The 6th International Conference on Life Cycle Management in Gothenburg 2013
USING WATER FOOTPRINTING ON A REGULAR BASIS -
PROBLEMS AND CURRENT DEVELOPMENTS FROM AN
INDUSTRY PERSPECTIVE
Andreas Uihlein* & Ellen Riise
SCA Hygiene Products GmbH
*Sandhofer Straße 176, 68305 Mannheim, Email:
[email protected] Keywords: water footprint; water inventory; supply chain; product portfolio
ABSTRACT
Water footprinting is still in its infancy. So far, an accepted general
framework and various water footprint impact assessment methods have
been developed. SCA performed a case study to evaluate the usefulness and
applicability of water footprinting in industry. For water footprinting
becoming a useful tool for decision making there are several problems: data
is lacking and of poor quality. In addition, some impact assessment methods
are not operational yet. Different operational impact assessment methods
might lead to different conclusions because of different pathways or
endpoints modelled or different characterisation factors. On-going
improvements will solve some of those problems and will hopefully allow
to use water footprinting as a robust basis for decision making in industry.
INTRODUCTION
SCA performed a case study on water footprinting for four selected tissue products. The main
focus of the project was the identification and analysis of existing and proposed water footprint
methods and their possible application in industry. The results of the study will be used
internally to identify hotspots and possible future improvement areas in terms of water use as
well as for educational purposes within the company. The study will also be used to identify the
advantages and challenges of water footprinting and the appropriateness and applicability of the
various methods related to SCA’s product portfolio.
MATERIALS AND METHODS
A screening of existing water footprint methods has been performed. The methods found in
literature range from simple indexes to sophisticated endpoint methods. While only the most
recent simpler methods have been selected, all midpoint and endpoint methods have been
applied in the case study.
Simple indexes can be based on water scarcity or vulnerability. The index methods applied in
this study are the water stress index (Pfister et al., 2009) and the water impact index (Veolia
water, 2010). Going beyond indexes, midpoint and endpoint methods for water footprinting
exist. As midpoint methods, we have chosen the WFN scarcity method (Hoekstra et al., 2009),
the ecological scarcity method (Frischknecht et al., 2009), and the methods according to Milà i
Canals et al. (2009), Bayart et al. (2010), and Boulay et al. (2011). As endpoint methods we have
selected: Pfister et al. (2009), Bayart et al. (2010), Boulay et al. (2011), Verones et al. (2010), Van
Zelm et al. (2011), Bösch et al. (2007), and Motoshita et al. (2008, 2009).
The 6th International Conference on Life Cycle Management in Gothenburg 2013
RESULTS
Water flows and water balance
A water balance including water inputs and outputs of each process of the life cycle of the
products was compiled. Typically, many different databases and data sources are used in
LCAs performed by industry and are mixed in a life cycle model (e.g. Ecoinvent database,
ILCD/ELCD database, data from industry associations, primary data). Nomenclature and the
way of how water is modelled vary between databases. Sometimes, only water withdrawals
are accounted for while water returns are documented only sometimes. Incorporation into
products and evaporation are usually not reported.
Often, water inputs and water outputs are not balanced in datasets with generic data. In
addition, some water flows are not specified (e.g. an input flow is just named ‘water’) which
makes the calculation of impacts impossible if we go beyond water volume accounting. In the
case study, for every individual process, an in-depth investigation had to be performed to
complete the water balance involving many assumptions and estimations. To perform a water
footprint assessment, it is also necessary to identify the location of water use for every
process. This is especially difficult for aggregated and generic datasets from databases that are
usually used for upstream data (e.g. “Diesel mix at refinery, EU-27”).
Water footprint assessment results
The results for the four products of the different water footprint methods have been compared by
establishing a ranking. Interestingly, all methods that are operational come to the same
conclusions regarding the ranking of the products. We also analysed the contribution of each life
cycle step to the results for each method applied (Figure 1). The results differ between the
methods. Most methods show highest impacts for raw material supply, followed by the
production process. For some methods, the water footprint is dominated by raw material supply,
for example the WFN scarcity (Hoekstra et al., 2009), Pfister (Pfister et al., 2009) and Boulay
endpoint (Boulay et al., 2011) methods. Those ‘outlier’ methods have very high characterization
factors for those countries where some of processes for raw material supply are located.
DISCUSSION
Data gaps and data quality
Data gaps exist for primary data but also in databases for secondary or generic data. Data
collection systems on water use upstream the supply chain are usually not in place. For
secondary data from databases, water inventories are not complete and not transparent. A
typical data gap is that only water withdrawals or water inputs are accounted for while
information on the fate of water is missing. Incorporation into products and evaporation are
usually not reported. Another problem of secondary data is that often the geographic location
of water flows are not known.
Data quality for water flows from existing databases is not very good. Data quality of primary
data collected by industry for own operations and from suppliers is better, however, in some
cases, questions on water flows cannot be answered since there are no measurements done.
Compared to energy there are fewer water meters installed, for example, split between
production lines cannot be done. So, there are many assumptions and estimates needed to get
The 6thh Internationnal Conferen
nce on Life Cycle Man
nagement in
n Gothenburrg 2013
a dataseet as compllete as posssible. In adddition, datta quality fo
or evaporattion from in
ndustrial
processees also has to be impro
oved.
Figure 11. Percentagge of life cy
ycle steps off total waterr footprint according
a too selected methods
m
Impact aassessment
Concernning the water
w footprrint methodds, many midpoint
m an
nd endpoinnt methods are not
operatioonal yet butt only prop pose a fram
mework (e.g. Bayart et al., 2010). Some auth hors like
Veroness et al. (20010) or Van n Zelm et al. (2011) only provid de characteerisation facctors for
selectedd case studiees (e.g. seleected waterssheds or cou
untries).
