The fate of nitrogen in an animal urine patch as affected by urine nitrogen loading rate and the nitrification inhibitor dicyandiamide

(1)

Lincoln University Digital Thesis

Copyright Statement

The digital copy of this thesis is protected by the Copyright Act 1994 (New Zealand).

This thesis may be consulted by you, provided you comply with the provisions of the Act and the following conditions of use:



you will use the copy only for the purposes of research or private study



you will recognise the author's right to be identified as the author of the thesis and due acknowledgement will be made to the author where appropriate



you will obtain the author's permission before publishing any material from the thesis.

(2)

The fate of nitrogen in an animal urine patch as affected by

urine nitrogen loading rate and

the nitrification inhibitor dicyandiamide

A thesis

submitted in partial fulfilment of the requirements for the Degree of

Doctor of Philosophy

at

Lincoln University by

Diana Selbie

Lincoln University

2014

(3)

ii

(4)

iii

Abstract of a thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy

Abstract

The fate of nitrogen in an animal urine patch as affected by

urine nitrogen loading rate and

the nitrification inhibitor dicyandiamide

by Diana Selbie

The animal urine patch is the main source of nitrogen (N) loss from agricultural grazed pasture systems. Losses include emissions of nitrous oxide (N2O), a potent greenhouse gas, and leaching of nitrate (NO3

-) into waterways, both of which contribute to environmental degradation. Urine patch N loss also represents an economic loss of N from a farm. The loading rate of N in the urine patch is primarily determined by the animal’s dietary N intake and the subsequent excretion of N in the urine.

Improving N use efficiency in the urine patch is therefore of critical importance, both environmentally and economically. There has been a considerable amount of research on the fate of N in a urine patch at a single N loading rate, as well as the fate of N from multiple urine rates on a single N pathway of loss or transformation, however there is a gap in current knowledge of the fate of N in multiple loss pathways from urine applied at varying N loading rates. The application of the nitrification inhibitor dicyandiamide (DCD) has been shown to reduce urine patch N losses and increase pasture N uptake however there has been little investigation into the effect of DCD at varying urine N loading rates.

The objective of the project was to determine the effect of urine N loading rate, and the effect of DCD at varying urine N loading rates, on the fate of N in grassland soils.

Two experiments were carried out in 2009-2010 (year one) and 2010-2011 (year two) using soil monolith lysimeters collected from a free-draining sandy loam soil under pastoral dairy grazing in south-east Ireland. Dairy cow urine was diluted with water or fortified with urea to produce a range of total N concentrations which corresponded to urine N loading rate treatments of 0, 300, 500, 700 and 1000 kg N ha^-1. Two litres of urine was applied in late autumn to the 0.2 m² surface area of each lysimeter to mimic a dairy cow urine patch deposited in the field. DCD was applied twice, to lysimeters receiving urine at 500 and 1000 kg N ha^-1, the day after urine and again in early spring. The

(5)

iv

DCD was applied in solution form at a rate of 15 kg DCD ha^-1 per application. Nitrous oxide emissions, N leaching in drainage water and pasture N uptake were measured periodically following urine application, using standard methods. A mass balance determined the apparent recovery of N from each urine N loading rate. In year two, urine in the 1000 kg N ha^-1 treatments (with and without DCD) received urea labelled with the isotope ¹⁵N which produced a mix containing 45 atom% ¹⁵N.

Additional measurement of di-nitrogen gas (N2) was carried out and lysimeters were destructively sampled at the end of year two to measure ¹⁵N recovery in the soil. A ¹⁵N balance was determined using the recovery of ¹⁵N in gaseous, drainage water, pasture and soil fractions.

Increasing the urine N loading rate resulted in an increase in the cumulative N2O emissions, N leaching and pasture N uptake in both experiments. In all cases, highly statistically significant curvilinear relationships were found, with the amount of N recovered diminishing at the higher N rates, except for pasture N uptake, where the curvilinear relationship was exponential. The reason for the diminishing curvature was hypothesised as extra N at the higher N rates being recovered in pathways other than N2O emissions, N leaching and pasture N uptake. This was confirmed in the ¹⁵N balance study carried out in year two, by the recovery of 23% and 26% of urine N applied in soil N immobilisation and N2 emissions, respectively. The large recovery of N2 emissions from the 1000 kg N ha^-1 urine treatment, was almost entirely derived from the process of co-denitrification, whereby the N in N2 is derived from both urine N and native soil N sources. This finding is important both for recognising the contribution of a relatively unrecognised process to denitrification in grazed grassland, and at a broader level, to closing the gap of ‘missing N’ in the grassland N budget. The application of DCD reduced N2O emissions, N leaching and increased pasture N uptake and dry matter yield;

however, the responses were variable. There was no consistent interaction found between urine N loading rate and the application of DCD on N2O emissions, N leaching or pasture N uptake. The most likely reason for the variable DCD response was the removal of the DCD by leaching or decomposition. DCD may be used as a mitigation strategy to reduce urine patch N loss in Irish grazed pastures, providing it remains in the soil at an effective concentration.

This work has clearly shown that an increase in the urine N loading rate applied to grassland soils increases the amount of N lost in N2O emissions, lost in N leaching and taken up by pasture plants.

Keywords: dicyandiamide, di-nitrogen gas, fate, grazed pasture, leaching, loading rate, mitigation, N- 15 isotope, N balance, nitrate, nitrification inhibitor, nitrogen, nitrous oxide, recovery, uptake, urine patch.

(6)

v

Project information

Funding for this project was provided by Ireland’s Department of Agriculture, Food and the Marine, in collaboration with Lincoln University and Teagasc under the Teagasc Walsh Fellowship, which are gratefully acknowledged. Ph.D supervision was provided through Lincoln University by Professor Keith Cameron, Professor Hong Di and Dr Jim Moir, and through Teagasc by Dr Karl Richards and Dr Gary Lanigan. This project included contributions from a large number of people, the majority of whom are acknowledged in Appendix G. Ph.D project components were carried out in three locations:

1. Lincoln University, Canterbury, New Zealand: Project proposal (March-June 2009) and part of thesis write-up (November 2011-November 2012).

2. Teagasc, Johnstown Castle, Wexford, Ireland: Experimental phase (June 2009-October 2011).

3. AgResearch, Ruakura, Hamilton, New Zealand: Thesis write-up (November 2012-July 2013).

Some of the results of this project have been presented in the following conference publications:

 D.R. Selbie, G.J. Lanigan, H.J. Di, J.L. Moir, K.C. Cameron, K.G. Richards (2010) Importance of urinary N content on nitrous oxide emissions from grassland soil lysimeters. Ecotrons & Lysimeters conference, 29-31 March 2010, Nancy, France.

