pefcr feed for food producing animals version 1 7. interpretation 41 7.1 robustness of the pef model...

PEFCR Feed for food producing animals

Version 1.5

Draft version 1.5 for public consultation after supporting studies

22 July 2016

Technical Secretariat for the Feed Pilot

2

Abstract

This section will be completed at a later stage

Authors

The Technical Secretariat is composed of EU and national feed associations (FEFAC, FEFANA, SNIA, ASSALZOO, DVT, NEVEDI, FHL, AIC, DAKOFO), feed companies (ForFarmers, AB Agri, Denkavit, Agrifirm, Sanders, DTC, Union Agricole, Evonik, DSM, Ewos, Ajinomoto Eurolysine, Agravis Raiffeisen, Elanco, Cargill Animal Nutrition), upstream and downstream partners of the feed industry (FEDIOL, UECBV, FEAP) and the Livestcok and Environmental Assessment and Performance Parntnership (LEAP) hosted by the Food and Agriculture Organisation of the United Nations (FAO). The Technical Secretariat is supported by Blonk Consultants.

3

Table of Contents

List of figures ...........................................................................................................................................5 List of tables ............................................................................................................................................5 Abbreviations and Units ..........................................................................................................................6 Definitions ...............................................................................................................................................7

1. Introduction ................................................................................................................................. 8

1.1 General information about the PEFCR ................................................................................ 9

2. Methodological inputs and compliance .................................................................................... 11

3. PEFCR review and background information .............................................................................. 12

3.1 PEFCR review panel ........................................................................................................... 12

3.2 Review requirements for the PEFCR document ................................................................ 12

3.3 Reasoning for development of PEFCR ............................................................................... 12

3.4 Conformance with the PEFCR Guidance ........................................................................... 12

4. PEFCR scope .............................................................................................................................. 13

4.1 Functional unit ................................................................................................................... 13

4.2 Representative product ..................................................................................................... 13

4.3 Product classification (NACE/CPA) .................................................................................... 14

4.4 Purposes of LCA assessment of feed supported by this PEFCR and related requirements 15

4.5 System boundaries – life-cycle stages and processes ....................................................... 15

4.6 Selection of the EF impact categories indicators .............................................................. 18

4.7 Additional environmental information ............................................................................. 19

4.8 Assumptions/limitations of the PEFCR .............................................................................. 20

5. Life cycle inventory analysis ...................................................................................................... 21

5.1 Definition of foreground and background data in relation to contribution and access ... 21

5.2 Data access and use of primary or secondary data and data quality requirements ......... 22

5.3 Primary data collection...................................................................................................... 24

5.4 Secondary data selection .................................................................................................. 35

5.5 Calculating the overall data quality score ......................................................................... 38

5.6 Reporting data gaps and proxies ....................................................................................... 38

5.7 Minimum data quality level for PEF communication ........................................................ 38

5.8 End-of-life stage ................................................................................................................ 39

5.9 Requirements for multifunctional products and multiproduct processes allocation ....... 39

6. Benchmark and classes of environmental performance ........................................................... 40

4

7. Interpretation ............................................................................................................................ 41

7.1 Robustness of the PEF model ............................................................................................ 41

7.2 Identification of hotspots .................................................................................................. 41

7.3 Estimation of uncertainties ............................................................................................... 42

7.4 Conclusions, limitations and recommendations ............................................................... 42

8. Reporting, Disclosure and Communication ............................................................................... 43

8.1 Context .............................................................................................................................. 43

8.2 Recommendation for communication of environmental information ............................. 44

9. Verification ................................................................................................................................ 45

10. Reference literature .............................................................................................................. 46

11. Supporting information for the PEFCR .................................................................................. 46

List of annexes ...................................................................................................................................... 47

5

List of figures Figure 4-1: System boundaries for the production of animal feed. All grey blocks are in the scope of the PEFCR, the dark grey block corresponds to the foreground processes. Note that transport of feed ingredients to the mill and transport from feed mill to farms is not always under the control of the feed business operator. ......................................................................................................................... 16 Figure 5-1: Data selection decision tree. Note that in some cases the DQR score for that given process need to be increased (see box on right) ................................................................................................ 36 Figure 8-1: Communication of environmental information along the feed and food supply chain. .... 43

List of tables Table 4-1: Impact categories to be quantified for feed for food producing animals. The column ILCD Quality provides the robustness according to ILCD Quality classification. ........................................... 18 Table 5-1: Most relevant flows, processes and activities for environmental impact of feed lifecycle . 21 Table 5-2: Access to primary / secondary data for the different life cycle stages ................................ 23 Table 5-3: Time period in relation to purpose and scope of the PEF study .......................................... 24 Table 5-5: Collection of activity data at the feed mill ........................................................................... 26 Table 5-6: Data Quality Rating system. ................................................................................................. 27 Table 5-7: Data collection for feed transport to farm if fuel use can be collected ............................... 28 Table 5-8: Data collection for feed transport if information on actual fuel use cannot be collected .. 28 Table 5-9: Data quality rating system for transport. ............................................................................. 29 Table 5-10: Example of transport data to be collected from suppliers of the feed materials per feed material. ................................................................................................................................................ 30 Table 5-11: Data quality rating system for feed ingredients production. ............................................. 34 Table 5-12: Data quality requirements. ................................................................................................ 38 Table 8-1: recommended communication vehicles, dependent on purpose. ...................................... 44

6

Abbreviations and Units BOM = Bill of Materials COD = Chemical Oxygen demand DE = Digestible Energy DM = Dry matter content DQA = Data Quality Assessment DQS = Data Quality Score EF = Environmental Footprint ELCD = European reference Life Cycle Database EPD = Environmental Product Declaration Feed Pilot = PEF pilot feed for food producing animals FEFAC = European Feed Manufacturers Federation GE = Gross Energy GHG = Greenhouse Gases GWP = Global Warming Potentials Ha = Hectare HH = Human health (used in ionizing radiation HH) ILCD = International Reference Life Cycle Data System IPCC = Intergovernmental Panel on Climate Change ISO = International Organization for Standardization kWh = kilowatt hour LCA = Life Cycle Assessment LCI = Life Cycle Inventory LCIA = Life Cycle Impact Assessment LHV = Lower Heating Value (or net calorific value) LUC = Land Use Change NACE = Statistical classification of economic activities in the European Community NPK = Nitrogen (N), Phosphorus (P) and Potassium (K) OEF = Organisation Environmental Footprint PCR = Product Category Rules PEF = Product Environmental Footprint PEFCR = Product Environmental Footprint Category Rules RER = Region Europe ReCiPe = Impact assessment method SC = Steering Committee TS = Technical Secretariat TS feed = Technical Secretariat of the pilot feed for food producing animals

7

Definitions Feed ingredient: These are either feed materials or feed additives. Ingredients are of plant, animal or aquatic origin, or other organic or inorganic substances and include:

- Feed materials1 - means products of vegetable or animal origin, whose principal purpose is to meet animals’ nutritional needs, in their natural state, fresh or preserved, and products derived from the industrial processing thereof, and organic or inorganic substances, whether or not containing feed additives, which are intended for use in oral animal-feeding either directly as such, or after processing, or in the preparation of compound feed, or as carrier of pre-mixtures;

- Feed additive2 means substances, micro-organisms or preparations, other than feed material

and pre-mixtures, which are intentionally added to feed or water in order to perform, in particular, one or more of the functions

Food producing animals refers to any animal that is fed, bred or kept for the production of food for human consumption, including animals that are not used for human consumption, but that belong to a species that is normally used for human consumption. It includes fish from aquaculture.

1 As defined in Regulation (EC) No 767/2009 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 13 July 2009 on the placing on the market and use of feed, amending European Parliament and Council Regulation (EC) No 1831/2003 and repealing Council Directive 79/373/EEC, Commission Directive 80/511/EEC, Council Directives 82/471/EEC, 83/228/EEC, 93/74/EEC, 93/113/EC and 96/25/EC and Commission Decision 2004/217/EC 2 as defined in Regulation (EC) No 1831/2003 of the European Parliament and of the Council of 22 September 2003 on additives for use in animal nutrition

8

1. Introduction This draft PEFCR shall be used in parallel with the PEF Guide (European Commission, 2013) and the latest version of the Guidelines from the European Commission (Version 5.2 - Guidance for the implementation of the EU Product Environmental Footprint (PEF)) (European Commission, 2016). It is drafted according to the PEF Guide requirements and the Template provided in Annex B to the guidance document (version 5.2). Where the requirements in this draft PEFCR are in line with but more specific than those of the PEF Guide, such specific requirements shall be fulfilled. This draft PEFCR integrates the feedback from the five supporting studies performed by the members of the Feed Technical Secretariat. The supporting studies are available in Annex 2 and the main comments from the supporting studies have been summarized in Annex 3. The draft PEFCR will be revised on the basis of the comments received during the final public consultation organised from 22 July 2016 to 9 September 2016. It will then undergo external review during Autumn 2016 and Spring 2017 before submission to the EF Steering Committee. The use of the present PEFCR is optional for PEF guide in-house applications, it is recommended for PEF-compliant external applications without comparison/comparative assertions, while it is mandatory for PEF-compliant external applications with comparisons/comparative assertions. Since feed is an intermediate product, this PEFCR can be used in different contexts. First there is the context of the LCA operator that conducts a PEF study for a food producing animal according to a specific PEFCR or the PEF guidelines itself if no PEFCR is available. For this use the PEFCR of feed for food producing animals provides the requirements of accurate transfer of LCI information. Then a second use of this PEFCR is to perform cradle to gate feed PEF studies used for in house-application or for external communication. So, this PEFCR supports the following purposes:

1) Provision of LCI information on compound feed in the context of food producing animals PEF studies ;

2) Cradle to gate PEF studies of compound feed for either internal or external use but without comparison

3) Cradle to gate PEF studies of compound feed for comparison, either between alternatives (e.g. sourcing, raw materials choices,..) or over time (e.g. trend monitoring)

The requirements set up by this draft PEFCR may vary according to the purpose of the study.

