Comparison

of the requirements of the

GHG Protocol Product Life Cycle standard
and the ISO 14040 series


Bo P. Weidema
2.-0 LCA consultants, Aalborg, Denmark
January 2022

Preface and acknowledgements
This report presents a comparison of the requirements of the GHG Protocol Product Life Cycle Standard
and the ISO 14040 series, with the purpose of assessing to what extent a Life Cycle Assessment study
performed according to ISO 14044:2006 and ISO 14067:2018 can be applied for reporting according to the
GHG Protocol, and what possible additions or modifications may be required for this purpose. The study
has been conducted for Mærsk A/S in January 2022 by Bo P. Weidema of 2.-0 LCA consultants.

Copenhagen, 17th January 2022

2.-0 LCA consultants, Denmark

2|P a g e
Contents
Executive summary ...................................................................................................................................4
1 Objective and scope...........................................................................................................................6
2 Attributional or consequential modelling? .........................................................................................7
3 Carbon accounting ........................................................................................................................... 10
4 Additional requirements of the GHG Protocol for the Product Life Cycle .......................................... 12
5 ISO 14044 series requirements not included in the GHG Protocol .................................................... 12
6 Conclusion ....................................................................................................................................... 13
References .............................................................................................................................................. 13

When citing this report, please use the following reference:

Weidema B P. (2022). Comparison of the requirements of the GHG Protocol Product Life Cycle Standard
and the ISO 14040 series. Aalborg, Denmark: 2.-0 LCA consultants.

3|P a g e
Executive summary
Both the ISO LCA standards (ISO 14044 and ISO 14067) and the GHG Protocol for Product Life Cycles are
open for different interpretations on how to link unit processes into products systems.

In the consequential approach activities are included in the product system to the extent that they are
expected to change as a consequence of a change in demand for the functional unit, tracing each required
product input, physical or monetary, through the marginal supplier(s) of each product, following verifiable
cause-effect relationships, including the consequences of excess supply of by-products from joint production.

In the attributional approach, inputs and outputs are attributed to the functional unit of a product system
by linking and/or partitioning the unit processes of the system according to a normative rule. In the ISO
14044 standard, the default approach for allocating to the products of joint production is allocation based
on their economic value (revenue).

At first sight, the GHG Protocol for Product Life Cycles is clearly declared as following an attributional
approach. However, a closer analysis reveals unclarities and contradictions that opens up for several
interpretations, including the specific consequential and attributional interpretations of ISO 14044.

The requirements of the GHG protocol goes beyond the requirements of ISO 14044 on several issues related
to carbon accounting, use of primary data for all processes under their ownership or control, reporting of
procedures and process maps, and separate reporting of different parts of the overall climate footprints.
These requirements are complementary and not in conflict with to those of ISO 14044. In a few cases,
requirements to exclude specific issues from the climate footprint results (the effects of offsetting and
delayed emissions) are modified by allowing separate reporting of these effects.

In conclusion, an ISO 14044 compliant LCA, whether attributional or consequential, can be used to provide
data for the GHG reporting according to the GHG Protocol, provided that 10 additional requirements from
the GHG Protocol are fulfilled. These requirements are listed in Box 1. However, when performing an ISO
14044 compliant consequential LCA that is intended to be used for reporting according to the GHG Protocol,
particular care is recommended in explaining the system boundaries, the exclusion of constrained processes,
and the underlying justifications, to directly address the unclarity in the GHG Protocol as to what is
considered ‘attributable processes’.

4|P a g e
Box 1. Additional requirements of the GHG Protocol compared to ISO 14044
• Collect primary data for all processes under their ownership or control;
• Report displaced emissions and removals separately from the studied product’s
end-of-life stage inventory;
• Report attributable processes in the form of a process map;
• Quantify and report biogenic and non-biogenic emissions and removals separately;
• Quantify and report land-use change impacts separately;
• Report when carbon contained in a product is not released to the atmosphere
during waste treatment and therefore is considered stored;
• For significant processes, report any efforts taken to improve data quality;
• Do not include offsets in the calculated climate footprint result (Separate reporting
is allowed);
• Apply the 100-year GWP factor to GHG emissions and removals to calculate the
inventory results in units of CO2 equivalent;
• Do not include weighting factors for delayed emissions when calculating the
climate footprint result (Separate reporting is allowed);
• Quantify and report the cradle-to-gate and gate-to-gate inventory results
separately or clearly state that confidentiality is a limitation to providing this
information;
• If life cycle stages are defined, quantify and report the percentage of total
inventory results by life cycle stage.

