 10% silt, 3% clay) and the other half containing a Haplic Cambisol with a sandyloam texture (27% coarse sand, 56% fine sand, 7% silt, 10% clay) (WRB, 2015). The main characteristics of the 0–200 mm soil layer at the beginning of the experiment are shown in Table 1. The soils had not received any fertilization in the preceding five years and the 0–20 m mm soil layer of each lysimeter was mechanically homogenized before the beginning of the experiment. The precipitation and minimum and maximum air temperature data recorded on-site during the experiment are shown in Fig. 1. 2.2. Raw and acidified slurry The raw cattle slurry used in this study was sampled twice a year (in autumn and spring) in the same commercial dairy farm located near Palmela (Portugal) during the 3-year experiment. The dairy cows were fed with ryegrass or maize silage and concentrated feed. Slurry was transported and kept at ambient temperature in plastic barrels (50 L) for approximately one week before application. In the 24 h before soil application of the treatments, raw slurry acidification was performed by addition of concentrated sulphuric acid (about 6 mL per L of slurry) to reach a final pH of 5.5, following the procedure described by Fangueiro et al. (2015b, 2015c). Representative samples of acidified and non-acidified cattle slurry were collected and then analysed for the characteristics shown in Table 2. The details of the standard analytical methods used to assess the physico-chemical properties of the soils and slurries studied are available in Fangueiro et al. (2015c). 88

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Table 1 Main soil properties at the beginning of the experiment (mean and standard deviation of 15 plots) and in each treatment considered after six slurry applications (n = 3). pH (H2O)1

OM content2 (g kg−1)

Extractable K3 (mg K kg−1)

Egner4

sandy

extractable S6 (mg SO4 kg−1)

Extractable P (mg P kg−1) Olsen5

Sandy Initial

7.1 ± 0.2

5.7 ± 0.4

26.8 ± 4.0

17.8 ± 1.6

/

/

Control IS SS AS ASS

7.4a 6.9ab 7.3a 6.4b 6.6b

5.0c 7.1b 12.8ab 10.0a 9.1ab

44.5bc 93.5ab 95.4ab 149.4a 55.9bc

25.0bc 31.3b 48.1a 50.6a 31.2b

5.9e 7.9d 14.8b 19.1a 9.3c

0.1b 0.5b 4.1b 52.2a 26.6ab

Sandy-loam Initial

6.1 ± 0.1

10.8 ± 0.6

94.6 ± 1.5

14.0 ± 1.9

/

/

Control

6.6a

7.2d

106.2c

24.8c

10.8b

IS SS AS ASS

6.3ab 6.1b 5.2c 5.2c

18.1c 18.3bc 22.8a 18.4ab

158.8b 265.0a 204.1b 185.9b

61.9ab 76.7a 57.7b 76.7a

19.8e 28.6 28.6d 38.8c 50.6a 42.5b

12.2b 14.9b 42.3a 40.4a

ANOVA Treatment (A) Soil type (B) A×B

*** *** ns

*** *** **

*** *** **

*** *** **

*** *** **

*** *** ***

ANOVA results presented referred to final soil properties. For each soil, different letters indicate values that differ statistically at p < 0.05 according to the Tukey test. ns, *, ** and *** mean that the interaction effects were, respectively, not significant (p > 0.05) and significant at p < 0.05, p < 0.01 and p < 0.001. 1 Soil:solution ratio - 1:2.5. 2 Dry combustion with near infra-red detector. 3 Ammonium acetate (1 M) pH 7. 4 Ammonium lactate pH 3.7. 5 sodium bicarbonate (0.5 M). 6 Calcium phosphate.

the soil, placing the raw slurry inside each row, and covering the slurry with soil. Band application was performed by hand: a watering can was used to apply slurry in four bands. Slurry incorporation was performed manually. Before each slurry application, the soil was ploughed to 100 mm. Slurry application was performed on the first days of November and May, for Oat and Maize, respectively. Oat (Avena strigosa cv. ‘Saia’) was sown seven days after slurry application, in four rows (interspace of 20 cm) at a density of 20 g m−2. Maize (Zea mays, cv. “Almagro”) was sown four days after slurry application, in two lines (interspace of 600 mm) with five plants per line. Oat was rain-fed only while maize was irrigated by sprinklers during the first two years and by drill irrigation in the third year.

2.3. Experimental details The experiment started in October 2012 and ended in August 2015, after the maize harvest. A meteorological station (Delta-T Devices, Cambridge, UK) located in the experimental site was used to collect rainfall and air temperature data during the experimental period. The experiment was a randomised block factorial design with two soil types and five treatments, with three replications each. The five treatments considered were: 1 Non-amended soil (Control); 2 Injection of raw cattle slurry (100 mm depth) (IS); 3 Band application of raw cattle slurry followed by soil incorporation (20 mm depth) (SS); 4 Band application of acidified (pH = 5.5) cattle slurry followed by soil incorporation (20 mm depth) (AS); 5 Band application of acidified (pH = 5.5) cattle-slurry without soil incorporation (ASS).

2.4. Crops and soil sampling Oat was harvested in the beginning of March and maize at the end of July during the 3 years experiment. At harvest, all the plants were cut and the fresh weight was assessed immediately. Plant drying was then performed at 65 °C until constant weight, to assess dry matter (DM) yield. The dry material was then ground (1 mm) and analysed for N, P, and K contents. Plant analyses were performed following methodologies described in Fangueiro et al. (2015c). The limitations inherent to the experimental device used (cubic lysimeters) meant that the crops did not reach maturity. Hence, harvest was performed between GS61 and GS92 for oat and close to the R1 stage for maize. The DM yield increase, relative to control treatment, was calculated for each amendment treatment as follows:

The traditional double-cropping forage system - growing oat from November to March, followed by hybrid maize between May and July was established and both crops were grown according to commercial practice but with no application of herbicides or insecticides. The rates of slurries applied in the assigned treatments were ca. 90 kg N ha−1 in autumn (oat crop) and ca. 170 kg N ha−1 in spring (maize crop) (see Table 2). Mineral P and K applications were also performed, when the plant-available P and K concentrations in the soil at the beginning of crop growth were lower than the recommended value. The control soil received only mineral P and K fertilizers, when necessary. The amounts of N, P, and K applied in each treatment are presented in Table A1 (Supplementary material). Slurry injection was simulated by opening rows (100 mm deep) in

DM yield increase (g g −1) =

DMtreatment − DMcontrol DMcontrol

(1)

With, DMtreatment: DM yield obtained with the amendment treatment and DMcontrol: DM yield obtained with the control treatment. 89

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Fig. 1. Monthly rainfall and maximum and minimum air temperature during the three seasons under study, compared with the long-term average.

The apparent N (ANR) and P (APR) recoveries were calculated as follows:

ANR (g g −1) =

APR (g

g −1)

(N uptake )treatment −(N uptake )control) N applied

((P uptake )treatment −(P uptake )control ) = P applied

Total ANR (g g −1) =

(∑

3 years

(N uptake )treatment −∑3 years (N uptake )control

)

∑3 years (total N applied)

(4) (2)

Total APR (g g −1) = (3)

(∑

3 years

(P uptake )treatment −∑3 years (P uptake )control

)

∑3 years (total P applied) (5)

Where, N and P applied refer to the manure and/or mineral fertilizers provided. The total ANR and total APR values were also assessed, to evaluate the global effects of the treatments on the three years of double cropping, and were calculated as follows:

The total N and P use efficiency (NUE and PUE, respectively) were calculated as described in Ovejero et al. (2016), using the following formulae:

90

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Table 2 Average physico-chemical characteristics of the acidified and non-acidified raw cattle slurry used and the amounts applied to the soil before each crop during the three years of the experiment (n = 3). Parameters

Raw slurry

Acidified raw slurry

oat

maize

oat

maize

Slurry composition pH (H2O) Dry matter (g kg−1) Electrical conductivity (mS cm−1) Total C (g C kg−1) Total P (g kg−1) Total K (g kg−1) Total N (g kg−1) NH4+-N (g N kg−1) NO3−-N (mg N kg-1) NH4+:total N ratio C:N

7.4 ± 0.2a 96.7 ± 14.0b 17.0 ± 5.4a 43.2 ± 5.8a 0.9 ± 0.0a 2.8 ± 0.5a 3.2 ± 0.3a 1.4 ± 0.2a 0a 0.43a 14a

7.6 ± 0.1a 150.7 ± 31.8a 10.8 ± 3.5b 49.1 ± 7.8a 0.7 ± 0.1b 2.7 ± 0.2b 3.7 ± 0.1a 1.5 ± 0.2a 0a 0.40a 13a

5.6 ± 0.1b 103.6 ± 16.7b 22.9 ± 9.7a 45.3 ± 5.9a 1.0 ± 0.1a 2.8 ± 0.5a 3.3 ± 0.3a 1.3 ± 0.2a 0a 0.41a 14a

5.6 ± 0.1b 158.8 ± 28.4a 11.2 ± 5.0b 51.7 ± 6.9a 0.7 ± 0.0b 3.0 ± 0.3b 3.6 ± 0.2a 1.5 ± 0.2a 0a 0.43a 15a

Application rate kg C ha−1 kg N ha−1 kg NH4+-N ha−1

1151b 85b 37b

2299a 172a 69a

1214b 90b 36b

2416a 167a 71a

Values presented with different superscripts within columns are significantly different (p < 0.05) according to the Tukey test.

