Tesi defintiva10 - international ph.d. in management

Transcrição

Integrated assessment of renewable energies for
decision making:
A two-case analysis
Thesis presented by:
Matteo Borzoni
to
The class of social sciences
for the degree of
Doctor of Philosophy
in the subject of
Management, competitiveness and development
Tutor: Prof. Marco Frey
Scuola Superiore Sant’Anna
A.Y. 2010-2011
La vida es una funcion derivada de las relaciones
entre energia individual y cosmo.
V = f( i, c).
José De Letamendi
Spanish doctor from the 19th century who
distinguished himself not only for his gallant fight
against the cholera epidemic in Barcelona but also
for his essays on law, economics and philosophy.
Contents
Acknowledgments .................................................................................................................i
INTRODUCTION ..............................................................................................................1
References ...........................................................................................................................7
A THEORETICAL REVIEW OF SUSTAINABILITY ASSESSMENT........................10
2.1 TRADITIONAL COST-BENEFIT ANALYSIS AND COMMENSURABILITY ...........................10
2.2 COMPLEXITY AND COMPLEX ADAPTIVE SYSTEMS.......................................................15
2.3 THE WAY OUT ..............................................................................................................17
References .........................................................................................................................21
MULTI-SCALE INTEGRATED ASSESSMENT OF SOYBEAN BIODIESEL IN
BRAZIL ............................................................................................................................25
3.1 INTRODUCTION ............................................................................................................26
3.2 SCENARIOS, GENERAL FRAMEWORK AND METHODOLOGY .........................................28
3.2.1 VARIABLES ........................................................................................................28
3.2.2 NET DELIVERY OF BIOFUELS..............................................................................30
3.2.3 SCENARIOS ........................................................................................................31
3.2.4 ENERGY BALANCE AND ENERGY ANALYSIS ......................................................33
3.2.5 INPUT-OUTPUT ANALYSIS: DATA AND DIRECT COEFFICIENT MATRIX ...............37
3.2.6 INPUT-OUTPUT ANALYSIS: IMPACT ASSESSMENT ..............................................41
3.3 DISCUSSION AND RESULTS...........................................................................................43
3.4 CONCLUSIONS ..............................................................................................................47
References .........................................................................................................................49
Appendix A3.1 – Additional results..................................................................................55
SOCIAL-MULTI CRITERIA EVALUATION OF ALTERNATIVE GEOTHERMAL
POWER SCENARIOS: THE CASE OF MT. AMIATA IN ITALY...............................56
4.1 INTRODUCTION ............................................................................................................57
4.2 METHODOLOGICAL FRAMEWORK ................................................................................59
4.3 HISTORICAL-INSTITUTIONAL ANALYSIS ......................................................................62
4.3.1 HISTORICAL CONTEXT .......................................................................................62
4.3.2 THE SCIENTIFIC DEBATE ....................................................................................64
4.3.3 CURRENT STATUS ..............................................................................................66
4.3.4 SOCIAL ACTORS .................................................................................................67
4.4 THE MULTI-CRITERIA MATRIX .....................................................................................71
4.4.1 GENERATION OF ALTERNATIVES .......................................................................71
4.4.2 CHOICE AND ESTIMATION OF CRITERIA .............................................................72
4.5 RANKING ALTERNATIVES ............................................................................................83
4.6 CONCLUSIONS ..............................................................................................................90
References .........................................................................................................................92
Appendix A4.1 – Summary of the interviews ...................................................................99
Appendix A4.2 – Cost structure ......................................................................................100
Appendix A4.3 – Additional results: sensitivity analysis................................................105
CONCLUSIONS.............................................................................................................108
References .......................................................................................................................113
Acknowledgments
In order to increase the chances of getting our work published, as part of a
seminar in the first year of my PhD course, we were advised to highlight what
differentiates our research by using words and phrases such as “contrary to”,
“innovative”, “surprisingly” etc. I do not think I have made much use of this wise
advice in my dissertation and so I am daring to apply it here.
Contrary to the most basic rules of logic, common sense and established
conventions, I want to start my acknowledgments by blaming somebody instead
of thanking them.
At the beginning of my PhD I used to play squash with a great friend of
mine, Sittáro. After our matches he used to dissimulate his profound happiness for
having consistently won by talking on topics that at that time sounded like pure
buzz words: MUSIASEM, MSIASM, SMCE, multi-criteria analysis, etc. Those
buzz words stimulated my curiosity and became the main methods of my thesis.
So Sittaro is the first person to be blamed for everything that I was unable to
accomplish. By the same token, he is the first person to be thanked. In addition, he
provided me with extremely useful suggestions while I was drafting the chapter
applying multi-criteria analysis, and finally he commented on the first draft of the
same chapter. This is why I want to thank him twice. By the way, I’m convinced
that had I won a couple more squash matches, this thesis would have been
completely different.
I’m grateful to my supervisor, Prof. Marco Frey, because he always
encouraged me to pursue my research interests including those that had little to do
with his background. Plus his comments on the early draft of the dissertation were
of great value.
In a certain sense, the research I carried out shares many things in common
with complex adaptive systems: there are legacy effects (from previous studies),
time lags (in submitting the thesis), a recombination of previous of events and
novelty. In 2007, a long before I decided to do a PhD program, I attended a
conference in Iceland. After many presentations I was tired and I went to the
conference cafeteria for a coffee, where, by chance, I met Prof. Decio
Zylbersztajn. Two years later he invited me to join his research group in Sao
Paulo. I’m grateful to him because he made it possible to collect the data I needed
for the biodiesel chapter. But even more, I would like to express my sincere
gratitude to him for having so clearly explained to me that whatever external
development intervention always ends up breaking social equilibria, consequently,
even the best intentioned actions can easily provoke disastrous effects.
i
I would like to thank Prof. Maria Sylvia Saes for her great support in
obtaining the data I needed but also for letting me taste what Brazilians consider
the best pizza in the world. On that occasion I had the first clear example of weak
comparability, an underlying theme of this thesis. In fact, before deciding if that
was actually the best pizza in the world, we needed to agree on the meaning of
pizza. We clearly had two very different ideas. Thanks are due to all the PENSA
group in Sao Paulo for their willingness to help me and for having made my stay
there so pleasant. For this I am indebted to Kassinha, Andrei, Flavia, Raquel,
Camila, Bruno, Nadia and Evandro.
