Content uploaded by Mark A J Huijbregts
Author content
All content in this area was uploaded by Mark A J Huijbregts on Jul 10, 2020
Content may be subject to copyright.
COMMENTARY AND DISCUSSION ARTICLE
ReCiPe2016: a harmonised life cycle impact assessment method
at midpoint and endpoint level
Mark A. J. Huijbregts
1,2
&Zoran J. N. Steinmann
1
&Pieter M. F. Elshout
1
&Gea Stam
1,3
&
Francesca Verones
4
&Marisa Vieira
1,5
&Michiel Zijp
3
&Anne Hollander
3
&
Rosalie van Zelm
1
Received: 29 November 2016 /Accepted: 29 November 2016
#Springer-Verlag Berlin Heidelberg 2016
Abstract
Purpose Life cycle impact assessment (LCIA) translates
emissions and resource extractions into a limited number of
environmental impact scores by means of so-called character-
isation factors. There are two mainstream ways to derive char-
acterisation factors, i.e. at midpoint level and at endpoint level.
To further progress LCIA method development, we updated
the ReCiPe2008 method to its version of 2016. This paper
provides an overview of the key elements of the
ReCiPe2016 method.
Methods We implemented human health, ecosystem quality
and resource scarcity as three areas of protection. Endpoint
characterisation factors, directly related to the areas of protec-
tion, were derived from midpoint characterisation factors with
a constant mid-to-endpoint factor per impact category. We
included 17 midpoint impact categories.
Results and discussion The update of ReCiPe provides char-
acterisation factors that are representative for the global scale
instead of the European scale, while maintaining the possibil-
ity for a number of impact categories to implement character-
isation factors at a country and continental scale. We also
expanded the number of environmental interventions and
added impacts of water use on human health, impacts of water
use and climate change on freshwater ecosystems and impacts
of water use and tropospheric ozone formation on terrestrial
ecosystems as novel damage pathways. Although significant
effort has been put into the update of ReCiPe, there is still
major improvement potential in the way impact pathways
are modelled. Further improvements relate to a regionalisation
of more impact categories, moving from local to global spe-
cies extinction and adding more impact pathways.
Conclusions Life cycle impact assessment is a fast evolving
field of research. ReCiPe2016 provides a state-of-the-art
method to convert life cycle inventories to a limited number
of life cycle impact scores on midpoint and endpoint level.
Keywords Characterisation factors .Ecosystem quality .
Endpoint indicator .Human health .Midpoint indicator .
Resource scarcity
1 Introduction
Life cycle impact assessment (LCIA) supports interpretation
of LCA studies by translating emissions and resource extrac-
tions into a limited number of environmental impact scores
(Hauschild and Huijbregts 2015). This is done by means of so-
called characterisation factors, which indicate the environ-
mental impact per unit of stressor (e.g. per kg of resource
Responsible editor: Mary Ann Curran
Electronic supplementary material The online version of this article
(doi:10.1007/s11367-016-1246-y) contains supplementary material,
which is available to authorized users.
*Mark A. J. Huijbregts
m.huijbregts@science.ru.nl
1
Department of Environmental Science, Institute for Water and
Wetland Research, Faculty of Science, Radboud University,
Nijmegen, The Netherlands
2
Dutch Environmental Assessment Agency (PBL), The
Hague, The Netherlands
3
National Institute for Public Health and the Environment, Centre of
Sustainability, Environment and Health (DMG),
Bilthoven, The Netherlands
4
Industrial Ecology Programme, Department for Energy and Process
Engineering, Norwegian University and Science (NTNU),
Tron dheim, Norw ay
5
PRé Consultants, Amersfoort, The Netherlands
Int J Life Cycle Assess
DOI 10.1007/s11367-016-1246-y
extracted or emission released). There are two mainstream
ways to derive characterisation factors, i.e. at midpoint level
and at endpoint level. Characterisation factors at the midpoint
level are located somewhere along the cause-impact pathway,
typically at the point after which the environmental mecha-
nism is identical for each environmental flow assigned to that
impact category (Goedkoop et al. 2009). Characterisation fac-
tors at endpoint level typically reflect damage at one of three
areas of protection which are human health, ecosystem quality
and resource scarcity. The two approaches are complementary
in that the midpoint characterisation has a stronger relation to
the environmental flows and comes in general with lower
parameter uncertainty, while the endpoint characterisation is
easier to interpret in terms of relevance of the environmental
flows (Hauschild and Huijbregts 2015).
Recently, Hauschild et al. (2013)reviewedalargenumber
of LCIA methods in order to provide recommended practice
for both midpoint and endpoint characterisation factors. This
consensus work had a significant influence on the establish-
ment of LCIA in the environmental policy arena in Europe,
e.g. via its testing in the Product and Organisational
Environmental Footprint initiative (EC 2013). The review,
however, also provided insight into a number of shortcomings
of the models used to derive characterisation factors recom-
mended by Hauschild et al. (2013). First of all, most models
have a continental focus, particularly focussing on Europe.
Moreover, for many impact categories at the endpoint level,
the best among existing characterisation models was still not
considered sufficiently mature for recommendation.
