User manual | Energy and nutrients from horse manure Life-cycle data inventory of horse

No 4

Energy and nutrients from horse manure

Life-cycle data inventory of horse manure management systems in

Gävleborg, Sweden

Jay Hennessy

Ola Eriksson

Research Report No 4 urn:nbn:se:hig:diva-20782

Research Report / University of Gävle

Research Reports are published electronically and are available from http://hig.se/Ext/En/University-of-Gavle/Research/Publications.html

Published by:

Gävle University Press [email protected]

Energy and nutrients from horse manure

Life-cycle data inventory of horse manure management systems in Gävleborg, Sweden

Jay Hennessy and Ola Eriksson

Faculty of Engineering and Sustainable Development

Department of Building, Energy and Environmental Engineering

Interim report 2 from the project

Hästkrafter och hästnäring - hållbara systemlösningar för biogas och biogödsel

Abstract

Management of horse manure is seldom subject to energy recovery. In the current project solutions for energy recovery of horse manure, with a focus on biogas production as the process not only recovers energy but also closes nutrient cycles, are identified and assessed from an environmental point of view. The number of horses in society is increasing.

Today, according to Statistics Sweden, there are more than 360,000 horses in Sweden, of which three-quarters are situated in urban or near-urban environments. With a dry matter content of 40 %, this equates to a quantity of 1,360 tonnes of horse manure per annum and corresponds to an annual biogas production of 641 GWh, which corresponds to almost 40 % of all biogas produced in Sweden in 2013. Although there are some practical limitations on how much of that potential can be exploited, this is still a significant potential for increased use of renewable energy. Collecting manure and anaerobically digesting it achieves three environmental benefits:

1. Emissions from conventional management, where the manure is piled and stored, or spontaneously composted or decomposed, are avoided.

2. Anaerobic digestion of manure produces biogas that can be utilised to generate electricity and/or heat or, after upgrading (purification and pressure increase), as vehicle fuel; thereby emissions from fossil fuels are reduced.

3. Following the process, the resulting digestate can be used in agriculture, thereby replacing chemical fertiliser and providing additional environmental benefits.

The aim of this project is to find a greater breadth of system solutions than previously, solutions that are proven to function technically and be economically feasible. If these systems are translated into practical reality, environmental gains are made, for example, through reduced environmental impact such as reduced eutrophication and reduced use of finite resources.

This report documents a data inventory made for the life-cycle assessment (LCA) of horse manure management systems in the Gävleborg region, Sweden. The overall result is that data are scarce for all parts of the system, from feedstock characteristics to waste treatment methods as well as utilisation of biofertiliser. There are few plants for solid state anaerobic digestion, at least using horse manure as substrate, and little is known about emissions from current manure practise. Moreover, as the number and location of horses are hard to estimate, the forthcoming systems analysis has to be made for a hypothetical amount of horse manure and emissions etc. have to be expressed per ton VS.

Given these uncertainties the systems analysis will just give indicative results. i

Sammanfattning

Hantering av hästgödsel är sällan föremål för energiåtervinning. I det aktuella projektet ska lösningar för energiåtervinning av hästgödsel, med fokus på biogasproduktion eftersom processen inte bara återvinner energi utan också sluter näringsämnenas kretslopp, identifieras och bedömmas ur ett miljöperspektiv. Antalet hästar i samhället ökar.

Idag finns det enligt SCB mer än 360 000 hästar i Sverige, varav tre fjärdedelar återfinns i städer eller stadsnära miljöer. Med en torrsubstanshalt på 40 %, motsvarar detta 1 360 ton hästgödsel per år vilket motsvarar en årlig biogasproduktion om 641 GWh, vilket utgör nästan 40 % av all biogas som producerades i Sverige under 2013. Även om det finns vissa praktiska begränsningar för hur stor del av denna potential som kan utnyttjas så är detta fortfarande en betydande potential för ökad användning av förnybar energi. Genom att samla in hästgödsel och röta det uppnås tre miljömässiga fördelar:

1. Utsläpp från konventionell hantering, där gödseln läggs på hög och lagras, eller genomgår spontan kompostering och bryts ned, undviks.

2. Rötning av gödsel producerar biogas som kan användas för (1) att generera elektricitet och/eller värme eller (2) efter uppgradering (rening och tryckökning) användas som fordonsbränsle. Därigenom kan utsläppen från fossila bränslen minskas.

3. Efter processen kan den biogödseln användas i jordbruket, och därmed ersätta konstgödsel vilket kan ge ytterligare miljöfördelar.

Syftet med detta projekt är att hitta en större bredd av systemlösningar än tidigare, lösningar som visat sig fungera tekniskt och ekonomiskt genomförbara. Om dessa system omsätts i praktisk verklighet kan miljövinster göras, till exempel genom minskad miljöpåverkan såsom minskad övergödning och minskad användning av ändliga resurser.

Denna rapport dokumenterar en datainventering inför en avslutande livscykelanalys

(LCA) av hästgödselhanteringssystem i Gävleborgsregionen i Sverige. Det övergripande resultatet är att data är knappa för alla delar av systemet, från råmaterialets egenskaper till behandlingsmetoder samt utnyttjande av biogödsel. Det finns få anläggningar för torrrötning, åtminstone som använder hästgödsel som substrat, och lite är känt om utsläpp från nuvarande gödselhantering. Dessutom, eftersom antalet och placeringen av hästarna

är svårt att uppskatta, så kommer den avslutande systemanalysen göras för en hypotetisk mängd hästgödsel och utsläpp etc. måste uttryckas per ton VS. Med tanke på dessa osäkerheter kommer systemanalysen endast ge vägledande resultat. ii

Preface

This is the second report out of three from a project funded jointly by Region Gävleborg and the University of Gävle. Previous report in Swedish (Hadin et al., 2015) explains details on the background of the project and qualitatively discusses various issues and aspects of horse manure management and considerations relating to the different technologies available to transform the manure.

The work documented in this report has been carried out during February-June 2015.

Expertise was provided by Åsa Hadin. The report has been peer-reviewed by members of the project reference group and finally by the assessor of the entire project. The third, and last, report from the project (Eriksson et al., 2015) shows results from an environmental assessment of different horse manure treatments based on data documented in this report.

The authors like to thank all people contributing to this report.

Gävle 2015-07-05

Jay Hennessy

Ola Eriksson iii

Glossary, prefixes and chemical symbols

In tables, ‘-‘ means the data was not specified/available anaerobic digestion carbon methane

CHP

dung

IVL

JTI

KTH

LCA

manure

N/A combined heat and power dry matter; see also TS fresh faeces from the animal

NMMO

n-methylmorpholine oxide nitrous oxide

-N

-N ammonia ammonium nitrogen oxidised nitrogen nitrate nitrogen

ORWARE an LCA computational model developed in Matlab

P phosphorous

SLU

SS-AD

VFA

VS fresh matter

IVL Swedish Environmental Research Institute (IVL Svenska Miljöinstitutet)

Swedish Institute of Agricultural and Environmental Engineering

(Institutet för jordbruks- och miljöteknik)

Royal Institute of Technology (Kungliga Tekniska högskolan)

Life-cycle assessment faeces, urine and bedding material collected from animal stall; fresh or after storage nitrogen

‘not applicable’ (data in tables)

Swedish University of Agricultural Sciences (Sveriges lantbruksuniversitet) solid-state anaerobic digestion total solids volatile fatty acid volatile solids iv

Abstract ................................................................................................................................ i

Sammanfattning .................................................................................................................. ii

Preface ................................................................................................................................ iii

Glossary, prefixes and chemical symbols .......................................................................... iv

Table of Contents ................................................................................................................ v

1 Introduction ....................................................................................................................... 1

1.1. Data sources ............................................................................................................ 1

1.2. Management systems considered ........................................................................... 1

2. The LCA method and model ........................................................................................... 4

2.1. The ORWARE model ............................................................................................... 4

2.2. Environmental impact assessment .......................................................................... 4

2.3. General assumptions for modelling.......................................................................... 5

3. Data inventory ................................................................................................................. 6

3.1. Validation .................................................................................................................. 6

3.2. Horse manure ........................................................................................................... 6

3.3. Transport .................................................................................................................. 9

4. Horse manure management systems ........................................................................... 10

4.1. Unmanaged manure

– natural decomposition ....................................................... 10

4.2. Aerobic digestion

– composting ............................................................................. 11

4.3. Incineration ............................................................................................................. 13

4.4. Anaerobic digestion (AD) ....................................................................................... 14

References ........................................................................................................................ 21

Appendix A ........................................................................................................................ 25

ORWARE vectors .......................................................................................................... 25

1 Introduction

This report details a data inventory made for the life cycle assessment (LCA) of horse manure management systems in the Gävleborg region, Sweden. Details of the aims of the project and the results of the initial literature study can be found in the preceding report,

Hadin et al. (2015). The horse manure management systems that have been considered, and for which data have been gathered, are:

1. Discarded piles (unmanaged composting) which should reflect current management practise

2. Managed composting and subsequent spreading of digestate on fields

3. Incineration with energy recovery (large scale and small scale)

4. Anaerobic digestion, both wet and solid state (single substrate and co-digestion of multiple substrates) and subsequent spreading of digestate on fields

The LCA approach uses a number of potential scenarios that include these systems as well as associated logistical systems such as transportation; the LCA system boundaries are described in more detail in Eriksson et al. (2015).