I gratefully acknowledge the influence Mario Giampietro from Universitat
Autonoma de Barcelona had on the entire work.
I would also like to thank to Prof. Maurizio Grassini from the University of
Florence for his clarifications on input-output analysis. The discussions with
Simone Bertini and Stefano Rosignoli from IRPET were also of great help.
I would like to express my gratitude to Franz. If he hadn’t introduced me to
the geothermal research area, the multi-criteria chapter would not have seen the
light. He also commented on the same chapter. The exchanges with Gonzalo
Gamboa on the problems arising during the multi-criteria exercise were also very
useful. Thanks are also due to all the CEGL staff for their technical clarifications
on geothermal exploitation and their assistance in collecting data: Fausto, Isabella,
Enrica and Giacomo.
Many people gave me their time for the interviews on Mt. Amiata.
I also need to thank Rolando for his willingness to help me with the
estimations of the taxes of the geothermal power plants.
And last, but not least, there is Cristina. I started this preface with the
concept of blaming. Thus following the common sense practice of recalling the
starting point of a piece of research in the conclusions, I would like to conclude
again by blaming someone. But as much as I try, I cannot find any single reason
to blame her. I simply have to thank her. She knows why.
ii
1
Introduction
The oil age began about 150 years ago and is still in place. According to many
scholars, an easily available energy form supporting major advances in
manufacturing, agriculture and transport has probably been the main enabler of
economic growth throughout human history (Boulding, 1966; Bullard and
Herendeen, 1975; Cleveland, 1991; Dung, 1992; Hall et al., , 1992; Hall et al.,
2008; Jorgenson, 1984; Küummel, 1982; Rosenberg, 1976). Today oil is still the
main source of energy providing about 30% of the world’s total primary energy
supply, while the entire set of fossil energies makes up more than 80% (IEA,
2010a). The pro-capita primary energy consumption in OECD countries is 195 GJ
with large differences between countries. The average American consumes 314
GJ while in Italy the average pro-capita consumption is 123 GJ (IEA, 2010a). The
energy contained in one barrel of oil is more than 6 GJ. Such heat content would
be generated by human muscles in about 2.5 years (Hagens and Mulder, 2008). In
a certain sense we are like the emperors of our modern times and fossil fuels are
our slaves.
The history of biological evolution is also the history of energy use. The
species that harvest and use high quality energy sources show higher survival
strategies (MacArthur and Pianka, 1966). The difference between how much
energy an organism receives for its efforts and how much it uses (i.e. net energy)
is a key element not only in the evolution of present day organisms (Lotka, 1922;
Odum et al., 1995), but also for the stabilization of structures and the functions of
modern societies (Giampietro et al., 1997). The history of human societies is one
of using the condensed energy of the sun. Since the beginning of sedentary
farming and the domestication of draft animals, traditional societies secured their
required mechanical energy by using human and animal muscles, and the thermal
energy needed for cooking and comfort by burning biomass. Buried plant matter
eventually decayed and became what we now call fossil fuels (oil, natural gas and
coal).
The average per-capita availability of all forms of energy remained low and
stagnant for a very long period of time. The U.S. consumption of fossil fuel
surpassed that of biomass only in the early 1880s. During the second half of the
1800s, the average per capita supply of all energy forms increased by only 25%
with the rise in coal consumption (Smil, 2003). In contrast, human advances
Charter 1 - Introduction
during the twentieth century were strongly linked with an unprecedented rise in
total energy consumption. This rise was characterized by a crucial change in the
dominant energy base as coal deposits, crude oil and natural gas became the
dominant form of energy (Hagens, 2010; Smil, 2000).
The result is that today, the vast majority of our energy supply is spent in
non-nutritive energy consumption (Price, 1995). In this regard, an important
distinction was introduced by Lotka, and later taken up by Georgescu-Roegen
(1971), between endosomatic and exosomatic energy. The former refers to energy
consumed in the form of food by human bodies to perform biological activities.
Consequently endosomatic instruments are the organs each individual is born with.
Conversely, exosomatic energy is the energy consumed by humans outside their
bodies. Today, exosomatic energy in developed and transitional countries consists
mainly in fossil energy. Thus, if over the last thousand years, the evolution of
human beings has been based on a slow adaptation of our endosomatic
instruments (i.e. organs), in less than 200 years, human evolution has shifted to
the rapid adaptation of exosomatic instruments.
The massive use of fossil fuels that characterizes modern human societies
cannot continue for long . Peak oil, the point in time when an oil field, a nation or
the world oil reaches its maximum oil production and then declines, has now been
acknowledged by an increasing number of scholars (Deffeyes, 2005; Duncan,
2000; Duncan and Youngquist, 1999; Hubbert, 1969; Ivanhoe, 1997; Strahan,
2007). Moreover, the continuous use of fossil fuels has dramatically increased
anthropogenic greenhouse gas (GHG) emissions. Currently the debate on the oil
peak is fundamentally about whether there are one, two or even 3.5 trillion barrels
of economically extractable oil left. One critical aspect in this debate (yet often
ignored) is the capital, operating and environmental costs, in terms of money and
energy, to exploit any oil fields that remain to be discovered and to generate
whatever alternatives we might decide to invest in (Hall et al., 2008). Natural gas
and coal can provide us a few more decades of easy accessible energy but their
peak will follow as long as the global demand for energy keeps increasing
(Energy Watch Group, 2007). The famous “Hirsch report” commissioned by the
US Department of Energy suggests that we need 10-20 years of lead time before a
global peak oil to prepare alternative energy systems (Hirsch et al., 2005).