5|P a g e
1 Objective and scope
This report compares the requirements of the GHG Protocol Product Life Cycle Standard (Bhatia et al. 2011)
and the ISO 14040 series. The objective is to assess to what extent a study performed according to ISO
14044:2006 and ISO 14067:2018 can be applied for reporting according to the GHG Protocol, and what
possible additions or modifications may be required for this purpose.

The overall international standards for Life Cycle Assessment (LCA) are ISO 14040:2006 and ISO
14044:2006, where the first provides the framework, while the second includes the specific requirements
and guidelines. ISO 14067:2018 provides a few more requirements for climate footprints (called ‘carbon’
footprints by ISO).

In addition to these LCA standards, the World Resources Institute and the World Business Council for
Sustainable Development has published a ‘GHG Protocol Product Life Cycle Accounting and Reporting’
(Bhatia et al. 2011) as part of a family of standards that include also ‘GHG Protocols’ for ‘Corporate
Accounting and Reporting’, ‘Corporate Value Chain (Scope 3) Accounting and Reporting’ and ‘Project
Accounting’. These standards are widely used for corporate GHG reporting.

It is of interest to companies that use the GHG Protocol for reporting, whether their LCAs made according
to ISO 14044 can also be used to provide data for the GHG reporting according to the GHG Protocol. This is
the core question to be answered by this report.

6|P a g e
2 Attributional or consequential modelling?
ISO LCA standards and the GHG Protocol for Product Life Cycles (Bhatia et al. 2011) have little guidance on
how to link unit processes into products systems, and this has given rise to two different interpretations of
the standards on this point, known as “attributional” and “consequential” approaches to modelling.

In the consequential approach activities are included in the product system to the extent that they are
expected to change as a consequence of a change in demand for the functional unit (Sonnemann & Vigon
2011). Consequential modelling seeks to model the potential environmental consequences of changes
resulting from a (potential) decision, or as expressed in Clause A.2 of ISO 14040:2006: “environmental
consequences of possible (future) changes between alternative product systems”. This implies modelling
marginal or incremental changes, as opposed to the average modelling implied in the attributional
approach. For both marginal and incremental modelling, Clause 6.4 of ISO 14049:2012, applies:
“The supplementary processes to be added to the systems must be those that would actually be involved
when switching between the analysed systems. To identify this, it is necessary to know:
• whether the production volume of the studied product systems fluctuate in time (in which case
different sub-markets with their technologies may be relevant), or the production volume is
constant (in which case the base-load marginal is applicable),
• whether (…) the inputs are delivered through an open market, in which case it is also necessary to
know:
o whether any of the processes or technologies supplying the market are constrained (in
which case they are not applicable, since their output will not change in spite of changes in
demand),
o which of the unconstrained suppliers/technologies has the highest or lowest production
costs and consequently is the marginal supplier/ technology when the demand for the
supplementary product is generally decreasing or increasing, respectively.”

A consequential product system thus ensures additionality, i.e., that processes are only included to the
extent that they are additional to a baseline without the studied change. In practice, the consequential
product system is identified by starting from the unit process(es) of the reference flow, tracing each
required product input, physical or monetary, through the marginal supplier(s) of each product, following
verifiable cause-effect relationships, including the consequences of excess supply of by-products from joint
production. The cost for the purchasing activity, is a revenue for the supplying activity. For each activity, a
part of the revenue leaks out as wages, taxes, and profits (together known as “value added”). In a closed
steady-state system, all the original revenue must eventually leave the system as value added, thus
providing a clear delimitation of the unit processes included in the system. The processes included are
limited to those that react to the change in revenue. The marginal suppliers and their technologies can be
divided in those that in response to a change in demand for the product will change their production
output immediately or within the short-term (i.e., within the current production capacity), and those that in
response to an accumulated change in demand for the product will change their production capacity in the
long-term. The impacts from the long-term changes in capacity will typically dominate the sum of the short-
term and long-term changes. The ILCD Handbook (JRC-IEA 2010, p. 37) recommends that when LCA is used
as decision-support, the LCI model should reflect the consequences of the decision. Nevertheless, the ILCD

7|P a g e
Handbook has contributed to much confusion by calling such modelling ‘attributional’, in contrast to the
definitions from UNEP (Sonnemann & Vigon 2011).