The model chosen for this task is ORWARE, which is an LCA tool for evaluating environmental impacts of waste management and has been in development since the early

1990s (Bisaillon et al., 2000). The model is described in more detail in section 2.1. LCA tools model material flows and, as such, require input data on the composition of materials and the material transformation processes involved in the waste management systems being considered.

The first step in building an LCA model is to make a data inventory of quality data sources. Where reliable data cannot be established it is also necessary to make assumptions. As an example, accurate data on the number and location of horses in the

Gävleborg region does not exist, but can be estimated from national statistics and qualitative data and surveys. Such assumptions are discussed in later sections. The ORWARE model itself includes a number of previous assumptions already documented in related literature and will not be included in this report. Once this data is established, scenarios may be proposed to consider the various waste management options. This process, along with the results of the modelling, is described in Eriksson et al., 2015.

1.1. Data sources

The data was collected from peer-reviewed journal papers, technical reports, questionnaires, personal communication, and in some cases relies on pre-existing values in the

LCA model, ORWARE. These data sources are referenced further below and the resultant data is listed in the main text or in Appendix A.

1.2. Management systems considered

Hadin et al. (2015) discussed the various horse manure management options, both in use today and as potential options that could be available in the short term. In Figure 1, a system view of possible horse manure management option/treatments is presented.

Figure 1. A system view of treatment options for horse manure and biogas generation.

In summary, the following systems are considered:

• Anaerobic (‘liquid-state’) digestion – the production of biogas with subsequent upgrading to vehicle gas and the resultant digestate, which can be used as fertiliser.

• Solid-state anaerobic digestion (SS-AD) – producing biogas (upgraded) and digestate as above.

• Aerobic digestion (composting) – transforming the manure into fertiliser through either:

- Piling and storage, leading to passive decomposition (this includes

‘unmanaged’ composting), or

- Through active mixing, aeration and turning.

• Combustion (incineration) – for heat and electricity production in either

- Large scale combined heat and power (CHP) plant. Includes landfill treatment of the resultant ash & slag.

- Small scale heat boiler at eg. a farm

• Sludge treatment – mixing with sewage sludge and composted before use as top soil cover on a landfill. This option have for long time been the current practise in

Gävle but will not be modelled due to a lack of data and also as this is not a future option as the landfill will be finalised within a couple of years.

With the exception of passive decomposition, the management systems above can not only prevent emissions from untreated horse manure, they can also potentially offset emissions from other energy uses, for example incineration can generate heat and electricity, and biogas can replace fossil fuels and possibly heat. Such offsetting can be considered in the LCA model, described below. Even if there are many plausible options, not all will be investigated in this project. For biogas baseline scenarios comprise upgrading to vehicle gas and using biogas in a CHP may be considered in sensitivity analysis. Pretreatment options for the management systems under study are also considered or included in the model, along with transportation and post-treatment such as sorting facilities. The model represents the current state of the systems being considered. As such, possible future changes in the market and policy measures are not considered.

The inventory in this report covers the main processes as described above. Further work related to the systems analysis including final definitions of system boundaries etc. is further described in Eriksson et al. (2015).

2. The LCA method and model

In this chapter, the technical systems covered by the model are discussed on a general level. A life-cycle assessment (LCA) model is used to study and compare the management options mentioned above. The systems modelled were identified through literature reviews, interviews with horse keepers and personal communication with technology experts.

2.1. The ORWARE model

The model chosen for this task is ORWARE, a computational LCA model for evaluating environmental impacts of waste management (Bisaillon et al., 2000). The ORWARE model has been under development since the early 1990s. It started as collaboration between KTH, SLU, JTI and IVL, and has led to a vast number of research papers, theses and major studies (detailed references are found in Eriksson et al., 2015). The model is currently used and developed mainly by the University of Gävle, Profu, SLU and JTI.

The basis for modelling waste management in ORWARE is that the handling of different wastes can be described at an elemental level, i.e., their composition of nutrients, carbon, water and contaminants such as heavy metals, etc. It can handle solid and liquid organic and inorganic waste from different sources. ORWARE is built from a number of modules that describe a process or treatment. To describe the elements that make up waste management requires a large amount of input data. Each project must decide how much information must be gathered for the case being studied. Once the composition of waste is established, the waste flows through the model from the point of collection, via transportation, to processing plants followed finally by consumption, new products or disposal. A major part of the model used here is built on previous work in a project within

Waste refinery (Holmström et al., 2013).

2.2. Environmental impact assessment

Life-cycle assessment generally describes categories of possible (potential) impacts on the environment as opposed to actual environmental impacts, which would require a more site-specific assessment of the impacts of emissions. Different emissions contribute to the different impact categories to varying degrees. Characterisation factors are therefore used to determine the effects of different emissions on each impact category. Each emission is multiplied by these characterisation factors so that its contribution can be summarised.

Impact categories may include the following:



Emissions that can lead to acidification (from sulphur dioxide, nitrogen oxides and ammonia), and eutrophication (nitrogen oxides and ammonia).



Emissions including greenhouse gases, which contribute to global warming: carbon dioxide (CO

), methane (CH

) and nitrous oxide (N

O).



Heavy metals and toxic levels of various compounds.

Besides the emissions from the core (in this case manure management) system, offsetting emissions from conventional production of products provided by the manure treatment

(eg. electricity, district heating, vehicle transport and nutrients) is also included. More details on LCA methodology are presented in Eriksson et al. (2015).

2.3. General assumptions for modelling

The amount of manure produced by horses varies considerably (see section 3.2). In addition, due to a lack of data regarding the exact number and location of horses (cf. Hadin,

Eriksson and Jonsson, 2015), the model calculates all impacts per tonne of horse manure treated. Thus, the model is based on the assumption that there is a fixed amount of horse manure, for example 10 000 tonnes, which includes a number of pre-set percentages of bedding material. The model is run using this fixed amount. In this way, it is possible to compare the effects of different management systems (anaerobic digestion, composting, etc.) for a given amount of manure and bedding material. Once the results have been established it can be estimated how many horses 10 000 tonnes may represent. Also because of a lack of data, costs are not included in the model. System-specific assumptions are discussed throughout this report.

3. Data inventory

Summaries of the data are included in this section. The raw data is collected in an Excel file that is available as a separate electronic appendix to this report.

3.1. Validation

The validity of the ORWARE model and its input data has been tested in numerous prior papers, theses and studies (cf. section 2.1). For this study, the ORWARE values for horse manure are brought up-to-date using the latest data collected (see also section 3.2). Having chosen values for the amounts and composition of horse manure, the outputs of the

ORWARE model are validated against results from the literature, which may include production-scale plants (‘Plant’ data), laboratory results (‘Lab’ data) and computational/LCA models (‘Model’ data). Therefore, although some data are only used for validation of the model, it is included in this document.

3.2. Horse manure

Pure horse manure (i.e. without bedding) is referred to as ‘horse dung’ in some literature

(Mönch-Tegeder et al., 2013), meaning horse faeces (and sometimes urine). The term

‘horse dung’ will be used in this report to differentiate it from ‘horse manure’, which will denote horse dung and urine mixed with bedding material and subsequently removed from stalls.

A study made during 2002-2003 of horse keepers in the county of Gävleborg, who responded to a survey, showed that 81 % use wood shavings, 19 % use straw and 15 % use peat as a bedding material (Femling, 2003). This agrees with Lundgren and Pettersson

(2009) who state that “wood shavings” (assumed to mean sawdust) are generally used in northern Sweden, due to their accessibility and therefore lower cost.

The bedding materials that are modelled in this study are:

• Sawdust/woodchips (of similar chemical composition and hence “equal” in model terms)

• Straw

• Peat

• Waste paper pellet (manufactured by e.g. Laxå Bruk)

Other beddings such as hay, straw pellets, reed and different mixes of different bedding materials are not included due to low biogas production (straw pellets) or lack of data

(hay, reed).

3.2.1. Physical properties/chemical composition

The ORWARE model uses a dataset for each material being investigated, known as a

‘vector’. Vectors are created for each of the bedding materials above and for pure horse dung. By combining the horse dung vector with a bedding material vector, the model is able to approximate the combination of horse dung and bedding material.

