The fate of nitrogen in lactose-depleted dairy factory effluent irrigated onto land

(1)

THE FATE OF NITROGEN IN LACTOSE-DEPLETED DAIRY FACTORY EFFLUENT IRRIGATED ONTO LAND

A thesis submitted in partial fulfilment of the requirements for the degree of

Master of Science

Bye.D. Ford

Lincoln University

2008

(2)

LINCOLN

U N I V E R S I T Y

Te Wh are W~ na l a 0 Ao ral i

Nameofcandidate:_ C _ \L .:...J.. l( .;;... H ;....;,. n .:....-..., ] ~_ l l-- ()_ (~ ....:...-

_ _ _ _ _ _ _ _ _ _

DECLARATION

This thesis is submitted in

Degree of

M ~ ~ t ~ ( () f

^partial ^_fulfilments of the requirements for the Lincoln University

St\U't~ (ri'\~ · \(~~m~nto.\)

The regulations for the degree are set out in the Lincoln University Calendar and are elaborated in a practice manual known as Houserules for the Study of Masters/PhD Degrees at Lincoln University.

Supervisor's Declaration

I confirm that, to the best of my knowledge:

• the research was carried out and the thesis prepared under my direct supervision;

• except where otherwise approved by the Academic Administration Committee of Lincoln University, the research was conducted in accordance with the degree regulations and houserules;

• the thesis represents the original research work of the candidate;

• the contribution made to the research by me, by other members of the supervisory team, by other members of staff of the univ~ and by others was consistent with normal supervisory practice.