In the attributional approach, inputs and outputs are attributed to the functional unit of a product system
by linking and/or partitioning the unit processes of the system according to a normative rule (Sonnemann &
Vigon 2011), “typically as an account of the history of the product” (Clause A.2 of ISO 14040:2006). In
Clause 4.3.4.2 of 14044:2006, the default approach for allocating to the products of joint production is
allocation based on their economic value (revenue). Therefore, in parallel to the consequential product
system, the attributional product system is identified by starting from the unit process(es) of the reference
flow, tracing each cost item input to the next upstream unit process. The revenue generated by the original
demand must eventually leave the product system as value added, thus providing a clear delimitation of
the processes included in the product system. This can be seen as a ‘value chain’ perspective on LCA
(Weidema et al. 2018). Note that unless there is complete proportionality between the physical properties
and the revenues for the joint products, the resulting allocated product systems will not be physically
balanced (Weidema 2018). In general, a value chain will therefore not reflect the physical causalities of
purchasing a product.

Both the consequential and the attributional ‘value chain’ approach to linking are implemented in the
ecoinvent 3 database, applying publicly available algorithms on the same set of unit process data.

At first sight, the GHG Protocol (Bhatia et al. 2011) declares itself clearly as following an attributional
approach: “A GHG product inventory shall follow the life cycle and attributional approaches” and in Box
5.1: “Although not followed in this standard, the consequential approach can provide valuable insight in
certain applications such as evaluating reduction projects or making public policy decisions”. However, the
GHG Protocol has its own definition of the ‘attributional approach’, namely: “An approach to LCA where
GHG emissions and removals are attributed to the unit of analysis of the studied product by linking
together attributable processes along its life cycle”. ‘Attributable processes’ are in turn defined as “Service,
material, and energy flows that become the product, make the product, and carry the product through its
life cycle”. Since processes are not flows, this opens up for three different interpretations, namely that
what is meant could be either of the following:
1) “Attributable processes are processes that are affected by the demand for the “service, material,
and energy flows that become the product, make the product, and carry the product through its
life cycle”. This would be an ISO 14044 conform consequential interpretation and is supported by
the requirements of the GHG Protocol for dealing with co-products, which is identical to that of ISO
14044, namely “Companies shall avoid allocation wherever possible by using process subdivision,
redefining the functional unit, or using system expansion” and the more specific system expansion
procedures “For allocation due to recycling, companies shall use either the closed loop
approximation method or the recycled content method”, and only when this is not possible to
apply allocation that reflects physical, economic, or other (causal) relationships between the
products. Avoiding allocation through system expansion is not meaningful in a pure attributional
interpretation.
2) “Attributable processes are the processes in the value chain of the “service, material, and energy
flows that become the product, make the product, and carry the product through its life cycle”.

8|P a g e
 This would be a ‘value chain’ interpretation, conform with the attributional interpretation of ISO
14044.
3) “Attributable processes are the processes that supply the physical “service, material, and energy
flows that become the product, make the product, and carry the product through its life cycle”.
This is what has been called a ‘supply chain’ perspective on LCA (Weidema et al. 2018), which does
not find support in the ISO 14044 and requires a further delimitation of the assessment to specific
physical properties of the “service, material, and energy flows that become the product, make the
product, and carry the product”, a complication that is elaborated in the following paragraph.

The physical supply chain of materials and parts may be very different from the supply chain of
information, energy, or other services to the same activity. For an unambiguous identification and to
provide a physically consistent and balanced system, it is therefore necessary to specify a supply chain in
terms of the physical (as opposed to economic) properties it supplies, e.g., mass, energy, or a specific
elemental content. The supply chain of an activity is identified by tracing the investigated property
backwards in the chain. The input of one activity is either the output of a supplying activity or an external
input to the system. In a closed steady-state system, all the input must eventually enter the system as
what, in LCA jargon, are called “elementary” flows, i.e., flows that come from the environment without
previous human transformation, thus providing a clear delimitation of the activities included in the system.
Taking the property mass as example, the sum of all mass inputs over a mass supply chain will equal the
mass of the analysed product and all wastes and by-products from the system. When joint production
activities are partitioned to obtain the supply chain for a single product, the mass balance for each single-
product activity is maintained by partitioning each input proportionally to the share that each joint product
has in the overall mass output. In LCA jargon this is known as “mass allocation”. If two properties are not
intrinsically linked, the supply chain of one property may at some point disengage from the supply chain of
the other property, and it then becomes impossible to say which of the two chains should be followed.
Thus, each specific supply chain analysis can only trace one specific property (or several intrinsically linked
properties). Obviously, it is possible to overlay several supply chains, depending on the purpose of the
analysis, and how many properties the decision maker wishes to take responsibility for.