9

1.1 General information about the PEFCR

1.1.1 Technical Secretariat The Technical Secretariat of the feed pilot consisted during the drafting of this PEFCR of the following members:

• AB AGRI • AGRAVIS Raiffeisen AG • Agrifirm Group • AIC – Agricultural Industries Confederation • Ajinomoto Eurolysine • Assalzoo - Associazione Nazionale tra i Produttori di Alimenti Zootecnici • Blonk consultants • Cargill Animal Nutrition • Dakofo, The Danish Grain- and Feed Trade Association • DENKAVIT • Deutsche Tiernahrung Cremer GmbH & CO.KG • DSM Nutritional Products AG • DVT - Deutscher Verband Tiernahrung e. V. • Elanco Animal Health • Evonik Industries AG Nutrition and Care Division • EWOS AS • FAO, Food and Agriculture Organisation of the United Nations • FEAP – Federation of European Aquaculture Producers • FEDIOL, the EU Proteinmeal and Vegetable Oil Industry • FEFANA, EU association of Specialty Feed Ingredients and their mixtures • FEFAC, European Feed Manufacturers Federation • NSL – The Norwegian Seafood Federation • ForFarmers N.V. • Nevedi - Dutch Feed Industry Association • Sanders • SNIA, Syndicat National de l’Industrie de la nutrition Animale • UECBV - European Livestock And Meat Trades Union • Union Agricole Holding AG

1.1.2 Consultation and stakeholders The development of this PEFCR can be followed on the dedicated wiki page of the EU pilots’ website: https://webgate.ec.europa.eu/fpfis/wikis/display/EUENVFP/PEFCR+Pilot%3A+Feed+for+food-producing+animals The Technical Secretariat of the PEF pilot on feed for food producing animals has on several occasions invited relevant stakeholders to participate in the PEFCR development. The relevant stakeholders for the PEFCR development include representatives from feed material suppliers, farm and trade associations, compound feed producers, consumers, government representatives, non-governmental organizations (NGOs), public agencies and independent parties

https://webgate.ec.europa.eu/fpfis/wikis/display/EUENVFP/PEFCR+Pilot%3A+Feed+for+food-producing+animals

https://webgate.ec.europa.eu/fpfis/wikis/display/EUENVFP/PEFCR+Pilot%3A+Feed+for+food-producing+animals

10

and certification bodies. The identified relevant stakeholders were proactively informed by the Technical Secretariat about the opportunity to take part in the different public consultations. The following public consultations were organised during the development of this PEFCR

- A first virtual consultation was organised in October 2014 on the scope and representative product of the Feed pilot (Technical Secretatiat for the Feed pilot, 2014). This consultation phase also included a physical consultation which took place on 28 October 2014. The list of participants in this consultation, the comments received and the description of how they have been addressed are available on the wiki page

- The second public virtual consultation was organised from 4 September 2015 to 3 October 2015. The purpose of this consultation was to gather feedback on the screening report (Technical Secretariat for the Feed Pilot, 2015) and the first draft PEFCR prepared by the Feed Pilot Technical Secretariat. Likewise, the list of participants in this consultation, the comments received and the description of how they have been addressed are available on the wiki page. The updated first draft PEFCR was then approved by the EF Steering Committee on 18 November 2015 (Technical Secretatiat for the Feed pilot, 2015).

- Final public consultation from 22 July 2016 to 9 September 2016, on final draft PEFCR taking into account the feedback from the supporting studies.

- The Technical Secretariat of the PEF pilot on feed for food producing animals produced a document describing the major comments received and how they have been addressed. This document is available in the EF virtual consultation Forum.

All along the pilot phase, Technical Secretariat created and maintained a log of the stakeholders that have been communicated with and responded to.

1.1.3 Date of publication and expiration Version number: 1.5 (final draft for public consultation) Date of publication/revision: July 2016 Date of expiration: N/A

1.1.4 Geographic region The PEFCR is valid for all EU feed mill operations and the supply chains that provide these operations.

1.1.5 Language(s) of PEFCR This PEFCR has been written in English. It is not foreseen at this stage to make this document available in other languages. Should this PEFCR be translated, the English version supersedes translated versions in case of conflicts.

11

2. Methodological inputs and compliance This draft PEFCR s has been developed according to the requirements of the PEF Guide (Annex II to Recommendation (2013/179/EU) and the Product Environmental Footprint Pilot Guidance (version 5.2). Where relevant, the recommendations of the Cattle Model Working Group (CMWG) are implemented in this draft PEFCR (JRC & European Commission, 2015). These recommendations are available on the EU pilot wiki page. The CMWG involved the red meat, dairy, pet food, leather and feed for food producing animal pilots and was operational in the second half of 2014. Where relevant, the recommendations from the various issue papers approved by the Technical Advisory Board have also been implemented in the draft PEFCR. The Guidelines for assessment of environment performance of animal feed supply chains, released in April 2015 by the FAO-led Livestock Environmental Assessment and Performance partnership (LEAP) (FAO LEAP, 2015) were also an important methodological input for the development of this draft PEFCR. These guidelines are less prescriptive than what is needed in PEFCR. Many of the suggestions on how calculations should be done in the LEAP guidelines are therefore translated to requirements that shall be fulfilled in this PEFCR.

12

3. PEFCR review and background information

3.1 PEFCR review panel The Feed Technical Secretariat is currently working on the composition of the review panel and this section will completed at a later stage.

3.2 Review requirements for the PEFCR document The critical review is essential for ensuring that the PEFCR:

• is consistent with the guidance provided in the PEF Guide and the PEFCR guidance (version 5.2);

• complements the PEF guide requirements with additional requirements specific to the particularities of the life cycle of compound feed;

• provides guidance to conduct a compliant PEF study enables comparisons and comparative assertions in all cases when this is considered feasible, relevant and appropriate.

3.3 Reasoning for development of PEFCR This PEFCR aims at providing guidance on how to assess the environmental performance of compound feed in a harmonised way. Considering the relative importance of compound feed in the environmental footprint of animal products, it is justified to harmonize the feed-specific aspects of the methodology across all food-producing animals.

3.4 Conformance with the PEFCR Guidance This PEFCR has been developed in compliance with the “Guidance for the implementation of the EU PEF during the Environmental Footprint (EF) pilot phase – Version 5.2”.

This section will be completed after the critical review of the PEFCR.

13

4. PEFCR scope

4.1 Functional unit

Feed is an intermediate product which means that no functional unit is considered as such. The declared unit / reference flow is considered instead. The reference flow is 1 tonne of animal feed product delivered to the livestock farm (or fish farm) entry gate. What? Animal feed for food-producing animals

How much? 1 tonne as fed

How well? To be able to conduct meaningful LCAs for food producing animals, i.e.

- With upstream feed emissions until delivery to the farm - With relevant additional information to enable modelling of feed use

on the farm (for instance , the dry matter (DM), gross energy (GE), Carbon (C), Ash, Phosphorus (P), Nitrogen (N), Zinc (Zn), Copper (Cu), content based on the feed supply to the animal species reared).

How long? Fulfil its function well until the expiry date, when stored under specified conditions. Normally feed is consumed in a short period after delivery. Losses during storage are uncommon and can usually be neglected.

The majority of the feed is sold in bulk, however if feed is sold in smaller packed units the packaging materials should be included in the analysis. The quantity of packaging material will be additional on top of the 1000 kg of feed ingredients. The different declared units or reference flows should not be compared, unless the comparison takes place in the context of a cradle to grave PEF study of animal products. The conditions to establish such a comparison are described in section 6.

4.2 Representative product The representative product is a virtual compound feed product and consists of the average composition of feed ingredients consumed by the EU compound feed industry in the time period 2009-2013. As feed is an intermediate product, the representative product does not correspond to a functional benchmark (see section 6 for more information). The representative product is further described in the screening report available on the wiki page. It will be integrated as annex I of the final version of this PEFCR.

14

4.3 Product classification (NACE/CPA)

The scope of the PEFCR is compound feed produced in feed mills and provided as a partial or complete ration to food-producing animals. It belongs to the CPA 10.91 product group “Manufacture of prepared feeds for farm animals (Eurostat ISSN 1977-0375)”

The total CPA 10.91 includes:

• manufacture of prepared feeds for farm animals • preparation of unmixed (single) feeds for farm animals • treatment of slaughter co-products to produce animal feeds and explicitly excludes:

o production of fishmeal for animal feed, see CPA 10.20 o production of oilseed cake, see CPA 10.41 o activities resulting in by-products usable as animal feed without special treatment

The PEFCR for feed focuses on compound (mixed) feed produced in a feed mill because it is the predominant industrial product that farmers buy as an external input. Moreover, the majority of feed products sold by EU feed manufactures are compound feeds. Code 10.91 is a close reference, but the scope of this PEFCR is actually narrower.

Following this reasoning, the following products do not formally belong to the scope of this PEFCR, although there are no methodological reasons for treating them differently when assessing their impact as part of a feed ration:

1. Single feed materials products, i.e. products that originate at a specific food, drink or biofuel processing plant and are sold directly to farmer (e.g. rapeseed meal, wet gluten feed and distillers supplied to dairy farms).

2. Feed materials that are produced on (or under the control of) the animal farm such as grass (silage), maize (silage) or grains fed directly to farm animals.

The feed PEFCR provides consistent methodological requirements for the entire upstream cradle to gate LCA of feed materials. Therefore, the feed PEFCR can also be used by the operators that produce single feed products, either industrially or on the farm. The PEFCR is therefore useful for other CPA codes, such as 10.20; 10.41 and 10.61, but it is not intended to be “the” PEFCR for these sectors. In other words, in the absence of PEFCR for home-mixing and straight-purchased feed ingredients, the feed PEFCR can be used for these products.

15

4.4 Purposes of LCA assessment of feed supported by this PEFCR and related requirements This PEFCR can be used as follows: 1) Provision of LCI information of feed to PEF studies for food producing animals 2) Cradle to gate PEF studies on compound feed for food-producing animals, without comparison 3) Cradle to gate PEF studies on compound feed for food-producing animals, with comparison

a) comparison between alternatives: such as, different manufacturing methods, evaluation of alternative feed configurations to the same or different nutrient profiles, feed ingredient sourcing and their manufacturing methods, on the basis of upstream life cycle emissions of feed ingredients and feed formulation;

b) comparison in time: monitoring trends/progress in environmental impact of feed products related to measures aimed at reducing environmental impact.