Total NUE (g DM g N−1) =

(∑

3 years

(DM yield)treatment −∑3 years (DM yield )control

)

analysed per crop and for the whole experiment (3 years) considering an RCBD for factor “treatments”, with factor “soil” as a split-plot on “treatments” and factor “year” as a split-plot on factor “soil” (Little and Hills, 1978). Tukey comparisons of means (p < 0.05) were carried out for the factors and their interactions.

∑3 years total N applied

(6) Total PUE (g DM g P −1) =

(∑

3 years

(DM yield)treatment −∑3 years (DM yield)control

)

∑3 years total P applied

3. Results and discussion (7) 3.1. Weather conditions

Soil samples were taken at the end of each crop growth period and analysed for the parameters described in Table 1. The potential of nitrification was determined at the end of each crop cycle following the methodology described in Fangueiro et al. (2015c).

Air temperature and precipitation are two key parameters determining oat and maize crop growth; in this case, precipitation had a particular impact on oat since it was grown on a rain-fed basis. The minimum and maximum air temperatures as well as the precipitation values recorded in the 2012–2015 time period are presented in Fig. 1. During oat growth, the total rainfall between October and March was similar in the 2012–2013 (∼870 mm) and 2013–2014 (∼750 mm) growing seasons - being significantly higher than in 2014–2015, when only 215 mm were recorded. The records of the last 30 years indicate an average value of 570 mm of precipitation during the same time period (Fig. 1). So, the 2014–2015 growing season can be considered a dry year, with potential impact on oat growth. During maize growth, 93 and 83 mm of precipitation were recorded in 2013 and 2015, respectively, against 175 mm in 2014 - a value close to the 30-year average of 155 mm (Fig. 1). Since the maize crop was irrigated, no significant impacts of precipitation on DM yields were expected, even if some NO3− leaching might have occurred in 2014. No significant differences (p < 0.05) in the temperatures were observed among the 3-years of the experiment, with both the minimum and maximum values being very close to the 30-year average (Fig. 1).

2.5. Gas flux measurements The NH3 fluxes were measured by the dynamic chamber technique during almost 72 h after soil amendment, while the N2O, CO2 and CH4 fluxes were measured by the closed chamber technique during the whole growing period (from slurry application till harvest) (Harrison et al., 1995). A detailed description of the methods used to assess gas fluxes can be found in Fangueiro et al. (2015c, 2017). The cumulative gaseous losses were determined by considering the mean gas flux of two sequential sampling dates and then multiplying by the time interval between samplings. The global warming potential (GWP) for each treatment and forage crop was determined using the GWP coefficients for N2O (265) and CH4 (28) and following a procedure similar to those described by Fangueiro et al. (2017), using Eq. (8).

yield _scaled GWP =

(265 × ∑ N2 O) + (1 × ∑ CO2) + (28 × ∑ CH4 ) yield

3.2. Dry matter yield

(8) Where, yield scaled GWP is the net GWP per unit of oat or maize yield (g CO2-eq g−1), ΣN2O, ΣCO2 and ΣCH4 are the accumulated amounts of N2O, CO2 and CH4 released during oat or maize cultivation (g CO2-eq m-2) and yield is the oat or maize yield (g m-2).

The individual effects of treatment, soil, and year of growth - as well as their interactions - on oat, maize, and total DM yield were statistically significant (p < 0.001) (data not shown). Some significant differences in the DM yields of both oat and maize were observed between growing seasons in all treatments. In amended soils, the oat DM yield varied from 0.74 to 2.93 ton ha−1 while the maize DM yield varied from 1.29 to 11.09 ton ha−1, over the 3-years of the experiment (data not shown). As expected, the lowest DM yield - for all treatments - was obtained for oat during the first growing season. No fertilization had been performed over the previous 5-years and intense rainfall occurred

2.6. Calculations and statistical analysis Analysis of variance was performed using the statistical software package STATISTIX 7.0 (USA) to assess the effect of the soil type, treatment, and soil × treatment interaction. The data collected were 91

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Fig. 2. Increase in dry matter yield (g g−1) relative to the Control (non-amended treatment) (n = 3). For each soil and each crop, different letters indicate values that differ statistically at p < 0.05.

treatments while the opposite was observed for IS and AS. It is well known that the soil characteristics, namely the texture and OM content, have a significant effect on DM yield (Fig. 2). In the sandy soil, the NH4+ not lost by NH3 volatilization might be converted to NO3− and then easily lost by leaching. In the sandy-loam soil, with higher clay content, part of the NH4+ can be adsorbed (Scherer, 1993) while another fraction can also be lost by leaching. Acidification led to better results in the sandy-loam soil, probably because NO3− leaching was lower in this soil. Indeed, measurements of NO3− leaching performed in the same experiment during oat growth indicate that AS treatment did not increase NO3− losses relative to SS in both soils. Nevertheless, higher NO3− losses occur when the acidified slurry is not incorporated in soil following application (ASS), (data not shown). When considering the total DM yield over the 3-years, significantly higher DM yield increases were observed in the sandy-loam soil relative to the sandy soil (Fig. 2). Furthermore, differences between treatments can be highlighted: in the sandy soil, values followed the order IS≈ASS > AS > SS > CTR, while in the sandy-loam soil, IS, AS, and ASS led to similar values that were significantly higher (p < 0.05) than in SS (Fig. 2). It can then be concluded that band application of acidified slurry might be a good strategy (and a good alternative to slurry injection) for the double cropping system, with no production break. Our findings are in agreement with previous studies dealing with acidified slurry. Indeed, the first trials considering acidified slurry application to grassland reported DM yield increases of about 10%, relative to non-acidified slurry (Pain et al., 1987; Kiely, 1988). Later, Frost et al. (1990) applied acidified cattle slurry to perennial ryegrass in Ireland and observed an average increase in DM yield of 54%, relative to non-acidified slurry. More recently, Schils et al. (1999) also observed increases in DM yield and ANR for two different soils (sandy and loamy) following application of cattle slurry previously treated with nitric acid.