However, the proposed solutions such as the exploitation of tar sands, oil shale
and coal-to-liquid fuels present tremendous environmental impacts (Hagens,
2010; Jaramillo et al., 2008).
Today energy policies face extremely difficult and contrasting challenges
such as ensuring adequate amounts of energy to satisfy human needs and wants,
reducing GHG emissions, minimizing the use of other natural resources, avoiding
the detrimental effects on human health for current and future generations, and
promoting technically feasible and economically viable alternatives. Basically we
require energy policies to provide sustainable alternatives.
2
At a global level, renewable energy makes up just 10% of the entire
exosomatic energy consumption (IEA, 2010a). If renewable energies are to
contribute to solve the current global energy and environmental problems, their
contribution to the energy mix needs to increase dramatically. In this regard, the
International Energy Agency (IEA) estimates that the delivery of energy from
renewables will increase from 840 Mtoe to between 1,900 (more than twice the
current level) and 3,250 Mtoe (almost four times the current level) in 2035
depending on the scenario considered (Current Policy scenario for the lower
estimated renewable energy production, and more aggressive GHGs abatement
policy scenario for the higher estimation). Specifically, IEA estimates that the
share of renewables in the generation of global electricity will increase from 19%
in 2008 to almost a third in 2035. Moreover, the share of renewables in heat is
expected to increase from 10% to 16%, and the demand for biofuels will grow
four-fold in the same period, thus meeting 8% of the global demand for road
transport fuel (IEA, 2010b). These increases in production and use of renewables,
can only be achieved through relevant policies. Thus the European Union has set
binding targets of GHG reductions and renewable energy uses to be achieved by
2020 such as 20% of renewable energy use in the energy mix and a 20% reduction
in GHG emissions in comparison to 1990. Additionally, more ambitious targets
were recently suggested by the European Commission to achieve a 80-95%
reduction in GHG emissions (compared to 1990 levels) in order to keep climate
change below 2°C (EU, 2011).
If alternatives to fossil fuels need to increase we must be able to properly
evaluate the sustainability of the proposed alternatives in order to avoid a
pointless and harmful waste of time and resources. Evaluating sustainability is
certainly not an easy task. Indeed, the practical business of evaluating the
sustainability of real decisions seems in many cases to be impaired by the
polemics, ambiguities and expediencies associated with many mainstream
commercial and political interests in “sustainable development”. The development
of sustainable evaluation techniques and indicators is a crucial issue in the overall
strategy for effective decision making for sustainable alternatives. However, in
spite of the institutional enthusiasm, there may be a tendency to forget that the
design of evaluation methodologies for sustainability is only a means to an end,
rather than an end in itself. Indicators and evaluation techniques are only as good
as the decisions they enable (Stirling, 1999).
Over the last few decades there has been a proliferation of evaluation
techniques both with descriptive and normative purposes: different forms of costbenefit, cost-effectiveness and multi-criteria analyses, life cycle assessments,
material flow accounting, comparative risk analyses, among many others.
Unfortunately, all this rapid reproduction of sustainability assessment techniques
does not necessarily suggest any institutional enlightenment. In fact, a common
problem is that sustainability, at least in terms of its most important aspect, is
regarded as if it were an objective determinate quantity. If this were actually the
3
case, the purpose of an evaluation would simply be to identify the best option
from an array of alternatives (Stirling, 1999). But sustainable development is a
multi-dimensional concept. This implies that policies promoting sustainable
development must deal with conflicting points of views. In Norgaard’s words
(1994, p.10 ) “Environmentalists want environmental systems and the diversity of
species sustained, […] consumers want consumption sustained. Workers want
jobs sustained. Capitalists and socialists have their “isms,” while aristocrats,
autocrats, bureaucrats, and technocrats have their cracies”. When we discuss
sustainability, we cannot escape questions such as: Sustainability of what?
Sustainability for whom? Sustainability for how long? Sustainability at what cost
(Allen et al., 2003)? Economic instruments are better suited to answer only the
last question. Thus, they need to be complemented with other approaches if we
want to deal with sustainability in a comprehensive way (Munda, 2008).
In the current context of the demand for environmental health and economic
stability, it is clear that energy and environmental policies will have to face more
complex goals at global, regional and local levels. In this sense, assessing our
energy and environmental policies across different scales and different
dimensions becomes crucial for the evaluation of sustainable alternatives.
The acknowledgement of the importance of scale for the analysis of
sustainability has grown considerably over the last few decades. In this regard, the
Millennium Ecosystem Assessment (2005) clearly recognized the crucial role of
multi-scale and multi-dimensional analyses to allow decision-making to identify
policy options which take critical interactions between human beings and
ecosystem services into account.
The issue of scale is crucial because the scale at which an assessment is
undertaken dramatically influences the structuring of the problem, the definition
of relevant attributes, and consequently the results of the assessment. In addition,
there are different social groups and stakeholders involved in the decision-making
process according to the scale of analysis. Finally, once a given scale is chosen,
several attributes and problems (those not relevant for the specific point of view
reflected by the analyst) are automatically left outside the modeling exercise. Thus,
if the different points of view reflected by the different possible scales are to be
taken simultaneously into account, the advantages and disadvantages of the policy
options which may be important for other points of view and for other
sustainability aspects may be completely lost. Furthermore, also in terms of a
single scale analysis, the multi-dimensional nature of sustainability assessments
requires the comprehensive use of indicators related to different scientific
dimensions and to different points of view. Of course, applying multi-scale
approaches and multi-dimensional analyses poses serious methodological and
practical challenges.