Supervisor

~~~

^Date

<7-;/06

Candidate's Declaration I confirm that:

• this thesis represents my own work;

• the contribution of any up isors and others to the research and to the thesis was consistent with normal supervisory practice.

Candidate Date

Pre-Publication of Parts of this Thesis Either:

1 We confirm that no p thesis for examination

is thesis has been submitted for publication in advance of submission of the

Candidate

~ \I~

^Date

Supervisor Date

Or:

2 Parts of this thesis have been submitted and/or accepted for publication in advance of submission of the thesis for examination.

In this case, please set out on a separate page information on:

1

• which sections have been submitted, which have been accepted and which have appeared;

• which journals they have been submitted to;

• who are the co-authors.

Candidate Date

(3)

II avra pel Kal ovl5eV J.1eVel

Everything flows, nothing stands still.

Heraclitus of Ephesus (c.535 - 475 Be)

(4)

Abstract of a thesis submitted in partial fulfilment of the requirements for the Degree of Master of Science (Environmental), Lincoln University, Canterbury, New Zealand.

The Fate of Nitrogen in Lactose-Depleted Dairy Factory Effluent Irrigated onto Land

ByC.D. Ford

A two-year lysimeter study was undertaken to compare the environmental effects (e.g. nitrate leaching and nitrous oxide emissions) of soil applied lactose-depleted dairy factory effluent (LD-DFE) with lactose-rich DFE. The aim of this experiment was to determine the fate of nitrogen from LD-DFE and dairy cow urine applied to a Templeton fine sandy loam soil (Udic Ustrochrept), supporting a herbage cover of ryegrass (Lolium perenne) and white clover (Trifolium repens). Measurements were carried out on the amount of nitrogen lost from the soil via leaching, lost by denitrification, removed by the pasture plants, and immobilized within the soil organic fraction. Further, a comparison between the fate of nitrogen in LD-DFE irrigated onto land under a "cut and carry"

system, as opposed to a "grazed" pasture system was undertaken. Lactose-depleted dairy factory effluent was applied at three-weekly intervals during the summer months at rates of 25 and 50 mm, until nitrogen loading targets of 300 and 600 kg N ha-¹yr-¹had been achieved. Measured leaching losses of nitrogen averaged 2 and 7 kg N ha-¹yr-¹for Control 25 and Control 50 treatments; 21, 20 and 58 kg N ha-¹yr-l for 25 and 50 mm "cut and carry" treatments respectively; and 96 kg N ha-¹yfl for the 25 mm "grazed"

treatment. The range of nitrate-N leaching loss from LD-DFE plus urine is no different from the lactose-rich DFE nitrate leaching loss. Uptake of nitrogen by the growing pasture averaged 153, 184,340,352,483, and 415 kg N ha-¹yfl for Control 25, Control 50, LD-DFE 25 and LD-DFE 50 mm "cut and carry" treatments, and the LD-DFE 25 mm

"grazed" treatment, respectively. Denitrification losses were 0.06, 4.4, 1.69, 19.70, and 7.4 kg N ha-¹yr-¹for Control 25, the LD-DFE 25 "cut and carry" treatments, the LD-DFE 25 mm "grazed" treatment, and calculated "paddock losses", respectively. Isotopic nitrogen studies found that 29.4 and 25.8% of applied LD-DFE nitrogen was immobilised in the LD-DFE 25 and LD-DFE 50 "cut and carry" treatments. The results of this experiment confirm the findings of the previous lactose-rich DFE study, in that the effects of grazing stock are of greater environmental concern than the removal of lactose from the effluent waste stream.

Keywords: Nitrogen, lactose-depleted dairy factory effluent, nitrate leaching, denitrification, immobilisation.

(5)

(6)

:My (J)earest of Prient!s

It is witli a{most immeasura6fe efation ant! re{ief tliat one su6mits tliis tliesis regarding Cactose-tfepfetea aairy factory eJffuent, ant! tlie fate of tlie nitrogen containea witliin, fo{{owing its aeposition onto soi{ in tliese fine CandS. 'Wlien em6arRjng on sucli a tasF<. one cannot imagine liow immensefy 6usy one wi{{ 6ecome - so mucli so tliat one's mina 6ecometli occupiea 6y a{{

matters tliesis-reCatea, witli neitlier siglit nor souna of pfeasure or merriment! It is tlierefore of tlie utmost importance tliat one tal?!s tlie time to tlian{ a{{ tlie fine fatfies ant! gent{emen of tliese parts wlio liatli contri6utea to ensuring tliat sucli a friglitfu{fy grana unt!ertalijng is carriea out witli tlie utmost of ease.

Pirstfy, one wouU fil?! to affora tlie most lieartfe{t tlianF<§ to Professor Cameron, ant! Professor CDi. :More {nowfeagea6fe ant! unweariea supervisors, one couU not imagine oneself 6eneatli tlie direction of. Sad'fy, one fears tliat one is fil?!fy to 6e recaffea as a ratlier arstractea apprentice. 'Wlien saUl Professors are talijng tea in years ten ana two fortli, ana consUfering stuaents of tlieir tuteCage gone, one regrets tliat tlie amount of time spent gaffivanting in tlie lii{{s of tliis fine country wi{{ 6e remem6erea prior to tlie difigence of one's stutfy lia6its. Purtlier, one couU not liave unt!erta~n sucli a commission witliout tlie financia{ espousa{ of tlie Centre for Soi{ anaP.nvironmenta{ Quafity.

}l{{ persons of tlie Soi{ ant! Pliysica{ Sciences group liave in some sma{{ manner easea tlie 6uraen of one's worn.., In particuCar, one feefs tliat tlie assistance in tlie foU aefiverea 6y :Messrs :Moore ana :Masters was far 6eyont! tlie rearms of any contractua{ 06figations, ana for tliat one consiaers oneself etemaffy inae6tea. :Miss }lmanaa CfiJfora is to 6e most liiglify commenaea for tlie rapUf ant! efficient aefiverance of any liaraware requirea from Stores. SimiCarfy, :Messrs Cresswe{{ ana cJ3reitmeyer of a{{ Services }lnafytica{ liave 6een of tlie most tfe{iglitfu{ nature to aea{ wit Ii. One cannot accentuate sufficientfy liow important it is tliat one aeafs witli aefiglitfuC-naturea gent{emen in tlie Ca60ratory. Of most wortliy mention afso, is tlie spiritea acquaintance tliat :Ms Lynne C{ucas liatli afforaea auring one's time in tliese parts. :Not onfy liave lier powers of conversation 6een consUfera6fe, 6ut slie liatli lid tlie generous foresiglit to provitfe a most agreea6fe assem6Cage of music wliicli liatli proven to 6e most cont!ucive in tlie creation of a pfeasing environment in wliicli one toifs. (J)r Cfougli liatli afso on occasion provitfea most su6stantia{ dvice regarding a mu{tipficity of issues. One must a{ways 6e of tlie most receptive nature wlien one is surrount!ea 6y sucli enufite persons. 'Ilie puroeyance of copious amounts of carrot cal?! liatli int!eea proven to 6e a most wortliy means of acquiring a variety of {nowfeagea6fe facts.

:Mem6ers of tlie Linco{n Vniversity C'WI are to 6e congratuCatea on tlieir liancficrajt ant! 6alijng slij{{s. :Miss Piona Slianliun (Scone }lficionaao) ana :Ms Janet cJ3ertram (Lace-:Malijng P.ntliusiast) liatli provitfea a wonaeifuffy caring environment for one to unaerta~ tlie improvement of one's competence in sucli slij{{s. }lfas, it pains one aeepfy to tliin{ tliat aistance liatli now tlirust its liarsli aagger tlirougli tlie ae{icate ties tliat 6ina us. J{ow, pray teff, wi{{ tliou suroive tlie aay of tlie week/ourtli if one is no fonger a6{e to anticipate tlie arriva{ of tlie mysterious strangerwlio gamislies tliese liaffs?

Con.fo£ants ana clierisliea friendS 6eyont! tlie wa{{s of tliis institution liatli afso contri6utea to tlie successfu{ comp{etion of tlie aforementionea tasF<, :Ms Sanara 'Tuffy ana:Mr Jolin J{u{me; :Mr CR,pa VardY; (J)r Sliaron P.ngfisli ant!:Mr Len Smytli; (J)r CRpfana :Meyer ant! :Mrs Vaferie :Meyer; :Ms Susan giffora; :Mr Lars }lnt!ersson; :Mr :Martin Coffins ant!:Mrs J{efen Coffins.