When loooking at thhe methods that were aapplied in this t case stuudy, the folllowing con nclusions
can be drawn: thee WSI acco ording to Pffister et al. (2009) is an index thhat is very easy to
understaand. It can be used as a first screeening indicator as alsoo proposed bby the auth hors. The
Boulay midpoint inndicator (B Boulay et al ., 2011) is very easy tot use. The ecological scarcity
method (Frischkneecht et al., 2009) can be easily applieda but is probablyy best to use when
water fo
footprint is integrated into a full LCA with h also the other
o impacct categoriees being
calculatted using thhe ecologicaal scarcity m
method. Thee endpoint methods
m thaat have been
n proven
to be moost operatioonal are the Pfister et all. (2009) an
nd Boulay ett al. (2011) endpoint methods.
m
CONCL
LUSIONS
Several conclusionss can be drrawn from tthis case stu tudy. The first
f findinggs indicate that
t data
quality ffor water floows from seecondary datta is not verry good. Maany data gapps exist. Forr primary
data, usuually, data quality
q is beetter, howeveer, data quaality should be
b improvedd by installing water
meteringg systems orr improving them (e.g. aallowing forr disaggregatted measureements).
We exppect that thee accounting
g of water fflows will beb improved in seconddary databases soon
and somme progress can be seeen already nnow. If we want
w to imp
prove data qquality and to close
data gaaps, it is important to consider that demaands on daata differ ffor variouss impact
assessm
ment methodds. It is thus often diffficult to deccide what data to colleect. What we
w would
need is an agreemeent on whatt data to coollect (e.g. what
w types of water, w
what informmation on
locationn and waterr quality, which
w tempporal disagg gregation too use). We hope that the ISO
The 6th International Conference on Life Cycle Management in Gothenburg 2013
standard 14046 on water footprinting will help in defining what should be the minimum
requirements regarding data collection.
Concerning the impact assessment methods, many midpoint and endpoint methods are not
operational yet but only propose a framework while others provide characterisation factors for
selected case studies. In addition, the case study has shown that different operational impact
assessment methods might lead to different conclusions.
From our case study, we have seen that water footprinting is possible, however, not at all
straightforward. A lot of manual data handling, assumptions and estimates are needed that all
have to be done on a case-by-case level. This will hinder water footprinting to be applied (or
added to more traditional impact categories) in industry for larger product portfolios, or for
regular performance tracking on assortment level. Data quality issues and the differences
between IA methods make it extremely difficult to use water footprinting as a robust basis
for decision making in industry.
REFERENCES
Bayart, J.B., Bulle, C., Deschênes, L., Margni, M., Pfister, S., Vince, F., Koehler, A. (2010). A framework for
assessing off-stream freshwater use in LCA. The International Journal of Life Cycle Assessment, 15, 439-453
Bösch, M.E., Hellweg, S., Huijbregts, M.A.J., Frischknecht, R. (2007). Applying Cumulative Exergy Demand
(CExD) Indicators to the ecoinvent Database. The International Journal of Life Cycle Assessment, 12, 181-190
Boulay, A.M., Bulle, C., Bayart, J.B., Deschênes, L., Margni, M. (2011b). Regional Characterization of Freshwater
Use in LCA: Modeling Direct Impacts on Human Health. Environmental Science & Technology, 20, 8948-8957
Frischknecht, R., Steiner, R., Jungbluth, N. (2009). The Ecological Scarcity Method Eco-Factors 2006. Bern,
Switzerland: Federal Office for the Environment
Hoekstra, A.Y., Chapagain, A.K., Aldaya, M.M., Mekonnen, M.M. (2009). Water Footprint Manual. State of the
Art 2009. Enschede, The Netherlands: Water Footprint Network
Milà i Canals, L., Chenoweth, J., Chapagain, A., Orr, S., Antón, A., Clift, R. (2009). Assessing freshwater use
impacts in LCA: Part I - inventory modelling and characterisation factors for the main impact pathways. The
International Journal of Life Cycle Assessment, 14, 28-42
Motoshita, M., Itsubo, N., Inaba, A. (2008). Development of impact assessment method on health damages of
undernourishment related to agricultural water scarcity. Proceedings of the 8th International Conference on
EcoBalance, Tokyo, 10 -12 December
Motoshita, M., Itsubo, N., Inaba, A. (2009). Development of damage assessment model for infectious diseases
arising from domestic water consumption. Proceedings of the SETAC Europe 19th Annual Meeting, Gothenburg,
31 May - 4 June
Pfister, S., Köhler, A., Hellweg, S. (2009). Assessing the Environmental Impacts of Freshwater Consumption in
LCA. Environmental Science & Technology, 43, 4098-4104
Van Zelm, R., Schipper, A.M., Rombouts, M., Snepvangers, J., Huijbregts, M.A.J. (2011). Implementing
groundwater extraction in life cycle impact assessment: characterization factors based on plant species richness.
Environmental Science and Technology, 45, 629-635
Veolia Water (2010). The Water Impact Index and the First Carbon-Water Analysis of a Major Metropolitan Water
Cycle. Veolia Water
Verones, F., Hanafiah, M.M., Pfister, S., Huijbregts, M.A.J., Pelletier, G.J., Koehler, A. (2010). Characterization
Factors for Thermal Pollution in Freshwater Aquatic Environments. Environmental Science & Technology, 44,
9364-9369