The focus of this thesis is on integrated assessments for decision making. In
the following chapter the reasons why an assessment of sustainable alternatives
cannot be based on a measurement of a single determinate quantity will be
4
explained further. After a thorough review of the traditional evaluation approaches
and a brief introduction to the complexity theory applied to the analysis of
coupled environmental and socio-economic systems, I will address to what extent
specific energy alternatives can be considered sustainable in multi-criteria and
multi-scale assessments. I will show how integrated assessments with specific
desired characteristics can be applied. The definition and the relevance of these
characteristics is then the subject of the subsequent chapter.
There are two cases proposed for the application of the specific methods:
biodiesel in Brazil and geothermal power in the south of Tuscany (Italy). The
former is the case of a very large country with the enormous potential for
agricultural and biomass based activities. The latter represents a case of a small
area with abundant geothermal resources. These two cases are strongly contested
and at the same time advocated by different stakeholders carrying different
legitimate perspectives. In both cases, decision makers need to take difficult
decisions and consequently need appropriate tools to assist the decision-making
processes.
I approach the biodiesel case using a Multi-Scale Integrated Assessment of
Societal and Ecosystem Metabolism (MuSIASEM), while the geothermal case is
explored by means of a Social Multi-Criteria Evaluation (SMCE). Both cases and
both methodologies seem very significant in terms of capturing the multidimensional nature of energy policies.
MuSIASEM builds on societal and industrial metabolism concepts (Ayres
and Simonis, 1994; Fischer-Kowalski, 1998; Fischer-Kowalski and Haberl, 1993)
thus capturing the biophysical aspects of the economy and facilitating an analysis
of the interaction between human societies and their natural environment. In
addition MuSIASEM explicitly addresses the issues of scale by providing a
coherent framework for parallel assessments at different scales and related to
different indicators originating from different disciplines. MuSIASEM makes it
possible to simultaneously analyze how flows of energy, money and material are
generated and exchanged among the different scales that make up a given societal
organization which, in turn, is embedded in a larger ecosystem. Practical
applications of MuSIASEM are mainly parallel economic and biophysical
historical analyses of trajectories of development of specific regions or countries
(Eisenmenger et al., 2007; Falconi-Benitez, 2001; Gasparatos et al., 2009;
Iorgulescu and Polimeni, 2009; Kuskova et al., 2008; Ramos-Martín, 2001;
Ramos-Martín et al., 2009; Ramos-Martín et al., 2007). I attempt a further step
forward by applying MuSIASEM as a scenario feasibility tool and integrating
MuSIASEM with an input-output analysis (IOA). This is because the application
of IOA makes it possible to generate the economic flows implied by the given
scenario. As will be explained in Chapter 3, MuSIASEM builds on the
Georgescu-Roegen fund-flow model (1971). However, although the fund-flow
model is a close cousin of input-output models (as Georgescu-Roegen has shown),
5
to the best of my knowledge, this is the first time that MuSIASEM has been
coupled with IOA.
SMCE is a robust methodological approach for the analysis of multidimensional attributes of alternatives and scenarios. It facilitates the identification
of compromise solutions from contrasting objectives. Thus, while MuSIASEM
can be classified as a descriptive tool, SMCE represents a normative approach.
As Stirling (1999) stresses, in the field of multi-criteria evaluation, very
often analyses attempt to express performances in terms of aggregated numerical
values and unitary sets of ranking. However if an integrated assessment of energy
alternatives cannot be reduced to a single discrete numerical result, one possible
solution is to think of it as a pattern of sensitivities. In this way, an integrated
assessment aims to map the sensitivity of results under different assumptions. It is
curious that sensitivity analyses are so widespread in deterministic disciplines
such as engineering, and yet so little used in integrated assessments of
environmental and energy problems (which are probably less determinate). In this
thesis, SMCE is applied as a form of “political sensitivity analysis” explicitly
showing how the rankings of options change under different social perspectives.
This thesis is structured as follows:
Chapter 2 introduces the problem of assessing alternative options in energy
and environmental problems. Traditional approaches are revised and their
practical, ethical, and epistemological problems are explained. The chapter ends
by describing the desired characteristics of methodological tools of integrated
assessment for supporting decision making.
Chapter 3 shows how MuSIASEM can be applied as a tool for the scenario
feasibility analysis to evaluate biodiesel in Brazil. The specific case offers a
significant insight since biofuels are promoted as clean and green energy sources,
especially in developing and transitional countries. Brazil presents excellent
conditions for the development of an energy matrix where a significant
contribution comes from biomass and biofuel sources. Thus after having replaced
more than 50% of gasoline with ethanol, Brazil has recently launched a new
biodiesel program. It is hoped that the very particular characteristics of Brazil
make it an enlightening case study. This chapter was also published in Ecological
Economics (Vol. 70, pp. 2028 – 2038).
Chapter 4 assesses the development of geothermal power by means of
SMCE in the region where geothermal power originated: Tuscany (Italy). The
case study describes a context characterized by strong uncertainty concerning key
issues such as the effects of geothermal power on water conservation and its
impact on human health. The scientific research addressing these issues has been
highly contested. In these conditions, geothermal power is giving rise to strong
opposition. SMCE is applied here to explore possible scenarios using a “political
sensitivity analysis”.
Chapter 5 draws some conclusions.
6
References
Allen, T.F.H. Tainter, J.A. and Hoekstra, T.W. (2003). Supply-Side Sustainability.
Columbia Univerity Press, New York
Ayres, R.U. and Simonis, U.E. (1994). Industrial Metabolism: Restructuring for
Sustainable Development. United Nations University Press, Tokyo, New
York, Paris
Boulding, K.E. (1966). The economics of the coming spaceship earth. In Jarrett, H.
(Ed.) Environmental quality in a growing economy. Johns Hopkins
University Press, Baltimore
Bullard, C.W. and Herendeen, R.A. (1975). The energy cost of goods and services.
Energy Policy, 3 (4), 268-278.