To further progress LCIA methods beyond the current con-
sensus state of the art, we updated the ReCiPe2008 method to
its version of 2016. ReCiPe provides a harmonised implemen-
tation of cause-effect pathways for the calculation of both
midpoint and endpoint characterisation factors (Goedkoop
et al. 2009). In order to make a step forward in overcoming
the shortcomings mentioned above, the update of ReCiPe fo-
cused on (1) providing characterisation factors that are repre-
sentative for the global scale, while maintaining the possibility
for a number of impact categories to implement characterisa-
tion factors at a country and continental scale and (2) improv-
ing the methods applied to model midpoint-to-endpoint fac-
tors. Compared to ReCiPe2008, we added the following extra
damage pathways in ReCiPe2016:
–Impacts of water use on human health, freshwater ecosys-
tems and terrestrial ecosystems
–Impacts of climate change on freshwater ecosystems
–Impacts of tropospheric ozone formation on terrestrial
ecosystems
For a number of impact categories, we also provide mid-
point and endpoint characterisation factors on a country level,
i.e. for photochemical ozone formation, particulate matter
formation, terrestrial acidification, freshwater eutrophication
and water use. This paper provides an overview of the key
elements of the ReCiPe2016 method.
2 Methods
2.1 Framework
We followed the model framework proposed in ReCiPe2008
with human health, ecosystem quality and resource scarcity as
areas of protection. The unit for human heath damage, DALYs
(disability adjusted life years), represents the years thatare lost
or that a person is disabled due to a disease or accident. The
unit for ecosystem quality is local relative species loss in ter-
restrial, freshwater and marine ecosystems, respectively, inte-
grated over space and time (potentially disappeared fraction of
species∙m
2
∙year or potentially disappeared fraction of
species·m
3
· year). To aggregate the impacts of terrestrial,
freshwater and marine ecosystems into one single unit
(species.year), we included species densities for these three
types of ecosystems in the same way as proposed by
Goedkoop et al. (2009). The unit for resource scarcity is dol-
lars ($), which represents the extra costs involved for future
mineral and fossil resource extraction. Endpoint characterisa-
tion factors (CFe) are derived from midpoint characterisation
factors (CFm) with a constant mid-to-endpoint factor per im-
pact category:
CFex;a¼CFmxFM→;E;að1Þ
Where adenotes the area of protection, i.e. human health,
(terrestrial, freshwater and marine) ecosystems or resource
scarcity, xdenotes the stressor of concern and F
M→E,a
is the
midpoint-to-endpoint conversion factor for area of protection
a. These mid-to-endpoint factors are constant per impact cat-
egory, because environmental mechanisms are considered to
be identical for each stressor after the midpoint impact loca-
tion on the cause-effect pathway.
Figure 1shows the link between the environmental mech-
anisms, i.e. the 17 midpoint impact categories, and the three
areas of protection, i.e. the endpoints, as included in
ReCiPe2016.
2.2 Model selection criteria
The selection criteria for the environmental models in
ReCiPe2016 were
–The models should refer to the global scale.
–The models should reflect the current state of the
art in science.
Int J Life Cycle Assess
–The models should maintain consistency between the
modelling of different impact categories, particularly rel-
evant for toxicity.
–In case of multiple suitable global models, we prefer models
that can be run in-house by the ReCiPe consortium.
We performed a review of the existing literature within and
outside the field of LCIA to select a preferred combination of
environmental models and databases per midpoint impact cat-
egory. We also selected models that were able to quantify
damage pathways relevant for the mid-to-endpoint factors.
For a number of impact categories, notably fine particulate
matter formation, photochemical ozone formation, land use
and water use, there is a fast increasing number of global
models published in the literature. Here, we pragmatically
selected the models that we were able to run within the con-
sortium without claiming that ReCiPe2016 is necessarily su-
perior compared to other global models out there.
2.3 Scenario analysis
Different sources of uncertainty and different methodological
choices were grouped into three scenarios. This means that
ReCiPe2016 does not provide one set, but three sets of midpoint
and endpoint characterisation factors and users are encouraged to
use all three of them for a sensitivity check of their LCA results.
One prominent choice is the time horizon for long living
pollutants. We included a 20-year, 100-years and 1000-year-
infinite time horizon in each scenario, respectively. The time
horizon for the third scenario was not always infinite, as not all
the environmental models provided sufficient information to
model steady-state conditions. Another important value choice
included was the level of evidence available for the environmen-
tal effects considered for the impacts related to ozone depletion,
ionising radiation, toxicity, fine particulate matter formation and
water use. We coupled effects with only a very high level of
evidence to a 20-year time horizon and all reported effects to a
1000-infinite time horizon to construct two extreme scenarios.
The third scenario refers to a middle ground with a 100-year time
horizon consistently included and effects with a level of evidence
that is considered acceptable by international bodies, such as the
World Health Organisation. More details on the scenario build-
ing can be found in Huijbregts et al. (2016).
3Results
3.1 Midpoint indicators
Impact categories and their indicators at the midpoint level are
summarised in Table 1and briefly explained below. The full
list of midpoint characterisation factors is available in spread-
sheet format (see Electronic Supplementary Material).
3.1.1 Climate change
The midpoint characterisation factor selected for climate
change is the widely used global warming potential (GWP),
which quantifies the integrated infrared radiative forcing in-
crease of a greenhouse gas (GHG), expressed in kg CO
2
-eq
(IPCC 2013;Joosetal.2013).
3.1.2 Stratospheric ozone depletion
The ozone depleting potential (ODP), expressed in kg CFC-11
equivalents, was used as characterisation factor on the midpoint
level. ODPs refer to a time-integrated decrease in stratospheric
ozone concentration over an infinite time horizon (WMO 2011).
3.1.3 Ionising radiation
The collective dose resulting from the emission of a radionuclide
is the point where the characterisation factor at midpoint level
was derived. The midpoint characterisation factor, called ionis-
ing radiation potential (IRP), is reported in Cobalt-60 eq to air.