However, it should be noted that Cui, Shi and Li (2011) have demonstrated that anaerobic digestion of bedding material in its raw form gives different results compared to anaerobic digestion of the same material that was used as bedding. For example, in the case of wheat straw, the latter is more easily digested and can produce up to 56 % greater me-

thane yield kg

¹ volatile solids (VS). The amount of nitrogen, phosphorous and potassium in the horse waste will depend partly on whether the horse is sedentary or exercising

(Westendorf and Krogmann, 2006). It is also unclear from the literature what amount of urine is included, if any.

The ORWARE vectors for horse dung and bedding materials (except waste paper pellet where a previous vector for waste paper have been applied) are included in Appendix A, including justifications for the values used and any associated references. From the

ORWARE perspective the composition of sawdust and woodchips is the same, since both are sourced from wood. From a horse keeping perspective however, it should be noted that these materials have different properties, and absorb different amounts of urine. This is not accounted for in the model.

3.2.2. Amounts of horse manure generated

The data below (Table 1) shows that total waste per horse is in the range 9-29 m³ horse

¹ annum

¹. The Swedish statutory requirement for storage of deep litter manure is 9.9 m³ horse

¹ annum

¹ according to Swedish Board of Agriculture, where standard values include faeces and urine, bedding material, decomposition losses, cleaning water, water spillage and precipitation. However not all manure will be collected, since horses are not always in the stall.

The values of horse waste production that are assumed for the model, based on the

‘mean’ values in the Table 1: 15 kg faeces horse

¹ day

¹ and 9 kg urine

horse

¹ day

(Westendorf and Krogmann, 2006; Wheeler and Zajaczkowski, 2009; Wartell et al.,

2012), totalling 24 kg waste horse

¹ day

¹, equivalent to 8.7 tonnes waste horse

¹ annum

¹.

From Table 1, values for bedding material used per horse per day are: straw 9.9 kg, sawdust 5.3 kg, and woodchips 16.3 kg (Airaksinen, Heinonen-Tanski and Heiskanen, 2001;

Häußermann, Beck and Jungbluth, 2002), which would therefore correspond to 29 %, 18

% and 40 % of the total waste by weight, respectively

. A horse using sawdust for bedding would thus

produce approximately 10.7 tonnes annum

¹ of manure, assuming it is in the stall all year. There is a large difference between sawdust and woodchips due to the different absorption properties. However, no consistent data on the relationship between type of bedding material, urine and faeces have been found, which in turn means that sawdust and wood chips are treated equally in modelling.

Steineck et al. (2000 cited in Baky et al., 2012), agree that generated faeces and urine is

8-10 tonnes horse

¹ annum

¹, and that bedding material adds from a few tonnes up to

7 tonnes horse

¹ annum

¹, depending on the size of the horse

(Jordbruksverket, 2006 cited in Baky et al., 2012). However Jordbruksverket (2013) claims that the straw content of horse manure can be up to 90 percent. Cui, Shi and Li (2011) state that bedding make up

25 % of the wet weight of manure.

Assuming 1 L urine equals 1 kg.

For example, for straw, 9.9 kg straw is 29 % of 15 kg faeces + 9 kg urine + 9.9 kg straw.

(24 kg waste + 5.3 kg sawdust) × 365 days = 10 695 kg.

They assume an average horse weighs 500 kg.

Table 1. Average properties of horses and horse manure generated per day or per annum

Reference

Wartell et al.

(2012)

Westendorf and

Krogmann

(2006)

Weight of horse [kg]

Feces [kg/day]

Urine [litre/day]

Total waste incl. bedding [kg/day]

Total waste incl. bedding [m³/annum]

454

27 a) Woodchips b) Wood granules c) Straw d) Rough wood shavings e) Sawdust / fine wood shavings f) Peat and sawdust (3:1)

Bedding (*excluding urine)/bedding manure [kg/day]

a) Unspecified 9* b) Woodchips c) Wood granules d) Straw e) Rough wood shavings f) Sawdust / fine wood shavings g) Peat and sawdust (3:1)

Notes regarding the data in Table 1 can be found in Appendix A.

454

4-7*

Lundgren and

Pettersson

(2009)

–29

Airaksinen,

Heinonen-

Tanski and

Heiskanen

(2001)

16.3

9.1

14.9

12.4

19.5

12.4

Häußermann,

Beck and

Jungbluth (2002)

Wheeler and

Zajaczkowski

(2009)

49.7

454

9.0

10.8

10.6

5.3

65.0

130.0

83.7

62.4

Mean

16.3

9.0

9.9

10.6

5.3

14.9

The values for bedding should therefore be considered uncertain and liable to variation depending on the horse-keeping practices and bedding materials in use. To account for this, the systems analysis will investigate the importance of different proportions of bedding in separate scenarios. This is being discussed further in Eriksson et al. (2015).

3.3. Transport

Transport modelling includes site specific data on distances between stables and treatment plants and type of vehicle being used as well as data related to each vehicle type such as energy efficiency, fuel type and related emission factors. No new data or specific modelling of transports is made in this study. Transport distances between waste source

(stable) and waste treatment plant are set to 15, 50 or 80 km, thereby testing the importance of transport distance. The distance between treatment plants are set to 15 km.

For biogas transported to a gas station and digestate and compost going to arable land, an average distance of 50 km is used. All these figures are assumptions used in the national study of a future biogas market (Holmström et al., 2013). Most vehicles are truck and trailer transport. For details on fuel and emissions, see Eriksson et al. (2015).

4. Horse manure management systems

The sections below describe in more detail the various treatments/management systems considered. See Table A-1 (Appendix A) for a system view of the different treatment methods under consideration.

4.1. Unmanaged manure

– natural decomposition

For the comparison of management systems, unmanaged horse manure represents the base case. This is manure that is left in piles or dumped in nature and left to decompose naturally. In this situation natural composting can occur, but often under sub-optimal conditions; uncontrolled emissions of gases can be higher and contamination occurs due to leaching of nutrients into water systems (Rodhe et al., 2015). Alternatively the manure does not compost, due to a lack of nitrogen or water, and is leached of nutrients.

Few data were found on average emissions from unmanaged horse manure. The IPCC

(2006) estimates the methane emission factor for horse manure in cool climates

to be

1.56 kg CH

horse

¹ year

¹.

However, there is an overlap between emissions from unmanaged manure and those from composting (cf. section 4.2), since unmanaged manure can still compost naturally. Rodhe et al. (2015) measured methane emissions from a manure mix – 90 % horse manure and bedding

and 10 % food waste – before, during and after drum composting

. The predrum-composting manure (stored in a pile indoors) produced 72-85 % of the CO

emissions of all three stages combined, even though it was stored for only 3 days.

Their results also showed that for the manure stored prior to drum composting, methane emissions (CH

) averaged 0.8 % of the total carbon (C) content, compared with an average 3.5 % from a literature review of cattle, prig and poultry manure (Webb et al., 2011 cited in Rodhe et al., 2015, p. 46). For a horse producing 24 kg waste day

¹ (see section

3.2), and assuming 22 % C (see Table A-2 in Appendix A), 0.8 % of C is 15.4 kg CH

4 horse

¹ annum

¹.

Rodhe et al. (2015) also provide measurements of emissions of nitrous oxide (N

O) from the manure mix, ranging from zero to 0.23 % of total nitrogen (N), depending on the stage of decomposition, compared with a standard IPCC value of 0.5 % (Rodhe et al.,

2015, p. 46). Ammonia (NH

) emissions were under 3 % of N content during all composting stages (Rodhe et al., 2015, p. 47).

Steineck et al. (2001 cited in Rodhe et al., 2015) state that windrow composting over about 1.5 months released ammonia emissions of 11.2 % of the total nitrogen (N) for horse manure with straw bedding and 0.2 % of the total N from horse manure with peat as bedding. Karlsson and Torstensson (2003) also tested emissions from windrow compost-

Cool climates are defined by the IPCC as those with an average annual temperature below 15 °C, whereas the average annual temperature for the Gävleborg region is around 5 °C (SMHI, 2015), so emissions may be lower.

With bedding material of 50% sawdust, 35% straw, 10% peat, and 5% straw pellets.

Figures taken from Rodhe et al. (2015, tables 7, 10, 24) Wiggeby study: averages of values for food combinations 1, 2, 3 and 4 (‘compost A’ and ‘compost B’).

Although the 0.8 % methane emissions include emissions from the 10 % proportion of food and the bedding material, and the indoor conditions meant there was a limited air flow.

ing, finding 0.3 % of N and 0.7 % of phosphorous (P) were lost during composting (6-8 weeks) and 2 % and 3 %, respectively, during storage (4.5-5 months). 6-8 % of total N was lost through NH

emissions during composting.

The values chosen to represent unmanaged manure are based on a number of data sources, including those mentioned above. The data is summarised in the Table 2, which shows results from Rodhe et al. (2015) for two situations: 1) emissions for all three composting stages (pre-composting, drum composting and post-composting) and 2) values specific to the 3.5-day pre-composting stage; Garlipp, Hessel and van den Weghe (2011) for deep bedding manure during 19 days; and miscellaneous values from other sources.