The requirements set up in this PEFCR may vary according to the purpose of the study. A summary of the main variations in relation to the objective of the study is presented in annex IV.

4.5 System boundaries – life-cycle stages and processes The system boundaries are described in Figure 4-1. The Figure shows all the different routes for feed production, the grey fields relate to the production of compound feed and are in the scope of the PEFCR.

The majority of feed ingredients used in compound feed originate from crop cultivation in its broad sense. The cultivation of crops requires the input of manure and fertilisers as well as energy carriers, water, crop protection products and auxiliary materials. The full lifecycle of the production of these products, including transport and depreciation of capital goods is in the scope of this PEFCR. The cultivation of crops can yield one main product, or yield two or more co-products. The crop (co-)product goes through one or more processing stages to separate fractions of the product and refine or purify for use in the animal feed. Processing usually requires energy, water and auxiliary materials (e.g. hexane for oilseed extraction). Waste water from processing will require treatment. Other sources for feed are co-products from animal products processing and feed additives which partly originate from industrial processes. Minerals are also used as feed ingredients.

Feed compounding is the next phase, in which multiple feed ingredients are mixed together to produce a compound feed for food producing animals. In terms of access to information for a compound feed company, the activities taking place in the feed mill are considered as foreground processes.

The delivery of the feed to the farm also belongs to the scope of this PEFCR but the following life cycle stages are considered to be outside the system boundaries in this PEFCR: feed utilisation in the animal production system, processing of animal products, retail and end of life. However, manure from the animal husbandry phase can re-enter the lifecycle at the cultivation of crops stage. Other products also re-enter the lifecycle at the compounding of feed as feed ingredients from the processing of animal products from the slaughterhouse for instance, e.g. plasma protein or the dairy processing industry such as whey powders. It should be noted that these products should be modelled using ‘average’ data as an attributional approach as prescribed in the PEF (thus using an ‘average’ LCI of the animal product).

16

Figure 4-1: System boundaries for the production of animal feed. All grey blocks are in the scope of the PEFCR, the dark grey block corresponds to the foreground processes. Note that transport of feed ingredients to the mill and transport from feed mill to farms is not always under the control of the feed business operator.

17

For some purposes the system boundaries need to be extended. This can be the case for studies where the impact of modifications in the feed formulation is assessed. Such modifications can indeed have consequences on feed utilisation digestion (formally outside the scope of this PEFCR) which require to be captured.

A comparative PEF study can be used for evaluation of alternative feed configurations. This could support decisions in changing the feed composition to improve environmental performance. However the system boundaries of the cradle to gate assessment can be inappropriate in several situations. Annex V gives an overview of these situations and to how to extend the system boundaries in the PEF study to come to definitive conclusions. These solutions shall be applied.

18

4.6 Selection of the EF impact categories indicators Since feed is an intermediate product all default impact categories being selected in the PEF guide shall be included in the assessment, see Table 4-1 .

Table 4-1: Impact categories to be quantified for feed for food producing animals. The column ILCD Quality provides the robustness according to ILCD Quality classification. ILCD

Quality Impact category

Description Source

High

robu

stne

ss

I Climate change Global Warming Potential calculating the radiative forcing over a time horizon of 100 years

IPCC 2007

I Ozone depletion

Ozone Depletion Potential (ODP) calculating the destructive effects on the stratospheric ozone layer over a time horizon of 100 years

World Meteorological Organization (WMO)

I Particulate matter

Quantification of the impact of premature death or disability that particulates/ respiratory inorganics have on the population in comparison to PM2.5. It includes the assessment of primary (PM10 and PM2.5) and secondary PM (incl. creation of secondary PM due to SOx, NOx and NH3 emissions) and CO

Rabland Spadaro 2004

Med

ium

robu

stne

ss

II Ionizing radiation (human health)

Quantification of the impact of ionizing radiation on the population, in comparison to Uranium 235

Frischknecht et al. 2000

II Photochemical ozone formation

Expression of the potential contribution to photochemical ozone formation. Only for Europe. It includes spatial differentiation.

Van Zelm et al. 2008

II Mineral, Fossil and Renewable Resource depletion

Scarcity of mineral resource, with scarcity calculated as ‘Reserve Base’. It refers to identified resources that meet specified physical and chemical criteria related to current mining practice. The reserve base may encompass those parts of the resources that have a reasonable potential for becoming economically available within planning horizons beyond those that assume proven technology and current economics

Van Oers et al. 2002

II Acidification Accumulated Exceedance (AE) characterizing the change in critical load exceedance of the sensitive area of terrestrial and main freshwater ecosystems, to which acidifying substances deposit. European country dependent

Seppӓlӓ et al. 2006 and Posch et al. 2008

II Terrestrial Eutrophication

Accumulated Exceedance (AE) characterizing the change in critical load exceedance of the sensitive area, to which eutrophying substances deposit. European country dependent

Seppӓlӓ et al. 2006 and Posch et al. 2008

II Marine Eutrophication

Expression of the degree to which the emitted nutrient reaches the marine compartment (nitrogen considered to be the limiting factor in marine water). European validity. Averaged characterisation factors from country dependent characterisation factors.

ReCiPe v1.05

II Freshwater Eutrophication

Expression of the degree to which the emitted nutrient reaches the freshwater end compartment (phosphorus considered to be the limiting factor in freshwater). European validity. Averaged characterisation factors from country dependent characterisation factors.

ReCiPe v1.05

Low

robu

stne

ss

II / III Human toxicity, cancer effects

Comparative Toxic Unit for humans (CTUh) expressing the estimated increase in morbidity in the total human population per unit mass of a chemical emitted (cases per kilogramme)

USEtox

II / III Human toxicity, non-cancer effects

Comparative Toxic Unit for humans (CTUh) expressing the estimated increase in morbidity in the total human population per unit mass of a chemical emitted (cases per kilogramme)

USEtox

II / III Freshwater ecotoxicity

Comparative Toxic Unit for ecosystems (CTUe) expressing an estimate of the potentially affected fraction of species (PAF) integrated over time and volumes per unit mass of a chemical emitted (PAF m3 year / kg)

USEtox

III Land use Soil Organic Matter (SOM) based on changes in SOM measured in (kg C/m2/a). Biodiversity impacts not covered by the dataset.

Mila i Canals et al. 2007

III Water resource depletion

Freshwater scarcity: Scarcity-adjusted amount of water used. Swiss ecoscarcity 2006

19

In addition the impact on climate change, as implemented in the ILCD method, shall be reported, with emissions from land use change excluded. This will be reported additionally as “Climate change ex LUC” according to the recommendations of the ENVIFOOD Protocol (Food SCP RT, 2013) and the LEAP feed guidelines (FAO LEAP, 2015).

The difference of robustness of the impact categories (categories I, II and III in the table above, from the more to the less robust) shall be taken into account when communicating PEF results. This is further explained in section 8.

The calculation and reporting of midpoint impact categories is the same for all defined Feed PEF purposes. However the PEF studies on food producing animals may only use a part for their external communication.

4.7 Additional environmental information The TS feed recognizes the importance of quantifying the impact on biodiversity that the feed lifecycle may have. In the screening study, the possibility to assess impact on biodiversity was tested using ReCiPe (see screening report for more information). This was further tested in the supporting studies and the outcome is reported in the supporting study annexes of the second draft PEFCR. The indicator translates the following midpoint impacts to a pressure indicator on biodiversity on the basis of an underlying model (Goedkoop et al., 2009), into species lost for a period of time (species.year).

• Climate change Ecosystems • Terrestrial acidification • Freshwater eutrophication • Terrestrial ecotoxicity • Freshwater ecotoxicity • Marine ecotoxicity • Agricultural land occupation • Urban land occupation • Natural land transformation

Although the indicator can be improved the results seem to in line with the scientific knowledge on the major threats on biodiversity as being defined by the UNEP in the global biodiversity outlook (Secretariat of the Convention on Biological Diversity, 2010). Therefore, impacts on biodiversity may be reported in PEF profiles as additional information, by applying this methodology.

One of the challenges when assessing the impacts on biodiversity of animal products, is to capture both on-farm and off-farm impacts. The off-farm impacts are mainly associated with the feed purchased by the farmer and ReCiPe is seen as a valuable method to capture those. The UNEP Setac initiative on biodiversity also provides useful information. The recommendations of the animal PEF pilots regarding biodiversity are complementary to the recommendations described in this PEFCR.

20

4.8 Assumptions/limitations of the PEFCR

This PEFCR assumes that the user has access to information from a specific feed business operator. This PEFCR can only be used if sufficient information is available on feed ingredients, and nutritional analysis data. This PEFCR is technology neutral from the perspective of the production of feed ingredients. If applied properly with sufficient access to data, differences between production techniques may become visible, but those are not distinguished based on a prejudice.

21

5. Life cycle inventory analysis

This section describes how to conduct the life cycle inventory for a Feed PEF study. To follow the order in which tasks are performed during a PEF study, this section is organised in the following manner: for each item of data collection, the recommendations on which data to collect are immediately followed by guidance on how to assess data quality.

5.1 Definition of foreground and background data in relation to contribution and access This PEFCR is based on the outcomes of the screening study and supporting studies. These studies gave insight into the main causes (due to emissions and resource use) of the environmental impact of the cradle to gate compound feed lifecycle. They are listed in Table 5-1 (see screening report for more information). These impacts are associated with the life cycle of compound feed production and are not related to the consumption of compound feed Table 5-1: Most relevant flows, processes and activities for environmental impact of feed lifecycle Most relevant elementary flows or mechanisms

Most relevant processes Most relevant life cycle stage

N relevant for eutrophication, acidification and climate change impact

Fertilizer and manure application

Raw material acquisition and primary processing

P relevant for eutrophication and resource depletion

Fertilizer and manure application


Land transformation Soy bean growing in Argentina & Brazil


Land occupation All crops Raw material acquisition and primary processing

Water depletion Irrigated crops in regions with water stress


Energy determines the impacts (photochemical smog formation, PM, Ionizing radiation) and has a relevant contribution to Climate Change, Acidification and Eutrophication

Energy production for, and/or fossil fuel use in processing, transport, field operations

Raw material acquisition and primary processing Distribution and transport Production of the main product

Metals (Zn, Cu, Cr, Pb, Hg) relevant for human toxicity and eco-toxicity.