Cleveland, C.J. (1991). Natural resource scarcity and economic growth revisited:
economic and biophysical perspectives. In Costanza, R. (Ed.) Ecological
Economics: The Science and Management of Sustainability. Columbia
University Press, New York, Pp 289–317
Deffeyes, K. (2005). Beyond oil: The view from Hubbert’s Peak. Farrar, Straus
and Giroux, New York
Duncan, R.C. (2000). Peak oil production and the road to the Olduvai Gorge.
Keynote paper presented at the Pardee Keynote Symposia. Geological
Society of America, Summit.
Duncan, R.C. and Youngquist, W. (1999). Encircling the peak of world oil
production. Natural Resources Research, 8 (3), 219–232.
Dung, T.H. (1992). Consumption, production and technological progress: a
unified entropic approach. Ecological Economics, 6 (3), 195–210.
Eisenmenger, N. Ramos-Martín, J. and Schandl, H. (2007). Análisis del
metabolismo energético y de materiales de Brasil, Chile y Venezuela. Revista
Iberoamericana de Economía Ecológica, 6, 17-39.
Energy Watch Group. (2007). Coal: resources and future production. [Online]
http://www.energywatchgroup.org/files/Coalreport.pdf
Available
from:
[Accessed 20 February 2010]
EU. (2011). A Roadmap for moving to a competitive low carbon economy in 2015.
COM(2011) 112 Final. Communication from the Commission to the EU
Parlament, the Council, the European Economic and Social Comitee and the
Comitee of the Region. European Commission. [Online] Available from:
http://ec.europa.eu/clima/documentation/roadmap/docs/com_2011_112_en.p
df [Accessed 29 August 2011]
Falconi-Benitez, F. (2001). Integrated assessment of the recent economic history
of Ecuador. Population and Environment: a Journal of Interdisciplinary
Studies, 22 (3), 257-280.
Fischer-Kowalski, M. (1998). Society's metabolism. The intellectual history of
material flow analysis, Part I: 1860–1970. Journal of Industrial Ecology, 2
(1), 61-78.
7
Fischer-Kowalski, M. and Haberl, H. (1993). Metabolism and colonization.
Modes of production and the physical exchange between societies and nature.
Innovation in Social Sciences Research, 6 (4), 415-442.
Gasparatos, A. El-Haram, M. and Horner, M. (2009). Assessing the sustainability
of the UK society using thermodynamic concepts: Part 1. Renewable and
Sustainable Energy Reviews, 13 (5), 1074-1081.
Georgescu-Roegen, N. (1971). The Entropy Law and the Economic Process.
Harvard University Press, Cambridge, MA
Giampietro, M. Bukkens, S. and Pimentel, D. (1997). Linking technology, natural
resources and socio economic structures of human societies: examples of
application. In Fresse, L. (Ed.) Advances in human ecology. JAI Press,
Greenwich (CT), Pp 129-199
Hagens, N.J. (2010). Towards an applied net energy framework. Doctoral thesis.
University of Vermont.
Hagens, N.J. and Mulder, K. (2008). A framework for energy alternatives: net
energy, Liebig’s law and multi-criteria analysis. In Pimentel, D. (Ed.)
Biofuels, Solar and Wind As Renewable Energy Systems Benefits and Risks.
Springer, New York, Pp 295-319
Hall, C.A.S. Cleveland, C.J. and Kaufmann, R. (1986). Energy and Resource
Quality. John Wiley & Sons, New York
Hall, C.A.S. Powers, R. and Schoenberg, W. (2008). Peak oil, EROI, investments
and the economy in an uncertain future. In Pimentel, D. (Ed.) Biofuels, Solar
and Wind as Renewable Energy Systems: Benefits and Risks. Springer, New
York, Pp 109-132
Hirsch, R. Bezdek, R. and Wendling, R. (2005). Peaking of world oil production:
impacts, mitigation, and risk management. US Department of Energy,
Washington D.C. [Online] Available from: http://www.netl.doe.gov/energyanalyses/pubs/Oil_Peaking_NETL.pdf [Accessed 2 August 2011]
Hubbert, M.K. (1969). Energy Resources. National Academy of Sciences.
National Research Council. Committee on Resources and Man. W. H.
Freeman, San Francisco. Pp. 157–242
IEA. (2010a). Key World Energy Statistics. International Energy Agency.
[Online] Available from: http://www.iea.org [Accessed 17 May 2011]
IEA. (2010b). World Energy Outlook. International Energy Outlook, Paris.
Iorgulescu, R.I. and Polimeni, J.M. (2009). A multi-scale integrated analysis of
the energy use in Romania, Bulgaria, Poland and Hungary. Energy, 34 (3),
341-347.
Ivanhoe, L.F. (1997). Get ready for another oil shock! The Futurist, 31 (1), 20–27.
Jaramillo, P. Griffin, M.W. and Matthews, H. (2008). Comparative analysis of the
production costs and life-cycle ghg emissions of ft liquid fuels from coal and
natural gas. Environmental Science Technology, 42 (20), 7559-7565.
Jorgenson, D.W. (1984). The role of energy in productivity growth. The American
Economic Review, 74 (2), 26–30.
8
Kuskova, P. Gingrich, S. and Krausmann, F. (2008). Long term changes in social
metabolism and land use in Czechoslovakia, 1830–2000: An energy
transition under changing political regimes. Ecological Economics, 68 (1-2),
394-407.
Küummel, R. (1982). The impact of energy on industrial growth. Energy - The
International Journal, 7 (2), 189–203.
Lotka, A.J. (1922). Natural selection as a physical principle. Proceedings of the
National Academy of Sciences, 8, 151-154.
MacArthur, R.H. and Pianka, E.R. (1966). On the optimal use of a patchy
environment. American Naturalist, 100 (916), 603-609.
Millennium Ecosystem Assessment. (2005). Ecosystems and Human Well-Being.
Island Press, Washington D.C.
Munda, G. (2008). Social Multi-Criteria Evaluation for a Sustainable Economy.