}lna finaffy: :Mr Pau{ 'Wi{{s, (J)r c.R...liys cJ3ums, :Mr :Mifton cJ3{oomfo[tf, ant! (J)r }lnarew cJ3eafe - a finer assem6fy of jovia{

gent{emen wlio unaerta~ unusuaffy farge assignments in tlie countrysUle, a fad] couU not liope to fint! in a {ifetime of searcliing. 'Iliou liatli a{{ 6een of tlie most munificent nature, ant! one prays tliat over time our friendSliips wi{{ most surefy 6fossom as ao tlie first springflowers tliat 6urst fortli in wliat are tlie trufy magnificent P.state garaens.

Pina{fy, one must mention tlie unencum6erea generosity ant! support of one's famify. 'Ilie fate :Mr Percy Pora - tliou liatli triea friglitfu{fy liar a to 6e precise. :Mrs Is06e{ Pora - one's painstalijng work. is finisliea at fast! 'Iliy fatlier, :Mr Ian Pora, is to 6e congratufatea for teacliing liis cliiUren to 6e of tlie most tliouglitfu{ ant! inaepent!ent nature - if onfy on occasion! 'Iliy aear motlier, :Mrs Caro{ Pora, must 6e appfauaea for lier patient teacliings at tlie typewriter {eys auring one's formative years. It is of tlie liigliest certainty tliat tlie a6ifity to toucli-type liatli maae tliy tas{easier. 'Iliy wont!etju{ si6fings:

:Mr }lfastair Pora ana :Mrs J{efen :Mo{{oy. }lna :Mr :Mar{ Coffins. J{ow, pray te{{, miglit one liave finisliea witliout tliy seemingfy ineJ(liausti6fe Cove, encouragement, ant! aavice?

Inopportunefy, time presses on, ana one must apofogise profusefy for tlie premature cfosure of tliis epistfe. Vnwittingfy, one findS oneself witli a muftituae of imminent tasli.§ - a{{ of wliicli require immeaiate attenaance. 'Witli great liaste one prays tliat you are a{{ 6fessea witli tlie most eJ(fe{fent of liea{tli anajoyfu{ liearts. 'Ilie quire tliat one cfutclies is aeteriorating at a most arsagreea6{e rate, ant! so one 6egs of you:

'You are a guest of nature. cJ3eliave.

:Most graciousfy yours, :Ms 1(ofeigline Pora

(7)

Figure 3.1 Cross section of a large soil monolith lysimeter 20 Figure 3.2 Lysimeter facility layout and treatment allocation 25 Figure 4.1 Water inputs and drainage - 25 mm treatments 39 Figure 4.2 Water inputs and drainage - 50 mm treatments 39 Figure 4.3 Mineral nitrogen concentration of drainage water from "cut and carry"

treatments 44

Figure 4.4 Mineral nitrogen concentration of drainage water from "grazed"

treatments 45

Figure 4.5 Cumulative nitrogen leaching losses from "cut and carry" treatments 48 Figure 4.6 Cumulative nitrogen leaching losses from "grazed" treatments 49 Figure 4.7 Nitrate nitrogen concentration of drainage water from "cut and carry"

treatments 51

Figure 4.8 Nitrate nitrogen concentration of drainage water from "grazed"

treatments 52

Figure 5.1 Daily nitrous oxide flux (Winter) 58

Figure 5.2 Daily nitrous oxide flux (Spring) 59

Figure 5.3 Daily nitrous oxide flux (Summer) 60

Figure 5.4 Daily nitrous oxide flux (Autumn) 61

Figure 5.5 Daily total nitrogen flux (Winter) 63

Figure 5.6 Daily total nitrogen flux (Spring) 63

Figure 5.7 Daily total nitrogen flux (Summer) 64

Figure 5.8 Daily total nitrogen flux (Autumn) 64

Figure 6.1 Pasture production during the 2 years of the LD-DFE experiment 68 Figure 6.2a Nitrogen content of pasture (%) during the experiment - 25 mm

treatments 71

Figure 6.2b Nitrogen content of pasture (%) during the experiment - 50 mm

treatments 71

(12)

Chapter 1: General Introduction

This study was undertaken in order to measure and compare the environmental effects of lactose-depleted dairy factory effluent (LD-DFE) irrigated onto land, with the results of a previous study that determined the fate of nitrogen from lactose-rich dairy factory effluent (DFE) and dairy cow urine in land treatment systems (Reijnen, 2002).

Developments in the industrial processing of milk and its by-products are ongoing, and as a result, waste streams from such processing are continually changing.

One of these changes is the recovery of lactose from the waste stream at some milk processing factories. The removal of organic carbon from the waste may have an impact on the fate of nitrogen in the effluent when applied to land. As yet, there has been no published data on the fate of nitrogen in LD-DFE irrigated onto pastoral land. Hence, a detailed research programme is required in order to provide reliable information to support environmental management decisions. Because land treatment areas are often used for grazing stock, or the mechanical conservation of pasture (e.g. hay, baleage, silage), it is also necessary to examine the effect of animal urine returns.

The following hypothesis was developed and tested: "That the irrigation of LD- DFE onto grazed pasture soil is a sustainable method of treating dairy factory waste water, which does not increase nitrate leaching, or nitrous oxide emissions."

Consequently, the objectives of this investigation were:

1. To determine the fate of nitrogen in LD-DFE irrigated onto land, by measuring the amount of nitrogen lost from the soil via leaching, lost by denitrification, removed by the pasture plants, and immobilised within the soil organic fraction.

2. To compare the fate of nitrogen in LD-DFE irrigated onto land under a "cut and carry" system (no cow urine returns), as opposed to a "grazed" pasture system (with cow urine returns).

The following chapters outline the aims of each component of the research, and how they contribute to the testing of the hypothesis.

A review of the literature (Chapter 2) was undertaken to highlight the possible pathways and transformations that the applied nitrogen may take as a result of the removal of lactose from the waste stream.

Chapter 3 details the materials and methods used throughout the course of the lysimeter experiment, and the laboratory techniques adopted during sample analyses.

(13)

The possibility that nitrogen, either from the raw product or its constituents, will leach into the groundwater is of greatest environmental concern following the land application of LD-DFE. Therefore, Chapter 4 examines the effect of LD-DFE and cow urine on nitrogen leaching.

In New Zealand, the soil processes of nitrification and denitrification are generally accepted as the main sources of nitrous oxide (de Klein et aI., 2001).

However, it is unknown how the chemical composition of LD-DFE will affect the rate of nitrous oxide emissions. Therefore, Chapter 5 explores the effect of LD-DFE and cow urine on denitrification.

Plant growth on a land-treatment site is a key feature of the system as it removes nutrients and water from the waste-treated soil, as well as protecting the soil from damage during waste application (Cameron et aI., 1997). Chapter 6 investigates the effect ofLD-DFE and cow urine on pasture production and pasture nitrogen harvested.

In the previous DFE study (Reijnen, 2002), a large amount of the applied nitrogen was immobilised into the soil organic fraction. Chapter 7 describes the effect of LD-DFE on the immobilisation of nitrogen in the soil system.

Results from Chapters 4-7 enable the production of a full nitrogen budget (Chapter 8) for each treatment following the land application of LD-DFE. As in the earlier DFE experiment, many measurements were extrapolated to calculate annual losses.

Finally, Chapter 9 details the mam conclusions drawn from this LD-DFE lysimeter study, and suggests possible directions for future research.

(14)

Chapter 2: Literature Review

2.1 Introduction