In the GHG Protocol (Bhatia et al. 2011), there is no requirements on which property (or properties) that
should be applied to define the attributional chain of processes. A ‘value chain’ interpretation appears to
be in conflict with the statement of the GHG Protocol that “companies are not required to include non-
attributable processes” and mentions as examples of these a large number of processes that do provide
services to ‘make the product’, and which would be part of the value chain:
• Capital goods (e.g., machinery, trucks, infrastructure)
• Overhead operations (e.g., facility lighting, air conditioning)
• Corporate activities and services (e.g., research and development, administrative functions,
company sales and marketing)
• Transport of the product user to the retail location
• Transport of employees to and from works
This seems to point at a narrower ‘supply chain’ delimitation to the processes that supply the physical
materials and energy, which would nevertheless leave the outstanding challenge to separately trace the
many different material properties of the product, a task for which there is currently no database and

9|P a g e
software support. Against this interpretation also speaks the statement at the top of the ‘Requirements’
Section 5.2 of the GHG Protocol, that “the accounting methodologies and requirements presented in this
standard follow the life cycle approach as established by ISO LCA standards 14040 and 14044.”

Furthermore, Clause 7.3.5 of the GHG Protocol opens up for the companies to define the inclusion of
activities based on relevance, e.g., according to the consequential perspective of reduction potentials:
“However, companies should include non-attributable processes in the inventory if they cannot be
separated from attributable process data, or if the company determines that the process is relevant to the
studied product. Relevance is determined by the company and may be based on many different factors
including business goals and reduction potentials, product rules or sector guidance, and relative impact in
relation to the rest of the inventory.”

Finally, as already mentioned, both the ‘value chain’ and the ‘supply chain’ interpretation of the GHG
Protocol is challenged by the requirement of the GHG Protocol to deal with co-products as in ISO 14044,
namely whenever possible to avoid allocation through process subdivision, redefining the functional unit or
using system expansion. If the ‘value chain’ or ‘supply chain’ interpretation was intended, there would not
have been any reason to use substantial amounts of text on describing system expansion and add specific
requirements on its specific implementations in the form of ‘closed loop approximation’ and ‘recycled
content’ procedures.

In conclusion, in spite of defining its approach as ‘attributional’, the further definitions of this approach and
the further requirements and allowances in the GHG protocol appears to support both an attributional and
a consequential interpretation.

3 Carbon accounting
For obvious reasons, the GHG Protocol is limited to accounting for GHGs, more precisely emissions and
removals of CO2, CH4, N2O, SF6, PFCs, and HFCs. A distinction is made between emissions and removals from
biogenic sources, non-biogenic sources, and land-use change impacts.

A requirement of the GHG Protocol (Bhatia et al. 2011) that goes beyond the ones of ISO 14044 and ISO
14067 is that “Companies shall report when carbon contained in a product or its components is not
released to the atmosphere during waste treatment and therefore is considered stored”. ISO 14067 allows
separate reporting of stored biogenic carbon, but not the inclusion of this in the climate footprint.

Requirements that go beyond those of ISO 14044 (but not beyond those of ISO 14067) are that:
• “Companies shall apply a 100-year GWP factor to GHG emissions and removals data to calculate
the inventory results in units of CO2 equivalent (CO2e)” *)
• “Companies shall quantify and report the (…) biogenic and non-biogenic emissions and removals
separately when applicable”

*) The term ‘inventory result’ is used differently in ISO (‘outcome of a life cycle inventory analysis that catalogues the
flows crossing the system boundary and provides the starting point for life cycle impact assessment’), i.e., before
impact assessment, and in the GHG Protocol (“The GHG impact of the studied product”), i.e., after impact assessment.

10 | P a g e
 • “Companies shall quantify and report the (…) land-use change impacts separately when applicable”
• “Companies shall not include the following when quantifying inventory results: weighting factors
for delayed emissions; offsets; and avoided emissions” *)†

With respect to the first of these four requirements, it is relevant to be aware that the 100-year GWP
metric gives 4-5 times less weight to CH4 emissions than if using direct radiative forcing that better reflect
the speed of change and thus the main impacts on nature. The tendency in the LCA community is to work
with separate metrics for the rate of change and the longer-term temperature effects mainly relevant for
human health, as recommended by UNEP (Levasseur et al. 2016): “To represent the complexity of climate
change impacts, more than one impact category is needed. Therefore, in LCA application, we recommend
considering two separate impact categories for climate change (shorter-term related to the rate of
temperature change, and long-term related to the long-term temperature rise).”

