Fertiliser N applications are subject to ammonia (NH3) volatilisation losses, particularly upon urea application, which may reduce N use efficiency and represent a substantial economic loss of N from agriculture. NH3 volatilisation also contributes to indirect nitrous oxide emissions (Martikainen, 1985) and is related to the deterioration of regional air quality, as well as eutrophication and acidification of natural ecosystems (Asman, 1998). As a result, a number of European countries have been set annual emissions ceilings for NH3 under the National Emission Ceilings Directive (European Commission, 2001). Meeting these ceiling obligations presents a challenge for Irish agriculture, which accounts for 98% of national NH3 emissions.
NH3 emissions from urea fertiliser on grassland in Ireland and the UK have been found to be quite variable, ranging from 8 to 68% of applied N (Chambers and Dampney, 2009; Forrestal et al., 2015) and may be due to differences in temperature, precipitation and wind speed following urea application (Black et al., 1987; Hatch et al., 1990; Sommer et al., 1991, 2003; Sanz-Cobena et al., 2011). Another important factor that may contribute to this range in emissions is the variation of soil types, with different physical and chemical characteristics (Stevens et al., 1989; Watson et al., 1994; He et al., 1999). Direct comparisons of NH3 emissions from different soil types are limited to laboratory-based studies (McGarry et al., 1987; Watson et al., 1994), while the majority of field-based studies on NH3 emissions from N fertilisers have been conducted on only one or two sites (Chambers and Dampney, 2009; Forrestal et al., 2015). NH3 emission factors (EFs) used for fertiliser N in national NH3 inventories are based on field studies and do not take soil type into account. Thus, there is a need for field-based comparisons of NH3 emissions from different soil types. The objective of this study was to compare NH3 emissions from six different soil types under similar environmental conditions in the field following urea fertiliser application.
Materials and methods
This experiment was conducted on a lysimeter unit at Teagasc, Johnstown Castle Research Centre, County Wexford, Ireland (52°180’N, 6°300’W; 62 m above sea level). The 30-yr mean annual precipitation and air temperature at the site are 1,037.5 mm and 10.4 °C, respectively. The establishment of the lysimeter unit is described in full by Ryan and Fanning (1996). The lysimeter unit consisted of replicated (n = 9) undisturbed monoliths of soils (diameter: 0.6 m; depth: 1 m) encased in rigid fibreglass cylinders, which were collected from six grassland sites across the Republic of Ireland in 1990. The soils were chosen to encompass a range of typical soil types, drainage characteristics and soil parent material (Table 1). Each lysimeter has a perennial ryegrass (Lolium perenne L.) sward, which was reseeded in April 2015. In the year before the commencement of this experiment, the lysimeters were fertilised with a total of 160 kg N/ha in four equal applications, the last being in June 2015. Herbage was cut to a height of 4–5 cm using a pair of hand shears and removed at least 1 wk before the commencement of the experiment.
Soil classification and properties of the six soils used in this experiment
|Site||Soil type1||Parent material||Texture||Field capacity2||CEC3||pH4||Sand5||Silt||Clay|
|Oakpark||Haplic Cambisol||Fluvioglacial gravels||Sandy loam||19.5||46.0 ± 9.3||6.56 ± 0.1||67||23||11|
|Castlecomer (Gortacla-reen)||Albic Gleyic Lixsol (Humic)||Fine loamy drift with siliceous stones||Fine loam||31.6||55.7 ± 7.2||6.34 ± 0.1||24||48||28|
|Clonroche||Haplic Cambisol||Glacial drift||Loam||24.8||48.8 ± 11.3||6.64 ± 0.1||44||39||17|
|Elton||Cutanic Luvisol||Glacial drift||Loam||24.1||66.5 ± 3.4||6.27 ± 0.0||48||35||17|
|Rathangan (Kilrush)||Luvic Stagnosol||Glacial sea drift||Loam||25.3||51.6 ± 6.7||6.10±0.1||44||37||19|
|Johnstown||Luvic Gleysol||Coarse loamy drift with siliceous stones||Loam||21.3||25.8 ± 0.0||6.46 ± 0.0||58||30||12|
The experiment was based on a completely randomised design with six soil types (treatments) and three replicates per soil type. The six soil types were named after the locations from which they were collected in Ireland and represent a range of drainage classes from light to heavy as follows: Oak Park, Clonroche, Elton, Rathangan, Castlecomer and Johnstown. The characteristics of each soil are presented in Table 1. The lysimeter soils had similar parent material and, to some extent, similar World Reference Base (WRB) classifications (WRB, 2014), but contrasting texture, structure and drainage properties (Table 1) (Kramers et al., 2012).