 D.R. Selbie, K.G. Richards, G.J. Lanigan, H.J. Di, J.L. Moir, K.C. Cameron, M.I. Khalil (2010) Manipulating N excretion – effect on N₂O emissions from grassland soil. A Climate for Change conference, 24-25 June 2010, Dublin, Ireland.

 D.R. Selbie, G.J. Lanigan, H.J. Di, J.L. Moir, K.C. Cameron, K.G. Richards (2011) Improving nitrogen efficiency using a nitrification inhibitor on urine-affected soil – a grassland lysimeter study.

Agricultural Research Forum, 14-15 March 2011, Tullamore, Ireland.

 D.R. Selbie, K.C. Cameron, H.J. Di, J.L. Moir, S. Whelan, K. Pierce, G.J. Lanigan, K.G. Richards (2011) Improving nitrogen efficiency from urine applied to grassland lysimeters in Ireland. Nitrogen and Global Change conference, 11-14 April 2011, Edinburgh, Scotland.

 D.R. Selbie, K.C. Cameron, H.J. Di, J.L. Moir, G.J. Lanigan, K.G. Richards (2012) The effect of urinary nitrogen content and DCD nitrification inhibitor on nitrogen emissions from grassland lysimeters in Ireland. Joint SSA and NZSSS Conference, 2-7 December 2012, Hobart, Australia.

 D.R. Selbie, K.C. Cameron, H.J. Di, J.L. Moir, G.J. Lanigan, R.J. Laughlin, K.G. Richards (2012) The fate of urine nitrogen with use of a nitrification inhibitor. 17th International Nitrogen Workshop, 26-29 June, Wexford, Ireland.

 D.R. Selbie, K.C. Cameron, H.J. Di, J.L. Moir, S. Whelan, K. Pierce, G.J. Lanigan, K.G. Richards (2013) Improving nitrogen efficiency from urine applied to grassland lysimeters in Ireland. Greenhouse Gases and Animal Agriculture Conference, 23-26 June 2013, Dublin, Ireland.

(7)

vi

List of Tables

Table 2.1: Examples of nitrate-N leaching losses under grazed pasture systems (Cameron et al.,

2013) ... 30

Table 2.2: A summary of nitrous oxide emissions from pastoral agriculture (Cameron et al., 2013) .. 33

Table 2.3: N inputs and outputs from intensive dairy farm systems in NZ receiving N fertiliser at nil or 410 kg ha^-1 year^-1 and data from a range of farm systems in western Europe (Ledgard et al., 2009) ... 34

Table 2.4: Annual excretal returns from grazing animals (Jarvis et al., 1995) ... 37

Table 2.5: Total N concentration in the urine of ruminant animals grazing grass-based diets ... 39

Table 2.6: Concentrations of some nitrogenous constituents of cattle and sheep urine. Adapted from Bristow et al., (1992) ... 40

Table 2.7: Characteristics for average dairy cattle and sheep urine patches (Haynes and Williams, 1993; Whitehead, 1995b) ... 42

Table 2.8: Summary of four urine patch ¹⁵N balance studies (data is percentage recovery of ¹⁵N applied in urine where total recovery equals 100%) ... 45

Table 2.9: Summary of research experiments using dietary manipulation to reduce urinary N excretion from dairy cows ... 54

Table 2.10: Summary of research experiments on the effect of the nitrification inhibitor DCD on urine N loss and pasture uptake from pastoral soils ... 58

Table 2.11: Decomposition of DCD in soil as affected by temperature (Amberger, 1989) ... 61

Table 2.12: Schematic presentation of a typical grass-based Irish dairy farm (40 ha)... 62

Table 2.13: Suggested nitrogen fertiliser application schedule for cattle grazed swards at various stocking rates in Ireland. Adapted from Teagasc (Teagasc, 2008) ... 63

Table 3.1: Profile description of the Moorepark soil. ... 68

Table 3.2: Key physical properties of the Moorepark soil. ... 70

Table 3.3: Concentrations of major nutrients in the Moorepark soil. ... 70

Table 3.4: Key chemical properties of the Moorepark soil. ... 71

Table 3.5: Description of treatments applied to lysimeters in years one and two. ... 75

Table 3.6: Nitrogen fertiliser application schedule in years one and two. ... 79

Table 4.1: Summary of treatments applied to lysimeters in year one and year two. ... 83

Table 4.3: Cumulative N2O emissions and emission factors in year one and year two as affected by urine N rate and DCD ... 99

Table 5.1: Summary of treatments applied to lysimeters in year one and year two. ... 112

Table 5.3: Cumulative amount of nitrate-nitrogen (NO3 --N) leached as affected by urine N rate and DCD application. ... 127

Table 5.4: Cumulative amount of ammonium-nitrogen (NH4 +-N) leached as affected by urine N rate and DCD application. ... 130

Table 5.5: Cumulative amount of nitrite-nitrogen (NO2 --N) leached as affected by urine N rate and DCD application. ... 131

Table 5.6: Cumulative amount of dissolved organic N (DON) (predominantly urea-N) leached as affected by urine N rate and DCD application. ... 133

Table 5.7: Cumulative amount of total N leached as affected by urine N rate and DCD application. 135 Table 6.1: Summary of treatments applied to lysimeters in year one and year two. ... 152

Table 6.3: Dates of the pasture harvests in year one and year two. ... 154

Table 6.4: Cumulative total pasture dry matter yield from individual harvests as affected by urine N rate and DCD application. ... 161

Table 6.5: Average pasture N content (%) in year one and year two as affected by urine N rate and DCD application. ... 164

Table 6.6: Cumulative total pasture N uptake as affected by urine N rate and DCD application. ... 167

Table 6.7: Average annual concentrations of major nutrients in the lysimeter pasture. ... 173

Table 7.1: Mass balance recovery of nitrogen from lysimeters in year one and year two. ... 182

Table 7.2: Summary of treatments applied to lysimeters. ... 191

(13)

xii

Table 7.3: Percentage ¹⁵N recovery from urine and DCD applied to lysimeters. ... 210 Table 7.4: Source and quantity of nitrous oxide emissions from ¹⁵N labelled urine and DCD

applied to lysimeters. ... 211 Table 7.5: Quantity of di-nitrogen gas emissions from ¹⁵N labelled urine and DCD applied to

lysimeters. ... 212 Table 7.6: Source and quantity of the total N leached from ¹⁵N labelled urine and DCD applied to

lysimeters. ... 212 Table 7.7: Source and quantity of pasture N uptake from ¹⁵N labelled urine and DCD applied to

lysimeters. ... 213 Table 7.8: Percentage recovery of ¹⁵N applied in urine from each lysimeter soil depth, as affected

by DCD. ... 214

(14)

xiii

List of Figures

Figure 2.1: The nitrogen cycle in grazed pasture systems (Cameron, 1992) ... 24

Figure 2.2 Typical volatilisation loss pattern and soil pH following urea broadcast onto pasture (McLaren and Cameron, 1996). ... 27