between October 2012 and March 2013, leading to significant N losses by NO3- leaching. By contrast, the highest DM yields - independently of the treatment - were obtained for maize in 2014. A significant increase (p < 0.05) in oat DM yield was observed for all amendment treatments, relative to the Control, during the three years (Fig. 2). The effect of slurry acidification on oat DM yield was relatively poor, with no significant differences (p > 0.05) between SS and AS in most cases (Fig. 2). When acidified slurry was not incorporated into the soil (ASS), the increase in oat DM yield was almost similar to that obtained for SS - except in 2012 and 2013 in the sandyloam soil, where the DM yield was significantly higher (p < 0.05) in ASS relative to SS (Fig. 2). The reason for this might be the fact that acidified slurry incorporation might promote N losses by NO3− leaching. Slurry injection led to an increase in oat DM yield relative to SS during the first year, but in the two following years no significant differences (p > 0.05) were observed (Fig. 2). Application of acidified slurry led to significant increases (p < 0.05) in the maize DM yields relative to SS, in both soils, and no DM yield decreases relative to the Control were observed over the 3years (Fig. 2). The effect of the soil incorporation of acidified slurry on maize DM yield is not clear, even if significantly lower (p < 0.05) values were obtained in ASS relative to AS on two occasions (Fig. 2). Acidified slurry incorporation in soil will always be performed before sowing, except in no-tillage systems, in order to obtain a more homogeneous slurry distribution in soil and avoid potential NH3 losses due to an increase in slurry pH after soil application. Although, the compulsory incorporation of slurry in the four hours following soil application should be revoked when slurry is acidified. This is particularly important when the time available for slurry application is limited due to machinery availability and it can also give farmers the opportunity to contract only slurry application and to perform slurry incorporation themselves (Webb et al., 2010). A significant increase (p < 0.05) in maize DM yield was observed in IS relative to SS, in both soils, over the 3-years, except in 2014 in the sandy-loam soil (Fig. 2). Treatment IS led to maize DM yield increases similar to or lower than those obtained in AS and ASS - except in 2013 and 2015 in the sandy soil, where the values in the AS treatments were significantly lower (p < 0.05) than in IS (Fig. 2). The oat DM yield increases in the sandy-loam soil were always significantly higher (p < 0.05) than in the sandy soil over the three years of the experiment, except in 2013 for the SS treatment (Fig. 2). However, the differences in maize DM yield between the two soils were not so clear: in the second and third years, maize DM yields were significantly higher (p < 0.05) in the sandy-loam soil than in the sandy soil, in all amendment treatments, but, in the first year, yields were higher in the sandy soil than in the sandy-loam soil for the SS and ASS

3.3. Apparent N recovery The values presented in Fig. 3 for the oat and maize crops are the average values over the 3-years of the experiment. The values of ANR varied from 4.4 (AS) to 9.2% (IS) of the applied N in the sandy soil and from 7.7 (SS) to 15.1% (ASS) in the sandy-loam soil during oat growth, while in maize N uptake varied from 4.9 (SS) to 11.5% (AS) in sandy soil and 6.7 (SS) to 15.2% (AS and ASS) in sandy-loam soil (Fig. 3). The effect of soil on ANR was significant (p < 0.001) over the whole experiment, except for maize in 2013 (Fig. 3). Similarly, the treatment applied and the interaction soil × treatment had a significant (p < 0.001) effect on ANR during the 3-years, except for oat in 2014 where the treatment effect was not significant (p > 0.05) (Fig. 3). The ANR values obtained in the sandy soil during oat growth with the AS and ASS treatments were significantly lower (p < 0.05) than for 92

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Fig. 3. Apparent N recovery, expressed as percentage of applied N (n = 3). For each soil and each crop, different letters indicate values that differ statistically at p < 0.05.

treatments IS and SS (Fig. 3). However, in the sandy-loam soil, IS, AS, and ASS led to similar values, significantly higher (p < 0.05) than in SS, indicating that acidified slurry application or slurry injection had no or a positive impact on ANR (Fig. 3). During maize growth, the ANR values in the sandy-loam soil for the IS, AS, and ASS treatments were always significantly higher (p < 0.05) than for SS (Fig. 3). In the sandy soil, the ANR values for treatments AS and IS were significantly higher (p < 0.05) than for SS and ASS (Fig. 3). Furthermore, treatment ASS led to ANR values higher than for SS. When considering the total ANR, the lowest value was obtained with treatment SS, in both soils (Fig. 3). In the sandy-loam soil the IS, AS, and ASS treatments led to similar values (p > 0.05), while in the sandy soil the value of total ANR was lower (p < 0.05) for treatment ASS than for IS and AS. Slurry acidification had a clear positive effect on ANR during maize growth (Fig. 3). The reason for this seems to be the decrease of N losses by NH3 emissions rather than any increased mineralization of organic N (Webb et al., 2010). Treatment ASS led to a lower (p < 0.05) ANR value relative to AS, which highlights the importance of AS incorporation for maize (Fig. 3). Indeed, AS incorporation prevents both losses of NH3 (due to later increase of slurry pH) and crusting of slurry - that limits N availability - and ensures a homogeneous spatial distribution of nutrients, a key factor for plant growth, particularly for maize (Federolf et al., 2016). Sørensen and Eriksen (2009) also reported that slurry acidification increased the mineral N fertilizer equivalence (MFE) relative to raw slurry, with more significant differences observed for winter wheat than for spring barley. Similarly, Kai et al. (2008) also reported an increase in MFE with surface application of acidified slurry to wheat. The effect of soil type on ANR followed the trend observed with the DM yields (Figs. 2 and 3). In the sandy soil, slurry acidification led to a decrease in ANR, relative to SS, during oat growth, probably due to leaching. In the sandy-loam soil, AS and ASS led to similar or higher values of ANR relative to SS indicating that acidification minimized NH4+ losses but also stimulated mineralization. Nitrogen immobilization is commonly observed following slurry application to soil and can limit the amount of N available to plants (Sørensen and Amato, 2002). Recently, Fangueiro et al. (2016) showed that the remineralization of this immobilized N relies mainly on soil properties. The reduced soil-slurry interaction and potentially lower N immobilization for treatments IS and ASS can also be seen as an advantage relative to traditional band application followed by incorporation (Federolf et al., 2016). The values of NUE during oat production varied from 4.5 to 11.0 g g−1 in the sandy soil and from 6.7 to 21.7 g g−1 in the sandy-loam soil (data not shown). The effects of both treatment and soil type on NUE were significant at p < 0.001 (Table 3). However, the effect of the interaction soil × treatment was significant only at p < 0.05 (Table 3).

Table 3 Total nitrogen (NUE) and phosphorous (PUE) use efficiency estimated in the treatments considered here over the three-year experiment (n = 3). Treatment

NUE (g g−1)

PUE (g g−1)

Sandy soil IS SS AS ASS

14.4a 7.9c 14.8a 11.8b

419.9a 232.2d 360.0b 287.8c

Sandy-loam soil IS SS AS ASS

22.7a 14.2b 22.4a 27.5a

662.8a 415.6b 545.3ab 667.4a

ANOVA results Soil Treatment Soil × treatment

*** *** **

*** *** **

ANOVA results presented referred to final soil properties. For each soil, different letters indicate values that differ statistically at p < 0.05 according to the Tukey test. ns, *, ** and *** mean that the interaction effects were, respectively, not significant (p > 0.05) and significant at p < 0.05, p < 0.01 and p < 0.001.

The total NUE values obtained with the IS, AS, and ASS treatments were similar and significantly higher (p < 0.05) than the value for SS; this trend was observed in both soils (Table 3). It can then be concluded that AS and ASS are as efficient as IS to provide N to plants during oat and maize growth. 3.4. Apparent P recovery The values presented in Fig. 4 for the oat and maize crops are the average values over the 3-year experiment. The APR values measured during oat growth varied in the sandy soil, from 4.8% in ASS to 8.6% in IS, and in the sandy-loam soil, from 6.1% in SS to 12.8% in IS (Fig. 4). During maize growth higher APR values were obtained, in a range of 11.9–21.4% in the sandy soil and 10.5–26.0% in the sandy-loam soil (Fig. 4). The effect of soil on APR was significant at p < 0.05 during the whole experiment, while the effect of the treatments on APR was significant (p < 0.05) only during the first 2-years. The interaction soil × treatment had a significant effect (p < 0.05) on APR during the whole experiment except for maize in 2015. The APR values obtained during oat growth were - on average, over the 3-years - significantly higher (p < 0.05) for IS than for AS and ASS in the sandy soil but no significant differences (p > 0.05) were observed in the sandy-loam soil (Fig. 4). It is of note that, in the sandyloam soil, the APR values were higher for ASS and AS relative to SS, 93

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Fig. 4. Apparent P recovery, expressed as percentage of P applied (n = 3). For each soil and each crop, different letters indicate values that differ statistically at p < 0.05.