Fertilizer and manure application at farm


Crop protection products mainly relevant for eco-toxicity scores

Farming Raw material acquisition and primary processing

Most of the determining emissions and resource use of the compound feed lifecycle happen at processes (mainly cultivation) in the supply chain that are commonly not under the control of feed companies. For these processes often only secondary data can be used. The following processes are considered to be always run by the compound feed company:

1. purchase of feed ingredients, 2. formulation of the compound feed, 3. operations in the feed mill.

22

Therefore, the feed business operator has specific knowledge of: 1) list of feed ingredients; 2) nutritional analysis data of the compound feed, 3) activity data of the compound feed mill including use of energy carriers and potential on-site energy production, feed ingredients, packaging and auxiliary materials. All these processes are considered to be foreground processes. The list of feed ingredients and the compound feed nutritional analysis (1 and 2) are very important in terms of lifecycle impact contribution. The list of feed ingredients is required for determining the impact of production of feed ingredients and the nutritional analysis is relevant for determining the impact of feed use downstream of the feed mill. Transport to the farm where the feed is consumed, is sometimes run by the feed company, sometimes outsourced and sometimes the feed is being collected by farmers at the factory (especially the smaller quantities). This PEFCR gives guidance on how to operate if primary data are available and when defaults can and shall be used. Transport of feed ingredients from the producer to the feed mill is in most cases not run by the feed business operator and also in most cases transport data such as distances and mode of transportation are not collected at the moment. This PEFCR gives guidance on which data to collect for inbound transport and when to use defaults. Feed ingredient production is hardly ever run by the feed business operator and commonly secondary data needs to be used here. To make this PEFCR also applicable for those situations where the operator conducting the PEF study has both access to primary feed mill data and production data of feed ingredients requirements are set on how to model feed ingredients production.

5.2 Data access and use of primary or secondary data and data quality requirements Table 5-2 gives an overview of the processes involved in the cradle to gate feed lifecycle for which primary and/or secondary data needs to be collected. Table 5-2 is the result of implementing the data needs matrix (Guidance Document 5.2) for the operator that has access to the feed mill data that are essential to conduct a PEF study (Process no. 1, 2 and 3 in Table 5-2). Without the data of these processes a PEF study compliant to this PEFCR is not possible. The data quality requirements depend on the purpose of the study.

The data quality shall be assessed using the data quality rating (DQR) system as described in the PEF Guidance document. The higher requirements (i.e. lower DQR score) apply for studies involving comparison. Less strict requirements (i.e. higher DQR score) apply for studies not aimed for public comparison.

23

The table should be read as follows:

- The column “process” described the different life cycle stages for which data shall be collected

- The column “access to primary data” refers to the sections of this PEFCR which provide guidance on primary data collection

- The column “no access to primary data” provides guidance in case of absence of primary data and refers to the appropriate sections of this PEFCR.

- Finally the column “DQR related to the scope of the study” specifies the applicable DQR for the different processes or life cycle stages, in relation to the purpose of the study.

Table 5-2: Access to primary / secondary data for the different life cycle stages Process Access to primary

data No access to primary data

DQR related to the scope of the study

Info

rmat

ion

of

proc

esse

s ru

n by

th

e co

mpa

ny

1 Feed formulation BOM section 5.3.1 No PEF possible Not applicable 2. Feed formulation Nutritional analysis data

section 5.3.2 No PEF possible Not applicable

3. Feed mill operation section 5.3.3. No PEF possible DQR <3 for purpose 1,2 DQR < 1,6 purpose 3, 4 if relevant for comparison

Info

rmat

ion

of p

roce

sses

(pa

rtia

lly)

run

by t

he

com

pany

info

rmat

ion

is pa

rtia

lly o

r not

ava

ilabl

e 4. Outbound transport (to feed user)

Primary data on fuel use or on transport means and distance section 5.3.4

Apply default values on transport distance and transport means section 5.3.4

DQR <3 purpose 1, 2 or default values. DQR < 1,6 for purpose 3, if relevant for comparison

5. Inbound transport (from final producer in Supply chain)

Primary data on fuel use or on transport means and distance section 5.3.5

Apply default values on transport distance and transport means section 5.3.5

DQR <3 for purpose 1,2 or default values DQR < 1,6 for purpose 3, if relevant for comparison

6. Feed ingredient production process lifecycle information

Supplier specific data derived according to (non-existing) PEFCR or LEAP guidelines section 5.3.6

Select processes from secondary database on feed ingredients production EC dataset GFLI dataset section 5.4.1

DQR <3 for purpose 1,2 DQR < 1,6 for purpose 3, if relevant for comparison DQR < 3

24

5.3 Primary data collection

List of feed ingredients (Bill of Materials, BOM) 5.3.1 The list of feed ingredients entails the following data:

• Types and quantities of feed materials • Types and quantities of feed additives

The reference to define the feed materials is the EU Catalogue of feed materials3 and to define the feed additives it is the EU Register of Feed Additives4.

For crops and processed feed materials used in the feed mill the country of origin shall be recorded if this information is provided in the transaction to the feed business operator. See 5.4.1 for further guidance on how to deal with missing or incomplete information on origin of production.

The feed ingredients list shall be the weighted average composition of a feed reflecting the practice of producing feed for the farms of food producing animals in scope. The weighted average shall be determined by taking time related variation and the variation of geographical origin for supply into account. Combining different geographical origins of supply in the calculation of the weighted average does not negatively affect the data quality rating. The scope of the PEF study determines the time period for deriving the weighted average as indicated in Table 5-3.

Table 5-3: Time period in relation to purpose and scope of the PEF study Purpose Scope Time period for deriving weighted

averages Reporting requirements

1. Cradle to gate PEF compliant information for studies on food producing animals

Determined by PEF study on food producing animals

According to specifications of PEF study on food producing animals

Record and communicate time period of setting weighted average.

2.Cradle to gate PEF study on compound feed, without comparison

Feeds on the market with fixed nutritional specifications for more than a year (such as basic dairy and pig fattening feed)

1 year average or longer up to 3 years if longer term market cycles occur in feed materials production

Report a time period of 1 year or longer if applicable and if not motivate the use of a longer time period.

Feeds on the market with nutritional specifications fixed for period shorter than a year.

Use the longest applicable time period Report and motivate the chosen time period.

3a. Cradle to gate PEF study including product comparison

To show if an innovative feed performs better than the alternative

Use a time period to derive feed composition for making a fair comparison

Report and motivate the chosen time period.

3b. Cradle to gate PEF study for performance tracking

To show developments and/or improvements in performance over time

Two options (can be combined): No averaging if trend analysis aims to show actual fluctuations (also related to market fluctuations and not to actual changes in composition). Rolling averages to correct for market fluctuations.

Report on trends and changes in feed materials composition and nutritional analysis data.

3 Commission Regulation (EU) No 68/2013 of 16 January 2013 on the Catalogue of feed materials 4http://ec.europa.eu/food/food/animalnutrition/feedadditives/docs/comm_register_feed_additives_1831-03.pdf

http://ec.europa.eu/food/food/animalnutrition/feedadditives/docs/comm_register_feed_additives_1831-03.pdf

http://ec.europa.eu/food/food/animalnutrition/feedadditives/docs/comm_register_feed_additives_1831-03.pdf

25

Nutritional analysis data 5.3.2 The nutritional analysis data is especially relevant for PEF studies for food producing animals. The nutritional analysis data needed for the purpose of the PEF study are:

• Nitrogen (N), Phosphorus (P) content in g/kg • Ash • Copper (Cu), Zinc (Zn) content in g/kg • Gross Energy (MJ/kg net calorific value or LHV) and digestible energy fraction5 (% of gross

energy) • Biogenic and Fossil carbon content, listed separately

The provision of the biogenic and fossil carbon content – although not strictly a nutritional property - is required to complete the assessment of carbon related emissions in later life cycle stages.

When a feed additive or a mixture of feed additives is incorporated in the compound feed, this information shall be part of the nutritional analysis data, if its addition has an impact on the environmental footprint of the farm (by e.g. improving the animal performance or reducing the emissions). With the indication of the presence of the feed additive or the mixture of feed additives, an estimation of the potential effect on the relevant criteria should be provided. The latter should be expressed as a range of improvement, in % of the current performance / emission on the farm and appropriately substantiated.

The data for Nitrogen (N), Phosphorous (P) can be obtained by calculation where the share of feed ingredients is multiplied with the typical or actual composition of the feed ingredients. Typical nutritional analysis data can be found in country datasets or if not available at http://www.feedipedia.org/. Actual composition data are those measured by the feed company during the year.

The data for Zinc and Copper can be generated by calculation where the share of feed materials (so not the additives) is multiplied with the typical or actual Zn and Copper composition of the feed ingredients. On top of this the addition of Zinc and Copper from mineral additives shall be added. Here the actual additions shall be used. The gross energy values, biogenic and fossil carbon shall be calculated following guidance provided in annex 6 or using a more detailed method.

The method of nutritional analysis, using typical or actual values and applying the default or more detailed method for gross energy, shall be reported.

The digestible energy as a percentage of the gross energy shall be reported. However, rather than providing a single value, it is allowed to provide the digestible energy as a range. The boundaries of this range should be meaningful, and depend on the animal type for which the feed is intended.

5 The digestible energy varies per animal species.

http://www.feedipedia.org/

26

5.3.3 Feed mill operation data The data mentioned in Table 5-4 shall be collected. The data should be recorded according to the format in the table. In the fourth column, the method of measurement should be explained. This includes the sources of information and any conversion of information and related assumptions.

Table 5-4: Collection of activity data at the feed mill Activity data Unit per ton of feed

out Quantity Source and method

of measurement (if relevant)

Electricity use kWh Gas use MJ LHV Heat use MJ LHV Other energy inputs MJ LHV

(specify type)

Water m3 (specify type)

Packaging (only in case of feed sold in small units e.g. 25 kg bags of calf feed )

Kg (specify type)

Data can be derived on different levels of accurateness which needs to be determined in relation to the scope of the study.