Figure 2.3 Generalized diagram showing plant growth in relation to the nutrient content in plant tissue (Whitehead, 1995a) ... 28

Figure 2.4: Gaseous nitrogen is emitted from nitrification and denitrification processes in the soil (Saggar et al., 2004) ... 31

Figure 2.5: Denitrification in (a) gley soil, (b) pan and (c) within a soil aggregate (McLaren and Cameron, 1996) ... 32

Figure 2.6: Nitrate-N leaching from grazed pasture systems as affected by total N input from fertilisers and/or clover N2 fixation. Data are a summary of studies from NZ, France, UK and Denmark (Ledgard et al., 2009) ... 35

Figure 2.7: Distribution of urine through the soil profile under sheep and cattle urine patches (Williams and Haynes, 1994) ... 39

Figure 2.8: Effect of dietary N concentration on the excretion of N by sheep. Total excretion (faeces plus urine) and excretion in faeces alone are shown (Barrow and Lambourne, 1962). ... 41

Figure 2.9: Concentrations of urinary N and the percentage of urea-N in the urine of dairy cows grazing a white-clover ryegrass mix and a high N (304 g cow^-1 day^-1) or low N (139 g cow^-1 day^-1) diet fed as concentrates (Petersen et al., 1998) ... 42

Figure 2.10: Mean daily pH at four soil depth increments in a urine patch over a 2-week period (Haynes and Williams, 1993) ... 44

Figure 2.11: Concentrations of urea, ammonium and nitrate in the soil profile below sheep and cattle urine patches following urine application (Williams and Haynes, 1994) ... 44

Figure 2.12: Relationships between urine application rate and total nitrate-N leaching loss with and without DCD nitrification inhibitor (Di and Cameron, 2007) ... 46

Figure 2.13: The relationship between total N application rate and total N2O emission from grassland, A = single spring CAN application, B = split CAN application. Percentage of applied fertiliser N emitted as N2O presented above bars (Velthof et al., 1997) ... 47

Figure 2.14: Pasture dry matter yields as affected by urine N application rate and DCD nitrification inhibitor (Di and Cameron, 2007) ... 48

Figure 2.15: Nitrous oxide emissions by sector in Ireland 1990-2005 (Environmental Protection Agency, 2005). ... 50

Figure 2.16: Nitrification inhibitors reduce the enzyme activity of nitrifying bacteria in the first stage of nitrification, the conversion of ammonium to nitrite. ... 55

Figure 2.17: (a) Relationship between the amout of DCD leached and cumulative drainage; (b) Literature comparison of DCD loss and drainage with the relationship found in (a), including extrapolation (broken line) (Shepherd et al., 2012) ... 60

Figure 3.1: Treatment allocation for 30 lysimeters in year one (left, green) and for 32 lysimeters in year two (right, brown). Each circle shows the lysimeter number and treatment name. ... 76

Figure 4.1: Schematic diagram of chamber and gas ring fitted to a lysimeter for nitrous oxide measurement ... 85

Figure 4.2: Daily and cumulative rainfall in the year one experiment. ... 91

Figure 4.3: Daily average air temperature in the year one experiment. ... 92

Figure 4.4: Daily and cumulative rainfall in the year two experiment. ... 92

Figure 4.5: Daily average air temperature in the year two experiment. ... 93

Figure 4.6: Temporal N2O emissions following urine application to lysimeters at varying urine N rates in year one. (Control is no urine, U300 is 300 kg N ha^-1 urine, U500 is 500 kg N ha^-1, U700 is 700 kg N ha^-1, U1000 is 1000 kg N ha^-1) (Bars display 5% LSD). ... 94 Figure 4.7: Temporal emissions following urine and fertiliser application to lysimeters at varying

urine N rates in year two from a 360 day measurement period. (Control is no urine,

(15)

xiv

U300 is 300 kg N ha^-1 urine, U500 is 500 kg N ha^-1, U700 is 700 kg N ha^-1, U1000 is 1000 kg N ha^-1) (Bars display 5% LSD). ... 95 Figure 4.8: Temporal N2O emissions following urine application to lysimeters at varying urine N

rates and DCD application in year one. (Control is no urine or DCD, U500 is 500 kg N ha^-1 urine, U1000 is 1000 kg N ha^-1 urine, U500+DCD is 500 kg N ha^-1 urine plus 30 kg DCD ha^-1, U1000+DCD is 1000 kg N ha^-1 urine plus 30 kg DCD ha^-1) (Bars display 5% LSD). ... 96 Figure 4.9: Temporal N2O emissions following urine application to lysimeters at varying urine N

rates and DCD application in year two over a 360 day measurement period.

(Control is no urine or DCD, U500 is 500 kg N ha^-1 urine, U1000 is 1000 kg N ha^-1 urine, U500+DCD is 500 kg N ha^-1 urine plus 30 kg DCD ha^-1, U1000+DCD is

1000 kg N ha^-1 urine plus 30 kg DCD ha^-1) (Bars display 5% LSD). ... 97 Figure 4.10: The relationship between urine N rate and cumulative N2O emissions in year one. ... 101 Figure 4.11: The relationship between urine N rate and cumulative N2O emissions in year two over

an 80 day measurement period. ... 101 Figure 4.12: The relationship between urine N rate and cumulative N2O emissions in year two over

a 360 day measurement period. ... 102 Figure 5.1: Daily average air temperature in year one. ... 119 Figure 5.2: Daily average air temperature in year two. ... 120 Figure 5.3: Daily and cumulative rainfall inputs, evapotranspiration and drainage losses in year

one. ... 121 Figure 5.4: Daily and cumulative rainfall inputs, evapotranspiration and drainage losses in year

two. ... 121 Figure 5.5: Year one nitrate-N concentration (±SE) in the drainage water of lysimeters as affected

by varying urine N application rate. (Control is no urine, U300 is 300 kg N ha^-1 urine, U500 is 500 kg N ha^-1, U700 is 700 kg N ha^-1, U1000 is 1000 kg N ha^-1) (1 PV indicates one pore volume of drainage which was 260 mm). ... 122 Figure 5.6: Year two nitrate-N concentration (±SE) in the drainage water of lysimeters as affected

by varying urine N application rate. (Control is no urine, U300 is 300 kg N ha^-1 urine, U500 is 500 kg N ha^-1, U700 is 700 kg N ha^-1, U1000 is 1000 kg N ha^-1) (1 PV indicates one pore volume of drainage which was 290 mm). ... 123 Figure 5.7: Year one nitrate-N concentrations (±SE) in the drainage water of lysimeters as affected

