IntroductionTop

Material and methodsTop

Site and experimental design

The trial was conducted at the experimental field ‘Soto Lezcano’ (middle Ebro Valley, Spain) during two wheatgrowing seasons (2016/17 and 2017/18) under semiarid Mediterranean-continental irrigated conditions. The climate is characterised by mean annual air temperature of 14.6 ºC and mean annual precipitation and reference evapotranspiration of 318 mm and 1,243 mm, respectively (period 2004-2019). The trial was established on a deep (>120 cm) Typic Xerofluvent soil (Soil Survey Staff, 2014; Table 1) where bread wheat (Triticum aestivum L. cv. ‘Rimbaud’) was cultivated under sprinkler irrigation. Crop water needs were calculated weekly from the reference evapotranspiration (ETo) estimated with the Penman-Monteith equation and the locally adapted crop coefficients (Kc) according to FAO procedures (Allen et al., 1998). Thereby, the crop received a total of 435 mm and 428 mm of water (rain plus irrigation) to supply the estimated crop ET of 429 mm and 383 mm, respectively, during the two cropping seasons.

The experiment had a randomised block design with four replicates and four treatments with a plot size of 2.0 × 3.6 m. The experiment was replicated three times: presowing 2016, side-dressing 2017 (both in the same crop cycle), and side-dressing 2018. Slurry from fattening pigs was used in the three trials (Table 2) and the four evaluated additives were provided by the manufacturing companies. The slurry was applied by hand to the experimental plots at a target rate of 120 kg NH4 + -N ha-1 in all treatments (actual rates in Table 2). Nitrogen fertiliser treatments were: a) non-N application (Control); b) pig slurry (PS); c) pig slurry mixed with the urease inhibitor monocarbamide dihydrogen sulphate (PS-UI); d) pig slurry mixed with a soil microbial activator (PS-A) or with nitrification inhibitor 3,4-dimethylpyrazole phosphate (Vizura®; PSNI). At side-dressing application in 2018, the PS-A was substituted by the PS-NI treatment. The urease inhibitor MCDHS was included as PS slurry additive treatment to assess the micro-acidification effect on ammonia losses declared by the manufacturing company that patented the inhibitor. The additives were applied according to the rate recommended by the manufacturing company: 2.5 L of MCDHS in 1 Mg of pig slurry, 2.5 kg of soil microbial activator in 1 Mg of pig slurry, and 3 L of Vizura® per hectare. Pig slurry was applied at presowing (14th November 2016) and at side-dress at tillering (7th April 2017 and 19th March 2018). At presowing in the two cropping seasons, 70 kg P2O5 ha-1 and 150 kg K2O ha-1 were applied to avoid limitations of these two nutrients.

The crop was managed according to standard practices in the region. Wheat was sown on 30th December 2016 and 16th November 2017 at a plant density of 200 and 175 kg seed ha-1, respectively. The crop was harvested at wheat maturity (4th July 2017 and 6th July 2018). Straw was hashed and incorporated to the soil before subsequent wheat seeding. Weeds, diseases, and pests were controlled and no special problems were detected during the experimental period.

 

 

Table 1. Physicochemical characteristics of the soil at the beginning of the experiment.

 

 

Table 2. Physicochemical characteristics of the pig slurry and amount of nitrogen applied at presowing 2016, side-dressing 2017, and side-dressing 2018.

 

 

Ammonia volatilisation

Ammonia volatilisation was measured at presowing 2016 and side-dressing 2017 using semi-opened free static chambers (SOC), similar to those of Araújo et al. (2009). Two polyethylene terephthalate chambers per plot (31.0-cm height and 10.3-cm inner diameter, 2-L volume bottle with the bottom cut) were located 2 cm above the soil surface and with a 2-cm diameter upper hole that ensured airflow. The removed bottom was situated 2 cm above the upper hole to avoid the entrance of rain and irrigation water into the chamber. Inside the chamber, there was an absorbent Spontex® Origin foam strip (Mapa Spontex Ibérica SA, Sant Cugat del Vallès, Spain) of 25.0 × 2.5 × 0.5 cm (length × wide × thickness) and 0.087 g cm-3 (density) previously impregnated in acid solution (60 mL H2SO4, 1 mol dm-3 + glycerine (2% v/v)). The bottom end of the strip was immersed in a 100-mL plastic jar with 50 mL of the acid solution in permanent contact with the strip. The trapped ammonia in the foam strip was extracted with 250 mL of 2 M KCl. The extracts were analysed to determine ammonium concentration by colourimetry using a segmented flow analyser (AutoAnalyser 3, Bran+Luebbe, Germany).

Samplings started the day of slurry application, foam strips were changed with a frequency ranging from twice a day during the first two days to once a week during the last weeks. Samplings were performed until 28 days after presowing fertilisation in 2016 (11 samplings) and 14 days after side-dressing fertilisation in 2017 (10 samplings) since spring weather conditions and irrigation management promoted that ammonia volatilisation stooped in a shorter timespan than in autumn. Ammonia volatilisation was not measured after side-dressing application in 2018.