acidified slurry application to soil is the potential decrease in soil pH. A significant decrease (p < 0.05) in soil pH was observed for treatments AS and ASS, in both soils (about 0.9 and 1.4 pH units, relative to the Control, in the sandy and sandy-loam soil, respectively), relative to CTR and SS, while no significant differences (p > 0.05) were found among IS, CTR, and SS (Table 1). Frost et al. (1990) also reported that the main effect of acidified slurry application to a sward installed on a clay-loam soil was a decrease in soil pH from 6.18 to 5.76 after three consecutive applications (equivalent to 75 kg NH4+ ha−1) between May and August. As referred to previously, no negative effects on DM production averaged over the 3-years were noticed when compared with SS or IS (Fig. 2), notwithstanding the drops in soil pH. However, the soils under study originally presented pH values in the neutral range (Table 1) and both of the graminaceous species grown are known to show moderate to low sensitivity to soil acidity (Tanaka, 1983), although intra-specific differences regarding tolerance of soil acidity/Al toxicity have been reported (Foy, 1984). Therefore, longerterm experiments are needed to study potential soil over-acidification situations and their agronomic and environmental consequences. Slurry application to soil is seen primarily as a way to provide N and P to plants. Nevertheless, the OM applied simultaneously with the nutrients also improves or at least maintains soil quality and productivity (Biau et al., 2012). As expected, a significant increase (p < 0.05) in the OM content was observed in the amended soils relative to Control (Table 1), in agreement with previous studies that support the benefits of manure application with regard to soil C sequestration (Fan et al., 2014; Zhang et al., 2016). Interestingly, the highest increases occurred with treatment AS in the sandy soil and with AS and ASS in the sandyloam soil (Table 1). Previous studies indicated that C mineralization is slower in soil amended with acidified slurry, relative to non acidified slurry (Fangueiro et al., 2013) - which can explain our findings. It is of note that, after 3-years of slurry application, the OM content was lowest for the IS treatment, in both soils (Table 1). As expected, a significant increase (p < 0.05) in soil extractable S was observed for treatments AS and ASS, due to slurry acidification using sulfuric acid (Table 1). Such increases might be problematic over time if high values are reached, namely in sandy soils where SO42− leaching could occur, with potential impact on water quality. Nevertheless, the low levels of S in sandy agricultural soils imply that, generally, some S applications could be avoided by using acidified slurry.

even if this increase was significant (p < 0.05) only for ASS. It has been reported that slurry acidification solubilized several mineral elements such as P - and increased P availability to plants (Fangueiro et al., 2009). However, following soil application, part of this available P can be leached or immobilized in soil - which can explain the lower APR values obtained for AS and ASS in the sandy soil during oat growth. During maize growth, treatments IS, SS, and ASS led to similar APR values in both soils (Fig. 4). Curiously, AS gave the highest and lowest values of APR in the sandy and sandy-loam soil, respectively (Fig. 4). In the sandy soil, more P was solubilized in acidified slurry and soil incorporation of acidified slurry increased P availability around the maize roots; however, in the sandy-loam soil, a larger fraction of plantavailable P from acidified slurry was probably immobilized after soil incorporation. Wen et al. (2016) showed that the combined application of N and P with manure stimulated maize root growth and consequently N recovery by plants. These authors attributed this effect to the enriched soil structure resulting from OM application via manure - which, in turn, allowed the roots to explore wider soil areas and recover more N and P. When considering the total APR values, in the Arenosol, IS and AS led to similar values, significantly higher (p < 0.05) than for SS and ASS. But, in the Cambisol, ASS and IS gave similar values, significantly higher than for SS and AS (Fig. 4). The values of PUE obtained during maize growth (average of 33.7 g g−1) were significantly higher (p < 0.05) than those obtained during oat growth (average of 14.2 g g−1) (data not shown). The treatment applied had a significant effect (p < 0.05) on PUE, in both soils, and the mean value obtained for SS was always lower those obtained for IS, AS, or ASS (Table 3). The type of soil also had a significant effect (p < 0.001) on PUE, while the interaction treatment × soil type was significant at p < 0.05 (Table 3). Considering the overall PUE, the following order was observed in the sandy soil: IS > AS > ASS > SS, but in the sandy-loam soil IS, AS, and ASS led to similar values and SS to the lowest value (Table 3). 3.5. Soil properties at the end of the experiment The treatments considered and the soil type had significant effects (p < 0.05) on most soil parameters, but their interactions were only of weak significance (Table 1). One of the main concerns related to 94

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

slurry application, relative to surface broadcasting. Besides, the soil application of acidified slurry has been found to reduce NH3 emissions (> 80%) relative to raw slurry (Frost et al., 1990; Kai et al., 2008; Fangueiro et al., 2015a,b,c). During oat growth, the daily fluxes of N2O were low after sowing (November) and peaked from January until oat harvest (March) in all amendment treatments (see Data in Brief article). A reason for this is the high soil water content, due to rainfall events occurred in this time period that promotes the denitrification process (Kool et al., 2011). During maize growth, the daily fluxes of N2O peaked few days after slurry application to soil (May) until June, followed by lower values close to those in the Control - until the end of maize growth (see Data in Brief article). The one exception was during 2013 when a peak was observed in July. The peak of N2O emissions observed few days after slurry application has been reported in previous studies and might be due mainly to nitrification since the NO3−content of the soil before slurry application was residual (Fangueiro et al., 2015c). Soil texture is known to influence the physical and biological processes associated with N2O production and emission in agricultural fields (Gaillard et al., 2016). Overall, the soil type had little or no effect on N2O emissions, except in treatment IS and during maize growth where the N2O emissions were significantly higher (p < 0.05) in the sandy loam soil than in the sandy soil (Table 4). This higher N2O emission in the former soil might be explained by lower oxygen diffusion, which favours denitrification processes and might have enhanced N2O emissions. Furthermore, the nitrification potential measured in the sandy soil were generally lower than those observed in sandy loam soil what can explain the lower N2O emissions observed in the sandy soil (Table 5). Higher cumulative N2O emissions were observed in IS, relative to all other treatments, during maize growth, while no significant differences (p > 0.05) existed between SS, AS and ASS - except in the sandy loam soil, where AS led to higher values than SS and ASS (Table 4). A similar increase of N2O emissions with slurry injection was already reported by Duncan et al. (2017). This might be due to two main factors: on the one hand, more NH4+ was available to be nitrified in treatment IS, relative to SS, because the injection reduced NH3 emissions to Control levels (Table 4); on the other hand, the formation of a wet, oxygen-depleted zone (anaerobic environment) in IS might have favoured the denitrification process and consequently N2O emissions (Petersen and Sommer, 2011; Fangueiro et al., 2015b). The lower N2O emissions observed with AS and ASS might be explained by the delay or inhibition of nitrification in the two soils (Fangueiro et al., 2016; Owusu-Twum et al., 2017), indicating that nitrification was diminished due to lower activity of bacterial nitrifiers and diminution of their populations (Gandhapudi et al., 2006; Fangueiro et al., 2015a, 2016). Such hypothesis is strongly supported by the significant decrease of the nitrification potential in AS and ASS relative to SS treatment (Table 5). Recently, Fangueiro et al. (2016) reported that the effect of slurry acidification on nitrification relies on the soil buffer capacity directly linked to its organic matter content: the lower the soil organic matter, the stronger the effect of slurry acidification. So, more significant differences between the two soils considered, in terms of N2O emissions following AS application, were expected. Nevertheless, the differences in crop yield (and consequent N removal) between the two soils might have masked some of these effects. During oat growth, the amounts of total N applied that were lost as N2O did not differ significantly (p > 0.05) among the amendment treatments (Table 4). However, during maize growth, the AS treatment reduced significantly (p < 0.05) - by ca. 65% - the loss as N2O of the total N applied, when compared with the IS treatment (Table 4). It allows one to conclude that band application of acidified slurry has a strong potential to minimize N losses following slurry application, not only in terms of NH3 but also due to decreased N2O emissions relative to IS during maize growth.