Weather and soil conditions
Meteorological parameters including air temperature (in degrees Celsius), rainfall (in millimetres) and wind speed (in metres per second) were recorded on an hourly basis at the nearest automatic weather station “Johnstown Castle” by the Irish Meteorological Service (Met éireann) (approximately 500 m distant from the study site). Additionally, volumetric soil moisture at 0–5 cm depth was determined daily during the experiment using a handheld soil moisture probe (Delta-T Devices, Cambridge, UK).
Application of urea fertiliser and simulated rainfall event
On 7 September 2015, 6.71 L of water (the equivalent of 25.4 mm of rainfall) was applied to each of the 18 lysimeters using a watering can with a rosette attachment to accentuate the drainage capacity of each soil type, which would be manifested in terms of soil surface moisture at the time of urea application. This simulated rainfall event also ensured that soil moisture was sufficient to promote urea hydrolysis after urea application, given the fact that rainfall had to be excluded from the lysimeters for the duration of the study due to the configuration of the NH3 measurement equipment on the lysimeter units. Urea fertiliser was subsequently applied to each lysimeter by hand at a rate of 80 kg N/ha.
NH3 emission measurements
A system of 18 wind tunnels (Lockyer, 1984) was used to measure NH3 volatilisation. Each wind tunnel unit consisted of (i) a canopy (0.5 m × 2 m) made of polycarbonate, (ii) a galvanised sheet steel duct and (iii) a control box. The wind tunnel canopy was placed over each lysimeter immediately after urea application and it stayed in place until the end of the experiment. Wind speed through the wind tunnels was set at 1 m/s (air flow rate of 0.229 m3/s), which was chosen to mimic atmospheric wind speed above the soil surface. Air entering and leaving the wind tunnel canopy was sampled and pumped through two individual conical absorption flasks (i.e. acid traps), which contained 100 mL of 0.02 M orthophosphoric acid (H3PO4, 85%; Merck, Darmstadt, Germany). Emissions were measured continuously for a period of 14 d after application. The acid traps were replaced every approximate 24 h until the 11th day after application, and final samples were collected after 72 h (Day 14) on the final day of the experiment. The acid trap samples were analysed for their ammonium-N content, and the NH3-N loss (in kilograms per hectare) was calculated as described by Fischer et al. (2016).
All data were statistically analysed using SAS 9.3 (2011; SAS Institute Inc., Cary, NC, USA). Data were checked for normality by assessing residual normality and variance. The effects of soil type and measurement date, as well as their interaction, on daily NH3 emissions and volumetric soil moisture were analysed using analysis of variance (ANOVA), with measurement date included as a repeated measure in the model. The treatment effect on cumulative NH3 loss was analysed using ANOVA. Post-hoc least significant difference (LSD) multiple-comparison tests were carried out to determine differences between treatment means. Pearson’s correlation and multiple linear regression analyses were conducted using individual lysimeter data (n = 18) to test for relationships between NH emissions and soil variables. A statistical probability of P < 0.05 was considered significant for all statistical tests.
Weather and soil conditions
The mean daily air temperature ranged from 11 to 15 °C (mean: 12.9 °C), mean daily wind speed was 3.68 m/s and the cumulative precipitation was 58.8 mm during the experiment. There was a significant (P < 0.01) interaction between soil type and measurement date on soil moisture during the study (Figure 1). All soil types had similar volumetric soil moisture on the date of urea application, with the exception of Oakpark, which had lower moisture. Soil moisture declined from the start to the end of the experiment for all soil types; however, the decline in soil moisture on the Johnstown soil type was more variable and lower than that of all other soil types (Figure 1).
Daily NH3 emissions ranged from 0 to 5.58 kg N/ha and there was no significant difference in daily emissions across soil types (Figure 2). However, daily NH3 emissions varied significantly across measurement dates (P < 0.001), with emissions peaking within 3 d after fertiliser application on each soil type and then declining slowly until the end of the measurement period (Figure 2). The majority (>85%) of the cumulative emissions during the experiment occurred for each soil type within 1 wk (7 d) of fertiliser application. There was no significant difference between soil types in terms of cumulative emissions (kg NH3-N/ha) or the percentage of N applied lost as NH3-N at the end of the experiment (Table 2). NH3 EFs (percentage of N applied) for the six soil types ranged from 12.8% to 21.5% and averaged (mean ± s.d.) 17.6 ± 6.2 across soil types. There were no significant relationships between NH3-N emissions (percentage of N applied) and soil variables such as pH, cation exchange capacity (CEC) and volumetric soil moisture content following Pearson’s correlation and multiple linear regression analyses.