by varying urine N application rate and DCD application. (Control is no urine or DCD, U500 is 500 kg N ha^-1 urine, U1000 is 1000 kg N ha^-1 urine, U500+DCD is 500 kg N ha^-1 urine plus 30 kg DCD ha^-1, U1000+DCD is 1000 kg N ha^-1 urine plus 30 kg DCD ha^-1) (1 PV indicates one pore volume of drainage which was 260 mm).124 Figure 5.8: Year two nitrate-N concentrations (±SE) in the drainage water of lysimeters as affected

by varying urine N application rate and DCD application. (Control is no urine or DCD, U500 is 500 kg N ha^-1 urine, U1000 is 1000 kg N ha^-1 urine, U500+DCD is 500 kg N ha^-1 urine plus 30 kg DCD ha^-1, U1000+DCD is 1000 kg N ha^-1 urine plus 30 kg DCD ha^-1) (1 PV indicates one pore volume of drainage which was 290 mm).125 Figure 5.9: The relationship between urine N rate and the total amount of nitrate-N leached in the

lysimeter drainage. ... 128 Figure 5.10: The relationship between urine N rate and the total amount of ammonium-N leached

in the lysimeter drainage. ... 129 Figure 5.11: The relationship between urine N rate and the total amount of nitrite-N leached in the

lysimeter drainage. ... 132 Figure 5.12: The relationship between urine N rate and the total amount of DON leached (kg N

ha^-1) in the lysimeter drainage. ... 134 Figure 5.13: The relationship between urine N rate and the total amount of N leached in the

lysimeter drainage. ... 136 Figure 5.14: Concentration of each form of N leached (±SE) in the lysimeter drainage in year one

where urine was applied at 1000 kg N ha^-1 without DCD. (Total N is the sum of nitrate-N, ammonium-N, nitrite-N and dissolved organic N) (1 PV indicates one pore volume of drainage which was 260 mm). ... 137 Figure 5.15: Concentration of each form of N leached (±SE) in the lysimeter drainage in year two

where urine was applied at 1000 kg N ha^-1 without DCD. (Total N is the sum of

(16)

xv

nitrate-N, ammonium-N, nitrite-N and dissolved organic N) (1 PV indicates one

pore volume of drainage which was 290 mm). ... 137

Figure 5.16: Concentration of bromide (±SE) in the lysimeter drainage by treatment in year one. (1 PV indicates one pore volume of drainage which was 260 mm)... 138

Figure 5.17: Concentration of bromide (±SE) in the lysimeter drainage by treatment in year two. (1 PV indicates one pore volume of drainage which was 290 mm)... 139

Figure 6.1: Year one average daily air temperature. The arrow indicates the date of urine application. ... 158

Figure 6.2: Year two average daily air temperature. The arrow indicates the date of urine application. ... 158

Figure 6.3: Year one daily rainfall and cumulative rainfall and evapotranspiration (ET). The arrow indicates the date of urine application. ... 159

Figure 6.4: Year two daily rainfall and cumulative rainfall and evapotranspiration (ET). The arrow indicates the date of urine application. ... 160

Figure 6.5: The relationship between urine N rate and pasture dry matter yield in year one. ... 162

Figure 6.6: The relationship between urine N rate and pasture dry matter yield in year two. ... 162

Figure 6.7: The relationship between urine N rate and the pasture N content in year one. ... 165

Figure 6.8: The relationship between urine N rate and the pasture N content in year two. ... 165

Figure 6.9 The relationship between urine N rate and the total amount of N taken up by the pasture in year one. ... 168

Figure 6.10: The relationship between urine N rate and the total amount of N taken up by the pasture in year two. ... 168

Figure 7.1: The effect of urine N rate on the net N recovery from lysimeters in year one (Back- transformed treatment means Table 7.1) (Net N recovery = Mass N recovery – N recovery in control). ... 183

Figure 7.2: The effect of urine N rate on the net N recovery from lysimeters in year two (Treatment means Table 7.1) (Net N recovery = Mass N recovery – N recovery in control). ... 184

Figure 7.3: The effect of urine N rate on the N recovery from lysimeters in year one and year two. (The regression was fitted for year one, where the relationship was significant P < 0.05; Table 7.1). ... 184

Figure 7.4: Schematic representation of the difference in the N concentration per unit volume of soil when the urine distribution is constant. ... 186

Figure 7.5: Schematic representation showing the effect of soil texture on the concentration of urine N per unit volume of soil, lower N concentration on the lighter textured soil per unit volume of soil. ... 187

Figure 7.6: The relationship between the percentage recovery of urine N taken up by the pasture and leached, from each lysimeter (n=1), in year one and year two. ... 188

Figure 7.7: Average daily air temperature and 100 mm soil temperature. The arrow indicates the date of urine application. ... 208

Figure 7.8: Daily rainfall and cumulative rainfall and evapotranspiration (ET). The arrow indicates the date of urine application. ... 209

(17)

xvi

List of Plates

Plate 2.1: The nitrogen loading rate under a dairy cow urine patch within the range 400-1200 kg N

ha^-1 (Photo courtesy of A. Judge). ... 37

Plate 2.2: Animal urine patches in a grazed pasture ... 38

Plate 3.1: Lysimeter casing was pushed down while soil was chipped away and the base of the monolith was cut from the surrounding soil. ... 72

Plate 3.2: The lysimeter was lifted from the trench and the cutting plate removed and replaced with the circular drainage plate. ... 72

Plate 3.3: Liquefied petroleum was poured down the internal edge of the lysimeter which was then loaded onto a trailer. ... 73

Plate 3.4: Lysimeters were installed into a prepared facility using a digger. ... 74

Plate 3.5: Lysimeters with gas rings and drainage vessels attached and the complete lysimeter facility. ... 74

Plate 3.6: Urine was collected by hand, using buckets, from dairy cows during the morning and afternoon milking periods. ... 77

Plate 3.7: The DCD solution was applied to the lysimeter surface using a spray bottle. ... 78

Plate 3.8: The Johnstown Castle weather station was situated adjacent to the lysimeter facility. ... 80

Plate 4.1: The gas ring trough was filled with water before the chamber was carefully inserted ... 86

Plate 4.2: Chamber headspace air was flushed twice before a sample was extracted and transferred to a pre-evacuated glass exetainer using a hypodermic needle and plastic syringe ... 86

Plate 4.3: Nitrous oxide analysis was carried out using a gas chromatograph coupled to an electron capture detector and auto-sampler. The N2O peak was identified and the area under the peak calculated. ... 87