3.1.4 Fine particulate matter formation
For the midpoint characterisation factors of fine particulate
matter formation, the human population intake of PM
2.5
was
Fig. 1 Overview of the impact categories that are covered in the
ReCiPe2016 method and their relation to the areas of protection. The
dotted line means there is no constant mid-to-endpoint factor for fossil
resources
Int J Life Cycle Assess
considered. Particulate matter formation potentials (PMFP)
are expressed in kg primary PM
2.5
-equivalents. The change
in ambient concentration of PM
2.5
after the emission of a
precursor, i.e. NH
3
,NO
x
,SO
2
and primary PM
2.5
, was pre-
dicted with the emission—concentration sensitivities matrices
for emitted precursors from the global source-receptor model
TM5-FASST (Van Zelm et al. 2016).
3.1.5 Photochemical ozone formation
For the midpoint characterisation factors of photochemical
ozone formation related to human exposure, the human pop-
ulation intake of ozone was considered. Human health ozone
formation potential (HOFP) is expressed in kg NO
x
-eq. The
change in ambient concentration of ozone after the emission
of a precursor (nitrogen oxides (NO
x
) or non-methane volatile
organic compounds (NMVOC)) was predicted with the emis-
sion—concentration sensitivities matrices for emitted precur-
sors from the global source-receptor model TM5-FASST (Van
Zelm et al. 2016). The ecosystem ozone formation potential
(EOFP), also expressed in kg NO
x
eq, relates to the sum of the
differences between the hourly mean ozone concentration and
40 ppb during daylight hours over the relevant growing season
in ppm∙h(AOT40;VanZelmetal.2016).
3.1.6 Terrestrial acidification
For the midpoint characterisation factors of acidifying emis-
sions, the fate of a pollutant in the atmosphere and the soil as
calculated by Roy et al. (2014) were taken. Acidification
Table 1 Overview of the midpoint impact categories and related indicators
Midpoint impact
category
Indicator CF
m
Unit Key references
Climate change Infrared radiative forcing
increase
Global warming potential
(GWP)
kg CO
2
-eq to air IPCC 2013; Joos
et al. 2013
Ozone depletion Stratospheric ozone decrease Ozone depletionpotential (ODP) kg CFC-11-eq to air WMO 2011
Ionising radiation Absorbed dose increase Ionising radiation potential (IRP) kBq Co-60-eq to air Frischknecht
et al. 2000
Fine particulate matter
formation
PM2.5 population intake
increase
Particulate matter formation
potential (PMFP)
kg PM2.5-eq to air Van Zelm et al. 2016
Photochemical oxidant
formation: terrestrial
ecosystems
Tropospheric ozone increase Photochemical oxidant
formation potential:
ecosystems (EOFP)
kg NOx-eq to air Van Zelm et al. 2016
Photochemical oxidant
formation: human
health
Tropospheric ozone
population intake increase
Photochemical oxidant
formation potential: humans
(HOFP)
kg NOx-eq to air Van Zelm et al. 2016
Terrestrial acidification Proton increase in natural soils Terrestrial acidification potential
(TAP)
kg SO
2
-eqtoair Royetal.2014
Freshwater
eutrophication
Phosphorus increase in
freshwater
Freshwater eutrophication
potential (FEP)
kg P-eq to freshwater Helmes et al. 2012
Human toxicity: cancer Risk increase of cancer disease
incidence
Human toxicity potential (HTPc) kg 1,4-DCB-eq to urban air Van Zelm et al. 2009
Human toxicity:
non-cancer
Risk increase of non-cancer
disease incidence
Human toxicity potential
(HTPnc)
kg 1,4-DCB-eq to urban air Van Zelm et al. 2009
Terrestrial ecotoxicity Hazard-weighted increase in
natural soils
Terrestrial ecotoxicity potential
(TETP)
kg 1,4-DCB-eq to industrial
soil
Van Ze lm et a l. 2009
Freshwater ecotoxicity Hazard-weighted increase in
freshwaters
Freshwater ecotoxicity potential
(FETP)
kg 1,4-DCB-eq to freshwater Van Zelm et al. 2009
Marine ecotoxicity Hazard-weighted increase in
marine water
Marine ecotoxicity potential
(METP)
kg 1,4-DCB-eq to
marine water
Van Ze lm et a l. 2009
Land use Occupation and time-integrated
land transformation
Agricultural land occupation
potential (LOP)
m
2
× yr annual cropland-eq De Baan et al. 2013;
Curran et al. 2014
Water use Increase of water consumed Water consumption
potential (WCP)
m
3
water-eq consumed Döll and Siebert
2002;
Hoekstra and
Mekonnen 2012
Mineral resource
scarcity
Increase of ore extracted Surplus ore potential (SOP) kg Cu-eq Vieira et al. 2016a
Fossil resource
scarcity
Upper heating value Fossil fuel potential (FFP) kg oil-eq Jungbluth and
Frischknecht 2010
Int J Life Cycle Assess
potentials (AP) are expressed in kg SO
2
-equivalents. Changes
in acid deposition, following changes in air emission of NO
x
,
NH
3
and SO
2
, were calculated with the GEOS-Chem model
(Roy et al. 2012a). Subsequently, the change in acidity in the
soil due to a change in acid deposition was derived with the
geochemical steady-state model PROFILE (Roy et al. 2012b).
3.1.7 Freshwater eutrophication
The fate of phosphorus forms the basis of the midpoint char-
acterisation factors for freshwater eutrophication. Freshwater
eutrophication potentials (FEP) are expressed in kg P to fresh-
water-equivalents. Global fate factors for phosphorus emis-
sions to freshwater were taken from Helmes et al. (2012).
For emissions to agricultural soils, it was assumed that typi-
cally 10% of all P is transported from agricultural soil to sur-
face waters (Bouwman et al. 2009).