Values for the following emissions are chosen for these reasons:

• Nitrous oxide (N

O) gas: value chosen to be between Rodhe et al. (2015) and

IPCC (2006).

• Ammonia (NH

) gas: value

chosen between Steineck et al. (2001 cited in Rodhe et al., 2015) and Karlsson & Torstensson (2003).

• Methane (CH

) gas: Rodhe et al. (2015) showed reduced methane production due to forced aerobic drum composting. The value is also chosen to be lower than emissions reported in Webb et al. (2012 cited in Rodhe et al., 2015, p. 46).

• Total nitrogen in leachate: value 10

chosen from Karlsson & Torstensson (2003) since it is the closest to unmanaged manure.

• Ammonium nitrogen (NH

-N) in leachate: average of source values used.

• Nitrite nitrogen (NO

-N) in leachate: value of zero used since the values reported are considered insignificant.

• Nitrate nitrogen (NO

-N) in leachate: value known for wheat straw and woodchip used, since it is assumed they are the most prevalent bedding materials.

• Phosphorous (P) in leachate: value from Karlsson & Torstensson (2003) is the only known data.

4.2. Aerobic digestion

– composting

The C/N ratio also affects composting (Karlsson & Torstensson, 2003). Similar to anaerobic digestion, aerobic digestion (composting) requires a C/N ratio (weight ratio) of 30:1 at start-up to be optimal (Poincelot, 1975 cited in Rodhe et al., 2015). Similar to AD, the mix of horse dung and bedding material chosen for use in the model is pre-validated in

Excel as having a C/N ratio in the range 20-30.

4.2.1. Windrow composting

Windrow composting is assumed to have similar emissions to unmanaged manure and is therefore not compared in the model.

4.2.2 Drum composting

A drum composter accelerates the composting process by physically turning the manure and providing a constant flow of air. In the commercially available Quantor XL®, it takes about 24 hours for the substrate to pass through the drum composting process and reaches a compost temperature of 52 °C during at least 13 hours (Ecsab, 2015), fulfilling Swedish regulations on hygienisation

Does not apply to peat.

May not apply to peat.

Discussed in more detail in Hadin et al. (2015).

Table 2. Data sources to determine unmanaged horse manure emissions

Reference

Rodhe et al.

(2015)

IPCC,

2006 cited in

Rodhe et al.

(2015, p.

46)

Comments

All composting stages combined: pre-, drum- and postcompost

Webb et al.

(2012 cited in Rodhe et al., 2015, p.

46)

Emissions from chicken, pig and cow manure

Manure mix

90 % horse manure & 10% organic food waste

- -

Rodhe et al.

(2015)

Pre-compost during 3.5 days

90 % horse manure & 10% organic food waste

Garlipp, Hessel and van den

Weghe (2011)

Deep bedding manure during 19 days

Horse manure: 60% horse faeces,

60 L urea, 25 kg bedding

Bedding

Air emissions

- N

₂O, Nitrous oxide [% of total N]

- NH

₃, Ammonia [% of total N]

- CH

₄, methane [% of total C]

- Leachate

- Nt, total nitrogen [% of total N]

- NH

₄-N, ammonium nitrogen [% total N]

- NO

₂-N, nitrite nitrogen [% of total N]

- NO

₃-N, nitrate nitrogen [% of total N]

- P, phosphorous [% of total P]

50% sawdust,

35% straw,

10% peat, 5% straw pellets

0-0.23%

0.86%

0.01175%

(drum only)

0.5%

3.5%

50% sawdust,

35% straw,

10% peat, 5% straw pellets

0.01%

0.5%

0.8%

Wheat straw

Rye straw Sawdust

0.10%

0.004%

0.06%

0.005%

0.00015% 0.00017%

0.033% 0.010%

0.10%

0.004%

0.034%

Steineck et al.

(2001 cited in

Rodhe et al.,

2015)

Straw

11.2%

Karlsson

Torstenss on (2003)

Chosen value for model

Windrow composting during

1.5  months

Windrow 6-

8 weeks plus storage 4-

5.5 months

Peat

0.2%

Mostly straw, some sawdust

6-8%

0.3%

0.7%

0.25%

10.0%

2.0%

0.30%

0.004%

0.033%

0.7%

Work by Rodhe et al. (2015) provided measurements of emissions from the three stages in the Quantor XL® drum composting process: storage prior to drum composting

(‘pre-compost’), mechanical drum composting (‘drum’), and finally post-composting

(‘post-compost’). The following data (Table 3), from the Wiggeby farm experiment

(Rodhe et al., 2015), are used in the model:

Table 3. Emissions from drum composting including pre-compost and post-compost stages

Reference

Rodhe et al. (2015)

Manure mix

Horse manure (90%) & organic food waste (10%) - Wiggeby farm

Bedding

Stage

Duration

50% sawdust, 35% straw, 10% peat, 5% straw pellets

Pre-compost

(in)

3.5 days

Drum (in)

26 hours

Post-compost

(in)

4-6 weeks

Post-compost

(out)

N/A

Total solids (TS) [% FM]

Total nitrogen [% TS]

C/N ratio

Emission factor gas CH

₄,

methane [% of C]

Emission factor gas N

₂O, nitrous

oxide [% of N]

Emission factor gas NH

₃,

ammonia [% of N]

Emission factor leachate NH

₄,

ammonium [% of N]

31.10%

1.32%

0.80%

0.01%

0.50%

32.30%

1.33%

0.02%

0.00%

0.52%

0.01%

33.10%

1.12%

0.04%

0.26%

1.77%

38.80%

1.49%

4.3. Incineration

The Fuel Handbook (Strömberg & Svärd, 2012) does not recommend the use of horse manure in its pure form in biofuel plants, due to a risk of both sintering and corrosion caused by high moisture content and high levels of sulphur and chlorine. Baky (2012) describes sintering when the bedding material used was peat or straw, but not from manure with bedding consisting of sawdust and only a little straw.

Incineration of pure horse manure has been tested in small-scale plants (Pettersson et al.,

2006; Lundgren & Pettersson, 2009; Edström et al., 2011), and incineration of horse manure mixed with other biofuels and wastes has been tested in both small- and large-scale plants (Puustinen, 2009 cited in Edström et al., 2011; Baky, 2012; Henriksson et al.,

2015).

The ORWARE model includes a sub-model for large-scale incineration, which is used in this study. The large scale incineration uses emission factors for the most relevant emissions based on environmental reporting from the Sävenäs waste CHP in Gothenburg

(Holmström et al., 2013). Data on degree of efficiency, power-to-heat ratio and use of electricity is based on national averages. Ash and slag is subject to landfill disposal using appropriate landfill submodels.

However, for small-scale incineration, data is taken from experiments by Baky (Baky et al., 2012; Baky, 2013b). Pre-drying horse manure has been shown in tests to increase the lower heating value (LHV) and produce a net gain in energy of 1.9 MWh/tonne of dried manure, after deducting the energy used for drying (Baky, 2012). Therefore manure is dried prior to combustion, decreasing water percentage from 57.4 to 9.7 %. Drying equipment uses 0.40 MWh heat/ton wet manure for evaporating water. Due to aeration of the manure a slight composting takes place, leading to some losses of C and N. The furnace has a degree of efficiency of 80 %. All energy is released as heat and utilised (just

50 % utilisation in the report). The simplified model does not handle combustion residues. For further details, please see references above. In Table 4 some data on incineration collected from literature is displayed.

4.4. Anaerobic digestion (AD)

The mono-feedstock anaerobic digestion (AD) of horse dung and of horse manure has been tested in numerous lab-based and pilot plant trials, but so far there are no known large-scale implementations. From these trials the results of measuring carbon, nitrogen and nutrient content and other chemical compounds in the substrate are shown to vary considerably due, in part, to the manure’s freshness (Mattsson, Karlsson and Nilsson,

2015) and the bedding material used (Olsson et al., 2014).

Measurements of methane production from the digestion process of horse dung (i.e. without bedding material) are also very inconsistent, production is strongly influenced by feeding intensity and feed composition (Mönch-Tegeder, 2013, p.166 citing Møller et al.,

2004 and Amon et al., 2007).

The age of the manure can also have a strong influence on methane production. According to Mattsson, Karlsson and Nilsson (2015), fresh horse manure produced more methane when peat had been used as a bedding material, whereas for one-month-old horse manure, straw bedding produced more methane. After two months of storage, methane production was worse in both cases compared to fresher manure. Even a late harvest due to weather conditions can affect straw feed quality and in turn methane production

(Mönch-Tegeder, 2014). See also section 4.4.1.

Therefore, although the model has modules for AD, it can only provide an approximate calculation of biogas production and emissions. It is still possible to validate those values as being within the range collected in the inventory data. This data is summarised in section 4.4.5.