Also, the three issues in the last-mentioned requirement calls for commenting:
Delayed emissions: The wilful disregard for the effects of delayed emissions, which in ISO 14067 is explicitly
formulated as “All GHG emissions and removals shall be calculated as if released or removed at the
beginning of the assessment period without taking into account an effect of delayed GHG emissions and
removals” is in direct conflict with the recommendation of ISO 14044 that “spatial and temporal
differentiation of the characterization model relating the LCI results to the category indicator should be
considered”. The disregard for the effects of delayed emissions implies that no benefit is calculated for the
temporal postponement of land use change (also known as indirect land use change), and no benefit is
calculated for the temporal postponement of emissions when carbon-containing materials are kept in
capital stocks or recycled as a material after use, so that they remain unreleased longer than they would
have if they had been left in nature. Calculating CO2 emitted in year 100 with the same impact factors as
CO2 emitted in year 0, fails to consider that an emission now creates more damage than the same emission
later, because impacts on human health are expected to be mitigated through adaptation (Smith et al.
2014) and impacts on nature mainly depend on the speed of change, which is currently very high and
decreases with the expected stabilisation and eventual decline of overall emissions (Collins et al. 2013).
However, ISO 14067 allows that delayed emissions “may be documented separately” in the study report,
and the GHG Protocol similarly suggests that “Companies may show the impact of delayed emissions and
removals separately from the inventory results.”
Offsets: Both ISO 14067 and the GHG Protocol define ‘offsets’ (GHG Protocol) or ‘offsetting’ (ISO 14067) as
occurring ‘outside the boundary of the product’s life cycle’ (GHG Protocol) or ‘outside the product system’
(ISO 14067), which makes it obvious that it should not be included in the calculation results, and must be
declared separately from the LCA results, even for a product that is sold with an explicit declaration that
part of the revenue from the product goes to offset an explicit amount of the impacts from the product’s
life cycle. However, the current definitions do not provide any explanation of what offsetting is, as opposed
to the removals that occur as part of the product’s life cycle and which it is required to include. A more
precise and explicit definition of offsetting would be: “Activities that are performed with the explicit
purpose of counterbalancing the effects of the impacts caused by the product system, but are not required

*) See footnote on the previous page.

11 | P a g e
for, occurring as consequences of, or intended to affect the amount of inputs and outputs from the product
system itself.”
Avoided emissions: It is unclear if this is just another word for offsets or if other forms of avoided emissions
exist, considering that removals are required to be included in the inventory, and that the GHG protocol
explicitly states that “Avoided emissions as defined here are not the same as emissions reductions that
occur due to directly attributable reduction projects, or allocated emissions using the system expansion
allocation method”.

In spite of the above comments, there is nothing in the additional requirements of the GHG Protocol that
could not be met by an LCA according to ISO 14044.

4 Additional requirements of the GHG Protocol for the Product Life Cycle
In addition to the above-mentioned requirements, the GHG Protocol (Bhatia et al. 2011) contains the
following requirements that are not requirements in ISO 14044:
• “Companies shall collect primary data for all processes under their ownership or control”
• “For significant processes, companies shall report (…) any efforts taken to improve data quality”
• “When using the closed loop approximation method, companies shall report displaced emissions
and removals separately from the studied product’s end-of- life stage inventory”
• “Companies shall report attributable processes in the form of a process map”
• “Companies shall quantify and report the (…) cradle-to-gate and gate-to-gate inventory results
separately or a clear statement that confidentiality is a limitation to providing this information”

Additional requirements that go beyond those of ISO 14044 (but not beyond those of ISO 14067) are that
“Companies shall quantify and report the percentage of total inventory results by life cycle stage”.
However, there is no explicit requirement to have more than one life cycle stage.

While the mentioned additional requirements may involve some extra efforts, they are complementary and
not conflicting with those of ISO 14044.

5 ISO 14044 series requirements not included in the GHG Protocol

Besides the additional requirements mentioned in Chapter 4, the GHG protocol (Bhatia et al. 2011) is
generally less elaborate than ISO 14044 and ISO 14067. The additional or stricter ISO requirements are not
described in detail here, considering that the scope of the current report is to assess the applicability of an
ISO compliant LCA to fulfil the requirements of the GHG Protocol, and not vice versa.