3.1.8 Toxicity
The fate and effects of chemical emissions expressed in kg
1,4-dichlorobenzene-equivalents (1,4DCB-eq) was used as
characterisation factor at the midpoint level for human toxic-
ity, freshwater ecotoxicity, marine ecotoxicity and terrestrial
ecotoxicity. We used the global multimedia fate, exposure and
effects model USES-LCA 2.0, the Uniform System for the
Evaluation of Substances adapted for LCA (Van Zelm et al.
2009), as a basis for our calculations, updated to deal with
dissociating chemicals (Van Zelm et al. 2013) and using the
chemical data from the USEtox database (Rosenbaum et al.
2008). The ecotoxicological effect factor represents the
change in PDF of species due to a change in the environmental
concentration of a chemical. The human-toxicological effect
factors were derived for carcinogenic and non-carcinogenic
effects separately, reflecting the change in lifetime disease
incidence due to a change in intake of the substance. Note that
we did not select USEtox model (Rosenbaum et al. 2008)for
implementation in ReCiPe2016, as USEtox does not provide
characterisation factors for terrestrial and marine toxicity.
Another practical reason for preferring USES-LCA compared
to USEtox is that USEtox does not easily provide the possi-
bility to assess the influence of value choices on the charac-
terisation factors, such as the option to derive time horizon
dependent characterisation factors.
3.1.9 Water use
The characterisation factor at midpoint level is m
3
of water
consumed per m
3
of water extracted. For agriculture, the con-
sumptive part of the withdrawal was estimated with water
requirement ratios based on Döll and Siebert (2002). For in-
dustry and domestic water use, assumptions were made based
on Hoekstra and Mekonnen (2012).
3.1.10 Land use
The midpoint characterisation factors (in m
2
·yr annual crop
equivalents) refer to the relative species loss caused by a spe-
cific land use type (annual crops, permanent crops, mosaic
agriculture, forestry, urban land, pasture). Relative species loss
was determined by comparing field data on local species rich-
ness in specific types of natural and human-made land covers
(De Baan et al. 2013; Elshout et al. 2014). For land conver-
sion, passive recovery towards a (semi-)natural, old growth
habitat was assumed, based on average recovery times from
Curran et al. (2014).
3.1.11 Mineral resource scarcity
The midpoint characterisation factor for mineral resource scar-
city is Surplus Ore Potential (SOP), expressed as kg Cu-eq.
The primary extraction of a mineral resource will lead to an
overall decrease in ore grade, meaning the concentration of
that resource in ores worldwide, which in turn will increase
the amount of ore produced per kilogramme of mineral re-
source extracted. The SOP expresses the average extra amount
of ore produced in the future caused by the extraction of a
mineral resource considering all future production of that min-
eral resource (Vieira et al. 2016a).
3.1.12 Fossil resource scarcity
The midpoint indicator for fossil resource use, determined as
the Fossil Fuel Potential (FFP in kg oil-eq), is defined as the
ratio between the higher heating value of a fossil resource and
the energy content of crude oil (Jungbluth and Frischknecht
2010).
3.2 Mid-to-endpoint factors
The damage pathways considered to go from the midpoint to
the endpoint level in ReCiPe2016, sorted per environmental
problem, are summarised in Table 2and briefly explained
below. The midpoint-to-endpoint factors are available in
spreadsheet format (see Electronic Supplementary Material).
3.2.1 Climate change
The first step in the midpoint-to-endpoint model quantifies the
link between time-integrated radiative forcing and time-
integrated temperature increase for CO
2
(Joos et al. 2013).
Concerning human health damage, De Schryver et al. (2009)
was used to quantify the increase in risk of diseases (malnu-
trition, malaria, and diarrhoea) and increased flood risk. For
terrestrial ecosystems, the increase in potentially disappeared
fraction of species (PDF) due to an increase in global temper-
ature was derived from the review by Urban (2015). Finally,
Int J Life Cycle Assess
the influence of global temperature increase on river discharge
and subsequent expected changes in fish species occurrences
was taken from Hanafiah et al. (2011).
3.2.2 Stratospheric ozone depletion
The human health effect of a decrease in stratospheric ozone
concentration, as modelled in the mid-to-endpoint calculation,
was derived in two consecutive steps, following Hayashi et al.
(2006). The first step relates a change in ozone depletion to an
increase in UVB radiation and the second step couples this
increase in UVB radiation to an increase in burden of disease.
To calculate the damage to human health, the increased inci-
dence and related loss of DALYs of three types of skin cancers
(malignant melanoma, basal cell carcinoma and squamous
cell carcinoma) and cataract due to UVB exposure were
included.
3.2.3 Ionising radiation
In the mid-to-endpoint calculations, the human health effect of
the collective dose on the incidence of different cancer types
was assessed by first taking the fatal and non-fatal cancer
incidence per cancer type from Frischknecht et al. (2000).
Table 2 Damage pathways in ReCipe2016
Environmental problem Area of protection Damage pathways References
Climate change Human health Years of life lost and disabled related to increased
malaria, diarrhoea, malnutrition and natural disasters
due to increased global mean temperature
IPCC 2013;Joosetal.2013;
De Schryver et al. 2009
Ecosystems (terrestrial) Species loss related to changing biome distributions
due to increased global temperature
IPCC 2013;Joosetal.2013;
Urban 2015
Ecosystems
(freshwater)
Fish species loss due to decrease river discharge Hanafiah et al. 2011
Stratospheric ozone depletion Human health Years of life lost and disabled related to increased
skin cancer and cataract due to UV-exposure
WMO 2011; Hayashi et al.