4.4.1. Storage

Garlipp, Hessel, and van den Weghe (2011) showed that deep litter horse manure, placed in mostly anaerobic conditions, gave different results according to the bedding material, with wheat straw producing significantly higher NH

, N

O, CO

, CH

and H

O concentrations over 19 days compared to rye straw, indicating faster degradation. Both types of straw generated significantly higher concentrations of N

O, CO

, and CH

compared to wood shavings over 19 days.

Total Nitrogen

Higher heating value

(dry fuel, free of ash)

Lower heating value

(dry fuel, free of ash)

EMISSIONS

CO, Carbon monoxide

% TS

MJ/kg

MJ/kg mg/Nm³ mg/MJ

NOx, Nitrogen oxide

mg/Nm³ mg/MJ

Particles (dry gas)

ASH (horse manure)

mg/Nm³

Ash content

Al, Aluminium

% TS

% Ash

Ca, Calcium

Cu, Copper

% Ash

K, Potassium

% Ash

Mg, Magnesium

Na, Sodium

% Ash

Pb, Lead

Si, Silicon

Zn, Zinc

% Ash

Table 4. Heating values, emissions and ash contents from incineration of horse manure, both pure and mixed

Reference

Fuel

Pettersson,

Lundgren, and

Hermansson

(2006, pp.

396-400);

Lundgren &

Pettersson

(2009);

Edström et al.

(2011, p. 9).

Horse manure

Edström et al.

(2011, p.

8) citing www.swe

bo.com

Horse manure

Puustinen

(2009) cited in Edström et al. (2011, p. 4)

Horse manure

(40%) & sawdust

Henriksson et al. (2015).

Ash analysis from

Strömberg &

Herstad

Svärd (2012) cited in

Henriksson et al. (2015).

Horse manure

(10%) & household waste (90%)

Baky et al.

(2012)

Horse manur e

(dried)

Bedding material

Woodchip/ sawdust

Sawdust

(woodchi p)

Sawdust Straw

Sawdu st

Plant size

250 kW(th)

150 kW(th)

40 kW(th) -

350 kW(th)

Total solids, TS

kg/t FM 43.0% 51.0% 30.1% 37.0% 90.3%

170.6

395.6

0.9%

19.37

18.14

425.8

320

340

20.4

19.1

120

15.45

75.15

1.0%

18.84

7.3%

24.3%

4.3%

3.0%

0.3%

7.0%

4.3%

3.9%

1.2%

0.0%

7.3%

1.3%

9.1%

0.0%

11.2%

0.1%

0.7%

17.29

16.08

12.3%

2.5%

30.1%

7.2%

1.9%

1.0%

10.6%

Mattsson, Karlsson and Nilsson (2015) tested the anaerobic digestion of horse manure after different storage times: fresh manure, 5 days old and 14 days old. They found that there was a decrease in methane production in relation to the increase in age, by up to one-third for 14-day-old manure. However they also point out that temperature has a big effect, and losses from storage at 10 °C are smaller (Møller, 2012 cited in Mattsson,

Karlsson and Nilsson, 2015).

The emissions from storage are assumed to be the same for all management options, since the manure will always require a period of storage regardless of the final treatment. This is discussed further in Eriksson et al. (2015).

4.4.2. Substrates suitable for anaerobic digestion

The version of the ORWARE model being used includes a module for wet anaerobic digestion. This submodel has been adjusted to also represent dry anaerobic digestion, see section 4.4.4. For the substrate to be suitable for anaerobic digestion, certain conditions must be met. Many authors cite a requirement for the carbon/nitrogen (C/N) ratio to be in the range 20-30 (Karthikeyan, 2013). Mönch-Tegeder (2013, p.164) and Mao et al.

(2015) also state the optimum C/N ratio to be 20-30.

The mix of horse dung, manure and bedding material chosen for use in the model is prevalidated in Excel as having a C/N ratio in the range 20-30 to ensure that the results of anaerobic digestion are valid.

Concentrations of compounds like calcium (Karthikeyan, 2013, p. 275-277) and low pH caused by excessive concentrations of volatile fatty acids (VFAs) – which are in turn caused by high organic loading and C/N ratio (Brown et al., 2012) – are also inhibitory to some methanogens. Kusch, Oechsner and Jungbluth (2008, p.1289) state that ammonia concentrations of horse manure are well within acceptable limits, but it is not clear if urine or bedding was included. Baky (2013) removed some ammonia in order to reduce the nitrogen level for AD, whereas Böske et al. (2014) added ammonium carbonate in order to increase it. Mönch-Tegeder (2014, p.166) states “horse manure cannot provide sufficient amounts of trace elements for a stable biogas process.” Factors such as these

(calcium, pH) are not considered in the anaerobic digestion submodel.

Despite the reservations above, the results of many experiments (Nilsson, 2000 cited in

Edström et al., 2011, p. 4; Edström et al., 2005; Pettersson et al., 2006; Kusch, Oechsner and Jungbluth, 2008; Wartell et al., 2012; Baky, 2013b; Mönch-Tegeder et al., 2013;

Böske et al., 2014, Smith and Almquist, 2014; Mattsson, Karlsson and Nilsson, 2015) – summarised below in section 4.4.5 and listed in more detail in the electronic appendix

‘Manure Composition and Treatments.xlsx’ – show examples of successful anaerobic digestion of unadulterated horse manure. It is assumed in the model that the composition of the manure, represented by the vector, is appropriate for AD.

4.4.3. Codigestion

Codigestion provides the flexibility to simultaneously utilise more than one substrate, and also gives the ability to manipulate the C/N ratio to improve the stability of the digestion process (Karthikeyan, 2013, p. 272). In this model, data from a plant-scale study from

Olsson et al. (2014) is used to validate the codigestion of horse manure mixed with cow manure.

Table 5. Data sources for anaerobic digestion using codigestion of multiple substrates

Reference

Substrate

Bedding material

Olsson et al. (2014)

Horse manure (69%) & cattle dung

(31%)

Saw dust

Olsson et al. (2014)

Horse manure (56%) & cattle dung (44%)

Straw pellets and sawdust

Digestion time [days]

Total solids (TS) [% FM]

Specific methane production

[m³ CH4/t VS]

Process temperature [°C]

Notes

7.5%

125

30-41

Manure ‘FL1’ in reference

7.5%

131

29-40

Manure ‘FL2’ in reference

4.4.4. Solid-state anaerobic digestion (SS -AD)

Solid-state anaerobic digestion (SS-AD) of horse dung is possible without the addition of inoculum (Kusch, Oechsner and Jungbluth 2008, p.1288). The model uses the same vectors for manure and bedding materials, only the sub-model representing the anaerobic digestion process is changed for SS-AD.

Initially a mesophilic digestion process is applied as most solid state digestion plants we know of (and they are quite few) has a process temperature representing mesophilic conditions. The percentage of TS in the digester has been set to 30 % which is considerably higher compared to liquid state digestion and corresponds quite well with the TS percentage in the substrate (which depends on how much bedding of which type is mixed with pure manure). The initial retention time is set to 27 days, which corresponds to the plant for food- and garden waste in Mörrum, Sweden. Use of electricity is supposed to be lower due to less pumping and mixing etc, an assumption of 50 % compared to wet anaerobic digestion is made. A gas loss of 1 % is assumed, which is in line with default values in ORWARE with reference to on-site measurements by SP on several biogas plants in Sweden. Biogas production will be used as validation parameter where the model may be fed with food waste to meet a production rate of 85 Nm

/ton. Finally, the digestate will have a TS percentage of 30 % to ensure dry spreading equipment can be used. The final values for the model after validation are presented in Eriksson et al.

(2015).

4.4.5. Validation data for anaerobic digestion

This section shows a summary (Table 6-9) of data collected for the validation of anaerobic digestion of horse manure.

Table 6. Results of anaerobic digestion of horse manure with straw bedding

Reference

Nilsson (2000) cited in

Edström et al. (2011, p. 4)

Edström et al.

(2005)

Scale

Lab/batch Plant

Digestion time [days]

Total solids (TS) [% FM]

Volatile solids (VS) [% TS]

Total nitrogen [% TS]

Process temperature [°C]

C/N ratio

Specific methane production [m³ CH4/t VS]

Specific methane yield [m³ CH4/t VS]

30.00%

84.00%

0.35%

180.0

30.00%

84.00%

2.60%

200.0

Baky

(2013b)

Model

23.25%

75.53%

2.51%

146.7

Table 7. Results of anaerobic digestion of horse manure with straw pellets

Reference

Mattsson,

Karlsson and

Nilsson (2015)

Mattsson,

Karlsson and

Nilsson (2015)

Mattsson,

Karlsson and

Nilsson (2015)

Scale

Digestion time [days]

Total solids (TS) [% FM]

Volatile solids (VS) [% TS]

Total nitrogen [% TS]

Process temperature [°C]

C/N ratio

Specific methane production [m³ CH4/t VS]

Specific methane yield [m³ CH4/t VS]

Lab

27.10%

1.85%

22.6

24.0

Lab

20.20%

3.12%

12.5

12.7

Lab

25.00%

2.92%

13.9

14.1

Age [days]

0 5 14

NB: the authors claim that some of these results may not be reliable and methane production was derived from total production.