Furthermore, an increase in the P content in soils subjected to slurry applications may contribute to the desorption of soil native SO42− as a result of the strong competition of phosphate for chemical adsorption sites, leading to an increase in SO42− leaching and deficiency (Tabatabai, 2005). Soil extractable P and K were increased significantly (p < 0.05) by the amendment treatments, relative to Control, in the sandy-loam soil. In the sandy soil, a significant increase (p < 0.05) was observed only with treatments ASS and SS, for P, and ASS, for K (Table 1). Such differences might be mainly due to the lower leaching potential in the sandy-loam soil. Furthermore, the amount of extractable P correlated positively and negatively with the APR values in acidified and non acidified slurry treatments, respectively. In the sandy soil, more P remained extractable in AS than in ASS, despite the higher total APR value obtained with AS (Table 1). However, the opposite trend was observed in the sandy-loam soil. In the latter soil, greater P insolubilization might have occurred following the incorporation of AS, while in the sandy soil, more P might have been lost by leaching for ASS than for AS since, in this case, incorporation might have contributed to more efficient soil P sorption. 3.6. Nitrogen emissions The NH3 and N2O daily fluxes from the double-cropping system over the 3-years experiment are presented in a Data in Brief article and will be succinctly described here. The daily NH3 fluxes from treatments AS and ASS remained similar (p > 0.05) to those of IS and very close to the Control values during all the measurement period. Significantly higher (p < 0.05) daily NH3 fluxes were observed in SS, relative to all other treatments, during the whole measurement period, particularly in the first two days after soil amendment. Hence, it can be concluded that band application of acidified slurry without soil incorporation is as efficient as slurry injection at minimising NH3 emissions at the field scale. The effect of AS incorporation following band application was also tested, because the slurry pH could rise during the days following soil application; consequently, NH3 emissions could be observed later on. During the period of six days measurement presented here, NH3 emissions from ASS were always residual and, even seven days after application, they remained close to control levels. So, soil incorporation of AS can be omitted or delayed for a few days with no negative impact on NH3 emissions. In practice, the no-incorporation option could be of interest on farms where no-tillage is used as soil conservation practice, or to give flexibility to farmers to proceed with slurry incorporation into soil at the most convenient time rather than in the four hours following application as occurred here. The cumulative NH3 emissions from treatment SS were significantly higher (p < 0.05) than those from all other treatments, while no significant differences (p > 0.05) existed among IS, AS and ASS (Table 4). The loss of NH3 in treatment SS represented 38.8% of the applied NH4+-N in the sandy soil and 18.5% in the sandy loam soil, significantly more (p < 0.05) than for the other amendment treatments where less than 2% of the NH4+-N applied was lost (Table 4). This lower NH3 emissions from the sandy loam soil relative to the sandy soil can be attributed to higher NH3 adsorption in the former, which has a higher content of clay minerals (Sommer et al., 2003). In addition, the sandy texture enables NH3 exchange between the soil and the atmosphere (Tisdale et al., 1985), leading to higher NH3 emissions relative to heavier soil (Table 4). It is still of note that no differences between the two soils were observed in NH3 emissions from acidified slurry treatments (Table 4), since lowering the slurry pH to 5.5 strongly modified the NH4+:NH3 ratio - giving about 98.00 to 99.98% NH4+ and the release of residual amounts of NH3 (Fangueiro et al., 2015a). In agreement with our results, previous studies (Webb et al., 2010; Carozzi et al., 2013) have shown that injection of raw slurry into soil (> 50 mm depth) is one of the most efficient strategies at the farm scale (e.g. reference technique) for reduction of NH3 emissions (> 95%) after 95

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Table 4 Ammonia (NH3), nitrous oxide (N2O) and total N emissions observed during the 3-year experiment. The data presented are the 3-years average values of the cumulative emissions measured during oat and maize growth as well as the total values considering both crops (n = 3). Treatment

NH3 emissions

N2O emissions

kg N ha−1

% NH4+-N applied

NH3+N2O emissions

kg N ha−1

% total N applied

% total N applied

% total N applied

oat

maize

total

oat

maize

total

oat

maize

total

oat

maize

total

oat

maize

total

oat

maize

total

0.0c 0.1c 16.2a 0.4c 0.3c

0.0c 2.1c 24.2a 0.9c 1.3c

0.1c 2.1c 43.1a 1.5c 1.9c

0.1c 44.3a 0.9c 0.8c

2.9c 35.3a 1.3c 1.8c

1.9c 45.6a 1.3c 1.5c

0.1c 19.1a 0.4c 0.4c

1.2c 14.1a 0.6c 0.8c

0.8c 16.8a 0.6c 0.7c

0.2b 0.6a 0.5a 0.5a 0.7a

0.1c 0.6b 0.2c 0.3c 0.2c

0.4d 1.3a 0.8b 0.8b 1.0ab

0.66a 0.55a 0.57a 0.75a

0.33b 0.13c 0.16c 0.14c

0.49a 0.32b 0.32b 0.38ab

0.8c 19.7a 1.0c 1.1c

1.5c 14.2a 0.7c 0.9c

1.3c 17.1a 0.9c 1.1c

Sandy loam soil Control 0.0c IS 0.1c SS 3.7b AS 0.1c ASS 0.4c

0.0c 0.2c 15.8b 0.4c 0.9c

0.1c 0.2c 20.1b 0.6c 1.7c

0.1c 10.0b 0.1c 1.1c

0.1c 23.0b 0.6c 1.3c

0.2c 21.4b 0.4c 1.4c

0.1c 4.4b 0.1c 0.5c

0.1c 9.2b 0.3c 0.6c

0.1c 7.8b 0.2c 0.6c

0.2b 0.4a 0.4b 0.6b 0.5a

0.2c 0.8a 0.3c 0.6b 0.2c

0.4cd 1.3a 0.8bc 1.4a 0.8bc

0.46a 0.46a 0.71a 0.55a

0.46a 0.16c 0.34b 0.13c

0.50a 0.30b 0.52a 0.30b

0.6c 4.8b 0.8c 1.0c

0.5c 9.3b 0.6c 0.7c

0.6c 8.1b 0.7c 1.0c

p p p

*** ** **

*** *** ***

*** *** ***

*** ** **

*** *** ***

*** *** ***

*** ** **

*** *** ***

*** ns ns

*** *** ***

*** ns **

ns ns ns

*** ** **

*** ns *

*** *** ***

*** ** **

*** *** ***

Sandy soil Control IS SS AS ASS

treatment soil treatment×soil

*** *** ***

Values for the interaction treatment×soil presented with different superscripts within columns are significantly different (p < 0.05) according to the Tukey test. ns, *, ** and *** mean that the interaction effects were, respectively, not significant (p > 0.05) and significant at p < 0.05, p < 0.01 and p < 0.001.

accepted that the initial CH4 emissions following slurry application are due to volatilization of CH4 dissolved in the slurry and produced during storage (Chadwick et al., 2011). During oat growth, no significant CH4 emissions were observed except in the last year of experiment. The highest initial CH4 fluxes were generally observed for treatments with acidified slurry, except in the sandy soil during oat growth. The cumulative CH4 emissions did not differ significantly (p > 0.05) between treatments AS and ASS, being significantly (p < 0.05) reduced - by ca. 40% - when compared with IS (Table 6). The reason for this difference might be the fact that during the acidification process high amounts of CO2, and possibly CH4, are released which could explain lower emissions after soil application (Fangueiro et al., 2015a). Furthermore, the lower CH4 production, in both soil types, after amendment with acidified slurry could be related to the lower soil pH that might have inhibited methanogenesis (Fangueiro et al., 2015c). No significant differences (p > 0.05) were observed between the CH4 emissions from the two soil types (Table 6).Overall, it can be concluded that the band application of acidified slurry without soil incorporation reduced the CH4 emissions by 40% relative to slurry injection. A significant (p < 0.05) increase in the daily CO2 fluxes was observed in all amended treatments, relative to the Control (see Data in Brief article). During the oat cycle, the daily fluxes of CO2 were lower than 12 kg ha−1 day−1 for all the amendments, whereas during maize growth the daily fluxes of CO2 peaked after soil application (May), followed by constant values that reached about 25 kg ha−1 day−1 in all amended treatments. The cumulative CO2 emissions did not differ significantly (p > 0.05) between the amended treatments or the two soil types (Table 6). However, during maize growth, the amounts of applied C lost as CO2 emissions from the IS treatment were, on average, higher than in all other amended treatments (Table 6). It is to note that ∼30% of applied C was lost as CH4 and CO2 in all amended treatments. Previous studies (Stevens et al., 1989; Ottosen et al., 2009; Fangueiro et al., 2013, 2015b) reported lower CO2 emissions from soil amended with acidified slurry, compared to raw slurry, because most of the inorganic C is released as CO2 during the acidification process (pH = 5.5) and also the acidification reduces microbial activity and consequently oxygen consumption, sulphate reduction and methanogenesis. Nevertheless, in our study, no differences were observed between AS, ASS and SS treatments (Table 6).