Semi-open chambers are not able to measure absolute NH3 emissions as the acid strips are not able to trap all the NH3 emitted from the soil, so they need to be calibrated determining their efficiency trapping ammonia. Thus, the efficiency of the SOC trapping ammonia was determined in the laboratory using three solutions with known concentrations of ammonium (407 mg NH4 + -N L-1, 289 mg NH4 + -N L-1, and 237 mg NH4 + -N L-1). SOCs were placed (by triplicate) over 100-mm diameter Petri dishes containing 30 mL of each of the three solutions. After 24 hours, the NH4 + -N trapped by foam strips and the volume and NH4 + -N concentration of the remaining solutions were quantified. SOC efficiency was defined as the ratio between NH4 + -N trapped by the foam and the volatilised NH3 estimated as the difference between the initial amount of ammonium in the Petri dish and the remaining amount after 24 hours. Efficiency in the field could not be determined because the solution with the known concentration was systematically polluted.

Ammonia captured in the control treatment was subtracted from the ammonia captured at each PS treatment and then divided by the efficiency to obtain the amount of ammonia volatilised in the PS treatments.

 

Direct nitrous oxide emissions

Direct nitrous oxide emissions were measured in side-dressing 2017 and 2018 using a manual closed chamber methodology. The closed-chamber technique and the N2O flux measurement procedure were similar to those described by Mateo-Marín et al. (2020). Shortly, polyvinyl chloride upper cover of chambers (18.5-cm height and 30.0-cm inner diameter) wrapped in reflective insulation film were set on collars inserted 10 cm into the soil, creating 13.1-L chamber headspace. At each sampling, 15 mL of inner air were taken using a polypropylene syringe at 0 and 60 min after chamber closure. The samples were injected into 12-mL Exetainer borosilicate glass vials (Model 038W, Labco). Samplings started roughly at the hour with the mean temperature of the day (between 9:30h and 11:00h GMT; Alves et al., 2012). Samples were analysed by gas chromatography using an Agilent 7890B gas chromatography system with HP-Plot Q column and electron-capture, flame-ionisation and methaniser detectors.

Direct nitrous oxide emission rates were calculated as the linear increment in gas concentration (corrected for the air temperature) in the chamber headspace and multiplied by the ratio between the chamber volume and the soil area covered by the chamber (MacKenzie et al., 1998).

In both seasons, nitrous oxide measurements started just before the pig slurry application at tillering and ended at harvest. The samplings had daily frequency after fertilisation (during six days) and then measurements were spreading from once a week to once every two weeks, with a total of 12 and 17 sampling dates in 2017 and 2018, respectively.

The N2O emission factor (EF, %) was calculated as the ratio between the difference of cumulative N2O emissions in fertilised and unfertilised N plots, and the amount of the N applied in the fertilised plots and multiplied by 100. The yield-scaled N2O emission (YSN2O; g N kg-1 grain) is the ratio between the cumulative N2O emissions and the grain yield.

 

Soil mineral nitrogen

Soil (0-15 cm and 15-30 cm depths) was sampled in the three experimental periods (presowing 2016, side-dressing 2017 and side-dressing 2018) with daily frequency the first 5 days after fertilisation and decreasing the frequency later to reach once a week at the end of the sampling period (total of 11, 12, and 15 sampling dates in presowing 2016, side-dressing 2017, and side-dressing 2018, respectively). Two soil core samples were taken per plot to make a composite sample for each depth. Samples were sieved (3 mm) and two subsamples were obtained. A subsample was dried at 105 ºC until constant weight to determine gravimetric soil water content. Another subsample of 10 g of fresh soil was extracted with 30 mL of 2 N KCl, shaken for 30 min, and filtered through cellulose filter. Nitrate (NO3 - ) and ammonium concentration in extracts were analysed by colourimetry using a segmented flow analyser (AutoAnalyser 3, Bran+Luebbe, Germany).

 

Productive parameters and efficiency in the use of nitrogen

At wheat maturity, each plot was hand-harvested in two random areas of 0.54 m2 to obtain the grain (adjusted to 120 g kg-1) and aboveground biomass yield. Subsamples of grain and straw were dried at 65 ºC and milled to obtain the grain and straw N concentration by dry combustion (TruSpec CN, LECO, St. Joseph, MI, USA).

Two parameters were used to compare the efficiency in the use of N between treatments, the nitrogen use efficiency (NUE) and the apparent N recovery efficiency (REN). NUE is the relation between the total aboveground N uptake and the N applied by fertilisation. REN is the increment in the aboveground N uptake due to the N application per unit of N applied (Eq. 1):

 

where UT is the N uptake by aboveground biomass in the T treatment; U0 is the N uptake by aboveground biomass in the unfertilised control plot; and FT is the amount of N applied in the T treatment. The mineral N contained in the applied slurry (i.e., NH4 + -N) was used for the calculations since it was considered that the contribution of pig slurry organic N and its residual effect was not substantial during the period of the experiment.

 

Statistical analysis

Statistical analyses were performed using SAS® software (University Edition, SAS Institute Inc., Cary, NC, USA). Normal distribution and homogeneity of variance were checked by Shapiro-Wilk and Levene’s test, respectively, and variables were transformed when necessary (Box-cox transformation). Analysis of variance (MIXED procedure) was used to assess the existence of treatment effects and differences in treatment means were established with the Tukey’s test at the 0.05 significance level. In cases with measurements over time (ammonia volatilisation, nitrous oxide emission, and soil mineral nitrogen content), repeated measure analysis was used according to a first-order autoregressive structure model AR(1).