Even if straw pellets are not investigated further Table 8 clearly shows the low methane production from this bedding material.

Table 8. Results of anaerobic digestion of soft woodchips

Reference

Wartell et al. (2012)

Scale

Digestion time [days]

Total solids (TS) [% FM]

Volatile solids (VS) [% TS]

Total nitrogen [% TS]

Process temperature [°C]

C/N ratio

Specific methane production [m³ CH4/t VS]

Specific methane yield [m³ CH4/t VS]

Lab

32.00%

79.80%

35 ±1

111

Table 9. Results of anaerobic codigestion of horse manure with cow slurry

Reference

Kalia & Singh

(1998)

Olsson et al.

(2014)

Olsson et al.

(2014)

Substrate mix

Horse manure

(20%) & cattle dung (80%)

Horse manure

(69%) & cattle dung (31%)

Horse manure

(56%) & cattle dung (44%)

Bedding material

Unspecified Saw dust

Straw pellets and sawdust

Scale

Plant Plant Plant

Baky (2013b)

Horse manure

(44%) & cattle slurry (56%)

Straw

Model

Digestion time [days]

Total solids (TS) [% FM]

Volatile solids (VS) [% TS]

Total nitrogen [% TS]

Process temperature [°C]

C/N ratio

Specific methane production

[m³ CH4/t VS]

8-10%

88.8%

25-30 ± 1

162.5

7.5%

30-41

125.0

7.5%

29-40

131.0

16.8%

77.4%

141.2

4.4.6. Pretreatment fo r AD

Hadin, Eriksson and Jonsson (2015) explained that hygienisation must be used in Sweden when horse manure is collected for treatment from two or more different production locations. The model therefore assumes that hygienisation is used and this is included in the model.

Lignocellulosic material, such as straw in horse manure, can cause low bioavailability or low biodegradability (Carlsson et al., 2012). The results data gathering of horse manure pretreatment methods for AD did not provide sufficient data for modelling, however they suggest that an increase of biogas production and/or rates of generation are possible and the potential methods are briefly discussed below. The model does, however, include a pre-existing sub-model for steam explosion.

Potential pretreatments of horse manure for AD

The chemical treatment of lignocellulosic materials has been investigated in a number of studies, although no studies were found that tested chemical pretreatments of purely horse dung or horse manure. For example, treating straw from cattle and horse manure with Nmethylmorpholine oxide (NMMO) for 15 hours was shown to increase the methane yield by 51 %, through a decrease in the structural lignin content (Aslanzadeh et al., 2011).

Teghammar (2013) also showed that NMMO pretreatment of spruce woodchips for 15 hours resulted in an increased production of 202 NmL CH

/g. Lidner et al. (2015) warn that the high cost of chemicals and investment in equipment is an obstacle for chemical pretreatment in full-scale plants.

In contrast to chemical and thermal methods, mechanical pretreatment has no inhibitory or toxic byproducts formed during the disintegration process (Mönch-Tegeder et al.,

2014). There is no change to the total solids (TS), volatile solids (VS) or chemical compostition (Mönch-Tegeder et al., 2014, p.142). In addition, mechanical devices for decomposition are available and easy to implement in biogas plants (Mönch-Tegeder et al.,

2014, p.139). The main disadvantage of mechanical pretreatment is the high energy consumption (Bruni et al. 2010).

Chopping manure with a compost chopper was shown by Kusch, Oechsner and Jungbluth

(2008, p.1288) to increased yield, especially during the initial stages of digestion (up to

42 days), although the final methane yield was the same over long digestion times. However, no data was provided on the energy consumption of the chopping process or the net energy saving.

Mönch-Tegeder et al. (2014), on the other hand, demonstrated that 15 seconds of mechanical grinding of horse manure (with straw bedding) could lead to a positive energy balance of 12.7 kWh tonne

¹ FM, since methane yields were increased 10 % in comparison to untreated manure.

References

Airaksinen, S., Heinonen-Tanski, H., Heiskanen, M.-L. (2001) ‘Quality of different bedding materials and their influence on the compostability of horse manure’, Journal of Equine Veterinary

Science, 21(3), 125–130.

Aslanzadeh, S. (2014) Pretreatment of Cellulosic Waste and High-Rate Biogas Production

[online], available: http://hdl.handle.net/2320/12853 .

Baky, A. (2013a) Life Cycle Inventory & Assessment Report: Combustion of Horse Manure with

Heat Utilisation, Sweden, JTI - Swedish Institute of Agricultural and Environmental Engineering.

Baky, A. (2013b) Life Cycle Inventory Report : Co-Digestion of Horse Manure and Dairy Cattle

Slurry, Sweden, JTI - Swedish Institute of Agricultural and Environmental Engineering, available: http://www.balticmanure.eu/download/Reports/codigestion_web.pdf

[Accessed 31 Mar

2015].

Baky, A., Karlsson, E., Norberg, I., Tersmeden, M., Yngvesson, J. (2012) Förbränning Av Förtor-

kad Häst Gödsel På Gårdsnivå, JTI - Swedish Institute of Agricultural and Environmental

Engineering.

Bisaillon, M., Haraldsson, M., Sundberg, J., Eriksson, O.N. (2000) Systemstudie Avfall Borås (A

Systems Study of the Future Waste Management System in Borås) (In Swedish with English

Summary), SP Sveriges Tekniska Forskningsinstitut.

Brown, D., Shi, J., Li, Y. (2012) ‘Comparison of solid-state to liquid anaerobic digestion of lignocellulosic feedstocks for biogas production’, Bioresource Technology, 124, 379–386, available: http://dx.doi.org/10.1016/j.biortech.2012.08.051

Bruni, E., Jensen, A.P., Angelidaki, I. (2010) ‘Comparative study of mechanical, hydrothermal, chemical and enzymatic treatments of digested biofibers to improve biogas production’, Bioresource Technology, 101(22), 8713–8717, available: http://dx.doi.org/10.1016/j.biortech.2010.06.108

Böske, J., Wirth, B., Garlipp, F., Mumme, J., Van den Weghe, H. (2014) ‘Anaerobic digestion of horse dung mixed with different bedding materials in an upflow solid-state (UASS) reactor at mesophilic conditions’, Bioresource Technology, 158, 111–118.

Carlsson, M., Lagerkvist, A., Morgan-Sagastume, F. (2012) ‘The effects of substrate pre-treatment on anaerobic digestion systems: A review’, Waste Management, 32(9), 1634–1650, available: http://dx.doi.org/10.1016/j.wasman.2012.04.016

Cesaro, A., Belgiorno, V. (2014) ‘Pretreatment methods to improve anaerobic biodegradability of organic municipal solid waste fractions’, Chemical Engineering Journal, 240, 24–37, available: http://dx.doi.org/10.1016/j.cej.2013.11.055

Cui, Z., Shi, J., Li, Y. (2011) ‘Solid-state anaerobic digestion of spent wheat straw from horse stall’,

Bioresource

Technology, 102(20), 9432–9437, available: http://dx.doi.org/10.1016/j.biortech.2011.07.062

Ecsab (2015) Approved Drum Composting System for Manure, Sludge, Biowaste [online], available: http://www.ecsab.com/en_index.htm

[Accessed 30 Jun 2015].

Edström, M., Nordberg, Å., Ringmar, A. (2005) Utvärdering Av Gårdsbaserad Biogasanläggning

På Hagavik. JTI-Rapport Kretslopp & Avfall 31, JTI - Swedish Institute of Agricultural and

Environmental Engineering.

Edström, B.M., Schüßler, I., Luostarinen, S., Edström, M. (2011) Combustion of Manure: Manure

as Fuel in a Heating Plant, available: http://www.balticmanure.eu/en/knowledge_forum/reports/project_results/manure_energy/combustion_of_manure_manure_as_fuel_in_a_heating_plant.htm

[Accessed 20 Feb 2015].

Eriksson, O., Hadin, Å., Hennessy, J., Jonsson, D. (2015) Hästkrafter och hästnäring – hållbara

systemlösningar för biogas och biogödsel - Explorativ systemanalys med datormodellen

ORWARE, FoU-rapport 43, University of Gävle

Eskilsson, J. (2013) Gödsel Och Miljö 2014, Jordbruksverket, Jönköping.