While there is extensive literature on the cycling of nitrogen in the soil-plant- atmosphere system (e.g. Haynes, 1986; Stevenson and Cole, 1999), until recently there has been a lack of both knowledge and understanding on the fate of nitrogen from dairy factory effluent (DFE) in land treatment systems. A previous study on the fate of nitrogen from DFE and dairy cow urine (Cameron et aI., 2002; Reijnen, 2002), quantified the fate of a large proportion of the applied nitrogen. However, as lactose (organic carbon) is now removed from the DFE waste stream during the processing of milk, there is a new gap in the literature regarding the environmental impacts of applying DFE with a low lactose content.

The following review of the literature briefly describes the size of the New Zealand dairy industry in order to emphasize the importance of this issue; and describes the composition of DFE. The cycling and storage processes of soil nitrogen are then reviewed with the intention of highlighting the possible fate of land-applied lactose- depleted dairy factory effluent (LD-DFE) nitrogen. Similarly, losses of soil nitrogen are described in order to emphasize the possible environmental effects of applying such waste to pastoral land.

2.2 The New Zealand Dairy Industry

The New Zealand dairy industry processed over 14.6 billion litres of milk in the 2003/04 season (up from 13.9 billion in the 2002/03 season) from 3,851,302 cows in 12,751 herds (http://www.stats.govt.nz. 2004). Around 25 dairy factories throughout the country processed more than 1.25 billion kilograms of milk solids into products destined predominantly for export.

Recently, major investment in high-value lactose extraction has been made at several dairy factories in New Zealand. The lactose plant at Clandeboye dairy factory, near Timaru, in the South Island of New Zealand, has been operational since September 2001, and this has resulted in a waste stream that is now depleted in carbon compared to prevIOUS years.

(15)

2.3 Composition of Dairy Factory Effluent

While published literature on the composition of DFE is limited (Reijnen, 2002), the wide range of chemical characteristics reported is due to the nature of the different products undergoing manufacture. Typical characteristics of DFE produced at New Zealand dairy factories are covered comprehensively in Reijnen (2002). Table 2.1 details the chemical analysis of lactose-rich DFE.

Table 2.1 Chemical analysis ofDFE (from Cameron et ai, 2002).

Analysis pH

N03-N (mg N L-¹⁾ NH4-N (mg N L-1)

Total N (mg N L-¹⁾ Organic C (mg C L-¹)*

C:N ratio*

Na+ (mg Na L-¹⁾

* Measured on a limited number of samples

Range 4.1-11.9 0.0 - 37.8 0.0 - 67.9 20.0 -262.0 4329.0 - 4782.0

24.4 - 32.1 20.0 -161.0

Average 7.0 6.7 24.0 158.0 4555.5 28.3 85.3

LD-DFE comprises milk, wash water, residual product (from rinsing), process losses, lubricants of the processing equipment; both acid and alkali wash products, and dust and boiler ash. Characteristics of the waste change over time, and are determined largely by the type of product being manufactured at any given time. However, any type ofDFE has generally been described as being high in both organic and nutrient content, of varying pH, and affected by processing and cleaning operations such as water recycling and storm water diversion (Barnett et aI., 1994).

Of particular importance to this study however, is the average organic carbon content of the DFE shown above in Table 2.1 (4556 mg C L-^1);compared to that measured in the LD-DFE (Appendix A) at 1508 mg C L-^1.Further, the average C:N ratio of DFE was 28.3 (Cameron et aI., 2002), whereas during the two years of LD- DFE effluent application it averaged only 9.9 (Appendix A).

2.4 Soil Nitrogen Cycling and Storage Processes

Biological transformations such as the mineralisation or immobilisation of soil nitrogen determine the forms of nitrogen present in the soil. Also, the process of denitrification leads to the formation of nitrogen gases that are lost into the atmosphere.

The form of nitrogen in the soil at the centre of this study is nitrate and its concentration is influenced by these three processes. Each of these processes is extremely difficult to describe mathematically because they are spasmodic - depending on edaphic conditions

(16)

which in turn drive soil microbial activity - rather than steady state (Cameron and Haynes, 1986). Figure 2.1 below, illustrates how the different soil nitrogen cycling and storage processes are interrelated.

GAINS

Nitrogen Fertilizers

CLAY MINERAL FIXATION

. LOSSES

~EMO~ABLE

Legume N Animal Animal . Piant Gaseous Fixation' manure liptake uptake Loss~s

LEACHING

~

N2. N20 NO,NHa

Figure 2.1 The nitrogen cycle in pastoral systems (from MCLaren and Cameron, 1996)

(17)

2.4.1 Mineralisation and Immobilisation

Nitrogen mineralisation is the conversion of organic nitrogen into forms of mineral nitrogen that are more susceptible to leaching from the soil profile, if not incorporated into microbial tissue, or taken up by vegetation (Stevenson, 1982b). It is a key process which affects the amount of soil nitrogen made available for plant uptake or loss via the processes of ammonia volatilisation, nitrate leaching, or denitrification (Monaghan and Barraclough, 1997).

Soil nitrogen immobilisation refers to the process where mineral nitrogen is taken up by, and incorporated into, the bodies of the soil microbial population. Any actively growing microbe in the soil contributes to this incorporation of soil ammonium and nitrate into microbial biomass (Tate, 2000). If conditions favour microbial growth, nitrogen is immobilised. Theoretically, the balance between the two processes can fluctuate depending on the soil and environmental conditions, and both the quantity and the quality of the available organic substrate. The term "immobilisation" may also be used in situations where there is a long-term accumulation of soil organic nitrogen. An example would be that of an undisturbed grassland soil with returns of plant shoot and root residues to the soil system. The processes of mineralisation and immobilisation usually occur simultaneously in the soil.

Many factors affect the rate at which nitrogen mineralisation andlor immobilisation occur in grazed pasture soils (Haynes, 1986). A change from net nitrogen immobilisation to net nitrogen mineralisation occurs during decomposition as substrate quality changes, or the C:N ratio becomes progressively smaller. This is an important consideration when applying liquid wastes to the soil. Different wastes exhibit diverse chemical properties, and their environmental impact will be equally varied. Cameron et aI. (2002) illustrate the difference between pig effluent (C:N ratio of 1:1), and DFE (C:N ration of 28:1), and how this impacts on the amount of nitrogen leached from the soil. The critical C:N ratio above which nitrogen immobilisation occurs is usually about 20: 1 (MCLaren and Cameron, 1996).

Environmental parameters such as aeration, soil temperature and moisture, fertiliser nitrogen, soil texture, soil pH, and the abundance of heavy metals andlor pesticides also play an important role (Haynes, 1986). Cycles of wetting and drying, and freezing and thawing appear to be of particular importance as they cause a "flush" of microbial activity, and consequently nitrogen mineralisation. Cultivation affects also determine the degree of nitrogen immobilisation in the soil (Jarvis et aI., 1996).

(18)

Reijnen (2002) utilised isotopic studies to examine the immobilisation of nitrogen in the soil following the application of lactose-rich DFE. Results showed that between 32.3 and 38.5% of applied DFE-nitrogen was immobilised in "cut and carry"

treatments. This was attributed to the high soluble carbon (lactose) content of the DFE, and was regarded as being responsible for lower than expected nitrate leaching losses following DFE application to land (Reijnen, 2002).