Plate 5.1: Drainage water from each lysimeter was collected via a pipe to a 10 L collection vessel. 114 Plate 5.2: Lysimeter drainage water was poured from each collection vessel into sample tubes ... 114

Plate 6.1: The lysimeters were collected from an existing perennial ryegrass (Lolium perenne L.) pasture which was sown in 2008. ... 150

Plate 6.2: The pasture was cut to approximately 30 mm using hand clippers, prior to lysimeter collection. ... 151

Plate 6.3: The lysimeter facility at Teagasc Johnstown Castle, County Wexford, in year one. ... 152

Plate 6.4: Pasture was cut to approximately 20 mm using scissors or electric clippers and the harvested herbage removed. ... 154

Plate 6.6: An example of the low pasture growth which was typical for the lysimeters in year one. 172 Plate 7.1: Glass fibre filter paper discs were made using a hole punch and acidified using 2.5 M KHSO4. ... 200

Plate 7.2: Glass bottles contained 150 mL diffusion solution and a screw top lid with acidified filter discs attached. ... 201

Plate 7.3: Soil KCl extractions were carried out on field-moist soil from three lysimeter soil depths. ... 204

(18)

xvii

List of Equations

Equation 2.1: N balance equation for the amount of mineral N in the soil (Cameron et al., 2013) ... 23

Equation 2.2: Nitrification – Step 1 ... 25

Equation 2.3: Nitrification – Step 2 ... 25

Equation 2.4: Ammonia equilibrium with ammonium in soil (Cameron, 1992) ... 26

Equation 2.5: Urea hydrolysis (McLaren and Cameron, 1996) ... 26

Equation 2.6: Calculation for the N loading rate in a urine patch ... 42

Equation 2.7: Calculation for the temperature dependent half-life of DCD in soil (Kelliher et al., 2008). ... 60

Equation 3.1: Calculation for soil bulk density ... 69

Equation 3.2: Calculation for soil total porosity ... 69

Equation 4.1: Calculation for hourly nitrous oxide emissions ... 88

Equation 4.2: Calculation of the emission factor for nitrous oxide ... 89

Equation 5.1: Calculation for the volume of drainage in a collection vessel ... 113

Equation 5.2: Calculation for the annual amount of N leaching from each lysimeter ... 117

Equation 5.3: Calculation for cumulative amount of N leaching per hectare ... 117

Equation 5.4: Calculation for one pore volume of lysimeter drainage ... 117

Equation 5.5: Calculation for annual average field scale N leaching loss (Di and Cameron, 2000) .. 144

Equation 6.1: Calculation for total pasture dry matter yield ... 155

Equation 6.2: Calculation for total pasture N uptake ... 155

Equation 7.1: Calculation for the mass recovery of N from each lysimeter. ... 180

Equation 7.2: Calculation for the percentage recovery of N from each lysimeter ... 181

Equation 7.3: Calculation for the total moles of urea needed per lysimeter ... 191

Equation 7.4: Calculation for the moles of ¹⁵N enriched urea needed per lysimeter ... 192

Equation 7.5: Calculation for the moles of natural abundance urea needed per lysimeter ... 192

Equation 7.6: Calculation for the mass of ¹⁵N enriched urea needed per lysimeter ... 192

Equation 7.7: Calculation for the mass of natural abundance urea needed per lysimeter ... 192

Equation 7.8: Calculation for the percentage ¹⁵N recovery in a measured fraction (Cabrera and Kissel, 1989) ... 193

Equation 7.9: Calculation for the moles of N2O emitted per day ... 195

Equation 7.10: Calculation of the quantity of N2O emitted derived from urine, in units of kg N ha^-1 195 Equation 7.11: Calculation for the percentage of N2O emitted dervied from urine ... 196

Equation 7.12: Calculation for the enrichment of the source of ¹⁵N labelled N2 (Mulvaney, 1984) ... 196

Equation 7.13: Calculation for the fraction of total N2 attributable to true denitrification, adapted from Mulvaney (Mulvaney, 1984) ... 197

Equation 7.14: Calculation for the fraction of total N2 attributable to co-denitrification (Clough et al., 2001b) ... 197

Equation 7.15: Calculation for the amount of N2 in the chamber atmosphere in grams ... 197

Equation 7.16: Calculation for the total N2 evolved into the chamber during the enclosure period in grams, adapted from Mulvaney and Boast (Mulvaney and Boast, 1986) ... 198

Equation 7.17: Calculation for the flux of N2 from true or co-denitrification ... 198

Equation 7.18: Calculation for the percentage recovery of ¹⁵N in N2 gas ... 198

Equation 7.19: Calculation for the moles of total N leached for each lysimeter. ... 201

Equation 7.20: Calculation for the quantity of labelled N leached for each lysimeter. ... 202

Equation 7.21: Calculation for the percentage of N leached derived from urine ... 202

Equation 7.22: Calculation for the moles of N in pasture from each lysimeter ... 202

Equation 7.23: Calculation for the quantity of labelled N in the pasture for each lysimeter ... 203

Equation 7.24: Calculation for the percentage of N in the pasture derived from urine... 203

Equation 7.25: Calculation for the quantity in kilograms of dry soil from each lysimeter depth ... 205

Equation 7.26: Calculation for the mass in grams of N in each soil depth ... 205

Equation 7.27: Calculation for the moles of total or inorganic N in each soil depth ... 205

(19)

xviii

Acronyms and abbreviations

ANOVA Analysis of variance

Br Bromide

°C Degrees Celsius/Centigrade

C Carbon

CO2 Carbon dioxide

DCD Dicyandiamide

DM Dry matter

DON Dissolved organic nitrogen

EA-IRMS Elemental analyser-Isotope ratio mass spectrometry

EF Emission factor

GHG Greenhouse gas

IPCC Intergovernmental Panel on Climate Change

IRMS Isotope ratio mass spectrometry

KCl Potassium chloride (soil extractant) kg N ha^-1 Kilograms of nitrogen per hectare

LSD Least significant difference (statistical analysis) MIT Mineralisation-immobilisation turnover (nitrogen)

15N Nitrogen-15 isotope

N2 Nitrogen gas (di-nitrogen)

NH3 Ammonia

NH4

+ Ammonium

N2O Nitrous oxide

NO2

- Nitrite

NO3

- Nitrate

P P-value (statistical analysis)

TN Total nitrogen

TOC-TN analyser Total organic carbon-total nitrogen analyser

TON Total oxidised nitrogen

U Urine (treatment)

(20)

19

Chapter 1 General introduction

Nitrogen (N) is a naturally-occurring chemical element and an essential nutrient for the growth and reproduction of plants and animals. N fertilisers are manufactured and used to support food production, without which half of the world’s population would not be alive today (Erisman et al., 2008). The requirement for N fertiliser continues due to a continually growing global population (Gruber and Galloway, 2008). Countries such as Ireland and New Zealand generate a large proportion of their gross domestic product from agriculture, the majority of which is pasture-based production of meat and milk products. In grazed pasture systems, meat and milk are produced from ruminant animals which consume a mostly grass/clover diet by grazing outdoors. However, the N cycle in agricultural systems, which includes all of the inputs, outputs and transformations of N, is a ‘leaky’

system, and N can be unintentionally lost into the environment (Galloway et al., 2008). The consequences of environmental N loss can be negative, including emissions of N gases such as nitrous oxide (N2O) which contribute to global warming (IPCC, 2007), and pollution of waterways (Sutton et al., 2011). An overview of N and the issues associated with its use are described in a video clip provided by the European Nitrogen Assessment¹.