Femling, M. (2003). Hästen i Gävleborg. Förstudierapport 2003: 4. Lantbruks- och veterinärenheten. Länsstyrelsen Gävleborg

Garlipp, F., Hessel, E.F., van den Weghe, H.F. a. (2011) ‘Characteristics of Gas Generation (NH

, N

O, CO

, H

O) From Horse Manure Added to Different Bedding Materials Used in Deep

Litter Bedding Systems’, Journal of Equine Veterinary Science, 31(7), 383–395, available: http://dx.doi.org/10.1016/j.jevs.2011.01.007

Government of Saskatchewan (2008) Straw Rations – Wintering Cows [online], available: http://www.agriculture.gov.sk.ca/Default.aspx?DN=5b2ed401-5fd2-483e-bce1-92c8d1687ba4 .

Hadin, Å., Eriksson, O., Jonsson, D. (2015) Energi och växtnäring från hästgödsel – Förbehand-

ling, rötning och biogödselavsättning, FoU-rapport 42, University of Gävle

Henriksson, G., Johansson, I., Hedenstedt, A., Johansson, I. (2015) Benchmarking Av Gödselsam-

rötning Med Avloppsslam Mot Förbränning Av Häst- Och Djurparksgödsel, SP Sveriges Tekniska Forskningsinstitut, Borås.

Hermansson, R. Lundgren, J., Pettersson, E., Lindgren, A., Forsberg, M., Jansson, M (Unknown).

Miljövänlig eldning av hästspillning, AB Swebo Flis.

Holmström, D., Bisaillon, M., Eriksson, O., Hellström, H., Nilsson, K. (2013) Framtida mark-

naden för biogas från avfall (The future market for biogas from waste), project report WR 46,

WASTE REFINERY, SP Sveriges Tekniska Forskningsinstitut, ISSN 1654-4706 (In Swedish)

Häußermann, A., Beck, J., Jungbluth, T. (2002) ‘Litter in horse keeping’, Landtechnik, 57(1), 50–

51, available: https://www.landtechnikonline.eu/ojs2.4.5/index.php/landtechnik/article/viewFile/2002-1-050-051/2759 [Accessed 8

Jun 2015].

Intergovernmental Panel on Climate Change (IPCC) (2006) 2006 IPCC Guidelines for National

Greenhouse Gas Inventories.

Jordbruksverket (2013) Hästgödsel – En Naturlig Resurs, Jordbruksverket, available: http://www2.jordbruksverket.se/webdav/files/SJV/trycksaker/Pdf_jo/jo13_5.pdf

Kalia, A.K., Singh, S.P. (1998) ‘Horse dung as a partial substitute for cattle dung for operating family-size biogas plants in a hilly region’, Bioresource Technology, 64, 63–66.

Karlsson, S., Rodhe, L. (2002) Översyn Av Statistiska Centralbyråns Beräkning Av Ammoniakav-

gången I Jordbruket – Emissionsfaktorer För Ammoniak Vid Lagring Och Spridning Av Stall-

gödsel, JTI - Swedish Institute of Agricultural and Environmental Engineering.

Karlsson, S., Torstensson, G. (2003) Strängkompostering Av Hästgödsel, JTI - Swedish Institute of

Agricultural and Environmental Engineering.

Karthikeyan, O.P., Visvanathan, C. (2013) ‘Bio-energy recovery from high-solid organic substrates by dry anaerobic bio-conversion processes: A review’, Reviews in Environmental

Science and Biotechnology, 12, 257–284.

Kusch, S., Oechsner, H., Jungbluth, T. (2008) ‘Biogas production with horse dung in solid-phase digestion systems’, Bioresource Technology, 99, 1280–1292.

Lindner, J., Zielonka, S., Oechsner, H., Lemmer, A. (2015) ‘Effects of mechanical treatment of digestate after anaerobic digestion on the degree of degradation’, Bioresource Technology, 178,

194–200, available: http://linkinghub.elsevier.com/retrieve/pii/S0960852414013765 .

Lundgren, J., Pettersson, E. (2009) ‘Combustion of horse manure for heat production’,

Bioresource Technology, 100(12), 3121–3126, available: http://dx.doi.org/10.1016/j.biortech.2009.01.050

Mattsson, M., Karlsson, N., Nilsson, S.B. (2015) Biogas Från Hästgödsel I Halland, Högskolan i

Halmstad.

Moreno-Caselles, J., Moral, R., Perez-Murcia, M., Perez-Espinosa, A., Rufete, B. (2002) ‘Nutrient

Value of Animal Manures in Front of Environmental Hazards’, Communications in Soil Science

and Plant Analysis, 33(15-18), 3023–3032, available: http://www.tandfonline.com/doi/abs/10.1081/CSS-120014499 .

Møller, H.B. 2012. Nye foderstrategier giver mere biogas. FiB nr 42. December 2012 http://www.bioprocess.dk/PDF/FiB 42-2012.03.pdf

Mönch-Tegeder, M., Lemmer, A., Oechsner, H., Jungbluth, T. (2013) ‘Investigation of the methane potential of horse manure’, Agricultural Engineering International: CIGR Journal, 15(2),

161–172.

Mönch-Tegeder, M., Lemmer, A., Jungbluth, T., Oechsner, H. (2014) ‘Effects of full-scale substrate pretreatment with a cross-flow grinder on biogas production’, CIGR Journal, 16(3), 138–

147, available: http://www.cigrjournal.org/index.php/Ejounral/article/view/2927 .

Mönch-Tegeder, M., Lemmer, A., Oechsner, H. (2014) ‘Enhancement of methane production with horse manure supplement and pretreatment in a full-scale biogas process’, Energy, 73, 523–530, available: http://linkinghub.elsevier.com/retrieve/pii/S0360544214007506 .

Nilsson, B. (1994) Kolets Sammansättning Och Energiinnehåll I Växt- Och Avfallsmaterial, available: http://www.vaxteko.nu/html/sll/slu/ex_arb_vaxtnaringslara/EVN089/EVN089.HTM

NIST (US National Institute of Standards and Technology) (2015a) Material Details. RM 8493 -

Monterey Pine Whole Biomass Feedstock [online], available: https://wwws.nist.gov/srmors/view_detail.cfm?srm=8493 [Accessed 30 Jun 2015].

NIST (US National Institute of Standards and Technology) (2015b) Material Details. RM 8494 -

Wheat Straw Whole Biomass

Feedstock [online], available: https://wwws.nist.gov/srmors/view_detail.cfm?srm=8494 [Accessed 30 Jun 2015].

Olsson, H., Andersson, J., Rogstrand, G., Persson, P., Andersson, L., Bobeck, S., Assarsson, A.,

Benjaminsson, A., Jansson, A., Alexandersson, L., Thorell, K. (2014) Samrötning Av Hästgöd-

sel Med Nötflytgödsel - Fullskaleförsök Vid Naturbruksgymnasiet Sötåsen. Rapport 51,

Kretslopp Och Avfall, JTI - Swedish Institute of Agricultural and Environmental Engineering, available: http://www.jti.se/index.php?page=publikationsinfo&publicationid=1034&returnto=146

[Accessed 12 Feb 2015].

Pettersson, E., Lundgren, J., Hermansson, R. (2006) ‘Co-combustion of wood-shavings and horse manure in a small scale heating plant’, in World Bioenergy 2006, Swedish Bioenergy Association (SVEBIO): Stockholm, 396–400, available: https://www.etde.org/etdeweb/servlets/purl/20812170-k7Afgh/20812170.PDF

Preston, R.L. (2010) What’s The Feed Composition Value of That Cattle Feed? Beef Magazine

[online], available: http://beefmagazine.com/nutrition/feed-composition-tables/feedcomposition-value-cattle--0301 [Accessed 25 Jun 2015].

Puustinen, H., Kajolinna, T., Pellikka, T., Kouki, J. & Vuorio, K. 2009. Characterisation of air-

borne emissions from combustion of horse manure. Research report VTT-R-01295-09, VTT,

Finland. 18 p. (in Finnish)

Rodhe, L., Niklasson, F., Oostra, H., Gervind, P., Ascue, J. (2015) Kontrollerad Kompostering

Med Liten Klimatpåverkan - Emissioner Och Värmeåtervinning, available: http://www.jti.se/index.php?page=publikationsinfo&publicationid=1081&returnto=96

[Accessed 19 May 2015].

Smith, D.B., Almquist, C.B. (2014) ‘The anaerobic co-digestion of fruit and vegetable waste and horse manure mixtures in a bench-scale, two- phase anaerobic digestion system’,

Environmental Technology, 35(7), 859–867, available: http://www.tandfonline.com/doi/abs/10.1080/09593330.2013.854398

Steineck, S., Svensson, L., Tersmeden, M., Åkerhielm, H. & Karlsson, J. (2001). ‘Miljöanpassad hantering av hästgödsel.’ JTI-Rapport 280. Lantbruk & Industri. JTI-Institutet för jordbruks- och miljöteknik.