2006
Ionising radiation Human health Years of life lost and disabled related to an increase in
cancer and hereditary diseases due to exposure to
radiation
Frischknecht et al. 2000;
De Schryver et al. 2011
Particulate matter formation Human health Years of life lost related to an increase in
cardiopulmonary and lung cancer caused by exposure
to primary and secondary aerosols
Van Ze lm et a l . 2016
Photochemical ozone
formation
Human health Years of life lost related to an increase in respiratory
diseases caused by exposure to ozone
Van Ze lm et a l . 2016
Ecosystems (terrestrial) Loss of plant species due to increase in ozone exposure Van Zelm et al. 2016
Terrestrial acidification Ecosystems (terrestrial) Loss of plant species due to decrease in soil pH Roy et al. 2014
Freshwater eutrophication Ecosystems (aquatic) Loss of aquatic species due to increased phosphorus
concentrations
Helmes et al. 2012;
Azevedo et al. 2013a,b
Toxicity Human health Years of life lost and disabled due to cancer and
non-cancer effects due to ingestion and inhalation
of toxic substances
Van Ze lm et a l . 2009
Ecosystems (marine) Species loss due to chemical exposure in marine waters Van Zelm et al. 2009
Ecosystems (terrestrial) Species loss due to chemical exposure in soils Van Zelm et al. 2009
Ecosystems
(freshwater)
Species loss due to chemical exposure in freshwater Van Zelm et al. 2009
Water consumption Human health Malnutrition caused by water shortage Pfister et al. 2009
Ecosystems (terrestrial) Decrease in Net Primary Productivity because of water
shortage as proxy for total species loss
Pfister et al. 2009
Ecosystems (aquatic) Fish species loss due to decreased river discharge Hanafiah et al. 2011
Land use Ecosystems (terrestrial) Species loss due to different types of land use
(agriculture, forestry, built up). Species loss caused
by transformation of natural land to used land,
including the time it takes to back-transform to natural
land
De Baan et al. 2013; Curran
et al. 2014
Mineral resource scarcity Resource scarcity Cost increase due to mineral extraction increase Vieira et al. 2016b
Fossil resource scarcity Resource scarcity Cost increase due to fossil extraction increase Vieira et al. 2016c
Int J Life Cycle Assess
This information was combined with the disability weight per
cancer type (Frischknecht et al. 2000;DeSchryveretal.
2011).
3.2.4 Fine particulate matter formation
Starting from the intake fraction, human effect and damage
due to cardiopulmonary and lung cancer mortality of fine
particulate matter were determined by Van Zelm et al. (2016).
3.2.5 Photochemical ozone formation
Starting from the intake fraction, effect and damage factors of
respiratory mortality due to ozone exposure were determined
by Van Zelm et al. (2016). For damage to terrestrial ecosys-
tems, the effect factor describes the change in PDF of forest
and grassland species due to the change in ground level ozone
exposure over forest and grassland area (Van Goethem et al.
2013a,b).
3.2.6 Terrestrial acidification
An effect factor was added to the endpoint calculations, de-
scribing the absence of species due to acidity of soils (Roy
et al. 2014). The effect factor quantifies the change in the PDF
of vascular plant species due to a change in the H
+
concentra-
tion and was derived for specific biomes, such as temperate
broadleaf mixed forest, tundra and (sub)tropical moist broad-
leaf forest (Azevedo et al. 2013a).
3.2.7 Freshwater eutrophication
The effect factor, added to the midpoint calculations, describes
the absence of species due to phosphorus concentrations in
freshwater (Azevedo et al. 2013b,c). It reflects the change
in PDF of species due to a change in total P concentration
and depends on the freshwater type (rivers or lakes), species
group (heterotrophs and autotrophs) and climate type (warm,
temperate, xeric or cold).
3.2.8 Toxicity
Ecotoxicological damage factors, added to the midpoint cal-
culations, were considered to equal one, as the effects estimat-
ed with acute toxicity data may approximate toxic effects in
field conditions (Posthuma and De Zwart 2006). For human
health, damage factors for carcinogenic or non-carcinogenic
effects were included (Huijbregts et al. 2005).
3.2.9 Water use
Impacts of water consumption on human health refer to
DALYs due to malnutrition, as caused by water shortage in
low development countries (Pfister et al. 2009). Impacts of
water consumption on terrestrial ecosystems were taken from
Pfister et al. (2009), who quantifiedthem based on the damage
for vascular plant species using net primary productivity
(NPP) as a proxy. Impacts of water consumption on freshwa-
ter ecosystems were taken from Hanafiah et al. (2011), who
quantified them as the change in fish species lost associated
with a decrease in discharge.
3.2.10 Land use
The mid-to-endpoint modelling for land use does not add
further steps to the cause-impact pathway, as the midpoint
characterisation factors already refer to local species loss.
3.2.11 Mineral resource scarcity
The mid-to-endpoint factor for mineral resource scarcity refers
to the conversion from surplus ore to surplus costs.
Cumulative tonnage relationships for surplus costs of 12
metals, as developed by Vieira et al. (2016b), were used as
input in the calculations.
3.2.12 Fossil resource scarcity
Endpoint characterisation factors for the extraction of crude
oil, natural gas and hard coal, expressed as Surplus Cost
Potential (SCP), were based on cumulative cost-tonnage rela-
tionships for these three fossil resources (Vieira et al. 2016c).
Note that we were not able to arrive at a constant mid-to-
endpoint factor for fossil resources due to lack of understand-
ing about the full cause-effect pathway.
4Discussion
Although significant effort has been put into the development
of ReCiPe2016, there is still major improvement potential in
the way impact pathways are modelled. A number of improve-
ment options are discussed below.