Cumulative ammonia emission (mean ± s.d.) and the percentage of applied nitrogen lost as ammonia during the experimental period (n = 3)
|Oakpark||14 ± 4.6||17.0a|
|Castlecomer||17 ± 3.0||20.9a|
|Elton||17 ± 6.4||21.5a|
|Rathangan||13 ± 4.4||17.8a|
|Johnstown||10 ± 8.9||12.8a|
Temporal and cumulative NH3 emissions
The daily temporal NH3 emissions from each soil type followed a similar trend as in previous field studies, wherein emissions peaked on Day 2 or Day 3 following urea application and declined thereafter (Chambers and Dampney, 2009; Forrestal et al., 2015). Watson et al. (1994) also found peak NH3 emissions to occur on Day 3 (range 1.8–4.5 d), on average, following urea application to 16 different soils in a laboratory study. The NH3 emissions in this study are on the lower side of the reported urea-NH3 emissions in studies using wind tunnels on grassland (Table 3). These previous studies were conducted on field plots and generally over a larger number of applications, including summer applications, which would be subject to higher temperatures and solar radiation, which are conducive to higher volatilisation losses (Huijsmans et al., 2001).
Summary of literature reporting ammonia emissions following urea application to grassland
|Soil type||n1||Country||Method||N application rate (kg/ha)||NH3-N emission (% of N applied)||Reference|
|Sandy clay loam||1||UK||Wind tunnels||90-120||30||Van der Weerden and Jarvis, 1997|
|Clay loam||1||UK||Wind tunnels||90-120||28||Van der Weerden and Jarvis, 1997|
|Loam||2||Ireland||Wind tunnels||80||36||Forrestal et al., 2015|
|Silty clay loam||4||UK||Wind tunnels||100||31||Chambers and Dampney, 2009|
|Course sandy loam||1||UK||Wind tunnels||100||43||Chambers and Dampney, 2009|
|Sandy loam||1||UK||Wind tunnels||100||12||Chambers and Dampney, 2009|
|Sandy loam||1||Denmark||Wind tunnels||80-120||25||Sommer and Jensen, 1994|
|N.a.||1||Canada||bLs2||80||29||Sommer et al., 2005|
|N.a.||1||Chile||IHF3||100||12||Salazar et al., 2012|
|Loam||1||UK||Wind tunnels||200||15||Ryden and Lockyer, 1985|
|N.a.||1||UK||Wind tunnels||70-100||20||Ryden et al., 1987|
|Silty loam||1||Australia||Micro-met||40||30||Suter et al., 2013|
|Clay||1||Netherlands||Wind tunnels||80–120||23||Velthof et al., 1990|
N.a Not available
Soil type and NH3 emissions
Watson et al. (1994) compared NH3 volatilisation from 16 different grassland soils with various chemical and physical properties under controlled laboratory conditions following the application of 100 kg N/ha as urea. There were significant differences between the different soils, which ranged from 5.8% to 38.9% of the N applied. Multiple regression analysis revealed that the soil properties pH (by KCl extraction) (range: 4.9–7.4) and titratable acidity explained up to 95% of the variation in NH3 emissions across the 16 soils. Chambers and Dampney (2009) found NH3 emissions to range from 10% to 58% of N applied as urea at six different sites with varying soil types. However, these measurements were not conducted in parallel and were therefore subject to varying weather conditions following urea application. There was no significant difference in NH3 emissions between the different soil types in the current field-based study. This could be due to the relatively high spatial variability and low sample size associated with field-based measurements of such emissions. Van der Weerden and Jarvis (1997), in a field plot study investigating emissions from urea, reported similar NH3 emissions for sandy clay loam (pH: 6) and clay loam (pH: 5.6) soils in the UK (Table 3).