In grazed pasture systems, N is taken up from the soil by the pasture plants, consumed by the grazing animal, and utilised in animal products such as meat and milk. However, the majority of the N ingested is returned to the pasture in animal urine and dung, particularly in the urine. Because urine is excreted in patches, rather than distributed evenly over the paddock (Lantinga et al., 1987), the amount of N in a urine patch exceeds pasture requirements, and the extra N may be lost into the atmosphere in gaseous emissions, leached into drainage water, or converted into soil organic matter (Clough et al., 1998). The loading rate of N in a urine patch can be up to 1000 kg N ha^-1 (Haynes and Williams, 1993), mostly in urea form (Bristow et al., 1992; Petersen et al., 1998). Urine patches have been identified as the main source of N loss in grazed pasture systems (Di and Cameron, 2002b), which can have negative environmental impacts (Ledgard et al., 2009).

The importance of N in the urine patch was highlighted in reviews by Haynes and Williams (1993), Jarvis et al. (1995) and Lantinga et al. (1987). Targeted investigations have included urine N dynamics in soil (Haynes and Williams, 1992; Monaghan and Barraclough, 1992; Williams and Haynes, 1994) and the fate of urine N (Whitehead and Bristow, 1990; Fraser et al., 1994; Decau et al., 2003). The ‘average’ dairy cow urine patch was described by Haynes and Williams (1993) in a review of existing literature, as 2 L of 10 g N L^-1 urine returned to a surface area (wet) of 0.2 m², which

1 https://www.youtube.com/watch?v=uuwN6qxM7BU.

(21)

20

corresponds to a urine patch loading rate of 1000 kg N ha^-1. The urine N loading rate actually varies from 400 to over 1000 kg N ha^-1 (Haynes and Williams, 1993; Jarvis et al., 1995), which is primarily due to the amount of N ingested by the animal (Barrow and Lambourne, 1962). In a productive ryegrass-based pasture, the N content is ~4% which is high relative to other animal feeds (Misselbrook et al., 2005). Therefore animals grazing a diet with a high proportion as pasture excrete more N in urine than cows consuming a mixed diet (van Vuuren and Meijs, 1987; van Vuuren et al., 1993).

Studies have shown that dietary manipulation can reduce the amount of N excreted in the urine of grazing animals, by ~20% on a per cow per day basis (Misselbrook et al., 2005; Whelan et al., 2011).

Therefore the range in urine patch N loading rates is produced mainly by changes in the dietary N intake of grazing animals.

As N inputs from mineral N fertiliser, urine and dung in grazed pastures increases, the amount of N loss increases, and this has been shown in several studies (Ledgard et al., 1999; de Klein et al., 2001;

Monaghan et al., 2005; Rafique et al., 2011). Barraclough et al. (1992) showed an apparent break point between 400-500 kg N ha^-1year^-1 of mineral fertiliser applied, above which the amount of N leaching loss from a grazed pasture accelerated. Furthermore, studies have reported the effect of increasing application rates of mineral N fertiliser on N2O emissions (Velthof et al., 1997), N leaching (Scholefield et al., 1993) and pasture yield (Monaghan et al., 2005). Fewer studies have investigated the effect of increasing urine N loading rate on N2O emissions (van Groenigen et al., 2005b), N leaching and pasture N uptake (Di and Cameron, 2007; Moir et al., 2012a). With the exception of Di and Cameron (2007), Singh et al. (2009), and Moir et al. (2012a), trials which applied multiple urine N loading rates measured the effect on a single N pathway only. Detailed experiments have been conducted on the fate of N in the urine patch using ¹⁵N isotopic balance, reporting the percentage of N applied recovered in soil as 25%, pasture as 43%, drainage water as 11% and gaseous emissions as 3%

(Fraser et al., 1994; Clough et al., 1998; Di et al., 2002; Decau et al., 2003). However these studies described the fate of N in multiple pathways from a single urine N loading rate. Therefore, there is a gap in the literature with little research reported into the effect of a range of urine N loading rates on the fate of N in multiple pathways, and this needs to be investigated.

Mitigation strategies and technologies to reduce N losses from grazed pasture systems have been reviewed by several authors (Saggar et al., 2004; Monaghan, 2008; Stark and Richards, 2008). One mitigation strategy which directly targets urine patch N loss is the use of the nitrification inhibitor dicyandiamide (DCD). Research has consistently shown that the application of DCD following urine deposition on pastoral soils can reduce N2O emissions and N leaching (Di and Cameron, 2004b, 2008;

de Klein et al., 2011; Dennis et al., 2012; Welten et al., 2013). Pasture N uptake and dry matter production may also be increased using DCD, however the results have been variable (Moir et al., 2007; Monaghan et al., 2009; Zaman and Blennerhassett, 2010; Carey et al., 2012). Three studies have tested the effect of DCD at varying loading rates of urine N (Di and Cameron, 2007; Singh et al.,

(22)

21

2009; Moir et al., 2012a). (Di and Cameron, 2007) reported an increase in the percentage change by DCD in N leaching and pasture N uptake, where the urine N rates were increased in 0, 300, 700 and 1000 kg N ha^-1 treatments. In a lab study using small intact soil cores, Singh et al. (2009) showed that the percentage reduction in N2O emissions increased with increasing urine N rate, but there was not a similar trend in N leaching or pasture dry matter yield. These results are conflicting, and require further investigation.