4.1 Scenario analysis
Due to lack of data, the influence of time horizon and level of
evidence was not considered in the calculation of characteri-
sation factors for photochemical ozone formation, terrestrial
acidification, freshwater eutrophication, land use and fossil
resource scarcity. This needs to be improved in future updates
of ReCiPe, if more information on value choices becomes
available in the underlying models employed.
Int J Life Cycle Assess
4.2 Regionalisation
Country- or region-specific characterisation factors for mid-
points and endpoints were included for a number of impact
categories, including fine particulate matter formation, photo-
chemical ozone formation, acidification, freshwater eutrophi-
cation and water use. Country-specific characterisation factors
for midpoints and endpoints were included for a number of
impact categories, including fine particulate matter formation,
photochemical ozone formation, acidification, freshwater eu-
trophication and water use. Particularly for the global models
related to fine particulate matter formation and photochemical
ozone formation, a higher spatial resolution at the global scale
and with a closer spatial connection between fate, exposure
and effects can further improve the reliability of LCIA (see,
e.g. Apte et al. 2015;Braueretal.2016). For other impact
categories, spatial differentiation has not been considered at all
in ReCiPe2016 and major improvements are possible on this
point. Most prominent impact categories for providing
regionalised results are land use (Chaudhary et al. 2015)and
toxicity (Kounina et al. 2014). For toxicity, spatial differenti-
ation can be considered particularly relevant for the modelling
of ecological impacts of metals, if speciation is taken into
account in fate, exposure and effect calculations (see, e.g.
Dong et al. 2016).
4.3 Global species extinction
Damage to ecosystem quality in ReCiPe2016 refers to the
aggregated local loss of species over space and time. Global
species extinction risk may, however, also be considered as an
indicator for ecosystem quality in addition to local species
loss. For both water use and land use, there are already possi-
bilities to account for global species decline in life cycle im-
pact assessment (see Chaudhary et al. 2015; Verones et al.
2015). Further research is needed to expand this also to other
impact categories.
4.4 Missing pathways
With ReCiPe2016, we firstly focused on advancing the impact
modelling of categories that were classified as interim by
Hauschild et al. (2013). Not all exposure and damage path-
ways could, however, be modelled in ReCiPe2016. First, hu-
man exposure pathways related to indoor emissions to
chemicals and fine particulate matter (Rosenbaum et al.
2015; Hodas et al. 2016) and direct application of pesticides
to food items (Fantke and Jolliet 2015) were not included and
should be considered in future updates of ReCiPe. There are
also missing pathways in the endpoint modelling of existing
impact categories due to lack of global information, such as
the change in incidence of infectious diseases due to climate
change (see, e.g. Fan et al. 2015). For fossil resource scarcity,
we were not able to establish a mid-to-endpoint factor which
requires further improvement. Finally, additional impact cate-
gories should be considered, particularly related to the marine
environment, such as marine eutrophication, invasive species
and plastic debris (Woods et al. 2016). For human health,
noise is a potentially relevant impact category to be considered
in a future update (see, e.g. Cucurachi and Heijungs 2014).
Impacts from emerging activities and substances, such as im-
pacts from nanoparticles, are also potentially relevant for fur-
ther expansion (Pini et al. 2016).
5Conclusions
Life cycle impact assessment is a fast evolving field of re-
search. ReCiPe2016 provides a state-of-the-art method to con-
vert life cycle inventories to a limited number of life cycle
impact scores on midpoint and endpoint level. Three endpoint
categories (human health, ecosystem quality and resource
scarcity) and 17 midpoint categories were included with a
focus on providing characterisation factors that are represen-
tative on the global scale in line with the global nature of many
product life cycles.
Acknowledgements The research was supported by the Dutch
National Institute for Public Health and the Environment RIVM-project
S/607020, Measurably Sustainable within the spearhead Healthy and
Sustainable Living Environment, commissioned by the Director-
General of RIVM and run under the auspices of RIVM’s Science
Advisory Board.
References
Apte JS, Marshall JD, Cohen AJ, Brauer M (2015) Addressing global
mortality from ambient PM2.5. Environ Sci Technol 49:8057–8066
Azevedo LB, Van Zelm R, Hendriks AJ, Bobbink R, Huijbregts MAJ
(2013a) Global assessment of the effects of terrestrial acidification
on plant species richness. Environ Pollut 174:10–15
Azevedo LB, Henderson AD, van Zelm R, Jolliet O, Huijbregts MAJ
(2013b) Assessing the importance of spatial variability versus model
choices in life cycle impact assessment: the case of freshwater eu-
trophication in Europe. Environ Sci Technol 47:13565–13570
Azevedo LB, van Zelm R, Elshout PMF, Hendriks AJ, Leuven RSEW,
Struijs J, de Zwart D, Huijbregts MAJ (2013c) Species richness–
phosphorus relationships for lakes and streams worldwide. Glob
Ecol Biogeogr 22:1304–1314
BouwmanAF,BeusenAHW,BillenG(2009)Humanalteration
of the global nitrogen and phosphorus soil balances for the
period 1970–2050. Global Biogeochem Cy 23:GB0A04.