He et al. (1999), in a controlled laboratory study, found NH3 emissions to increase by 150% as the soil pH increased from 4.5 to 5.5 and by 10% from 5.5 to 6.5. Further increasing the soil pH to 7.5 and 8.5 had no significant effect on NH3 emissions compared to a soil pH of 6.5. The range of soil pH across the soil types in this study was lower than that in the studies mentioned herein and was most likely too small to influence NH3 emissions from the applied urea under field conditions. Chambers and Dampney (2009) also found no relationship between NH3 emissions and soil pH, CEC, clay content or organic carbon content.
Implications for national inventories
This study and other recent field-based studies suggest that urea-NH3 EFs do not differ between different soil types in cases where the soils are actively managed for soil pH through soil fertility planning. This finding may have implications for the calculation of urea-based NH3 emissions in national inventories in terms of the disaggregation of the urea-NH3 EF based on soil type. Further field-based studies are required across a larger range of soil types and seasonal weather conditions to improve the quantitative accuracy of the urea-NH3 EF used for the Irish national inventory.
NH3 emissions (% of N applied) did not differ significantly between the six soil types in this study and ranged from 5 to 29. The relatively small range observed in measured soil variables; pH, CEC and volumetric moisture were not able to explain any of the variation in measured NH3 emissions. The results of this study represent only the second study of NH3 emissions from urea fertiliser on Irish grasslands. Further field studies are required to improve the accuracy of urea-NH3 EF used for Irelands national inventory.
The authors would sincerely like to thank the laboratory and field staff at Teagasc Johnstown Castle for their assistance on this study, in particular Ms Joella Rutgers. This research was financially supported by the Irish Department of Agriculture, Food and the Marine (grant number RSF 13/S/430).
Asman W.A.H. 1998. Factors influencing local dry deposition of gases with special reference to ammonia. Atmospheric Environment 32:415–421.
Black A.S. Sherlock R.R. and Smith N.P. 1987. Effect of timing of simulated rainfall on ammonia volatilization from urea applied to soil of varying moisture content. Journal of Soil Science 38: 679–687.
Chambers B. and Dampney P. 2009. Nitrogen efficiency and ammonia emissions from urea-based and ammonium nitrate fertilisers. Proceedings No. 657 International Fertiliser Society York UK.
European Commission 2001. Directive 2001/81/EC of the European Parliament and the Council of 23 October 2001 on national emission ceilings for certain atmospheric pollutants. L 309/22.214.171.1241 Available from: http://eur-lex.europa.eu/Lex-UriServ/LexUriServ.do?uri=OJ:L:2001:309:0022:0030:EN:PDF [accessed 26 October 2016].
Fischer K. Burchill W. Lanigan G.J. Kaupenjohann M. Chambers B.J. Richards K.G. and Forrestal P.J. 2016. Ammonia emissions from cattle dung urine and urine with dicyandiamide in a temperate grassland. Soil Use and Management 32: 83–91.
Forrestal P.J. Harty M. Carolan R. Lanigan G.J. Watson C. Laughlin R.J. McNeill G. Chambers B.J. and Richards K.G. 2015. Ammonia emissions from stabilised urea fertiliser formulations in temperate grassland. Soil Use and Management 32: 92–100.
Hatch D.J. Jarvis S.C. and Dollard G.J. 1990. Measurements of ammonia emission from grazed grassland. Environmental Pollution 65: 333–346.
He Z.L. Alva A.K. Calvert D.V. and Banks D.J. 1999. Ammonia volatilization form different fertilizer sources and effect of temperature and soil pH. Soil Science 164: 750–758.
Huijsmans J.F.M. Hol J.M.G. and Hendriks M.M.W.B. 2001. Effect of application technique manure characteristics weather and field conditions on ammonia volatilization from manure applied to grassland. NJAS – Wageningen Journal of Life Sciences 49: 323–342.
- Export Citation
Huijsmans, J.F.M., Hol, J.M.G. and Hendriks, M.M.W.B. 2001. Effect of application technique, manure characteristics, weather and field conditions on ammonia volatilization from manure applied to grassland.)| false NJAS – Wageningen Journal of Life Sciences49: 323–342. 10.1016/S1573-5214(01)80021-X
Kramers G. Holden N.M. Brennan F. Green S. and Richards K.G. 2012. Water content and soil type effects on accelerated leaching after slurry application. Vadose Zone Journal 11: .
Lockyer D.R. 1984. A system for the measurement in the field of losses of ammonia through volatilization. Journal of the Science of Food and Agriculture 35: 837–848.