If the feed operation is not part of assessing differences in a comparison between alternatives or changes in time the minimum level of accurateness shall be average feed mill data determined for 1 year of normal operation. (Normal operation is data corrected for calamities).

If comparisons are made (between alternatives or in time) that include changes in the feed mill operation (e.g. pelleting or not, temperature, pressure etc.) specific feed mill processing data shall be collected. This can either be done on the basis of measurements or an analysis where use of energy and auxiliary materials is derived on technical specifications of equipment. Also if specific data are collected all use of energy and auxiliary materials of the feed mill shall be divided over the specific products.

Thus any estimate of specific energy and auxiliary materials use for a feed product shall be done on the basis of allocating the use of the complete factory to sub-processes, also the processes of a general nature such as lighting, heating, office equipment, etc. How this is done shall be motivated and recorded by the PEF operator.

27

Data Quality Scoring and minimum data quality level

Minimum data quality is 3 if no comparison is involved. If data are important for comparison minimum data quality is 1.6. See Table 5-5 for data quality rating system.

Table 5-5: Data Quality Rating system. Quality

level Quality rating

C TiR P TeR GR EoL (related to packaging, only if feed is sold in

small quantities) Very good

1 All data points as mentioned in Table 5.2

Data6 cover exactly the time period in the scope of the study

≤ 10% The technology(ies) is/are specific for the feed product(s) in scope based on measurements

The data concern the specific feed mill(s) of production in scope of study on the basis of weighted share

The EoL formula is implemented country specific

Good 2 All data points as mentioned in Table 5.2, except water

Data partially cover the time period in the scope of the study

10% to 20%

The technology(ies) is/are specific for the product(s) in scope based on assigning overall energy and materials use

The data concern the specific feed mill(s) of production in scope of study but unweighted

The EoL formula is implemented with European averages

Fair 3 Only electricity and gas

Data are not older than 2 years with respect to the time period in the scope of the study

20% to 30%

Average feed mill data is used

The data concern average company data (not being the scope of the study)

…

Poor 4 Only electricity use


30% to 50%

NO PEF NO PEF …

Very poor

5 No energy use data


> 50% NO PEF NO PEF The EoL formula is not implemented

The data quality score for the feed mill operation DQRfm is:

DQRfm6

EoLPCGRTeRTiR +++++=

TeR: Technological-Representativeness GR: Geographical-Representativeness TiR: Time-Representativeness C: Completeness P: Precision/uncertainty EoL: End-of Life Formula

6The reference time is the one when data have been originally collected and not the publication/calculation date. In case there are multiple data, the oldest is the one against which the calculation should be made.

28

5.3.4 Outbound transport to farm (feed user) This section will be updated to take into account data on energy and transport which became available recently. If actual fuel use data of outbound transport can be collected, because there is a suitable accounting system in place, these data shall be used. Fuel use data will be connected to secondary LCI data for fuel production and combustion7. See Table 5-6 for a format that can be used for data collection.

If no fuel use data is available, transport emissions can be estimated by collecting activity data on transport distance (based on average distance to the farms supplied by the mill, unless more accurate data is available), transport vehicles, load fractions, load fraction of the return trip and share of biofuel use. With this information transport inventories can be selected in the secondary database8.

Table 5-6: Data collection for feed transport to farm if fuel use can be collected Activity data Unit Quantity Technology

(e.g. EURO-class 1, 2, 3, etc.)

Source and method of measurement (if relevant)

Fuel use (type 1) unit/tonne delivered feed (specify unit)




Table 5-7: Data collection for feed transport if information on actual fuel use cannot be collected Activity data Unit Quantity Source and method

of measurement (if relevant)

Vehicle type 1 share % [-] • Load capacity Tonne • Technology EURO-class • Distance per trip Km • Load fraction % • Empty return

transport %

• % biofuel % Vehicle type 2 share % etc. The load capacity and the EURO-class are the basic parameters that define the emissions profile of the transport vehicle. The other parameters affect its efficiency.

7 This secondary data source has been acquired by the EC but is not available yet 8 The level of detail and PEF compliance of secondary data for transport differ. It depends on the acquired secondary dataset which LCI data for transport should be selected. In the final version of the PEFCR specific further guidance on secondary data selection for transport will be given.

29


Minimum data quality is 3 if no comparison is involved. If data are important for comparison minimum data quality is 1.6. If the data quality level of 3 is not achieved, default values shall be applied for transport. The data quality rating system for transport shall be conducted according to Table 5-8.

Table 5-8: Data quality rating system for transport. Quality


C TiR P (related to

distance)e

TeR (related to transport means and logistic

parameters)

GR related to transport distances

Very good9

1 Data on actual fuel use, fuel composition and technology (EURO 1-6)

Data10 cover exactly the time period in the scope of the study

≤ 10% The technology(ies) is/are specific for transport of feed product(s) in scope

The data concern feed transport scenarios specific for the feed and feed mill in scope based on weighted average.

Good 2 Data on actual fuel use

Data partially cover the time period in the scope of the study

10% to 20%

The technology(ies) is/are companies feed transport

The data concern feed transport scenarios specific for the feed in scope based on company average

Fair 3 Data on transport distance and transport means


20% to 30%

The technology(ies) is/are representing countries feed transport in scope

The data concern average company data (not being the scope of the study)

Poor 4 Data on transport distance


30% to 50%

The technology(ies) is/are representing countries transport of agricultural products

The data concern country averages

Very poor

5 No data Data are not older than 6 years with respect to the time period in the scope of the study

> 50% The technology(ies) is/are representing EU transport of agricultural products

The data concern European averages

The data quality score for outbound transport DQRo is:

DQRo5

PCGRTeRTiR ++++=

TeR: Technological-Representativeness GR: Geographical-Representativeness TiR: Time-Representativeness C: Completeness P: Precision/uncertainty

9In some cases referred to as “excellent” 10The reference time is the one when data have been originally collected and not the publication/calculation date. In case there are multiple data, the oldest is the one against which the calculation should be made.

30

5.3.5 Data of (inbound) transport to feed mill Feed business operators producers shall collect the following information of logistics from their supply of feed materials if they can know the origin of production.

• The last production location of the feed ingredient before transport to the feed mill and its distance to the feed mill (in case of a processed material this is the processing plant11, in case of a crop this is the location of cultivation).

• The average transport scenario of the feed ingredient differentiated per transport means. Table 5-9 gives an example how this could look like.

Table 5-9: Example of transport data to be collected from suppliers of the feed materials per feed material. Activity data Unit Quantity Source and method of measurement (if

relevant) Feed material A

• Supplier 1 Share 20% Internal procurement records (3 years average)

o Transport means x Truck 32 tonnes load capacity

Information from supplier

o Distance Km 150 Distance of fastest route determined by route planner software …

o Load fraction % 100% Information from supplier o Load fraction return % 100% There is no additional information collected

so worst case assumption is applied o Transport means y Barge 1000 tonnes Information from supplier o Distance Km 300 Distance of fastest route determined by

route planner software … o Load fraction % 100% o Load fraction return % 100%

• Supplier 2 o Transport means x Truck 32 tonnes Information from supplier o Distance Km 420 Distance of fastest route determined by

route planner software … o Load fraction % 100% Information from supplier o Load fraction return % 100% There is no additional information collected

so worst case assumption is applied Etc. If the feed business operator cannot determine the transport distances and modes, default data on distances and modes can be used (however production location still needs to be known), see annex 7. The data quality score for inbound transport DQRi can be determined using Table 5-8 and the following formula:

DQRi5

PCGRTeRTiR ++++=


11 For processed ingredient, the first step of transport, from the place of cultivation to the place of processing is covered by the secondary databases for feed ingredients, see section 5.4

31

5.3.6 Feed ingredient production process The environmental footprint of a cradle to gate feed product is mainly determined by the environmental footprint of its feed ingredients. In many cases secondary data will be used, as the process is not run or under the control of the company applying the PEFCR. However, when considered relevant and feasible, it is possible to model the production of feed ingredients and to use primary data instead of secondary data. Replacing secondary data for feed ingredients shall fulfil the same requirements described in the sections 5.3.6.1 and 5.3.6.2.

5.3.6.1 Cultivation The modelling requirements of this section shall apply to any primary cultivation data replacing default secondary data in background processes if considered feasible and relevant:

This section summarizes and translates the LEAP guidelines (FAO LEAP, 2015) to PEFCR requirements. Further guidance on how to do the assessment in practice can be found in the LEAP guidelines.

Cultivation includes all field and storage operations until the product is being sent for transport to the feed mill or further processing. Cultivation may also involve land use change. For land use change, the modelling guidelines of PAS 2050:2011 and the supplementary document PAS 2050-1:2012 for horticultural products shall be followed.

The following inputs shall be quantified per hectare of crop cultivation:

• Seeds, NPK-fertilizers, manure, fuels, irrigation water, crop protection products(), chemicals, auxiliary materials taking into account crop rotation and steady state of production (averaging over more years, see LEAP guidelines and PAS2050/1 (BSI, 2012) for further guidance in case of perennial crops)

• For the LCIs of production and logistics of agricultural inputs (fertilizers, crop protection products, fuels etc.) the EC acquired PEF datasets or the GFLI dataset shall be used.

The following economic outputs shall be quantified per hectare:

• Main crop product (mass, DM, financial value, gross energy content (LHV)) • Co-product(s) (mass, DM, financial value, gross energy content (LHV)) • Residual materials that remain on the field or in soil (mass, DM) • Residual materials that are burnt and associated emissions • Waste flows and destination

The following background information shall be collected on region of cultivation and farm management:

• Country of production • Blue water consumption in country/region of production • Land transformation in past 20 years according to PAS2050/1 (BSI, 2012) • Description of farm practices (as meta information)

o Farm rotation scheme o Tillage/ no tillage o Method of crop protection products application

32

o Method of manure/fertilizers application

The following outputs shall be quantified per hectare

1. Emissions from combustion of fuels 2. CO2 emissions related to application of fossil carbon containing products (lime, peat,

etc.) 3. N2O emissions related to manure and fertilizer application and to crop residues NH3

emissions related to manure and fertilizer application 4. NO3 emissions to water related to manure and fertilizer application 5. P emissions to soil and water related manure and fertilizer application 6. Heavy metals emissions related to manure and fertilizer application on basis of mass

balance approach 7. Crop protection products emissions

These emissions shall be calculated according to broadly accepted method of calculating emissions, such as methodologies outlines in IPCC 2006 Guidelines (IPCC, 2006), EEA (European Environment Agency, 2013) or other reputable source.