Table 5 Nitrification potential (μg N-NO2− g-1 soil 24 h-1) measured at the end of each crop during the 3-year experiment (n = 3). Treatment Oat

Sandy soil Control IS SS AS ASS Sandy loam Control IS SS AS ASS

Maize

2012/2013

2013/2014

2014/2015

2013

2014

2015

0.119c 0.163b 0.291a 0.144bc 0.177b

0.186e 0.919b 0.870c 1.012a 0.706d

0.218d 0.429b 0.780a 0.307c 0.303c

0.789d 1.673b 2.225a 0.812d 1.102c

1.358e 2.789b 3.639a 2.358c 2.075d

0.138e 0.471b 0.505a 0.342c 0.216d

soil 0.064d 0.098bc 0.112a 0.088c 0.106ab

0.465d 0.787b 1.319a 0.781d 0.756c

0.103e 0.929b 1.527a 0.395d 0.498c

0.885d 1.698b 3.281a 0.953d 1.594c

0.999c 2.236b 2.903a 1.057c 0.900d

0.392d 0.596b 1.401a 0.498c 0.643b

Values for the interaction treatment×soil presented with different superscripts within columns are significantly different (p < 0.05) according to the Tukey test.

Considering the total amount of N lost by NH3 and N2O emissions, no significant differences (p > 0.05) were observed between treatments IS, AS and ASS, where N losses by gaseous emissions were ca. 92% lower than in SS (Table 4). It is of note that in treatment SS more than 8% of the total N applied was lost by gaseous emissions, against less than 1.5% in all other treatments (Table 4). The total N losses from the different treatments in the sandy soil were, on average, higher than those in the sandy loam soil, but the difference was statistically significant (p < 0.05) only in the SS treatment (Table 4). 3.7. Carbon emissions The daily fluxes of CH4 and CO2 measured over the 3-years of experiment are presented in a Data in Brief article and will be succinctly described here. The CH4 daily fluxes observed during maize growth followed a similar pattern in all treatments and both soils: CH4 emissions peaked immediately after soil application and then decreased during the 4 following days to reach the Control values. It is generally 96

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

Table 6 Cumulated emissions of carbon dioxide (CO2) and methane (CH4), global warming potential (GWP) and yield-scaled global warming potential observed during the 3year experiment. The data presented are the 3-years average values measured during oat and maize growth as well as the total values considering both crops (n = 3). Treatment

CO2 emissions

CH4 emissions

GWP

Yield-scaled GWP

kg C ha−1

kg C ha−1

kg CO2-eq ha−1

kg CO2-eq kg−1

oat

maize

total

oat

maize

total

oat

maize

total

oat

maize

total

Sandy soil Control IS SS AS ASS

261c 351b 312bc 342b 378ab

326c 753a 713a 731a 662ab

638c 1179a 1129ab 1169ab 1163ab

0.3c 1.0a 0.7b 0.4bc 0.6bc

0.3e 2.7abc 2.0bcd 1.7bcd 1.5cde

0.7e 4.1ab 3.2abcd 2.3cd 2.5cd

1036c 1571ab 1375b 1483ab 1690a

1265d 3091ab 2781ab 2858ab 2583bc

2519d 5003a 4604ab 4721ab 4771ab

2.0a 1.1bcd 1.2bc 1.5b 1.4b

2.4a 0.9cd 1.4bc 0.7cd 0.9cd

2.4a 1.0cd 1.5bc 1.0cd 1.2cd

Sandy loam soil Control IS SS AS ASS

337ab 361ab 363ab 344ab 418a

324c 871a 431bc 816a 620ab

733c 1303a 918bc 1270a 1176ab

0.3bc 0.4bc 1.1a 0.6bc 0.6b

0.3e 3.4a 1.4de 1.0de 2.8ab

0.8e 4.1a 2.8bcd 2.0de 3.8abc

1315bc 1511ab 1547ab 1552ab 1761a

1279d 3651a 1748cd 3269ab 2471bc

2897cd 5478a 3796bc 5297a 4778ab

1.5ab 0.6e 0.9cde 0.7de 0.7de

1.8ab 0.7cd 0.5d 0.6cd 0.4d

1.9ab 0.7d 0.7d 0.7d 0.6d

p p p

*** *** ns

*** ns **

*** ns *

*** ns ***

*** ns **

*** ns *

*** ** ns

*** ns **

*** ns *

*** *** ns

*** *** ns

*** *** ns

treatment soil treatment×soil

Values for the interaction treatment×soil presented with different superscripts within columns are significantly different (p < 0.05) according to the Tukey test. ns, *, ** and *** mean that the interaction effects were, respectively, not significant (p > 0.05) and significant at p < 0.05, p < 0.01 and p < 0.001.

performed in specific conditions with a small plot area (1 m2) and in which the oat and maize were harvested prior to the end of the typical growing cycle - to the farm scale, the results presented here indicate clearly that the EF used currently has to be reviewed. Considering the amount of plant-available N that is saved by slurry acidification, lower amounts of N should be applied as a top-dressing fertilisation, namely during maize growth. Indeed, a significant increase of dry matter production and N uptake was observed in AS and ASS treatments relative to SS in both soils indicating that more N was available for plants in these two treatments. The cost associated with slurry acidification should therefore be balanced by the saving obtained with top dressing, leading to an economically and environmentally sustainable alternative to slurry injection. Thus, band application of acidified raw slurry should be recommended to farmers as a good alternative to slurry injection, to minimize NH3 (< 1.5% of the applied NH4+-N) and GHG emissions (65% and 40% reductions for N2O and CH4, respectively) at the field scale, in a Mediterranean double-cropping forage system.

3.8. Global warming potential and yield-scaled global warming potential The GWP did not vary significantly (p > 0.05) between amendment treatments during oat growth, in both soils, or during maize growth in the sandy soil (Table 6). However, in the sandy loam soil, IS and AS led to significantly higher (p < 0.05) values of GWP than did SS during maize growth. Besides, the GWP from the whole double-cropping forage system was ca. 65% during maize growth and ca. 35% during oat growth (Table 6). The yield-scaled GWP values were higher in the sandy soil, relative to the sandy loam soil, due only to the lower crop yield obtained in the sandy soil (Table 6). Our results indicate that all the treatments considered here led to similar values of yield-scaled GWP (independently of the soil considered). The only exception was band application of acidified slurry without incorporation into the sandy loam soil, for which significantly lower values were observed. Hence, our previous proposal to delay acidified slurry incorporation into sandy loam soil could have a positive impact on GWP mitigation. Overall, our results indicated that AS and ASS treatment allowed decreasing gaseous emissions relative to SS and IS with no negative impact on crop production.

4. Conclusions Slurry pre-treatment by acidification before band application and soil incorporation increased significantly oat and maize DM yields as well as N recovery, in both the sandy and sandy-loam soil. Furthermore, band application of acidified slurry was as efficient as raw slurry injection with respect to minimising NH3 emissions (< 1.5% of the applied NH4+-N) and there was no need for immediate soil incorporation. Slurry injection increased N2O emissions by 65% during maize growth, while band application of acidified slurry led to emissions similar to those of non acidified slurry. Acidified slurry incorporation immediately after soil application might contribute to the minimisation of CH4 emissions. The P recovery was also higher in soil amended with acidified slurry, compared to raw slurry, even if such an effect relied strongly on the soil considered. Furthermore, after six consecutive applications of acidified slurry, more plant-available P remained in the sandy-loam soil than with raw slurry application. Regarding the impact of acidified slurry application on soil properties, the main consequences were a decrease in soil pH and an increase in S content; but, for both parameters, the values at the end of the experiment were not of concern in terms of soil quality for crops growth.