Martikainen P.J. 1985. Nitrous oxide emission associated with autotrophic ammonium oxidation in acid coniferous forest soil. Applied and Environmental Microbiology 50: 1519–1525.
McGarry S.J. O’Toole P. and Morgan M.A. 1987. Effects of soil temperature and moisture content on ammonia volatilization from urea-treated pasture and tillage soils. Irish Journal of Agricultural Research 26: 173–182.
Ryan M. and Fanning A. 1996. Effects of fertiliser N and slurry on nitrate leaching – lysimeter studies on 5 soils. Irish Geography 29: 126–136.
Ryden J.C. and Lockyer D.R. 1985. Evaluation of a system of wind tunnel for field studies of ammonia loss from grassland through volatilization. Journal of the Science of Food and Agriculture 36: 782–788.
Ryden J.C Whitehead D.C. Lockyer D.R. Thompson R.B. Skinner J.H. and Garwood E.A. 1987. Ammonia emission from grassland and livestock production systems in the UK. Environmental Pollution 48: 173–184.
Salazar F. Martinez-Lagos J. Alfaro M. and Misselbrook T. 2012. Ammonia emissions from urea application to permanent pasture on a volcanic soil. Atmospheric Environment 61: 395–399.
Sanz-Cobena A. Misselbrook T. Camp V. and Vallejo A. 2011. Effect of water addition and the urease inhibitor NBPT on the abatement of ammonia emission from surface applied urea. Atmospheric Environment 45: 1517–1524.
Saxton K.E. Rawls W.J. Romberger J.S. and Papendick R.I. 1986. Estimating Generalized Soil-water Characteristics from Texture. Soil Science Society of America Journal 50: 1031–1036.
Sommer S.G. Génermont S. Cellier P. Hutchings N.J. Olesen J.E. and Morvan T. 2003. Processes controlling ammonia emission from livestock slurry in the field. European Journal of Agronomy 19: 465–486.
Sommer S.G. and Jensen C. 1994. Ammonia volatilization from urea and ammoniacal fertilizers surface applied to winter wheat and grassland. Fertilizer Research 37: 85–92.
Sommer S.G. McGinn S. and Flesch T. 2005. Simple use of the backwards Lagrangian stochastic dispersion technique for measuring ammonia emission from small field-plots. European Journal of Agronomy 23: 1–7.
Sommer S.G. Olesen J.E. and Christensen B.T. 1991. Effects of temperature wind-speed and air humidity on ammonia volatilization from surface applied cattle slurry. Journal of Agricultural Science 117: 91–100.
Stevens R.J. Laughlin R.J. and Kilpatrick D.J. 1989. Soil properties related to the dynamics of ammonia volatilization from urea applied to the surface of acidic soils. Fertilizer Research 20: 1–9.
Suter H. Sultana H. Turner D. Davies R. Walker C. and Chen D. 2013. Influence of urea fertiliser formulation urease inhibitor and season on ammonia loss from ryegrass. Nutrient Cycling in Agroecosystems 95: 175–185.
Van der Weerden T. and Jarvis S. 1997. Ammonia emission factors for N fertilizers applied to two contrasting grassland soils. Environmental Pollution 95: 205–211.
Velthof G.L. Oenema O. Postmus J. and Prins W.H. 1990. In situ field measurements of ammonia volatilization from urea and calcium ammonium nitrate applied to grassland. Proceedings of the 13th General Meeting of the European Grassland Federation II Banská Bystrica Slovakia pages 51–55.
Watson C.J. Miller H. Poland P. Kilpatrick D.J. Allen M.D.B. Garrett M.K. et al. 1994. Soil properties and the ability of the urease inhibitor N-(n-BUTYL) thiophosphoric triamide (nBTPT) to reduce ammonia volatilization from surface-applied urea. Soil Biology and Biochemistry 26: 1165–1171.
- Export Citation
Watson, C.J., Miller, H., Poland, P., Kilpatrick, D.J., Allen, M.D.B., Garrett, M.K.)| false et al. 1994. Soil properties and the ability of the urease inhibitor N-(n-BUTYL) thiophosphoric triamide (nBTPT) to reduce ammonia volatilization from surface-applied urea. Soil Biology and Biochemistry26: 1165–1171. 10.1016/0038-0717(94)90139-2
WRB. 2014. I.W.G. 2014. “World Reference Base for Soil Resources 2014 First Update 2015” World Soil Resources Reports 106. FAO Rome.