All economic inputs and elementary flows (resource use and emissions) per hectare shall be related to the net yields (after losses) per hectare.

Allocation in case of crop rotation and co-production (e.g. wheat and straw) shall be treated according to the decision-tree and recommendations mentioned in the LEAP guidelines.

The requirements of the tender for Feed LCI datasets recently published may trigger an update of this section.

33

5.3.6.2 Processing of crops When considered relevant and feasible, it is possible to model the production of feed ingredients and to use primary data instead of secondary. The modelling requirements of this section shall apply to any primary processing data replacing default secondary data in background processes. For further guidance see the LEAP guidelines (FAO LEAP, 2015).

The following inputs shall be quantified per tonne of feed ingredient input.

• fuels, electricity, auxiliary materials taking into account steady state of production (averaging over appropriate period)

• for the LCIs of production and logistics of these inputs the EC acquired PEF datasets or the GFLI dataset shall be used or any other commercial dataset if not available in the previous two.

The following outputs shall be quantified:

• Product of interest (mass, DM, financial value, gross energy content (LHV)) • Co-product (mass, DM, financial value, gross energy content (LHV)) • Residual materials that are considered to have zero value (mass, DM) • Waste flows and destination

The following background information shall be collected on region of production:

• Country of production • Blue water consumption in country/region of production

The following outputs shall be quantified

• Emissions from the combustion of fuels • Process specific emissions to water, air and soil

The requirements of the tender for Feed LCI datasets recently published may trigger an update of this section.

34


Minimum data quality is 3 if no comparison is involved. If data are important for comparison minimum data quality shall be 1.6. The data quality is measured with default metrics published in the guidance document 5.2, see Table 5-10.

Table 5-10: Data quality rating system for feed ingredients production. Quality


C TiR P TeR GR

Very good

1 All 15 PEF Impact Categories

Data12 are not older than 4 years with respect to the release date or latest review date

≤ 10% The technologies covered in the dataset are exactly the one(s)modelled

The processes included in the dataset are fully representative for the geography stated in the title and metadata

Good 2 14 PEF Impact Categories (and all 10 categories classified I or II in ILCD are included13)

Data are not older than 6 years with respect to the release date or latest review date

10% to 20% The technologies modelled are included in the mix of technologies covered by the dataset

The processes included in the dataset are well representative for the geography stated in the title and metadata

Fair 3 12-13 PEF Impact Categories (and all 10 categories classified I or II in ILCD are included)


20% to 30% The technologies modelled are representative of the average technology used for similar processes

The processes included in the dataset are sufficiently representative for the geography stated in the title and metadata

Poor 4 10-11 PEF Impact Categories (and all those covered are classified I or II in ILCD)


30% to 50% Technology aspects are different from what described in the title and metadata

The processes included in the dataset are only partly representative for the geography stated in the title and metadata

Very poor

5 Less than 10 PEF Impact Categories (and all those covered are classified I or II in ILCD)

Data are older than 10 years with respect to the release date or latest review date

> 50% Technology aspects are completely different from what described in the title and metadata

The processes included in the dataset are not representative for the geography stated in the title and metadata

12The reference time is the one when data have been originally collected and not the publication/calculation date. In case there are multiple data, the oldest is the one against which the calculation should be made. 13The 10 impact categories classified in ILCD Handbook as category I or II are : Climate change, Ozone depletion, particulate matter, ionizing radiation human health, photochemical ozone formation, acidification, eutrophication terrestrial, eutrophication freshwater, eutrophication marine water, resource depletion mineral fossil and renewable.

35

The data total quality score for the production of a feed ingredients production DQRfp is:

DQRfp5

PCGRTeRTiR ++++=


5.4 Secondary data selection Following the requirement of the PEF guidance, the recommended sources of secondary data are listed below:

- PEF compliant datasets in the International Life Cycle Data Network: the first source of data is the European Commission, which has decided to buy LCI datasets, to facilitate the implementation of the PEFCRs/OEFCRs developed during the Environmental Footprint pilot phase. This process includes feed LCI datasets, which will be made publically available through the ILCDN. This source of data is referred to as “EC feed database” in this document

- Global Feed LCA Institute (GFLI)14. The Global Feed LCA Institute is an independent feed industry initiative with the vision to develop a freely and publically available Feed Life Cycle Analysis (LCA) database and tool. The objective of the GFLI is to support meaningful environmental assessment of livestock products and to stimulate continuous improvement of the environmental performance in the feed industry. The GFLI data will be PEF compliant. This source of data is referred to as “GFLI database” in this document.

5.4.1 Select data on feed ingredients production related to BOM This section gives guidance on which secondary data to use depending on availability of datasets in the two secondary databases this PEFCR refers to: the EC feed database and the GFLI EU database. The detailed content of the two data sources will be described in an annex to this PEFCR when the EC process of buying feed LCI datasets is completed. The EC feed database is always the preferred option but may not contain all necessary data or not specific for certain countries. In those cases where more specific data is available in the GFLI EU database, this database shall be used. The figure below guides the user through the different steps leading to the choice of the secondary database.

This section is setup from the perspective of the PEF operator that collects a BOM (feed ingredients list) and has the task to match his feed ingredients as good as possible to the datasets available in the two secondary databases.

14 http://www.ifif.org/pages/t/Global+Feed+LCA+Institute+%28GFLI%29

36

Figure 5-1: Data selection decision tree. Note that in some cases the DQR score for that given process need to be increased (see box on right)

37

5.4.2 Identification of proxies In some cases no data is available in the EC or GFLI databases. In that case proxies need to be used. Using proxies always trigger lower data quality (i.e. higher DQR).

Case 1: the ingredient is available in the EC or GFLI database, but not for the appropriate origin (e.g. corn from Mexico. Data on corn may be available but not for corn from Mexico)

Proxy: Use the world average for the same ingredient and modify data quality according to Figure 5-1. Error! Reference source not found.

Case 2: for a given ingredient (processed or unprocessed), there is no data available in the EC database nor in the GFLI database

Proxy: Use the average of the appropriate group of ingredients according to GFLI classification and increase data quality according to Figure 5-1Error! Reference source not found.. The GFLI will provide these group average data sets, e.g. a proxy dataset for “Vegetable meals”.

The use of proxy data shall be reported.

5.4.3 Select secondary energy production data to connect to primary activity data

Further guidance will be given here after publication of the energy and transport data by the EC

5.4.4 Select secondary transport data to connect to primary activity data Further guidance will be given here after publication of the energy and transport data by the EC

38

5.5 Calculating the overall data quality score If all data quality scores per process fulfill the minimum requirements an overall score shall be calculated per environmental impact by

The overall data quality score DQRtotali is calculated per environmental impact on the basis of the following formula:

𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝑙𝑖 = �𝐷𝐷𝐷𝐷𝐷 ∗ 𝐷𝑝𝑖 +�𝐷𝐷𝐷𝐷 ∗ 𝑏𝑝𝑖 + �𝐷𝐷𝐷𝐷 ∗ 𝑐𝑝𝑖 + �𝐷𝐷𝐷𝐷𝐷𝐷 ∗ 𝑑𝑝𝑖

+ �𝐷𝐷𝐷𝐷𝐷𝐷 ∗ 𝑒𝑝𝑖

Where: • 𝐷𝑝𝑖 , 𝑏𝑝𝑖, 𝑐𝑝𝑖,𝑑𝑝𝑖 , 𝑒𝑝𝑖 = The relative contribution of process p to impact i in the respective. • DQRfm = DQR of feed mill operation data(s); a =contribution of feed mill operation(s) to

Environmental Impact (EI) score • DQRo= DQR outbound transport(s); • DQRi = DQR of inbound transport(s); • DQRfip = DQR of feed ingredient(s) primary data; • DQRfis =DQR of feed ingredient(s) secondary data;

And ∑ 𝐷𝑝𝑖

𝑝 + ∑ 𝑝 𝑏𝑝𝑖 +∑

𝑝 𝑐𝑝𝑖 + ∑ 𝑝 𝑑𝑝𝑖 + ∑

𝑝 𝑒𝑝𝑖 = 1 Meaning that the relative contributions of all processes combined should equal 1, for every impact category. The overall data quality score (DQRtotal) can then be calculated by calculating the average of the scores per impact category. In this calculation it is assumed that the data quality score of secondary databases for energy production and transport processes does not negatively affect any of the DQRs calculated according to this PEFCR. The DQR overall values and the breakdown to the 5 calculated DQRs shall be reported per impact category, and averaged to determine the overall DQR. The minimum data quality requirements per life cycle stage are listed in Table 5-11. Table 5-11: Data quality requirements. DQR

Total DQRfm

Feed mill DQRo

Outbound transport

DQR1 Inbound transport

DQRfip Feed

ingredient primary

DQRfis Feed

ingredient secondary

Required minimum level 3 3 or 1.6 3 - 1.6 3 - 1.6 3 -1.6 3

5.6 Reporting data gaps and proxies Proxies that are made in case of filling in data gaps shall be reported with their effect on the data quality score.

5.7 Minimum data quality level for PEF communication Results shall not be communicated externally as a PEF compliant study if the minimum data quality levels are not met.

39

5.8 End-of-life stage In case of packed feed ingredients or compound feed, the packaging materials will be discarded at the farm of use for which the PEF End of Life Formula should be applied by the LCA operator for the study. The default end of life formula shall be used and if there is no data collected the default worst case assumptions shall be applied: there is no secondary material used during production and no recycling of materials after use.

( ) *13

2,,,,3

*211

221

2221

DDelecSEelecERheatSEheatERER

P

SVoLrecyclingErecycledV

ERERREXLHVEXLHVER

QQ

EERERER

×−

−−+××−××−×

+

×−×+×+×

−

5.9 Requirements for multifunctional products and multiproduct processes allocation Allocation shall be conducted according to the LEAP guidelines. This implies;

Transport

Allocation of empty transport kilometers shall be done on the basis of the average load factor of the transport that is under study. If no supporting information is available, it shall be assumed that 100 percent additional transport is needed for empty return (check with default values in sections 5.3.4 and 5.3.5).