3.9. Considerations for slurry management The current Portuguese NH3 emission inventory uses an emission factor (EF) of 55% of the NH4+-N available in slurry for land spreading, and reductions of NH3 of 60 and 80%, respectively, for immediate incorporation by disc and deep injection (EMEP-EEA, 2016; PIIR, 2017). The results presented here were obtained in a small-plot experiment and cannot be used as references. Nevertheless, it appears clearly that the EFs used underestimate NH3 losses during slurry spreading and that some revised EF values should be proposed. Similarly, the current National N2O emission inventory uses a direct EF of 1% of the N input for all types of fertiliser (IPCC, 2006; PNIR, 2016), but Cayuela et al. (2017) found an average overall direct N2O EF for Mediterranean agriculture of 0.5% of the N input. The direct N2O EF in this study for slurry injection was ca. 50% lower than the EF of the National inventory, whereas the direct N2O emission EF for band application of acidified slurry without soil incorporation was ca. 70% lower than the EF of the National inventory. Despite the risks of the extrapolation of the data from this study 97

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

It can be concluded that band application of acidified slurry followed by soil incorporation is an efficient solution to provide nutrients to plants while minimising NH3 emissions, and can then be proposed as an alternative to non acidified slurry injection. Nevertheless, the impact of acidified slurry application on soil properties needs to be accurately monitored. The present study should contribute to a wider adoption of slurry acidification in countries from southern Europe that relies on the gathering and dissemination of more data relating to the agronomic value of acidified slurry and to the residual effects of such application in soil.

Environ. Manage. 149, 46–56. Fangueiro, D., Pereira, J., Bichana, A., Surgy, S., Cabral, F., Coutinho, J., 2015b. Effects of cattle-slurry treatment by acidification and separation on nitrogen dynamics and global warming potential after soil surface application to an acidic soil. J. Environ. Manage. 162, 1–8. Fangueiro, D., Surgy, S., Fraga, I., Cabral, F., Coutinho, J., 2015c. Band application of treated cattle slurry as an alternative to slurry injection: implications for gaseous emissions, soil quality and plant growth. Agric. Ecosyst. Environ. 211, 102–111. Fangueiro, D., Surgy, S., Fraga, I., Monteiro, F.G., Cabral, F., Coutinho, J., 2016. Acidification of animal slurry affects the nitrogen dynamics after soil application. Geoderma 281, 30–38. Fangueiro, D., Becerra, D., Albarrán, Á., Peña, D., Sanchez-Llerena, J., Rato-Nunes, J.M., López-Piñeiro, A., 2017. Effect of tillage and water management on GHG emissions from Mediterranean rice growing ecosystems. Atmos. Environ. 150, 303–312. Federolf, C.F., Westerschulte, M., Olfs, H.W., Broll, G., Trautz, D., 2016. Enhanced nutrient use efficiencies from liquid manure by positioned injection in maize cropping in northwest Germany. Eur. J. Agron. 75, 130–138. Foged, H.L., Flotats, X., Bonmati Blasi, A., Palatsi, J., Magri, A., Martin Schelde, K., 2011. Inventory of manure processing activities in Europe. Technical Report No. I Concerning “Manure Processing Activities in Europe” to the European Commission. Directorate-General Environment 138 pp. Foy, C.D., 1984. Physiological effects of hydrogen, aluminum and manganese toxicities in acid soils, In: Adams P. (Ed), Soil Acidity and Liming. Ed. Agron. Monogr. 12, 2nd ed., Am. Soc. Agron., Madison, Wisconsin, USA, pp. 57-97. Frost, J.P., Stevens, R.J., Laughlin, R.J., 1990. Effect of separation and acidification on ammonia volatilization and on the efficiency of slurry nitrogen for herbage production. J. Agric. Sci. 115, 49–56. Gaillard, R., Duval, B.D., Osterholz, W.R., Kucharik, C.J., 2016. Simulated effects of soil texture on nitrous oxide emission factors from corn and soybean agroecosystems in Wisconsin. J. Environ. Qual. 45, 1540–1548. Gandhapudi, S.K., Coyne, M.S., Angelo, E.M.D., Matocha, C., 2006. Potential nitrification in alum-treated soil slurries amended with poultry manure. Bioresour. Technol. 97, 664–670. Gómez-Muñoz, B., Case, S.D.C., Jensen, L.S., 2016. Pig slurry acidification and separation techniques affect soil N and C turnover and N2O emissions from solid, liquid and biochar fractions. J. Environ. Manage. 168, 236–244. Harrison, R.M., Yamulki, S., Goulding, K.W.T., Webster, C.P., 1995. Effect of fertilizer application on NO and N2O fluxes from agricultural fields. J. Geophys. Res. 100, 25923–25931. IPCC, 2006. IPCC guidelines for national greenhouse gas inventories. 2006. In: Eggleston, H.S., Buendia, L., Miwa, K., Ngara, T., Tanabe, K. (Eds.), IPCC National Greenhouse Gas Inventories Programme. IGES, Hayama, Japan Technical support unit. http:// www.ipcc-nggip.iges.or.jp. Kai, P., Pedersen, P., Jensen, J.E., Hansen, M.N., Sommer, S.G., 2008. A whole-farm assessment of the efficacy of slurry acidification in reducing ammonia emission. Eur. J. Agron. 28, 148–154. Kiely, P.V., 1988. Slurry application to grassland in relation to efficiency of nitrogen utilization. Proceedings of the 4th International CIEC (International Scientific Centre of Fertilizers) Symposium pp. 101–104. Kool, D.M., Dolfing, J., Wrage, N., Van Groenigen, J.W., 2011. Nitrifier denitrification as a distinct and significant source of nitrous oxide from soil. Soil Biol. Biochem. 43, 174–178. Langevin, B., Genermont, S., Basset-Mens, C., Lardon, L., 2015. Simulation of field NH3 and N2O emissions from slurry spreading. Agron. Sustain. Dev. 35, 347–358. Little, T.M., Hills, F.J., 1978. Agricultural experimentation. Design and Analysis. John Wiley and Sons, New York, USA 350 pp. Ottosen, L.D.M., Poulsen, H.V., Nielsen, D.A., Finster, K., Nielsen, L.P., Revsbech, N.P., 2009. Observations on microbial activity in acidified pig slurry. Biosyst. Eng. 102, 291–297. Ovejero, J., Ortiz, C., Boixadera, J., Serra, X., Ponsá, S., Lloveras, J., Casas, C., 2016. Pig slurry fertilization in a double-annual cropping forage system under sub-humid Mediterranean conditions. Eur. J. Agron. 81, 138–149. Owusu-Twum, M.Y., Loick, N., Cardenas, L.M., Coutinho, J., Trindade, H., Fangueiro, D., 2017. Nitrogen dynamics in soils amended with slurry treated by acid or DMPP addition. Biol. Fertil. Soils 53, 339–347. Pain, B.F., Thompson, R.B., De La, Lande, Cremer, L.C.N., Ten Holte, L., 1987. The use of additives in livestock slurries to improve their own properties, conserve nitrogen and reduce odours. In: Meer, H.G.v.d, Unwin, R.J., Van Dijk, T.A., Ennik, G.C. (Eds.), Animal Manure on Grassland and Fodder Crops. Martinus Nijhoff pp. 229–246. Petersen, S.O., Sommer, S.G., 2011. Ammonia and nitrous oxide interactions: roles of manure organic matter management. Anim. Feed Sci. Technol. 166–167, 503–513. PIIR, 2017. Portuguese Informative Inventory Report 1990–2015. Submitted Under the NEC Directive (EU) 2016/2284 and the UNECE Convention on Long-range Transboundary Air Pollution. Portuguese Informative Inventory Report (PIIR). Portuguese Environmental Agency, Lisbon, Portugal 449 pp. PNIR, 2016. Portuguese National Inventory Report on Greenhouse Gases, 1990–2014. Submitted Under the United Nations Framework Convention on Climate Change, Portuguese National Inventory Report (PNIR). Portuguese Environmental Agency, Portugal. http://cdr.eionet.europa.eu/. Roboredo, M., Fangueiro, D., Lage, S., Coutinho, J., 2012. Phosphorus dynamics in soils amended with acidified pig slurry and derived solid fraction. Geoderma 189–190, 328–333. Rodhe, L., Pell, M., Yamulki, S., 2006. Nitrous oxide, methane and ammonia emissions following slurry spreading on grassland. Soil Use Manage. 22, 229–237. Schils, R.L.M., van der Meer, H.G., Wouters, A.P., Geurink, J.H., Sikkema, K., 1999. Nitrogen utilization from diluted and undiluted nitric acid treated cattle slurry