2.4.2 Nitrification

Nitrification is the biological oxidation of ammonium to nitrite, and thence to nitrate (Haynes, 1986), and is illustrated below in Equation 2.1 :

Equation 2.1 This initial reaction is carried out mainly by the Nitrosomona bacteria, although other lesser known bacteria are also capable of completing the oxidation process.

The Nitrobacter genus of bacteria oxidise the nitrite formed in the initial stage of nitrification (above), to nitrate (Equation 2.2):

2.2 The oxidation of nitrite is more rapid than that of ammonium, so only rarely are there more than trace amounts of nitrite present in the soil (Schmidt, 1982). The rate at which nitrification occurs is sensitive to various soil conditions. These include soil pH, soil moisture content and aeration, soil temperature, nutrient status and fertiliser applications, vegetation, and the application of pesticides (MCLaren and Cameron, 1996).

2.4.3 Ammonium Adsorption

Ammonium ions (NH4 +) can be adsorbed by cation exchange reactions onto the surface of clays and organic matter in the soil (MCLaren and Cameron, 1996). Clays and organic matter have a predominantly negative charge, and are able to attract and hold positively charged cations (Cameron and Haynes, 1986). Cation exchange reactions hold NH4 + ions by electrostatic attraction and the NH4 + ions remain in an exchangeable form. These reactions protect the NH4 + ions from leaching and retain NH4 + in the soil for plant uptake. Leaching losses of NH4 ⁺therefore, are only likely to be a problem in

(19)

soils with an extremely low cation exchange capacity (CEC) (Cameron and Haynes, 1986).

2.4.4 Ammonium Fixation

The pool of ammonium that is held within the lattice of some 2: 1 type clay minerals (and therefore not readily exchangeable with other cations) is considered to be

"fixed" (Cameron and Haynes, 1986). It cannot be removed by leaching, and is generally considered to be unavailable to both micro-organisms, and plants (Stevenson, 1982a). Such clays include illite and vermiculite. They have negatively charged internal surfaces, and can expand and contract. During expansion of the clay, ammonium is attracted to these surfaces. They become "fixed" when the clay contracts. This is one process by which plant-available nitrogen can be removed from soil solution.

2.4.5 Plant Uptake

The objective of land application of wastes is to utilise the soil/plant system to assimilate the waste components and thus reduce the risk of releasing nutrients into water or air (Cameron et aI., 1997). Plant uptake of nutrients represents a very important part of this concept and is responsible for the removal of large quantities of nutrients in a land treatment system.

Nitrogen uptake by a growing crop provides a sink for the nitrate and ammonium present in soil solution. Therefore, plant uptake reduces the potential for nitrate leaching during the growing season. Reijnen (2002) reported that plant uptake of nitrogen following the application ofDFE ranged from 150-375 kg N ha-^lyr-l .

2.4.6 Nitrogen Returns via Grazing Stock and Plant Residues

Grazing livestock return nitrogen to the soil via the excretion of both dung and urine. The amount will depend on the type of animal, the kind of herbage consumed, and total dry matter intake (Whitehead, 1986). In an intensive grazed pasture system, it was calculated that more than half of the consumed nitrogen was excreted as urine (Haynes and Williams, 1993). The area over which a typical urination event occurs may receive up to the equivalent of 1000 kg N ha-^l(Ball and Ryden, 1984; Cameron, 1993), and cover an area of between 0.16 - 0.49 m²(Haynes and Williams, 1993). Such a high loading rate of nitrogen within a urine patch may potentially lead to large nitrogen losses from the grazed system via nitrogen leaching, denitrification, and ammonia

(20)

volatilisation (Ball et aI., 1979). Reijnen (2002) reported nitrogen leaching, denitrification, and volatilisation losses that were higher in "grazed" than "cut and carry" treatments following the application ofDFE to land.

The amount of nitrogen returned to the soil after the breakdown of plant residues will depend largely on the plant species grown in the system, and management practices. The importance of plant residue decomposition within a grassland nitrogen cycle is highlighted by (Floate, 1987).

2.5 Losses of Soil Nitrogen

The leaching of nitrate is a particular problem in cultivated agricultural ecosystems, where it is often the most important means of nitrogen loss from field soils (Cameron and Haynes, 1986). Several mechanisms lead to gaseous losses of nitrogen from the soil system. These include ammonia volatilisation, biological denitrification, and chemodenitrification. Both crop and animal product removal also lead to nitrogen loss from the system.

2.5.1 Leaching

Nitrogen is leached mainly as nitrate, largely because nitrate ions are highly soluble and are not retained by the soil's exchange complex (Hillel, 1998). Nitrate leaching occurs when there is an accumulation of nitrate in the soil profile that coincides with, or is followed by a period of high drainage (Di and Cameron, 2002). Factors such as rainfall, evaporation, soil type, and plant cover all contribute to the degree to which leaching may occur. Excessive application of waste or effluent that is high in nitrogen, and nitrogen returns in animal urine can therefore have a major impact on nitrate leaching from grazed pastures.

The leaching of nitrate from the soil system has a wide range of consequences environmentally, economically, and on the health of both humans and livestock.

Nitrogen - particularly dissolved organic nitrogen - can increase enrichment (eutrophication) in surface water bodies (e.g. rivers, lakes, or estuaries). This can lead to the proliferation of algae, oxygen depletion, pH variability, and changes in plant species quality with subsequent food-chain effects (MFE, 2001). In New Zealand, agricultural non-point sources account for 75% of the total nitrogen loading to surface waters. The principal sources of high nitrogen levels on farmland are urine and dung from grazing

(21)

livestock, the application of nitrogenous fertilisers and waste, and nitrogen fixation by clovers (MFE, 2001).

Leaching losses of nitrogen occur on both a nationwide scale, and at the individual farm level. Such losses are of concern as they result in an economic loss of nitrogen from the soil and can lead to contamination of the groundwater (Williams et aI., 2000).

A high concentration of nitrate in drinking water is considered harmful to human health, particularly to infants less than one year of age. It can interfere with oxygen transport in the blood, and may result in methemoglobinemia (blue-baby syndrome). In an attempt to protect human health, national and world health organisations have established drinking water standards that limit nitrate concentration to a maximum of 10-11.3 mg N03-N L-¹ (WHO, 2004). Elevated concentrations of nitrate in drinking water are also toxic to livestock, and can cause methemoglobinemia and abortions in cattle (Di and Cameron, 2002). Concentrations of between 40 and 100 mg N03-1-N L-¹ in stock drinking water are considered unsafe.

2.5.1.1 The Process of Nitrate Leaching

If steady-state water conditions exist in a homogenous non-aggregated soil and no interaction occurs between nitrate ions and the soil, then the process of nitrate movement can be described by three processes: convection, diffusion, and dispersion (Cameron and Haynes, 1986).