Stanton, T.L. (2014) Feed Composition for Cattle and Sheep [online], available: http://www.ext.colostate.edu/pubs/livestk/01615.html

[Accessed 25 Jun 2015].

Strömberg, B., Herstad Svärd, S. (2012) Bränslehandboken 2012, Värmeforsk.

Swedish Meteorological and Hydrological Institute (SMHI) (2015) Normal Årsmedeltemperatur

[online], available: http://www.smhi.se/klimatdata/meteorologi/temperatur/normalarsmedeltemperatur-1.3973

[Accessed 21 May 2015].

Teghammar, A. (2013) Biogas Production from Lignocelluloses. Pretreatment, Substrate Char-

acterization, Co-Digestion, and Economic Evaluation, available: http://bada.hb.se/bitstream/2320/12317/1/phd thesis teghammar.pdf

US National Institute of Standards and Technology (NIST) (2015a) Material Details. RM 8494 -

Wheat Straw Whole Biomass Feedstock [online], available: https://wwws.nist.gov/srmors/view_detail.cfm?srm=8494 [Accessed 30 Jun 2015].

US National Institute of Standards and Technology (NIST) (2015b) Material Details. RM 8493 -

Monterey Pine Whole Biomass Feedstock [online], available: https://wwws.nist.gov/srmors/view_detail.cfm?srm=8493 [Accessed 30 Jun 2015].

Wartell, B. a., Krumins, V., Alt, J., Kang, K., Schwab, B.J., Fennell, D.E. (2012) ‘Methane production from horse manure and stall waste with softwood bedding’, Bioresource Technology,

112, 42–50, available: http://dx.doi.org/10.1016/j.biortech.2012.02.012

Watts, K.A. (2005) ‘A review of unlikely sources of excess carbohydrate in equine diets’, Journal

of Equine Veterinary Science, 25(8), 338–344, available: http://www.safergrass.org/pdf/JEVS8-

05.pdf

Webb, J., Sommer, S., Kupper, T., Groenestein, K., Hutchings, N., Eurich-Menden, B., Rodhe, L.,

Misselbrook, T., Amon, B. (2012) ‘Emissions of Ammonia, Nitrous Oxide and Methane During the Management of Solid Manures’, in Lichtfouse, E., ed., Agroecology and Strategies for Cli-

mate Change SE - 4, Sustainable Agriculture Reviews, Springer Netherlands, 67–107, available: http://dx.doi.org/10.1007/978-94-007-1905-7_4 .

Wennerberg, P., Dahlander, C. (2013) Hästgödsel Som En Resurs. En Förstudie Om Olika Hante-

ringskedjor För Hästgödsel., available: http://www.hastforetagarnagoteborg.se/Hästgödselsomresurs_Tecnofarm_maj13.pdf

[Accessed

27 Mar 2015].

Westendorf, M., Krogmann, U. (2006) Horse Manure Management: Bedding Use [online], available: http://esc.rutgers.edu/fact_sheet/horse-manure-management-bedding-use/ [Accessed

8 Jun 2015].

Wheeler, E., Zajaczkowski, J.S. (2009) ‘Horse Stable Manure Management’, Horse Facilities 3, available: http://extension.psu.edu/publications/ub035/view .

Appendix A

An electronic version(Table A-1) of the data below (Table A-2) shows that total waste per horse is in the range 9-29 m³ horse

¹ annum

¹. The Swedish statutory requirement for storage of deep litter manure is 9.9 m³ horse

¹ annum

¹ according to Swedish Board of

Agriculture, where standard values include faeces and urine, bedding material, decomposition losses, cleaning water, water spillage and precipitation. However not all manure will be collected, since horses are not always in the stall.

Table A-1 is available as an electronic appendix - ‘Measurements Per Horse.xlsx’. The values stated for volume per year in Airaksinen, Heinonen-Tanski and Heiskanen (2001) are described as relating to 'bedding manure'. It is assumed this means bedding material that includes urine but is separated from faeces - the bedding material is taken from box stalls via a daily removal of faeces followed by the 'dirty' bedding with urine, while all clean bedding is left in place. Values for bedding [kg/day] are derived from table 3 & 4 in

Airaksinen, Heinonen-Tanski and Heiskanen (2001, p. 128f) - this can be seen in cells

F16-21 (Table A-1).

Wheeler and Zajaczkowski (2009) value for manure [kg/day] is derived from the annual value for a "full-time" occupant (see cell H6 in Table A-1).

Häußermann, Beck and Jungbluth (2002, table 2, p. 51) show the amount of 'dung' and

'dung volume' but it is inferred from 'density of dung' that these values are for faeces, urine and bedding combined. Figures in red are believed to be incorrect or use a strange measurement system.

ORWARE vectors

The known values of the ORWARE vector for horse dung (i.e. excluding bedding) are detailed below (Table A-2).

Table A-2. ORWARE vector for horse manure, including justifications for choices of values

Vector index

Property Value Justification

Total solids, TS [fraction of

Fresh Matter]

2.35E-01

Organic carbon total

[fraction of TS]

3.70E-01

Carbon-lignin [fraction of

TS]

2.23E-02

Fraction of FM. Chosen from Mönch-Tegeder et al. (2013), who made analyses of samples from

10 farms. This is the average of values ranging from 20.67 to 27.3. Contradicts Strömberg &

Herstad Svärd (2012, p.244), who state 40.1%

TS. Wartell et al. (2012, p. 44) states 37%, however it is not certain that this excludes bedding. Smith & Almquist (2014) states 27.8%.

Same as original ORWARE vector. Sum of vector indexes 2, 3, 4, 5 and 41 is 21.7%. The difference between this and the ORWARE value may be the insoluable lignin.

NOT total lignin: only the soluable part. Average of 10 farms with value range 0.87-1.17% of TS acid detergent lignin in Mönch-Tegeder et al.

(2013, p.165)

Carbon-starch & sugar

[fraction of TS]

0.00E+00

Carbon-fat [fraction of TS] 6.71E-03

Same as original ORWARE vector

Average of 10 farms with value range 1.8-4.1% of TS crude fat in Mönch-Tegeder et al. (2013, p.165)

Carbon-protein [fraction of

TS]

Volatile substance, VS

[fraction of TS]

8.87E-01

Dry substance 1.00E+00

O, Oxygen total [fraction of

(ash-free) TS]

3.93E-01

H, Hydrogen total [fraction of (ash-free) TS]

6.42E-02

N, Nitrogen total [fraction of (ash-free) TS]

1.48E-02

NH3/NH4+-N

1.93E-02

8.09E-04

S, Sulphur [fraction of

(ash-free) TS]

3.00E-03

P, Phosphorus [fraction of

(ash-free) TS]

1.24E-03

Cl, Chlorine [fraction of

(ash-free) TS]

2.80E-03

K, Potassium [fraction of

(ash-free) TS]

1.91E-03

Ca, Calcium [fraction of

(ash-free) TS]

1.00E-02

Pb, Lead [fraction of (ashfree) TS]

1.10E-06

Cd, Cadmium [fraction of

(ash-free) TS]

1.00E-07

Hg, Mercury [fraction of

(ash-free) TS]

1.00E-08

Cu, Copper [fraction of

(ash-free) TS]

1.40E-05

Cr, Chromium [fraction of

(ash-free) TS]

4.70E-06

Ni, Nickel [fraction of (ashfree) TS]

3.10E-06

Zn, Zinc [fraction of (ashfree) TS]

5.50E-05

Carbon-cellulose 1.69E-01

Average of 10 farms with value range 6.8-11.7% of TS crude protein in Mönch-Tegeder et al.

(2013, p.165)

Derived from values of % of FM from Mönch-

Tegeder et al. (2013), who made analyses of samples from 10 farms. This is the average of values of VS as % of FM ranging from 18.22 to

24.72.

Same as original ORWARE vector

Fraction of ash-free TS. From average value

39.31% in Hermansson et al. (Unknown) cited in

Strömberg & Herstad Svärd (2012, p.244).

Fraction of ash-free TS. From average value

6.42% in Hermansson et al. (Unknown) cited in

Strömberg & Herstad Svärd (2012, p.244).

Average of 10 farms with value range for total nitrogen 1.172-1.868% of TS in Mönch-Tegeder

(2013, p.165)

Missing NH3?? Average of 10 farms with value range for NH4 of 0.037%-0.233% of TS in

Mönch-Tegeder (2013, p.165)

Fraction of ash-free TS. From average value

0.3% in Hermansson et al. (Unknown) cited in

Strömberg & Herstad Svärd (2012, p.244).

Example from Hermansson et al. (Unknown) cited in Strömberg & Herstad Svärd (2012, p.244)

Fraction of ash-free TS. From average value

0.28% in Hermansson et al. (Unknown) cited in

Strömberg & Herstad Svärd (2012, p.244).

Fraction of ash-free TS. Example from