Nevertheless, it is relevant to mention the GHG Protocol requirement that “Companies shall disclose and
justify any exclusions of attributable processes in the inventory report”. The corresponding requirements in
ISO 14044 are that "decisions regarding the data to be included shall be based on a sensitivity analysis to
determine their significance" and "The deletion of life cycle stages, processes, inputs or outputs is only
permitted if it does not significantly change the overall conclusions of the study." Although the ISO 14044
requirements are stricter than the corresponding requirement of the GHG Protocol, the uncertainty of

12 | P a g e
what is to be considered ‘attributable processes’ warrants particular care in explaining the system
boundaries, the exclusion of constrained processes, and the underlying justifications, when performing an
ISO 14044 compliant consequential LCA that is intended to be used for reporting according to the GHG
Protocol.

6 Conclusion
In spite of the GHG Protocol framing its LCA approach as ‘attributional’, the closer analysis presented in
Chapter 2 of this report shows that the GHG protocol supports both an attributional and a consequential
interpretation. An LCA performed according to ISO 14044, whether consequential or attributional, can thus
also be seen as conform with the GHG Protocol and can therefore be used to provide data for the GHG
reporting according to the GHG Protocol, provided that the following additional requirements from the
GHG Protocol are fulfilled:
• Collect primary data for all processes under their ownership or control;
• Report displaced emissions and removals separately from the studied product’s end-of-life stage
inventory;
• Report attributable processes in the form of a process map;
• Quantify and report biogenic and non-biogenic emissions and removals separately;
• Quantify and report land-use change impacts separately;
• Report when carbon contained in a product is not released to the atmosphere during waste
treatment and therefore is considered stored;
• For significant processes, report any efforts taken to improve data quality;
• Do not include offsets in the calculated climate footprint result (Separate reporting is allowed);
• Apply the 100-year GWP factor to GHG emissions and removals to calculate the inventory results in
units of CO2 equivalent;
• Do not include weighting factors for delayed emissions when calculating the climate footprint
result (Separate reporting is allowed);
• Quantify and report the cradle-to-gate and gate-to-gate inventory results separately or clearly state
that confidentiality is a limitation to providing this information;
• If life cycle stages are defined, quantify and report the percentage of total inventory results by life
cycle stage.

References
Bhatia P, Cummis C, Brown A, Drauker L, Rich D, Lahd H. (2011). Greenhouse Gas Protocol Product Life
Cycle Accounting and Reporting Standard. Washington DC: World Resources Institute and World
Business Council for Sustainable Development.
Collins M, Knutti R, Arblaster J, Dufresne J-L, Fichefet T, Friedlingstein P, Gao X, Gutowski W J, Johns T,
Krinner G, Shongwe M, Tebaldi C, Weaver A J, Wehner M. (2013). Long-term Climate Change:
Projections, Commitments and Irreversibility. Pp. 1029-1136 in: Stocker et al.: Climate Change 2013: The
Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge University Press.

13 | P a g e
JRC-IEA. (2010). International Reference Life Cycle Data System (ILCD) Handbook – General guide for Life
Cycle Assessment – Detailed guidance. First edition March 2010. Publications Office of the European
Union, Luxembourg. Available at <http://lct.jrc.ec.europa.eu/>
Levasseur A, de Schryver A, Hauschild M, Kabe Y, Sahnoune A, Tanaka K, Cherubini F. (2016). Greenhouse
gas emissions and climate change impacts. Pp. 58-75 in Frischknecht & Jolliet: Global Guidance for Life
Cycle Impact Assessment Indicators. Volume 1. Paris: United Nations Environmental Programme.
Smith K R, Woodward A, Campbell-Lendrum D, Chadee D D, Honda Y, Liu Q, Olwoch J M, Revich B,
Sauerborn R. (2014). Human health: impacts, adaptation, and co-benefits. Pp. 709-754 in Field et al.:
Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects.
Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge University Press.
Sonnemann G, Vigon B. (2011). Global Guidance Principles for Life Cycle Assessment Databases.
Paris/Pensacola: UNEP/SETAC Life Cycle Initiative.
Weidema B P. (2018). In search of a consistent solution to allocation of joint production. Journal of
Industrial Ecology 22(2):252-262.
Weidema B P, Pizzol M, Schmidt J, Thoma G. (2018). Attributional or consequential Life Cycle Assessment:
A matter of social responsibility. Journal of Cleaner Production 174:305–314.

14 | P a g e