a) Convection

Convection refers to the movement of nitrate due to the mass flow of water. It can be described by a modified form of Darcy's Law:

J

c

= qc = -c [K dHldx]

2.3

where Jc = convective nitrate flux (g S-l); c = nitrate concentration (g m-3); q = water flux; K = hydraulic conductivity; and dH/dx = hydraulic gradient.

The water and nitrate move down the soil profile in response to a hydraulic gradient. The rate of movement is dependent on the magnitude of the hydraulic gradient, and the hydraulic conductivity of the soil (Cameron and Haynes, 1986).

The distance that a band of nitrate is transported per unit time depends on the average pore water velocity (U) where:

(22)

U

⁼

q/8

2.4

and 9 is the volumetric water content.

b) Diffusion

When there is an uneven distribution of nitrate in the soil solution there is a diffusive flux of nitrate from areas of high concentration to areas of low concentration (Cameron and Haynes, 1986), as described by Fick's Law:

Jd =

-Ds (8) dc/dx 2.5

where Jd

=

diffusive flux of nitrate; Ds

=

diffusion coefficient (depends on 9) and for nitrate in soil at -1.0 kPa this is approximately 10-⁶cm^{2 S-l,}or 0.5 cm d-¹(Cameron and Haynes, 1986); and dc/dx = nitrate concentration gradient.

Diffusion therefore, causes a spreading out, or equalisation of nitrate down the soil profile.

c) Hydrodynamic Dispersion

The mechanical action of a solution flowing through the soil causes mixing, and tends to equalise the nitrate distribution by a process commonly termed "hydrodynamic dispersion". The effect enhances nitrate spread due to diffusion, and often completely masks it. Hydrodynamic dispersion occurs because:

i) There is a large variation in pore sizes, and this causes an extremely wide range of pore water velocities.

ii) Flow velocity within a single pore is not uniform. It is faster at the centre of the pore, and slower closer to the walls of the pore (Figure 2.2), due to frictional drag.

_ _ _ _ _ _ _ • slow flow

-=d~ir=ec=t~io=n~o~fs=o~l=u=te~fi~o~w~ _______ ~ fast flow

Figure 2.2 Flow velocity gradients within a soil pore.

t

pore walls

(23)

iii) Tortuosity of the soil pore geometry results in a fluctuation of pore path length (Figure 2.3).

Figure 2.3 An illustration of tortuosity where the green area indicates solid particles, white indicates pore space, the red line illustrates the actual flow path, and the black arrow shows the overall direction of flow.

d) Combined Solute Flux

The combined effects of the convective-diffusive-dispersive transport mechanism can be described by the convective-dispersive equation (CDE) (Cameron and Haynes, 1986):

3c/3t 2.6

where Da = apparent diffusion coefficient; and is the sum of diffusion and dispersion, and:

Ds + mU

2.7

where m is the dispersivity of the soil. Thus, the value of Da depends on the flow velocity, and tends to increase with increasing values of U (Haynes, 1986).

The CDE provides a succinct description of nitrate transport through the soil, and describes the rate of change in nitrate concentration at any given point in the profile.

The CDE accounts for transient-state conditions, in which both fluxes and concentrations can vary in time and space. The various mechanisms of nitrate leaching are illustrated overleaf in Figure 2.4.

(24)

::z:

...

_Q.

....

o

....I

o (J)

SOIL NITRATE CONCENTRATION )

CoJv~ctive

tra\'s~ort a'oi_nel

I ,

.. ::: .. ^~:^....

r

I

Convect ion- diffusion-

Figure 2.4 The various components of nitrate leaching: a) convection, b) convection-diffusion- dispersion, c) anion exclusion, d) anion adsorption, and e) macropore bypass and macropore leaching. (From Cameron and Haynes, 1986).

2.5.1.2 Factors That Affect Nitrate Leaching

Two of the main factors controlling nitrate leaching from the soil profile are the quantity of water passing through the soil profile, and the concentration of nitrate in the soil profile at the time of leaching (Haynes, 1986). Season and climate, soil factors such as macropore effects, soil reaction, soil variability, biologically-driven transformations of nitrogen (immobilisation and mineralisation), and plant uptake also affect the amount of nitrate leaching. Land management, irrigation, and the application of organic wastes may also contribute to increased leaching losses of nitrogen (Di and Cameron, 2002).

Much recent work has focused on nutrient losses in both drainage water, and surface runoff from grazed pastures (e.g. Monaghan et aI., 2000), and following the application to land of farm dairy-shed effluent (Monaghan and Smith, 2004). While these studies highlight key on-farm management strategies that can be implemented in order to reduce the transfer of nutrients and faecal organisms from soil to waterways, as yet

(25)

there is no information reported on losses of nitrogen following the application to land ofLD-DFE.

2.5.1.3 Nitrate Leaching from DFE

Research results reported by Cameron et al (2002) and Reijnen (2002) show that following land application ofDFE, leaching losses of nitrogen ranged from 12.9 - 94.0 kg N ha-1yr-1 depending on the system (Table 2.2).

Table 2.2 Annual nitrogen leaching losses (kg N ha-¹yr-¹⁾(From Cameron et aI., 2002) Treatment Equivalent NappI. rate Yr1

(kg N ha-1 yfl)

Yr2 Yr3 Average

DFE25 300 5.5 10.0 25.2 13.6

DFE50 600 7.4 4.6 5.6 5.9

DFE 25 + Urine* 1300 65.1 123.3 93.8 94.0

DFE 50 + Urine* 1600 69.3 48.4 75.8 64.5

Control 25 mm 0 4.7 7.0 27.1 12.9

Control 50 mm 0 4.1 5.1 3.4 4.2

* Calculated as a paddock loss where 25% of the treatment area receives urine each year.

Table 2.2 shows that neither of the DFE treatments (DFE applied as 25 mm depth of solution; DFE applied as 50 mm depth of solution) caused a significant increase in the amount of nitrate leached compared to the Controls (25 mm or 50 mm water) despite the large amounts of N applied (Cameron et aI., 2002). Also, Table 2.2 illustrates that the "cut and carry" system (DFE 25 and DFE 50) proved to be a more efficient land treatment system than the "grazed" system involving animal urine inputs.

The DFE 50 treatment (either alone, or with urine inputs) did not cause a significant increase in the amount of nitrate leached, compared with the DFE 25 treatment.

Much less nitrate was leached from DFE applied to land than was expected at the rates of nitrogen applied. Over the 3-year period, leaching accounted for less than 1 % of the nitrogen applied. Current knowledge advocates that the amount of nitrate lost from soil treated with organic wastes and fertiliser generally increases with increasing rates of nitrogen application (Cameron et aI., 1997; Di and Cameron, 2002). Therefore, it was hypothesised that the unexpected results in this DFE experiment were caused by the high carbon (lactose) content of the effluent (Cameron et aI., 2002). It was suggested that the readily available carbon in the lactose stimulated the soil microbial pool and as a result, the amount of nitrogen immobilised or denitrified in the soil was increased.

(26)

2.5.2 Ammonia Volatilisation

Ammonia (NH3) gas can be lost from the soil and return to the atmosphere (Figure 2.1). This is referred to as ammonia volatilisation. Ammonia is a very soluble gas, and undergoes base hydrolysis in water as illustrated below in Equation 2.8:

2.8 The rate at which volatilisation occurs depends on the dispersion of ammonia in the soil atmosphere close to the surface, and the concentration of both ammonia and ammonium in soil solution. Reijnen (2002) reported very small ammonia volatilisation losses following the land treatment of DFE. It accounted for less than 1 % of the nitrogen applied in DFE treatments, and less than 7% in urine treatments.