doi:10.1029/2009GB003576
Brauer M, Freedman G, Frostad J, van Donkelaar A, Martin RV, Dentener
F, Dingenen RV, Estep K, Amini H, Apte JS, Balakrishnan K (2016)
Ambient air pollution exposure estimation for the global burden of
disease 2013. Environ Sci Technol 50:79–88
Chaudhary A, Verones F, De Baan L, Hellweg S (2015) Quantifying land
use impacts on biodiversity: combining species-area models and
vulnerability indicators. Environ Sci Technol 49:9987–9995
Int J Life Cycle Assess
Cucurachi S, Heijungs R (2014) Characterisation factors for life cycle
impact assessment of sound emissions. Sci Total Environ 468-469:
280–291
Curran M, Hellweg S, Beck J (2014) Is there any empirical support for
biodiversity offset policy? Ecol Appl 24:617–632
De Baan L, Alkemade R, Köllner T (2013) Land use impacts on biodi-
versity in LCA: a global approach. Int J Life Cycle Assess 18:1216–
1230
De Schryver AM, Van Zelm R, Humbert S, Pfister S, McKone TE,
Huijbregts MAJ (2011) Value choices in life cycle impact assess-
ment of stressors causing human health damage. J Indus Ecol 15:
796–815
De Schryver AM, Brakkee KW, Goedkoop M, Huijbregts MAJ (2009)
Characterization factors for global warming in life cycle assessment
based on damages to humans and ecosystems. Environ Sci Technol
43:1689–1695
Döll P, Siebert S (2002) Global modelling of irrigation water require-
ments. Water Resour Res 38:1037
Dong Y, Rosenbaum RK, Hauschild MZ (2016) Assessment of metal
toxicity in marine ecosystems: comparative toxicity potentials for
nine cationic metals in coastal seawater. Environ Sci Technol 50:
269–278
EC (2013) Commission recommendation of 9 April 2013 on the use of
common methods to measure and communicate the life cycle envi-
ronmental performance of products and organisations. 56:1–210
Elshout PMF, Van Zelm R, Karuppiah R, Laurenzi IJ, Huijbregts MAJ
(2014) A spatially explicit data-driven approach to assess the effect
of agricultural land occupation on species groups. Int J Life Cycle
Assess 19:758–769
Fan J, Wei W, Bai Z, Fan C, Li S, Liu Q, Yang K (2015) A systematic
review and meta-analysis of dengue risk with temperature change.
Int J Environ Res Public Health 12:1–15
Fantke P, Jolliet O (2015) Life cycle human health impacts of 875 pesti-
cides. Int J Life Cycle Assess 21:722–733
Frischknecht R, Braunschweig A, Hofstetter P, Suter P (2000) Human
health damages due to ionising radiation in life cycle impact assess-
ment. Environmental Impact Asses Rev 20:159–189
Goedkoop M, Heijungs R, Huijbregts MAJ, De Schryver A, Struijs J, van
Zelm R (2009) ReCiPe 2008: a life cycle impact assessment method
which comprises harmonised category indicators at the midpoint
and endpoint levels. First edition. Report i: characterization.
The Netherlands: Ruimte en Milieu, Ministerie van
Volkshuisvesting, Ruimtelijke Ordening en Milieubeheer
Hanafiah MM, Xenopoulos MA, Pfister S, Leuven RS, Huijbregts MAJ
(2011) Characterization factors for water consumption and green-
house gas emissions based on freshwater fish species extinction.
Environ Sci Technol 45:5572–5278
Hauschild MZ, Goedkoop M, Guinee J, Heijungs R, Huijbregts M, Jolliet
O, Margni M, De Schryver A, Humbert S, Laurent A, Sala S, Pant R
(2013) Identifying best existing practice for characterization model-
ing in life cycle impact assessment. Int J Life Cycle Assess 18:683–
697
Hauschild MZ, Huijbregts MAJ (2015) Introducing life cycle impact
assessment. In: Hauschild M, Huijbregt M (eds) Life cycle impact
assessment. Springer, Dordrecht Chapter 1
Hayashi K, Nakagawa A, Itsubo N, Inaba A (2006) Expanded damage
function of stratospheric ozone depletion to cover major endpoints
regarding life cycle impact assessment. Int J Life Cycle Assess 11:
150–161
Helmes RJK, Huijbregts MAJ, Henderson AD, Jolliet O (2012) Spatially
explicit fate factors of phosphorous emissions to freshwater at the
global scale. Int J Life Cycle Assess 17:646–654
Hodas N, Loh M, Shin H-M, Li D, Bennett D, McKone TE, Jolliet O,
Weschler CJ, Jantunen M, Lioy P, Fantke P (2016) Indoor inhalation
intake fractions of fine particulate matter: review of influencing fac-
tors. Indoor Air 26:836–856
Hoekstra AY, Mekonnen MM (2012) The water footprint of humanity.
PNAS 109:3232–3237
Huijbregts MAJ, Rombouts LJA, Ragas AMJ, Van de Meent D (2005)
Human-toxicological effect and damage factors of carcinogenic and
noncarcinogenic chemicals for life cycle impact assessment. Integr
Environ Assess Manag 1:181–244
Huijbregts MAJ, Steinmann ZJN, Elshout PMF, Stam G, Verones F,
Vieira MDM, Van Zelm R, (2016) ReCiPe2016. A harmonized life
cycle impact assessment method at midpoint and endpoint level.
Report I: characterization. RIVM Report 2016–0104. National
Institute for Human Health and the Environment, Bilthoven
IPCC (2013) Climate change 2013: the physical science basis.In: Stocker
TF, Qin D, Plattner GK, Tignor M, Allen SK, Boschung J, Nauels A,
Xia Y, Bex V, Midgley PM (eds) Contribution of working group I to
the fifth assessment report of the intergovernmental panel on climate
change. Cambridge University Press, Cambridge, p. 1535.