Springler, Berlin
Noorgard, R.B. (1994). Development Betrayed: the End of Progress and a CoEvolutionary Vision of the Future. Routledge, London and New York
Odum, W.E. Odum, E.P. and Odum, H.T. (1995). Nature's pulsing paradigm.
Estuaries, 18, 547-555.
Price, D. (1995). Energy and human evolution. Population and Environment: a
Journal of Interdisciplinary Studies, 16 (4), 301-319.
Ramos-Martín, J. (2001). Historical analysis of energy intensity of Spain: from a
conventional view to an integrated assessment. Population and Environment:
a Journal of Interdisciplinary Studies, 22 (3), 238-256.
Ramos-Martín, J. Cañellas-Boltà, S. Giampietro, M. and Gamboa, G. (2009).
Catalonia’s energy metabolism: Using the MuSIASEM approach at different
scales. Energy Policy, 37 (11), 4658-4671.
Ramos-Martín, J. Giampietro, M. and Mayumi, K. (2007). On China's exosomatic
energy metabolism: an application of multi-scale integrated analysis of
societal metabolism (MSIASM). Ecological Economics, 63 (1), 174-191.
Rosenberg, N. (1976). Perspectives on Technology. Cambridge University Press,
London, New York, Melbourne
Smil, V. (2000). Energy in the 20th century: Resources, conversions, costs, uses,
and consequences. Annual Review of Energy and the Environment, 25, 21–51.
Smil, V. (2003). Energy at the crossroads. The MIT Press, Cambridge
Stirling, A. (1999). The appraisal of sustainability: Some problems and possible
responses. Local Environment, 42 (2), 111-135.
Strahan, D. (2007). Open letter to Duncan Clarke. [Online] Available from:
http://www.davidstrahan.com/blog/?p=35 [Accessed 19 November 2008]
9
2
A Theoretical Review of Sustainability
Assessment
Abstract
The assessment of sustainability gives rise to important methodological and
ethical issues. Traditionally, environmental economics has relied on a cost-benefit
analysis (CBA) to evaluate alternatives. This chapter briefly reviews the
theoretical, ethical and practical problems of the application of cost-benefit
analyses for the appraisal of sustainability. Contrary to the reductionist approach
underlying CBA, this chapter suggests that the assessment of sustainable
alternatives should build on complexity theories and acknowledge the main
characteristics of complex adaptive systems as crucial inputs. The relevance of
post-normal science problem-solving strategies for addressing sustainability
problems is explained along with a presentation of the desired characteristics of
assessment methodologies. Finally, two methodological approaches are
introduced.
Keywords: Sustainability assessment,
incommensurability, post-normal science
cost-benefit
analysis,
complexity,
2.1 Traditional cost-benefit analysis and commensurability
A reduction in the use of fossil fuels and their replacement (at least partial) with
renewable energy forms is expected and advocated by a wide array of
international, national and local government institutions. This is because of
environmental concerns, peak oil arguments and political reasons. The transition
from fossil fuels to renewable energy raises important questions regarding the
supposed environmental sustainability of potential alternatives, their costs, their
technical feasibility, and their socio-economic implications.
Charter 2 – A theoretical review of sustainability assessment
Environmental economics has traditionally relied on a cost-benefit analysis
(CBA) to evaluate environmental policies including energy alternatives. This
rather simple decision-making procedure essentially consists in weighing up the
gains and costs of a decision, project or policy. It generally includes the following
phases (Bojö et al., 1990; Garrod and Willis, 1999; Pearce, 1971): i) a definition
of the project, which means the identification of benefits and costs; ii) an
evaluation (and quantification) of benefits and costs in terms of a common
monetary unit; iii) the choice of a social discount rate; iv) the introduction of a
time horizon; v) the construction of one dimensional indicators which put together
benefits and costs (typically the net present value); and finally vi) the application
of a decision rule. The underlying assumptions of this monetary evaluation are
rationality and optimization. In order to extend the application of CBA from
individual choices to social decisions, the assumptions of rationality and
optimization are maintained.
CBA has many advantages for deciding on economic alternatives related to
individual choices and clearly defined economic costs and benefits.
Environmental economists have developed a wide range of methodologies to deal
with the technical limits of CBA, mainly the absence of a market for
environmental goods. These techniques are mostly based on the creation of
artificial markets and include contingent valuations (the willingness to pay and to
accept), travel cost methods, hedonic prices, etc. Criticisms of such approaches
are related to ethical reasons, theoretical foundations of CBA, epistemological
arguments and empirical considerations contesting the validity of the results of the
specific methods.
Political philosophers stress that market boundaries should exist for specific
goods and services (Anderson, 1993; Kuttner, 1999; Lukes, 1996; Miller and
Walzer, 1995). For instance, trading in children, certain drugs, or "weapons of
mass destruction," is morally wrong. The artificial surrogate markets of CBA are
regarded as similarly unacceptable within these contexts. Thus many economists
acknowledge that there are clear limits to the use of CBAs in policy making,
however CBAs are considered still relevant whenever resource constraints are
involved (Beckerman and Pasek, 2001). However as Aldred (2006) stresses, such
a contention is unhelpful because almost all policies have important resource
implications, for example, the death penalty (versus more prisons), abortion
policies (clinics), and foreign affairs (armed forces). The mix of economics and
moral in these policies is just what critics of CBA claim characterizes
environmental policies (Holland, 1997; O'Neil, 1993; Vatn, 2004). In Aldred’s
words “if CBA is inappropriate for political decisions concerning say, abortion
policy, then it is argued to be inappropriate for much environmental policy too”
(2006 p. 142).