2.5.3 Denitrification

2.5.3.1 Biological Denitrification

One of the biological processes involving the reduction of oxidised forms of nitrogen (i.e. nitrate and nitrite) is dissimilatory nitrate reduction - more commonly termed denitrification (Stevenson and Cole, 1999). Gaseous losses of nitrogen can occur through the process of denitrification, and of those generated; nitrous oxide is the· most important. Besides being a greenhouse gas, nitrous oxide also plays another important role in the atmosphere since it is further oxidized into NOx in the stratosphere, where it acts to deplete ozone (Hillel, 1998). While nitrous oxide is present in the atmosphere at much smaller concentrations than carbon dioxide (C02), or methane (CH4), its radiative forcing is about 300 times greater than carbon dioxide and its mean lifetime in the atmosphere is about 120 years (Hillel, 1998). Soils are the major medium on planet Earth that generate nitrous oxide. Equation 2.9 illustrates how the production of nitrous oxide in soil is an intermediate stage in the pathway of nitrogen oxide reduction during denitrification:

N0

3- -+

N02-

^-+

[NO]

-+

N20

^-+

N2

2.9 nitrate nitrite nitric oxide nitrous oxide dinitrogen

gas gas gas

On a global scale, agricultural activities that contribute to nitrous oxide production include the use of nitrogenous fertilisers, animal urine returns to the soil, the

(27)

burning of fossil fuels and biomass, and the conversion of forests to pasture in tropical regIOns.

For denitrification to occur in soil, the essential requirements are the presence of nitrate (or other nitrogen oxides) to fill the role of terminal electron acceptors, the presence of the heterotrophic denitrifying bacteria which require organic carbon compounds as electron donors and as a source of cellular material, and anaerobic or nearly anaerobic conditions (Smith, 1990). Biological denitrification therefore, can be limited by a lack of readily available total soil organic carbon. Thus, rates of denitrification are highly correlated with available soil carbon as evaluated by extracting sugars, by water-soluble organic carbon, or by readily mineralisable carbon (Cameron and Haynes, 1986). Figure 2.5 shows the linear relationship between water-soluble carbon and the denitrification capacity of 17 soils.

,...

0 N 400

z

I/)

::::J

a.

z eN ell

,..

i

300

:>

'0 _~.

•

Z 1:1)

-

=t 200

UJ (I)

0 ...I

Z 0

~

¹⁰⁰

-

u ^LI.

-

^a:

!::

z w Q

o .

0 100 200 300

WATER-SOLUBLE ORGANIC 'C~'RBON (.ugC 9-' )

Figure 2.5 The relationship between denitrification capacity and water-soluble organic carbon in 17 soils (Burford and Bremner, 1975).

(28)

Denitrification is also affected by a number of primary (proximal) soil factors such as soil oxygen content, temperature, mineral nitrogen content, and pH (de Klein et aI., 2001). Factors that affect and increase the levels of available carbon in soils (e.g.

cycles of drying and wetting, and freezing and thawing) have also been shown to increase the capacity of soils to denitrify added nitrate (Cameron and Haynes, 1986).

Significant amounts of denitrification have been reported to occur following the application ofDFE to soil (Cameron et aI., 2002), with most denitrification taking place within five days of the effluent application. The amount of nitrogen lost as both nitrous oxide and dinitrogen varied seasonably with the amount lost during spring being higher than that recorded in both summer and autumn (Table 2.3). Between 10 and 30% of the denitrified nitrogen was lost as nitrous oxide, with the remainder as dinitrogen. The amount of nitrous oxide lost by denitrification (the 'emission factor') for DFE was higher than that recorded from any other type of organic waste or fertiliser applied to soil (de Klein et aI., 2001), as was the total denitrification nitrogen loss (c. 5-20% of applied nitrogen).

2.5.3.2 Chemodenitrification

Chemodenitrification is the term commonly used to describe various chemical reactions of nitrite ions within soil that results in the emission of a variety of nitrogenous gases (Haynes and Sherlock, 1986). Stevenson and Cole (1999) describe chemodenitrification as the process by which nitrogen gases are formed in soils by chemical reactions of nitrite with organic matter. These particular soils are not necessarily anaerobic, but they may have had large amounts of ammonium fertiliser added to them (MCLaren and Cameron, 1996). If high levels of ammonium are present in soil, this restricts the activity of Nitrobacter, and an accumulation of nitrite occurs.

Thus, losses of dinitrogen gas occur.

(29)

While few attempts have been made to measure directly the losses of NOx from soils via chemodenitrification, they are estimated to be in the order of <1 kg N ha-1 yr-l (Haynes and Sherlock, 1986), and although they may be of little agronomic importance (MCLaren and Cameron, 1996), they may be of environmental significance.

2.5.4 Crop and Animal Removal

Nitrogen can be removed from agricultural production systems as various products. Nutrient transfer out of the system occurs when "cut and carry" operations such as crop harvesting result in minimal plant residue being returned to the soil system.

Further, when grazing stock play an integral part in a production system, removal of either the animals, or their products, results in off-farm nutrient transfer. Ball and Field (1982) estimated nitrogen loss via product removal from an intensive dairy farm receiving fertiliser inputs of 450 kg N ha-1yr-1 to be 61 kg N ha-1yr-1.

2.6 Conclusion

This literature review highlights the fact that there is no published data on the environmental impacts of applying LD-DFE to land. This thesis will therefore compare nitrogen losses after LD-DFE application with that found after applying lactose-rich DFE (Reijnen, 2002).

(30)

Chapter 3: Materials and Methods

3.1 Introduction

This experiment utilised 24 existing lysimeters at the Lincoln University lysimeter facility. Twenty two of these large soil monolith lysimeters (500 mm diameter, 700 mm deep) were used previously by Reijnen (2002). By following this earlier study, replication of the waste disposal regime undertaken at the Clandeboye dairy factory (Temuka, South Canterbury) could be achieved. That is, three years of lactose-rich DFE application had previously occurred, followed now by the addition ofLD-DFE.

Lysimetry was used because it allows accurate quantification of nitrogen removals or losses by leaching, denitrification, immobilisation, and plant uptake following the application of organic waste to land.

3.2 Lysimeters

The collection and installation procedure used for these particular lysimeters has been detailed previously (Reijnen, 2002), and was based on the method described by Cameron et aI. (1992). Briefly, a metal cylinder casing was placed on the soil surface, and this was carefully dug around in order to minimise disturbance to the internal soil structure. The casing was gradually pushed down in small increments, until the desired depth of 700 mm was reached. The base of the soil monolith was cut with a cutting plate which was secured onto the lysimeter casing and lifted from the collection site. The lysimeters were then transported to the field trench lysimeter facility at Lincoln University using a specially designed trailer with enhanced suspension. As such, disturbance to the soil structure was minimised. The gap between the metal casing and the soil core was sealed using petrolatum. This eliminated the effect of edge-flow (Cameron et aI., 1992). A layer of soil at the base of the lysimeter (50 mm) was replaced with washed gravel to ensure a free draining soil monolith. Not only did this negate the requirement for a tension drainage system, but it replicated the presence of gravel at the base of many typical sedimentary soils throughout the Canterbury Plains. Finally, the lysimeters were installed in the trench facility. The surface of each lysimeter was positioned level with that of the surrounding soil surface in order to maintain typical pasture growing conditions.