Allocation of transport emissions to transported products shall be done on the basis of physical causality, such as mass share, unless the density of the transported product is significantly lower than average so that the volume transported is less than the maximum load.

Processing of farm products to feed materials

The baseline allocation for allocation at processing is Economic and two physical alternatives shall be calculated in the sensitivity assessment to ensure that the influence of allocation on the results is transparent. Economic allocation will be conducted on the basis of the method and default allocation fractions in Annex 10 unless it is clear that the process under study deviates from the default processes. Then economic allocation will be done according to the procedure as being explained in the LEAP guidelines for feed.

Allocation of co-products from a crop at the farm The baseline allocation for allocation at processing is Economic and two physical alternatives shall be calculated in the sensitivity assessment to ensure that the influence of allocation on the results is transparent. Economic allocation will be conducted on the basis of the method and default allocation fractions in Annex 10.

Allocation issues related to assigning company data to production systems The LEAP feed guidelines give detailed guidance in chapter 8 on how to assign company data (on farms and processing plants) to productions systems. These shall be followed.

40

6. Benchmark and classes of environmental performance Feed is an essential input for the production of food-producing animals. It is basically an intermediate product whose composition varies depending on the desired response of the animal and associated nutritional requirements depending and the availability and prices of feed ingredients. Measuring the impacts associated with the production of feed, as well as the feed performance on the farm is necessary in order to achieve meaningful LCAs of food-producing animals.

The different types of compound feed have different applications and different functions. Food-producing animals are fed differently according to various parameters such as animal species, stage of development, and other farm-specific conditions (e.g. dairy feed is formulated to balance the diet with forages produced on the farm).

The feed performance in terms of production per unit of feed is closely linked with farm management practices. In other words, the performance of the same feed used in two similar farms can vary significantly according to the farm’s specific conditions (breed, animal health status; etc.…) and management. This is extremely important to bear in mind when considering the improvement of the environmental performance of feed production. Reducing the environmental footprint of feed production without taking into account the potential consequences on its efficiency in the use phase would be very counterproductive.

Customers of compound feed are professional livestock farmers meaning that communication relative to compound feed is on a business to business basis only.

Professional livestock farmers do not require a simplified comparison with a representative product which does not fulfil the same function as the feed they have bought. The PEF profile of the representative product can nevertheless be useful to check whether the results of a PEF study are credible and match in terms of order of magnitude.

Following the modular approach implemented in this PEFCR, the livestock farmers (or other relevant downstream partner) will be provided with the full results of upstream emissions of feed up to the delivery to the farm (for all impact categories) and with the necessary information for LCA modelling of on-farm feed use.

The comparison of the PEF profiles of different compound feeds shall therefore:

• only take place when it is clear that they fulfil the same function, i.e. in the context of cradle to grave PEF study of identical animal products (e.g. one kilogramme of eggs on similar farms with two types of feed) and;

• only be interpreted as part of the complete interpretation of the PEF profile of the animal product at stake.

The possibility to develop a benchmark for animal products, and the potential limitations are described in the animal products PEFCRs.

41

7. Interpretation The interpretation of the PEF results shall integrate the following elements:

- Assessment of the robustness of the PEF model - Identification of hotspots - Estimation of uncertainty and - Conclusions, recommendations and limitations

7.1 Robustness of the PEF model The PEF results rely on impact assessment methods which are not equally robust. They can be classified as follows:

High robustness

- Climate change - Ozone depletion - Particulate matter

Medium robustness

- Ionizing radiation - Photochemical ozone formation - Resource depletion (fossil – mineral) - Acidification - Eutrophication (marine, terrestrial, freshwater)

Low robustness

- Human toxicity (cancer and non-cancer) - Freshwater ecotoxicity - Land use - Water resource depletion

During the EF pilot phase, particular attention was paid to the toxicity and ecotoxicity related impacts which rely on USEtox. The results for these impact categories are currently not considered robust enough to support benchmarking and public communication.

7.2 Identification of hotspots For each impact category, hot spots shall be identified concerning life cycle stages, process and elementary flow. In each case, the assessment shall account for the cumulative required to reach at least 50% to any impact category before normalisation and weighting (from the most contributing in descending order)

The procedure to identify hotspots is detailed in Annex XX

42

7.3 Estimation of uncertainties The robustness of the impact assessment methods shall be taken into account when estimating the uncertainty (see previous section).

In addition, the uncertainties associated with the data robustness shall also be reported:

- Uncertainty related to primary data collection - Uncertainty associated with use of secondary data, either from the EC database or from the

GFLI database.

7.4 Conclusions, limitations and recommendations The conclusions of the PEF study shall always be accompanied with a clear description on the limitations of the study.

The list below is a non-exhaustive compilation of limiting factors for a PEF study

- Assumptions made throughout the study - Data gaps in secondary database - To be expanded

43

8. Reporting, Disclosure and Communication

8.1 Context Compound feed is an intermediate product sold to professional livestock farmers. At this stage, further discussion is needed to clarify the target of the communication of the environmental performance of compound feed. The most straightforward target group are the farmers. In the situation where the results of a compound feed PEF study are used to support the communication of the environmental performance of animal products, the possibility to communicate directly from the feed company to the meat/dairy/egg/fish company could also be envisaged.

The communication of environmental information along the feed and food supply chain is described in the figure below.

Figure 8-1: Communication of environmental information along the feed and food supply chain.

The purple arrows describe the potential flows of communication of environmental information from a feed company to its partners. As direct customer, the livestock farmer is considered as the main target for communication. In the situation where the results of a compound feed PEF study are used to support the communication of the environmental performance of animal products, the possibility to communicate directly from the feed company to the meat/dairy/egg/fish company can also be envisaged. Finally, direct communication from feed company to retailer can also happen, although less common.

During the communication testing phase, the below-described PEF declaration will be tested, mainly with farmers but to the possible extent, also with animal products companies. Focus group discussions will also be organised with farmers and/or animal products companies, if possible.

44

8.2 Recommendation for communication of environmental information The current recommendations for communicating the PEF results are listed below. They may vary according to the purpose of the study.

Table 8-1: recommended communication vehicles, dependent on purpose. Purpose Communication vehicle Cradle to gate Feed PEF study to be used for animal products PEF study

PEF declaration supported by PEF external communication report

Cradle to gate Feed PEF study with external communication


Cradle to gate Feed PEF study including product comparison

PEF external communication report

Cradle to gate Feed PEF study for performance tracking

PEF performance tracking report

8.2.1 PEF external communication report A template for PEF external communication report is available in chapter 8 of the PEF Guide (may be integrated as an annex of the final PEFCR)

8.2.2 PEF declaration From an intellectual property perspective, the information relative to the nutritional analysis data (see section 5.3.2) may be sensitive. A stepwise approach is therefore recommended for the development of a Feed PEF declaration

o Step 1 PEF declaration, containing LCA results of cradle to gate Feed PEF study, but not meant for further LCA modelling of on-farm activities. This Step 1 PEF declaration would not contain sensitive information on the feed composition

o Step 2: “complete” PEF declaration, with additional information necessary for modelling of on-farm activities. This complete PEF declaration would be available upon request and the conditions of disclosure would be discussed in a business to business bilateral relationship.

8.2.3 PEF performance tracking report The main characteristics of a PEF performance tracking report are described in section 3.14.2 of the PEF Guidance (may be integrated as an annex of the final PEFCR).

45

9. Verification This section will be completed when the consolidated verification report of the supporting studies for all pilots will be available.

According to ISO 14044 and to the PEF Guide, any PEF study claiming to be in line with the PEF Guide and any PEF study for external communication shall be critically reviewed in order to ensure that

- The methods used to carry out the PEF study are consistent with the PEF Guide - The methods used to carry out the PEF study are scientifically and technically valid - The data used are appropriate, reasonable and met the defined data quality requirements - The interpretation of results reflects the limitations identified - The study report is transparent, accurate and consistent

The development of a PEFCR should simplify the review process. This however needs to be confirmed by the consolidated report on the verification of the supporting studies.

Likewise, when secondary data are used from a source which has been acknowledged as PEF compliant, these data shall not be verified further, especially regarding methodological consistency, completeness and uncertainty.

The Feed Technical Secretariat recommends focusing the verification (critical review) on the following aspects

- Is the list of feed ingredients representative for the feed under study, and does it accurately reflect the time related variability?

- Is the list of feed ingredients correctly connected to the available secondary data? - If proxies have been used, are these determined in accordance to the procedures described

in this PEFCR, and has the data quality been modified accordingly?

46

10. Reference literature European Commission. (2013). 2013/179/EU: Commission Recommendation of 9 April 2013 on the

use of common methods to measure and communicate the life cycle environmental performance of products and organisations. Official Journal of the European Union.

European Commission. (2016). Guidance for the implementation of the EU Product Environmental Footprint (PEF) during the Environmental Footprint (EF) pilot phase - draft version 5.2.

FAO LEAP. (2015). Environmental performance of animal feeds supply chains - Guidelines for assessment. Retrieved from http://www.fao.org/partnerships/leap/resources/resources/en/

Food SCP RT. (2013). ENVIFOOD Protocol Environmental Assessment of Food and Drink Protocol. Brussels, Belgium.

Goedkoop, M., Heijungs, R., Huijbregts, M., Schryver, A. De, Struijs, J., & Zelm, R. Van. (2009). ReCiPe 2008 A life cycle assessment method which comprises harmonised category indicators at the midpoint and the endpoint level. Report 1 Characterisation. Den Haag.

JRC, & European Commission. (2015). Baseline Approaches for the Cross-Cutting Issues of the Cattle Related Product Environmental Footprint Pilots in the Context of the Pilot Phase.

Secretariat of the Convention on Biological Diversity. (2010). Global diversity Outlook 3.

Technical Secretariat for the Feed Pilot. (2015). PEF screening report in the context of the EU Product Environmental Footprint Category Rules ( PEFCR ) Pilots - Feed for food producing animals.

Technical Secretatiat for the Feed pilot. (2014). Feed pilot - scope and representative product.