Therefore, the overall objectives of the research described in the thesis were:

1. To quantify the effect of urine N loading rate on cumulative N2O emissions, N leaching and pasture N uptake from grassland lysimeters in Ireland;

2. To quantify the effect of dicyandiamide (DCD) nitrification inhibitor and N loading rate on cumulative N2O emissions, N leaching and pasture N uptake from urine applied to grassland lysimeters in Ireland; and

3. To determine the fate of urine N applied at varying N loading rates.

Two experiments were conducted using intact soil monolith lysimeters collected from a dairy grazed pasture in Ireland, over a two year period. Each experiment consisted of an autumn urine application followed by measurements of N2O emissions, N leaching and pasture growth periodically for approximately one calendar year. Treatments consisted of dairy cow urine applied at urine N loading rates of 0, 300, 500, 700 and 1000 kg N ha^-1. DCD was applied in solution form at a rate of 15 kg DCD ha^-1 twice following urine application, at the urine N rates of 500 and 1000 kg N ha^-1. The treatment structure was a randomised block design with four replicates per treatment. Following analysis of the results from the year one experiment, a ¹⁵N isotope tracer was added with the urine applied at 1000 kg N ha^-1 (with and without DCD) in year two, in order to better understand the fate of urine N.

A review of the literature identifies the gaps in existing knowledge and establishes the research objectives and hypotheses in Chapter 2. General methods are described in Chapter 3. The effects of urine N loading rate (Objective #1) and DCD (Objective #2) are discussed for N2O emissions in Chapter 4, N leaching in Chapter 5, and pasture N uptake in Chapter 6. Objective #3, the effect of urine N loading rate on the fate of N, is investigated in Chapter 7 using mass balance and ¹⁵N balance methods. Chapter 8 includes an evaluation of the hypotheses, discussion of the limitations and implications of the research, and recommendations for future research.

(23)

22

Chapter 2 Literature review

2.1 Introduction

The main objective of this review was to identify gaps in current knowledge about nitrogen (N) losses from grazed pasture soils in Ireland and options to reduce those losses.

Some more specific objectives of this literature review were to:

 Understand some of the key N cycle processes occurring in pastoral soils;

 Describe the key drivers for N transformations and losses in pastoral soils, particularly N returned in animal urine;

 Describe some of the conditions in a urine patch and factors influencing the fate of urine N;

 Summarise the main issues associated with environmental N emissions;

 Identify and describe potential mitigation options for reducing environmental N losses;

 Provide context for pastoral systems in Ireland.

With this information, and the identified gaps in current knowledge, new research objectives and hypotheses were formulated.

(24)

23

2.2 Soil nitrogen cycle in pastoral agriculture

The N content of most soils is 0.1-0.6%, which is equivalent to approximately 2-12 t N ha^-1 depending on the soil type (Cameron et al., 2013). The majority of N is present in organic forms (plant residues, humus and microbial biomass) and mineral forms (ammonium NH4

+, nitrite NO2

- and nitrate NO3 -).

The inputs, outputs and transformations of N in the soil system influence the amount of N that is available for plants and transferred into the wider environment (Cameron et al., 2013) (Figure 2.1).

Mineral N contained in the soil solution is the central source of N from which all major soil processes are derived. A N balance equation may be used to estimate the amount of mineral N in the soil ( ) (Equation 2.1):

Equation 2.1: N balance equation for the amount of mineral N in the soil (Cameron et al., 2013) where is precipitation and dry deposition, is biological fixation, is fertiliser, is urine and dung returns to the soil, is mineralisation, is plant uptake, is gaseous loss, is immobilisation, is leaching loss and is erosion and surface runoff.

The soil N cycle in pastoral systems is described in further detail by Stevenson (1982), Haynes et al.

(1986), Whitehead (1995b), McLaren and Cameron (1996), Ledgard et al. (1999), and Cameron et al.

(2013). Some of the key transformations and losses of N in grazed pasture systems are described here:

mineralisation, nitrification, immobilisation, plant uptake, leaching, volatilisation and denitrification.

(25)

24

Figure 2.1: The nitrogen cycle in grazed pasture systems (Cameron, 1992) 2.2.1 Mineralisation, nitrification and immobilisation

Mineralisation is the process by which organic N is converted to mineral N, and involves the breakdown of complex proteins to amino acids through a series of reactions by microbes, eventually releasing ammonia (NH3). This final stage, where NH3 is released is termed ammonification, and provides microbes with energy and a source of N (Cameron, 1992).

Nitrification is a two-step oxidation reaction which involves the conversion of NH4

+ to NO3

- via NO2 -

(Figure 2.1). The first step, from NH4

+ to NO2

-, is carried out by the ammonia monooxygenase enzyme associated with ammonia oxidising bacteria (AOB), such as Nitrosospira and Nitrosomonas (Equation 2.2). In the second step, the NO2

- produced is oxidised to NO3

- by Nitrobacter (Equation 2.3).

Because the proportion of bacteria mediating nitrification is small, soil conditions such as pH, moisture content, temperature and nutrient status are important (McLaren and Cameron, 1996).

(26)

25

Nitrification is of particular significance in relation to potential Nleaching, because it is the sole production pathway of NO3

-, which is readily lost from the soil due to its negative charge (Section 2.2.4). The conversion of NH4

+ to NO3

- is usually rapid, and consequently, N addition to soil is vulnerable to loss if it is not utilised by plants, or immobilised into soil organic matter.

→

Equation 2.2: Nitrification – Step 1

→

Equation 2.3: Nitrification – Step 2

The reverse process of mineralisation is immobilisation, where N is converted into organic N forms.

The main pathway for immobilised N is the assimilation of N into microbial tissue. Mineralisation and immobilisation occur simultaneously in soil (Cameron, 1992), and the amount of mineral N resulting will depend on the net process taking place. Net mineralisation occurs where the amount of N released exceeds the amount of N assimilated by microbes, whereas net immobilisation describes the reverse. The carbon (C) to N ratio (C:N) of the material decomposing (organic N source) can be used to determine the net process occurring. When the material has a high N content (e.g. 3% N) then the C:N is low (e.g. a C:N ratio <25:1) and more mineral N is produced than is required by microbes, thus net mineralisation occurs. Net immobilisation occurs when the organic material has a low N content (e.g. a C:N ratio >25:1) and the N produced is insufficient for microbial N requirements. Net immobilisation results in the reduction of N available for plant uptake as well as for loss via leaching and gaseous emissions. The C:N ratio of some common soil components and plant materials are clover (20:1), grass (24:1), soil bacteria (5:1), and soil humus (10:1) (Whitehead, 1995i; McLaren and Cameron, 1996).

Some of the key factors affecting mineralisation, nitrification and immobilisation are: soil type (Monaghan and Barraclough, 1995; Decau et al., 2003), moisture and temperature (Haynes, 1986a), forms and amounts of C and N (Barrett and Burke, 2000; Bengtsson et al., 2003), cultivation (Silgram and Shepherd, 1999) and microbial factors (Lovell and Jarvis, 1998). Other key references describing mineralisation-immobilisation turnover (MIT) include: Jansson and Persson (1982), Jenkinson et al.