Acknowledgments This work was supported by: European InvestmentFunds by FEDER/ COMPETE/POCI Operational Competitiveness and Internationalisation Programme, under Project POCI-01-0145-FEDER006958, and National Funds by FCT - Portuguese Foundation for Science and Technology, under the project UID/AGR/04129/2013, PEst-OE/AGR/UI0528/2014 and project PTDC/AGR-PRO/119428/ 2010. David Fangueiro has received a grant from the FCT (SFRH/BPD/ 84229/2012). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.agee.2018.08.011. References Bhandral, R.S., Bittman, G., Kowalenko, K., Buckley, M.H., Chantigny, D.E., Hunt, F., Bounaix, A., Friesen, A., 2009. Enhancing soil infiltration reduces gaseous emissions and improves N uptake from applied dairy slurry. J. Environ. Qual. 38, 1372–1382. Biau, A., Santiveri, F., Mijangos, I., Lloveras, J., 2012. The impact of organic and mineral fertilizers on soil quality parameters and the productivity of irrigated maize crops in semiarid regions. Eur. J. Soil Biol. 53, 56–61. Bittman, S., Dedina, M., Howard, C.M., Oenema, O., Sutton, M.A. (Eds.), 2014. Options for Ammonia Mitigation: Guidance from the UNECE Task Force on Reactive Nitrogen. Centre for Ecology and Hydrology, Edinburgh, UK. Carozzi, M., Ferrara, R.M., Rana, G., Acutis, M., 2013. Evaluation of mitigation strategies to reduce ammonia losses from slurry fertilisation on arable lands. Sci. Total Environ. 449, 126–133. Cavalli, D., Cabassi, G., Borrelli, L., Geromel, G., Bechini, L., Degano, L., Marino Gallina, P., 2016. Nitrogen fertilizer replacement value of undigested liquid cattle manure and digestates. Eur. J. Agron. 73, 34–41. Cayuela, M.L., Aguilera, E., Sanz-Cobena, A., Adams, D.C., Abalos, D., Barton, L., Ryals, R., Silver, W.L., Alfaro, M., Pappa, V.A., Smith, P., Garnier, J., Billen, G., Bouwman, L., Bondeau, A., Lassaletta, L., 2017. Direct nitrous oxide emissions in Mediterranean climate cropping systems: emission factors based on a meta-analysis of available measurement data. Agric. Ecosyst. Environ. 238, 25–35. Chadwick, D., Sommer, S., Thorman, R., Fangueiro, D., Cardenas, L., Amon, B., Misselbrook, T., 2011. Manure management: implications for greenhouse gas emissions: a review. Anim. Feed Sci. Technol. 166–167, 514–531. Cocolo, G., Hjorth, M., Zarebska, A., Provolo, G., 2016. Effect of acidification on solidliquid separation of pig slurry. Biosyst. Eng. 143, 20–27. Duncan, E.W., Dell, C.J., Kleinmanm, P.J., Beegle, D.B., 2017. Nitrous oxide and ammonia emissions from injected and broadcast-applied dairy slurry. J. Environ. Qual. 46, 36–44. EMEP-EEA, 2016. EMEP/EEA air pollutant emission inventory guidebook 2013. EEA Report No 21/2016. European Environment Agency (EEA), Copenhagen, Denmark. https://doi.org/10.2800/247535. http://www.eea.europa.eu/publications/emepeea-guidebook-2016. Fan, J.L., Ding, W.X., Xiang, J., Qin, S.W., Zhang, J.B., Ziadi, N., 2014. Carbon sequestration in an intensively cultivated sandy-loam soil in the North China Plain as affected by compost and inorganic fertilizer application. Geoderma 230, 22–28. Fangueiro, D., Pereira, J., Coutinho, J., Moreira, N., Trindade, H., 2008. NPK farm-gate nutrient balances in dairy farms from Northwest Portugal. Eur. J. Agron. 28, 625–634. Fangueiro, D., Ribeiro, H., Vasconcelos, E., Coutinho, J., Cabral, F., 2009. Treatment by acidification followed by solid-liquid separation affects slurry and slurry fractions composition and their potential of N mineralization. Bioresour. Technol. 100, 4914–4917. Fangueiro, D., Surgy, S., Coutinho, J., Vasconcelos, E., 2013. Impact of cattle slurry acidification on carbon and nitrogen dynamics during storage and after soil incorporation. J. Plant Nutr. Soil Sci. 176, 540–550. Fangueiro, D., Hjorth, M., Gioelli, F., 2015a. Acidification of animal slurry–a review. J.

98

Agriculture, Ecosystems and Environment 267 (2018) 87–99

D. Fangueiro et al.

ten Hoeve, M., Gomez Muñoz, B., Jensen, L.S., Bruun, S., 2016. Environmental impacts of combining pig slurry acidification and separation under different regulatory regimes a life cycle assessment. J. Environ. Manage. 181, 710–720. Tisdale, S.L., Nelson, W.L., Beaton, J.D., 1985. Soil acidity and liming. In: Tisdale, S.L., Nelson, W.L., Beaton, J.D. (Eds.), Soil Fertility and Fertilizers, fourth ed. Macmillan, New York, USA pp. 484–526 (Chapter 11). Webb, J., Pain, B., Bittman, S., Morgan, J., 2010. The impacts of manure application methods on emissions of ammonia, nitrous oxide and on crop response–a review. Agric. Ecosyst. Environ. 137, 39–46. Webb, J., Sørensen, P., Velthof, G., Amon, B., Pinto, M., Rodhe, L., Salomon, E., Hutchings, N., Burczyk, P., Reid, J., 2013. An assessment of the variation of manure nitrogen efficiency throughout Europe and an appraisal of means to increase manureN efficiency. Adv. Agron. 119, 371–442. Webb, N., Broomfield, M., Brown, P., Buys, G., Cardenas, L., Murrells, T., Pang, Y., Passant, N., Thistlewaite, G., Watterson, J., 2014. UK Greenhouse Gas Inventory, 1990 to 2012: Annual Report for Submission under the Framework Convention on Climate Change. Department of Energy and Climate Change, London. Wen, Z., Shen, J., Blackwell, M., Li, H., Zhao, B., Yuan, H., 2016. Combined applications of nitrogen and phosphorus fertilizers with manure increase maize yield and nutrient uptake via stimulating root growth in a long-term experiment. Pedosphere 26, 62–73. WRB, 2015. World reference Base for soil resources 2014. World Soil Resources Reports 106. FAO, Rome pp 192. Zhang, Y., Li, C., Wang, Y., Hu, Y., Christie, P., Zhang, J., Li, X., 2016. Maize yield and soil fertility with combined use of compost and inorganic fertilizers on a calcareous soil on the North China Plain. Soil Till. Res. 155, 85–94.

following surface application to grassland. Nutr. Cycl. Agroecosyst. 53, 269–280. Scherer, H.W., 1993. Dynamics and availability of the non-exchangeable NH4+-N—a review. Eur. J. Agron. 2, 149–160. Schröder, J.J., de Visser, W., Assinck, F.B.T., Velthof, G.L., 2013. Effects of short-term nitrogen supply from livestock manures and cover crops on silage maize production and nitrate leaching. Soil Use Manage. 29, 151–160. Sørensen, P., Amato, M., 2002. Remineralisation and residual effects of N after application of pig slurry to soil. Eur. J. Agron. 16, 81–95. Sørensen, P., Eriksen, J., 2009. Effect of slurry acidification with sulphuric acid combined with aeration on the turnover and plant availability of nitrogen. Agric. Ecosyst. Environ. 131, 240–246. Sommer, S.G., Genermont, S., Cellier, P., Hutchings, N.J., Olesen, J.E., Morvan, T., 2003. Processes controlling ammonia emission from livestock slurry in the field. Eur. J. Agron. 19, 465–486. Sommer, S.G., Christensen, M.L., Schmidt, T., Jensen, L.S., 2013. Animal Manure Recycling: Treatment and Management. Wiley, UK 384 pp. Stevens, R.J., Laughlin, R.J., Frost, J.P., 1989. Effect of acidification with sulphuric acid on the volatilization of ammonia from cow and pig slurries. J. Agric. Sci. 113, 389–395. Tabatabai, M.A., 2005. Chemistry of sulfur in soils. In: Dick, W.A. (Ed.), Chemical Processes in Soils, Book Series 8. Am. Soc. Agron., Madison, Wisconsin, USA p 193–226. Tanaka, A., 1983. Physiological aspects of productivity in field crops. Potential Productivity of Field Crops Under Different Environments. International Rice Research Institute, Manila, Philippines p 61–80.

99