Technical Secretatiat for the Feed pilot. (2015). PEFCR Feed for food producing animals - Draft version 1.1 for EF Steering Committee, (October), 1–40.

11. Supporting information for the PEFCR This section will be completed at a later stage.

47

List of annexes Annex 1 – Representative product The screening report and its annexes are available from the stakeholders’ workspace of the wiki page. Annex 2 – Supporting studies Available from the stakeholders’ workspace of the wiki page. Annex 3 – Main lessons learnt from the supporting studies Annex 4 – summary of main variations of the requirements defined in this PEFCR, according to the purpose of the study Annex 5 – Situations which require a modification of the system boundaries Annex 6 – Guidance for the calculation of carbon content and gross energy This section will be completed at a later stage

Annex 7 – Default data for transport This section will be completed at a later stage, and will contain a list of default distances and modes between different locations (e.g. the average transport distance between Malaysia and Germany, and default transport modes or the average transport distance and mode for transport within France)

Annex 8 – Background data This section will be completed at a later stage, in relation with the tender procedure for LCI datasets

Annex 9 – Template for PEF declaration Annex 10 - Method and default allocation fractions for feed ingredients

This section will provide a list of default allocation fractions and guidance on how to use this to calculate the allocation percentages for a list of common feed ingredients.

48

Annex 3

Main lessons learnt from the supporting studies

- The historical approach, with requirements to calculate an average formula for a period of three years, is extremely difficult to implement. The concern was expressed that an average based on such a long period was not representative of the current situation. The fact that the composition of some feeds is more stable than others over time is another factor to take into account.

- The sequence of recommendations should be revised, since data collection and data quality assessment are often performed one after the other

- The benchmarking approach is not meaningful for an intermediate product such as compound feed

- The results related to toxicity and eco-toxicity are difficult to understand, interpret and thus communicate

- The types of transport proposed are difficult to understand for a non-expert

- Data gaps in databases, in particular for additives, vitamins and minerals

- Identifying the countries of origin for all the feed ingredients is also challenging, even with collaboration of suppliers. It is however likely that this information will be requested on a more and more regular basis. This will be challenging to achieve in the short to medium term as it is not a legal requirement and hence will have commercial implications.

- It is difficult to link energy consumption in a feed mill to a specific product (but this is currently not required in the PEFCR)

- Requirements should be defined in a way that is auditable

49

Annex 4: summary of main variations of the requirements defined in this PEFCR, according to the purpose of the study

Different requirements apply according to the purpose of the PEF study. The table below shows the relation between purposes and PEFCR requirements and the sections where they are described. The rationale of this distinction in purposes is that methodology, accuracy and effort for data collection should be proportionate to the purpose of the study.

Overview of potential purposes of PEF study on feed in relation to requirements

Purpose 1: Feed PEF study to provide information for animal PEF study

Purpose 2. Cradle to gate PEF study without comparison

3. Cradle to gate PEF study including a comparison: 3a comparison between existing or new alternative products

3b comparison in time (trend monitoring)

System boundaries [CH 4.5]

System boundaries can be adapted to goal of study

Default system boundaries



Additional modelling [CH 4.7]

To be determined per animal PEF study

NO Depending on comparison

Depending on comparison

Type of data [CH 5]

To be determined per animal PEF study

Primary data for feed mill operation. Rest can be secondary data

Primary data for feed mill operation; Primary data for other processes shall be collected if they are relevant for comparison

Primary data for feed mill operation; Primary data for other processes shall be collected if they are relevant for comparison

Communication [CH. 8]



PEF external communication report

PEF performance tracking report

50

Annex 5: Situations which require a modification of the system boundaries

Situation Solution 1. the nutritional value or composition of the

feed changes in a way so that it affects the production performance of food producing animals,

Full Lifecycle approach on the basis of measured or estimated use phase at production of food producing animals.

2. the chemical composition of the feed changes so that it affects the environmental performance of the farming systems where the feed is consumed (including digestion and manure management) or where the manure is applied,

Full Lifecycle approach applying parameterized models for digestion and manure management. Lifecycle shall be extended with animal manure application if relevant changes occur due to content of N, P, Zn and Cu in manure.

3. There is an intentional change: • in consumption of constrained /residual co-

products or • in origin of supply of feed material

production on a scale that requires additional LCA modelling (see below for further explanation). “If the change in the feed lifecycle involves more than 5% of the national supply of a raw material or area for cultivation then the need for additional LCA modelling should be considered.”

Cradle to gate life cycle but with additional modelling to calculate impacts related to the supply and demand of raw materials or reallocation of processes.

Additional explanation for situation 3

Introducing a threshold in relation to the scale of national production volume/area of a feed material follows the line of reasoning in the ILCD LCA guide where a split between micro and meso/macro level is made. The ILCD guide does not give concrete recommendations that help the LCA user the scale in quantitative terms. The threshold of 5% is arbitrary but it is more practical than the ILCD qualitative approach and gives the PEFCR user a tool to prove that additional modelling is not required for his study.

The 5% threshold is sometimes lower than for instance the annual fluctuations in European feed material availability. The yields of crops in a certain country and the related production volume of the processed crops can fluctuate 15% or more over the years [sugar and grain statistics]. This means that the availability of feed materials and their price also strongly fluctuate through the years. The 5% is a considerable part of the normal fluctuation but at the same time it is still far away from the more extreme fluctuations.

The introduction of rape seed cultivation in Europe for bio-fuels is an interesting example. This did not have big implications on land use and shifts in commodity flows right away but it got implications

51

when it grew to a certain scale. It triggered for instance the replacement of significant volumes of imported soybean meal by domestic rapeseed meal in compound feed

The biofuels case has been extensively assessed and could give a bit more information on if and how a threshold should be defined.

Example regarding using additional constrained feed materials.

A feed company reduces the environmental impact of its dairy feed by using in the composition of the dairy feed on average 10% dried sugar beet pulp and 15% palm kernel expeller. Normally the composition would on average be 6% dried sugar beet pulp and 10% palm kernel expeller. The company is located in the Netherland and sells 2 Mt of this dairy feed. The dried beet pulp is bought in the Netherlands and the additional 4% equals to 80.000 ton dried beet pulp which is nearly 10% of the Dutch market of available beet pulp. The palm kernel expeller is bought from Malaysia that produces 2.4 Mt of expeller per year. So the 5% additional expeller equals 100.000 ton which is 4% of Malaysian production. In this case, the feed company has to assess if its additional use of dried beet pulp leads to changes in the market of processing and use of beet pulp that changes the initial estimate of improved environmental performance. The increased demand of dried beet pulp will not affect the amount of sugar production due to the low contribution to revenues of a sugar factory. So there should then be a change in the processing of beet pulp or the destination of the pulp. There could be an increased drying of beet pulp at the cost of fresh beet pulp that was directly sold to (other) dairy farms. In this case energy use becomes higher (the impact of extra drying is bigger than reduced transport) and increases the environmental impact. (Also the feeding performance of wet and dry beet pulp can be different). So the additional assessment would show that the substantial increased use of dried beet pulp cannot be considered as an environmental improvement and should not be communicated as such.

52

Annex 7: Default activity data for transport

This annex will contain a list of default transport activity data and background datasets to be used when no primary data is available. How this annex will be filled, depends on the background data that will become available as a result of the EC tender.

53

Annex 9: Template for PEF declaration

Product Environmental Footprint (PEF) declaration for [name of the product] according to Product Environmental Footprint Category Rules Feed for food-producing animals [reference of the PEFCR]

Date of the declaration

Time Validity

Purpose of the study

Company introduction: identification of the organization making the PEF declaration Name of the company Address Other relevant information

Product identification Trade name Reference Other common name of the product Description of the product (e.g. broiler feed) Intended use of the product

Picture of production site

(optional)

Picture of the product

(recommended)

54

Description of the PEF study Functional unit Scope System boundaries Information on life cycle stages which are not considered, if the study deviates from the PEFCR Statement that environmental declarations from different programmes may not be comparable Additional PEF-related general information can be found here: http://ec.europa.eu/environment/eussd/smgp/communication/what_is_pef.htm For explanatory material and supporting information related to the calculations in this study, please contact: [add contact point]

Visual description of system boundaries (recommended)

http://ec.europa.eu/environment/eussd/smgp/communication/what_is_pef.htm

55

Life Cycle Impact Assessment results (for one tonne of feed delivered to the farm) Tables presenting characterised results in a written form with absolute figures, total results, and results per life cycle stage (recommended) Table 1 - category I impact categories: high robustness

Impact category Unit Total Raw material acquisition and pre-processing

Feed production

Delivery to farm

Climate change Kg CO2 eq Climate change excluding land use change

Kg CO2 eq

Ozone depletion kg CFC-11 eq

Particulate matter kg PM2.5 eq

Table 2 - category II impact categories: medium robustness


Feed production

Delivery to farm

Acidification molc H+ eq Terrestrial eutrophication molc N eq Fresh water eutrophication

kg P eq

Marine eutrophication kg N eq Photochemical ozone formation

kg NMVOC eq

Mineral fossil and renewable resource depletion

kg Sb eq

Ionizing radiation kBq U235 eq

Table 3 - category III impact categories: low robustness


Feed production

Delivery to farm

Human toxicity, cancer effects CTUh

Human toxicity, non-cancer effects CTUh

Freshwater ecotoxicity CTUe Water resource depletion m3 water eq Land use kg C deficit

Results for less robust impact categories (categories II and III) should be handled with care.

56

Characterised results presented in a visual form, showing the contribution of the different life cycle stages (recommended)

Optional: normalised results

57

Additional information: biodiversity score (optional)

Discussion needed to identify best format for communication

58

Part 2 of PEF declaration: information needed for further LCA modelling at farm level (optional, available upon request with conditions of disclosure discussed in B to B relationship)

Nutritional analysis data

Composition parameters All values per kg “as is”

Dry matter content

Gross energy value

Digestible energy (range)

Carbon biogenic

Carbon fossil

Ash

Phosphorous

Nitrogen

Zinc

Copper

Information on feed delivery

Distance

Transport means Type of truck, load capacity

Packaging material If applicable

pefcr feed for food producing animals version 1 7. interpretation 41 7.1 robustness of the pef model...

Documents