doi:10.1017/CBO9781107415324
Joos F, Roth R, Fuglestvedt JS, Peters GP, Enting IG, Von Bloh W,
Brovkin V, Burke EJ, Eby M, Edwards NR, Friedrich T, Frölicher
TL, Halloran PR, Holden PB, Jones C, Kleinen T, Mackenzie FT,
Matsumoto K, Meinshausen M, Plattner G-K, Reisinger A,
Segschneider J, Shaffer G, Steinacher M, Strassmann K, Tanaka
K, Timmermann A, Weaver AJ (2013) Carbon dioxide and climate
impulse response functions for the computation of greenhouse gas
metrics: a multi-model analysis. Atmos Chem Phys 13:2793–2825
Jungbluth N, Frischknecht R (2010) Cumulative energy demand. In:
Hischier R, Weidema B (eds) Implementation of life cycle impact
assessment methods. Ecoinvent centre, St Gallen, pp. 33–40
Kounina A, Margni M, Shaked S, Bulle C, Jolliet O (2014) Spatial anal-
ysis of toxic emissions in LCA: a sub-continental nested USEtox
model with freshwater archetypes. Environ Int 69:67–89
Pfister S, Koehler A, Hellweg S (2009) Assessing the environmental
impacts of freshwater consumption in LCA. Environ Sci Technol
43:4098–4104
Pini M, Salieri B, Ferrari AM, Nowack B, Hischier R (2016) Human
health characterization factors of nano-TiO2 for indoor and outdoor
environments. Int J Life Cycle Assess 21:1452–1462
Posthuma L, De Zwart D (2006) Fish community responses in the field
show the empirical meaning of the potentially affected fraction of
species as relative measure of mixture risk. Environ Toxicol Chem
25:1094–11 05
Rosenbaum RK, Bachmann TM, Gold LS, Huijbregts MAJ, Jolliet O,
Juraske R, Koehler A, Larsen HF, MacLeod M, Margni M, McKone
TE, Payet J, Schuhmacher M, Van de Meent D, Hauschild MZ
(2008) USEtox-the UNEP-SETAC toxicity model: recommended
characterisation factors for human toxicity and freshwater
ecotoxicity in life cycle impact assessment. Int J Life Cycle Assess
13:532–546
Rosenbaum R, Meijer A, Demou E, Hellweg S, Jolliet O, Lam N, Margni
M, McKone TEM (2015) Indoor air paper. Indoor air pollutant ex-
posure for life cycle assessment: regional health impact factors for
households. Environ Sci Technol 21:12823–12831
Roy P-O, Azevedo LB, Margni M, Van Zelm R, Deschênes L, Huijbregts
MAJ (2014) Characterization factors for terrestrial acidification at
the global scale: a systematic analysis of spatial variability and un-
certainty. Sci Total Environ 500:270–276
Roy PO, Huijbregts M, Deschenes L, Margni M (2012a) Spatially-
differentiated atmospheric source-receptor relationships for nitrogen
oxides, sulfur oxides and ammonia emissions at the global scale for
life cycle impact assessment. Atmos Environ 62:74–81
Roy PO, Deschenes L, Margni M (2012b) Life cycle impact assessment
of terrestrial acidification: modeling spatially explicit soil sensitivity
at the global scale. Environ Sci Technol 46:8270–8278
Urban MC (2015) Accelerating extinction risk from climate change.
Science 348:571–573
Int J Life Cycle Assess
Van Goethem T, Azevedo LB, Van Zelm R, Hayes RM, Ashmore MR,
Huijbregts MAJ (2013a) Plant species sensitivity distributions for
ozone exposure. Environ Pollut 178:1–6
Van Goethem T, Preiss P, Azevedo LB, Friedrich R, Huijbregts MAJ, Van
Zelm R (2013b) European characterization factors for damage to
natural vegetation by ozone in life cycle impact assessment.
Atmos Environ 77:318–324
Van Zelm R, HuijbregtsMAJ, Van de Meent D (2009) USES-LCA 2.0: a
global nested multi-media fate, exposure and effects model. Int J
Life Cycle Assess 14(30):282–284
Van Zelm R, Stam G, Huijbregts MAJ, Van de Meent D (2013) Making
fate and exposure models for freshwater ecotoxicity in life cycle
assessment suitable for organic acids and bases. Chemosphere 90:
312–317
Van Zelm R, Preiss P, Van Goethem T, Van Dingenen R, Huijbregts MAJ
(2016) Regionalized life cycle impact assessment of air pollution on
the global scale: damage to human health and vegetation. Atmos
Environ 134:129–137
Verones F, Huijbregts MAJ, Chaudhary A, de Baan L, Koellner T,
Hellweg S (2015) Harmonizing the assessment of biodiversity
effects from land and water use within LCA. Environ Sci Technol
49:3584–3592
Vieira MDM, Ponsioen TC, Goedkoop M, Huijbregts MAJ (2016a)
Surplus ore potential as a scarcity indicator for resource extraction.
J Indus Ecol. doi:10.1111/jiec.12444
Vieira MDM, Ponsioen TC, Goedkoop M, Huijbregts MAJ (2016b)
Surplus cost potential as a life cycle impact indicator for metal ex-
traction. Resources 5:1–12
Vieira MDM, Ponsioen T, Goedkoop M, Huijbregts MAJ (2016c) Fossil
resource scarcity. In: Huijbregts MAJ, Steinmann ZJN, Elshout
PMF, Stam G, Verones F, Vieira MDM, Van Zelm R (eds)
ReCiPe2016. A harmonized life cycle impact assessment method
at midpoint and endpoint level. Report I: characterization. RIVM
Report 2016-0104. National Institute for Human Health and the
Environment, Bilthoven. Chapter 13
WMO (2011) Scientific assessment of ozone depletion: 2010, Global
Ozone Research and Monitoring Project-report no.52. World
Meteorological Organization, Geneva
Woods JS, Veltman K, Huijbregts MAJ, Verones F, Hertwich EG (2016)
Towards a meaningful assessment of marine ecological impacts in
life cycle assessment (LCA). Environ Int 89-90:48–61
Int J Life Cycle Assess