(1985) and Jarvis et al. (1996). Ledgard et al. (1998) estimated daily mineralisation and immobilisation rates of 4.0 and 0.1 kg N ha^-1 day^-1, respectively, under grazed pasture receiving 200 kg N ha^-1 year^-1, and 3.0 and 1.0 kg N ha^-1 day^-1, respectively, under grazed pasture receiving 0 N.

Daily rates of up to 13.3 and 27.8 kg N ha^-1 day^-1 for mineralisation and immobilisation, respectively, were measured by Watson and Mills (1998), for soil under long-term grassland. The net effects of mineralisation, nitrification and immobilisation processes are notoriously difficult to measure at a field

(27)

26

scale except with the use of multiple assumptions (Murphy et al., 2003), thus it is an area that requires further methodological development.

2.2.2 Ammonia volatilisation

Volatilisation refers to the gaseous loss of NH3 from the soil surface. The concentration of NH3 in soil is dependent on an equilibrium with NH4

+ (Equation 2.4). Hydroxide ions (OH^-) increase the soil pH, where gaseous NH3 production is favoured and consequently N loss via volatilisation is enhanced.

⇔

Equation 2.4: Ammonia equilibrium with ammonium in soil (Cameron, 1992)

Ammonia volatilisation occurs as a result of fertiliser application, urine and dung deposition, and native soil N mineralisation. Soils with a naturally high pH (e.g. calcareous soils) are more likely to lose significant amounts of applied and native N by NH3 volatilisation. After urea application (fertiliser or urine), the soil pH is temporarily increased (Black et al., 1985b). This is because urea ((NH2)2CO) is converted to ammonium carbonate ((NH4)2CO3) which dissociates to produce OH^- ions, NH4+

, NH3 and carbon dioxide (CO2) (Equation 2.5). This temporary pH increase occurs immediately around each urea granule and so NH3 loss is greatest between 2 and 4 days after fertiliser application (Figure 2.2). Increases in temperature will also cause volatilisation by shifting the equilibrium in Equation 2.4 towards NH3. High concentrations of NH4

+ produced under a urine patch coupled with a spike in pH close to the soil surface result in high concentrations of NH3 and potential volatilisation losses (Haynes and Williams, 1993).

( ) ⇔ ( ) ⇔ Equation 2.5: Urea hydrolysis (McLaren and Cameron, 1996)

(28)

27

Figure 2.2 Typical volatilisation loss pattern and soil pH following urea broadcast onto pasture (McLaren and Cameron, 1996).

Ammonia losses from pastoral soils were recently summarised by Cameron et al. (2013), with on average 19% of N volatilised from 15-500 kg N ha^-1 applied as urea. Maximum volatilisation rates were found under high NH4

+ concentrations at the soil surface, coupled with warm temperatures and elevated pH (Black et al., 1985a; Black et al., 1985b). Thus the greatest N loss would be expected in the warmer months of the year. Ball et al. (1979) reported that 5 and 16% of N applied was volatilised in cool and warm conditions, respectively. Rainfall and irrigation reduce NH4

+ at the soil surface and thus the potential NH3 volatilisation loss by up to 80% (Black et al., 1987). Mitigation options are generally based on fertiliser application methods which reduce the concentration of NH4

+ at the soil surface, for example sub-surface injection, and fertiliser application prior to predicted rainfall or irrigation events (Cameron et al., 2013). Another option is the use of a urease inhibitor such as N-(n- butyl) thiophosphoric triamide (NBPT) to coat urea granules, which can reduce NH3 volatilisation by up to 95% (Watson et al., 1994). NH3 volatilisation is described in further detail by Whitehead (1995k), Haynes and Sherlock (1986), Saggar et al. (2004) and Sommer et al. (2004).

The extent of NH3 volatilisation loss is influenced by a range of environmental factors, and for this reason a number of studies have reported variable emissions over the short (Hatch et al., 1990) and longer term (Jarvis et al., 1989). Environmental factors include pH, soil moisture, temperature, wind velocity and soil organic carbon. pH is probably the most important factor influencing NH3 losses which are most serious: (1) on soils of pH greater than 7; or (2) with fertiliser forms that produce a high pH after addition to soil e.g. via urea hydrolysis. Black et al. (1985b) found that NH3

volatilisation was positively related to the maximum soil surface pH produced by each fertiliser type.

Hot, dry summer conditions favour volatilisation rather than cool, moist conditions (Haynes and Williams, 1993). Urine patch volatilisation losses were found to be 22% in summer, 25% in autumn, compared to only 12% in winter (Sherlock and Goh, 1984). At a mean air temperature of 8 °C, Ryden et al. (1987) measured urine N loss of only 10% (as NH3), compared to 22% at 16 °C.

2.2.3 Plant uptake

Over 90% of the N plants take up is in NH4+

and NO3-

forms (Whitehead, 1995a), which is absorbed from the soil solution, through the root, to the xylem. NH4

+ is rapidly converted to amino acids and amides, whereas NO3

- must first be converted to NH4

+ in a two-step process which requires considerable energy from photosynthesis (Haynes, 1986b). The capacity of pastures to take up N is high compared to other crops, and may be up to 500 kg N ha^-1 (Whitehead, 1995a), although typically 200-400 kg N ha^-1 (Goh and Haynes, 1986; Di et al., 1998). Figure 2.3 shows a generalized diagram

The fate of nitrogen in an animal urine patch as affected by urine nitrogen loading rate and the nitrification inhibitor dicyandiamide

Lincoln University Digital Thesis

Copyright Statement

The digital copy of this thesis is protected by the Copyright Act 1994 (New Zealand).

This thesis may be consulted by you, provided you comply with the provisions of the Act and the following conditions of use:

you will use the copy only for the purposes of research or private study

you will recognise the author's right to be identified as the author of the thesis and due acknowledgement will be made to the author where appropriate

you will obtain the author's permission before publishing any material from the thesis.

The fate of nitrogen in an animal urine patch as affected by

urine nitrogen loading rate and

the nitrification inhibitor dicyandiamide

A thesis

submitted in partial fulfilment of the requirements for the Degree of

Doctor of Philosophy

at

Lincoln University by

Diana Selbie

Lincoln University

2014

Abstract of a thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy

Abstract

The fate of nitrogen in an animal urine patch as affected by

urine nitrogen loading rate and

the nitrification inhibitor dicyandiamide

by Diana Selbie

Project information

Table of Contents

List of Tables

List of Figures

List of Plates

List of Equations

Acronyms and abbreviations

Chapter 1

General introduction

Chapter 2 Literature review

2.1 Introduction

2.2 Soil nitrogen cycle in pastoral agriculture