Another point often raised in the academic literature (a full treatment of this
issue can be found in Aldred, 2006) is related to the reluctance of rational agents
to attach a monetary value to some goods or services (including environmental
11
assets). In this case, the respondent behaviour is the so-called protest bid (zero or
implausibly large bids). It is widely acknowledged that incommensurability
problems are particularly evident and important in environmental evaluation
endeavours. According to the utilitarian view intrinsic in CBA, different values
attached to the environment can be traded off by resorting to one unique monetary
value. Willingness to pay (WTP) and willingness to accept (WTA) imply a tradeoff between environmental damage and monetary compensation or a payment to
avoid the environmental damage. It is worth mentioning that the refusal to make a
trade-off is found both among those scholars believing that monetary valuations
can be useful and among those who are against the use of monetary valuations of
environmental assets (Spash, 2000).
Sustainability is an inherently multi-dimensional concept. Many of the
disparate environmental, economic and social aspects involved in sustainability
assessments are mutually incommensurable. Issues such as child mortality,
occupational safety, future cancer risks, employment, biodiversity losses, regional
and social development, gender and global equity cannot be adequately measured
with the same yardstick. The different types of environmental and social impacts
associated with sustainability can be classified according to dimensions such as:
duration, familiarity or controllability, severity (the balance of mortality and
morbidity), immediacy (disease versus injury), demographic and geographic
distribution and gravity (spread over a number of relatively minor events versus
concentrated in serious single episodes) (Fischhoff et al., 1981; Stirling, 1999).
These dimensions are not commensurable in the sense that they cannot be
aggregated into any single common measure. Of course, it is possible to take
different - but equally reasonable - views of the relative importance of these
dimensions. As one of the pioneers of environmental economics, David Perce,
acknowledges “the issue of “incommensurables” grew to be the single most
controversial issue in CBA, and it remains so today (Pearce, 2000 p.51). As
Funtowicz and Ratvets stress, when the irreducible complexity of environmental
issues is acknowledged “the issue is not whether it is only the marketplace that
can determine value, for economists have long debated other means of valuation;
our concern is with the assumption that in any dialogue, all valuations or
“numeraires” should be reducible to a single one-dimension standard” (1994,
p.199). These concerns from the academic arena have so far had a limited effect
on the practice of sustainability evaluation (Stagl, 2009).
Another strand of criticism of CBA is related to the behavioural
assumptions which underlie neo-classical micro-economics and consequently
CBA. According to this view, CBA is at odds with empirical evidence and
modern psychology. The usual behavioural assumption of conventional policy and
economic analysis is that the valuation of losses and gains are essentially
equivalent. According to this view, the amount people would be willing to pay to
avoid damage (such as the cleaning up a site) should be the same as the
compensation they would be ready to accept to allow somebody to provoke the
12
damage, such as polluting the site (Coase, 1960; Willing, 1976; Zeckhauser and
Phillips, 1989). Thus it does not make much difference if WTP or WTA is used
because people would feel the same about these two options except for the limited
effect of income constraints. As a result, the most conveniently measured
willingness to pay has became the main means for estimating both gains and
losses (Knetsch, 1995). However there is growing evidence that people value
gains and losses asymmetrically. Specifically, the empirical evidence from many
controlled tests (Kachelmeier and Shehata, 1992; Kahneman et al., 1991;
Kahneman and Tversky, 1979; Knetsch and Sinden, 1984) and real life situations
(Frey and Pommerehne, 1987) consistently indicates that losses matter much more
than gains, and that reductions in losses are more valuable than foregone gains.
Moreover, the reported differences between WTA and WTP have been
demonstrated to be independent from the repetition of trade offers, transaction
costs, wealth constraints and income effects (Kahneman et al., 1990). As a
consequence, using payment measures to assess losses seriously underestimates
their magnitude (Kahneman et al., 1991; Knetsch, 1995).
One practical problem in the application CBA to sustainability problems is
that a sustainability analysis requires suitable long-term tools. It is well known
that extremely small increases in discount rates tend to exclude the long-term
effects of available options. This is perfectly consistent with the utility theory
which underlies CBA. In fact, the utility (or disutility) associated with today’s
benefits (or costs) is assumed to be higher than that of tomorrow. However such
an approach certainly raises serious issues regarding the burden of effects that
today’s choices will have on tomorrow’s generations. This important aspect was
also explicitly addressed by the Stern Review (2006), which suggests using a very
low rate of 0.1%.
The discount rate debate is one example of the distributional effects of CBA
(e.g. on future generations). However there are other distributional effects of a
more explicit nature. WTP, which underlies the different methods based on
artificial markets, depends on the ability to pay. The consequence is that
externalities have a much lower valuation when they are borne by poor people.
The famous World Bank internal memo (The Economist, 1992) suggesting that
dirty industries migrate more to less developed countries was exactly based on
this kind of reasoning. Thus, choices made on CBA easily imply intragenerational inequality. Accepting low values for negative externalities is a
political decision which is far from being ethically neutral (Munda, 2008). Also
inter-generational equity is seriously compromised. In fact, those who are not yet
born cannot bid in real or artificial markets. This is why, Martinez-Alier and
O’Connor (1999) emphasize that externalities should be considered as an instance
of cost-shifting success more than a case of market failure.
The appeal of CBA depends on a cramped and misguided interpretation of
the rationality concept, one that mistakenly identifies rational choices with those
that can be arrived by optimizing principle and algorithmic procedures. In CBA
13
there is one unique measure of value (i.e. monetary) through which policy options
can be ranked. Thus CBA assumes complete value commensurability. However,
commensurability can take a strong or a weak form. When options are strongly
commensurable, they are compared through a cardinal scale. Conversely, weak
commensurability implies comparison on an ordinal scale (O'Neil, 1994). When
options are weakly comparable, rational choices are still possible but “substantive
rationality” must be replaced by “procedural rationality” (Martinez-Alier et al.,
1998). According to Simon (1976), substantive rationality refers to the rationality
of the results irrespectively of the decision-making process, while procedural
rationality is about the decision process itself.