(31)

I+-Soll cover (gas sampling)

Treaehwall

Leachate Gravel (10

mml'~

. / ooWonport

Disposable leachate

--+--w

eo1Iectionbag .

Figure 3.1 Cross section of a large soil monolith lysimeter. Not to scale (from Reijnen, 2002).

20

(32)

The soil type has been described as a Templeton fine sandy loam (Immature Pallic Soil, Hewitt, 1998; Udic Ustrochrept, USDA, 1998). Pasture cover was a mixture of ryegrass (Lolium perenne) and white clover (Trifolium repens). The physical and chemical properties of this particular soil are summarised in Table 3.1.

Between the completion of the previous DFE experiment (Reijnen, 2002) and the commencement of this study (i.e. December 2000 - October 2002), the lysimeters received natural rainfall, and the pasture was cut and removed as required to simulate grazing or forage harvesting.

3.2.1 Lysimeter Maintenance

At the beginning of the study, any pre-existing bare patches of soil were re-sown by hand with a mixture of ryegrass (Lolium perenne) and white clover (Trifolium repens). In addition, hand weeding was undertaken to remove any weeds that were present in the herbage cover. Weeding continued as required throughout the course of the experiment.

Because the soil monolith was prone to shrinkage during the dry summer months, each lysimeter casing was closely inspected on an annual basis in early November. Any gaps were filled with warm liquefied petrolatum (c. 40°C) which, upon cooling, sealed the edge of each monolith to its casing, hence ensuring that "edge-flow"

was eliminated.

Diazinon was applied annually in January (30 g m-²⁾ in order to protect the pasture roots from grass grub (Costelytra zealandica) larvae attack. Lime was applied in August 2003 at 2 t ha-¹in order to maintain a typical soil pH for a pasture soil (PH of 5.8).

(33)

Table 3.1 Physical and chemical properties of the Templeton fine sandy loam (Udic Ustrochept) (from Reijnen, 2002). Figures in brackets indicate ± one S.E.M.

Soil Property Units 0-50mm 50-100 mm 100 - 200 mm 200 - 400 mm 400 - 600 mm 600 - 800 mm Physical

Bulk Density g cm-³ 1.27 (± 0.03) 1.32 (± 0.03) 1.45 (± 0.02) 1.50 (± 0.02) 1.50 (± 0.01) 1.46 (± 0.02) Particle Density g cm-³ 2.53 (± 0.05) 2.56 (± 0.03) 2.57 (± 0.02) 2.63 (± 0.02) 2.67 (± 0.03) 2.66 (± 0.03) Porosity % 0.50 (± 0.08) 0.52 (± 0.06) 0.56 (± 0.04) 0.57 (± 0.04) 0.56 (± 0.04) 0.55 (± 0.05) Chemical

pH 5.8 (± 0.1) 5.8 (± 0.1) 5.7 (± 0.1) 5.5 (± 0.1) 5.8 (± 0.1) 6.1 (± 0.1)

Cation Exchange Capacity cmole kg-¹ 28.4 (± 1.1) 22.1 (± 1.2) 20.3 (± 1.0) 11.2 (± 0.7) 8.5 (± 0.3) 6.5 (± 0.3) Sodium Absorption Ratio ratio 9.65 (± 0.02) 1.02 (± 0.02) 1.11 (± 0.03) 1.73 (± 0.02) 2.52 (± 0.04) 2.67(± 0.03) Exchangeable Sodium % 1.23 (± 0.03) 1.48 (± 0.05) 1.64 (± 0.04) 2.74 (± 0.03) 3.64 (± 0.06) 3.87 (± 0.07) Total Nitrogen g N kg-¹ 2.51 (± 0.1) 2.45 (± 0.1) 2.55 (± 0.1) NID N/D NID

Mineral Nitrogen mgN kg-¹ 8.3 (± 1.3) 8.2 (± 1.1) 7.6 (± 1.4) N/D N/D NID

Organic Nitrogen gNki¹ 2.50 (± 0.1) 2.44 (± 0.1) 2.54 (± 0.1) NID NID N/D

Mineralisable Nitrogen mgN kg-¹ 11.3 (± 1.1) 11.1 (± 1.0) 9.5 (± 1.4) NID NID NID

Total Carbon % 2.78 (± 0.08) 2.68 (± 0.07) 2.41 (± 0.08) NID NID NID

Microbial Biomass Nitrogen mgN kg-¹ 83 (± 7) 76 (± 5) N/D N/D NID N/D

22

(34)

3.3 Treatments

3.3.1 Lactose depleted-DFE (LD-DFE)

Lactose-depleted dairy factory effluent was collected from the Clandeboye dairy factory every three weeks during the milk processing season. Collection involved effluent samples being taken from the waste pipeline over a 24 hr period. These samples were bulked, and the resultant c. 200 litres was couriered overnight to the lysimeter facility at Lincoln University. Immediately prior to the application of LD-DFE, a sub- sample of the effluent was taken for chemical analysis. Total N was determined by Kjeldahl digestion (Blakemore et aI., 1987). Mineral N (NH/, N03-, and N02-) was determined by flow injection analysis (FIA) (ALPKEM FS3000; O-I-Analytical, USA).

Inorganic C and organic C were measured on a TOC 5000A Analyser (Shimadzu, Australia) in order to calculate total C. Cations (Na+, K+, Mg2+, Ca2+), and anions (Cr, Br-,

pol-,

SO/-) were determined by ion exchange chromatograph (DX-120; Dionex Corp., CA, USA). Also, the pH of the LD-DFE was measured (HANNA HI 9025C;

Hanna Instruments, Portugal).

Determination of the nitrogen content of the LD-DFE enabled the accurate application of effluent to the lysimeters at the three rates detailed below in Table 3.2 (equivalent nitrogen loading of 0, 300, and 600 kg N ha-¹yfl). Lysimeter LD-DFE treatments involved a single application of 25 mm (LD-DFE 25), or 50 mm (LD-DFE 50) equivalent depth, every 21 days during the milking season starting in October, and finishing in either March or April when the nitrogen loading targets had been reached (see Appendix A). These application rates were chosen to replicate the previous DFE study undertaken by Reijnen (2002), and were also similar to the rates used in the land treatment system at the Clandeboye dairy factory. Each rate of LD-DFE was applied in a single application on the same day, using a watering can.

(35)

Table 3.2 Treatment applications

Treatment LD-DFE, Urine Replicates

or water (kg N ha-1)

1 Control25 25 mm water 2

2 Control 50 50 mm water 2

3 LD-DFE + 15N(10%) 25mmLD-DFE 6

4 LD-DFE + Urine 25 mmLD-DFE 1000 (November) 4

5 LD-DFE + 15N(50%) 25 mmLD-DFE 4

6 LD-DFE + 15N(10%) 50 mmLD-DFE 6

As in the previous study (Reijnen, 2002), the Controllysimeters received either 25 or 50 mm of fresh irrigation water at each effluent application. This was done in order to balance the hydraulic loading between treatments.

3.3.1.1 Experimental Design

The two control treatments (Control 25 and Control 50) were established with two replicates each. Treatments 3 and 6 were replicated six times, with the remaining Treatments (4 and 5) each having four replicates. Layout of the lysimeters in the trench facility, and treatment allocation is illustrated in Figure 3.3.

Lysimeters 1-11 and 13-23 were used previously by Reijnen (2002). Treatments in the current study therefore, were allocated to these lysimeters based largely on parallel treatments applied in the previous experiment.

(36)

N

1. Control 25 13. LD-DFE 25 + U1000

2. Control 50 14. LD-DFE 25 + 15N (10%)

3. LD-DFE 50 + 15N (10%) 15. LD-DFE 25 + 15N (10%)

4. LD-DFE 50 + 15N (10%) 16. LD-DFE 25 + 15N (10%)

5. LD-DFE 25 + U1000 17. LD-DFE 25 + 15N(10%)

6. LD-DFE 50 + 15N(10%) 18. LD-DFE 25 + 15N(10%)

7. LD-DFE 50 + 15N (10%) 19. LD-DFE 25 + U1000

8. LD-DFE 25 + 15N (10%) 20. LD-DFE 25 + U1 000

9. Control 50 21. Control 25

10. LD-DFE 50 + _{15N (10%)} 22. LD-DFE 50 + 15N (10%)

11. LD-DFE 25 + 15N (50%) 23. LD-DFE 25 + 15N (50%)

12. LD-DFE 25 + 15N (50%) 24. LD-DFE 25 + 15N (50%)

Figure 3.2 Lysimeter facility layout, and treatment allocation.

(37)

The fate of nitrogen in lactose-depleted dairy factory effluent irrigated onto land