Weak comparability implies incommensurability, that is, the absence of a
common unit of measurement across different values. Munda (2004) further
distinguishes
between
technical
incommensurability
and
social
incommensurability. Technical incommensurability refers to the use of indicators
and models that cannot be reduced to each other. When different indicators related
to different benefits and costs defined at different scales (e.g.
improvement/worsening in my backyard versus improvement/worsening at global
level) or defined in different scientific disciplines (e.g. economic losses measured
in 1995 US$ versus biodiversity losses over a 100-year time frame), it is simply
not possible to devise an accounting system to substantially reduce these different
types of benefits and costs to a common numeraire. Social incommensurability
refers to the plurality of values in society which cannot be reduced to each other.
This implies that in a given social conflict a “freedom-fighter” on one side can be
seen as a “terrorist” on the other side (Giampietro et al., 2006).
When an integrated assessment of sustainability involves multi-dimensional
and multi-scale analyses, technical and social incommensurability must be dealt
with. If we accept that sustainable development is a multi-dimensional concept,
we also have to accept that policies promoting sustainable development have to
deal with conflicting values and points of views. In this regard, it is worth
recalling the Impossibility Theorem of the Nobel Price winning economist
Kenneth Arrow. Arrow (1963) showed that, given a set of minimal conditions, it
is impossible to aggregate individual preferences in a plural society in a consistent
and democratic way. Essentially any social preference order (or social welfare
function) would violate al least one of the minimal set of axiomatic conditions1
1
Such conditions are: 1) the Free Triple Condition, i.e. the ordering of social preferences for each
set of alternatives should be the same, irrespectively of the way sub-sets of these alternatives are
grouped together 2) the Non-Negative Association, i.e. any alternative that is increasingly
supported by all individuals should be increasingly supported in the expression of social ordering.
3) Independence of Irrelevant Alternatives, i.e. the introduction of a new alternative or the deletion
of an existing option should not cause changes in the ordering of preferences in other alternatives.
4) Non Imposition, i.e. it must always be the case that a social ordering between two alternatives is
possible if the individual members of a society are able to express their preferences over the two
alternatives. 5) Non Dictatorship, i.e. the social preferences should not be determined by the
preferences of any single individual no matter what other individual preferences are.
14
required to aggregate individual choices (a very sad conclusion of the theorem is
that the only political system satisfying all conditions is dictatorship!). No matter
how much information we have and how much consultation is involved, no
analytical procedure can fulfill the role of a democratic political process. In other
words, “in terms of the theoretical framework underlying the assessment
methodologies themselves, there can be no uniquely 'rational' way to resolve
contradictory perspectives or conflicts of interests in a plural society” (Stirling,
1998a p 103).
2.2 Complexity and complex adaptive systems
Decision-making in environmental policies is extremely challenging. It can easily
include masses of details related to many different issues thus requiring separate
management and analysis. In these conditions there is a natural temptation to
reduce the unavoidable complexity to simpler and more manageable elements.
Simple systems which can be captured by deterministic and linear relations can be
analyzed and managed by reductionist approaches. However the now global
energy and environmental problems caused by human action on the ecosystems in
which human societies are embedded, are not manifestations of simple systems.
Rather, they are exceptionally complex.
In order to introduce complex systems, Rosen’s definition comes into play.
He defines a complex system as one “for which we have at our disposal a large
number of subsets of measuring instruments, each of which gives rise to a
different mode of description of the system. Another way of saying this is that a
complex system is one which allows us to discern many subsystems…, depending
entirely on how we choose to interact with the system” (1977, p.229). As Rosen
himself stresses, this definition implies that complexity is not an intrinsic property
of the system, rather it depends on the way the analyst decides to interact with a
system. A complex system gives rise to different perspectives according to
selected representations of the same system. The other key element of the above
definition is that models can only capture one part of a complex system, the part
the analyst is interested in (Giampietro, 2002). Inevitably, every observer of a
complex system chooses to operate though certain selection criteria, with certain
values at a certain scale-level (Funtowicz et al., 1999). Here, it is worth recalling
the famous words of Schumpeter: “Analytical work begins with material provided
by our vision of things, and this vision is ideological almost by definition” (1954,
p.54).
Just to give an example of how the unavoidable arbitrariness of a modeling
endeavor can affect energy analyses, I would like to recall an amazing finding by
Stirling (1998b). His work reviewed a large number of government and industry
15
sponsored studies aimed at assessing the external environmental costs of new coal
power plants in industrialized countries. Reporting all values in 1995 US$ he
found that the difference between the lowest and the highest value was about
50,000 times!
This implies that looking for the right model that addresses all aspects of a
complex systems is a pointless exercise. Conversely, complexity requires the
ability to deal with an expanding set of perceptions and representations of unequal
observers (Giampietro and Ramos-Martin, 2005).
All natural systems of interest for sustainability (from socio-economic
systems to ecological systems) are dissipative systems (Prigogine and Stengers,
1981). These dissipative systems are “self-organizing”, open systems far from the
thermodynamic equilibrium, and that are “becoming in time” (Prigogine, 1978).
As such, they invoke a certain complexity because they are organized into nested
hierarchical levels, which operate in parallel. For instance, “the planet is
composed of interacting ecosystems made up of interacting species and individual
organisms. The organisms are composed of organs and the organs are composed
of cells. Human societies share the same nested hierarchical structure:
macroeconomic entities (e.g. the European Union) are made up of countries,
which are made up of smaller local administrative units, that are composed by
entities such as cities, communities and households. At the same time, they are
part of a global economy, which is embedded in larger biophysical processes at
the level of the planet” (Giampietro and Mayumi, 2000b, p.118). Such systems
show different identities when looked and represented at different hierarchical
levels: “the existence of different levels and scales at which a hierarchical system
is operating implies the unavoidable existence of non-equivalent ways of
describing it” (Giampietro, 2002, p.249). As Munda (2004) emphasizes, multiple
identities of complex systems are a consequence not only of epistemological
plurality (non equivalent observers) but also of ontological characteristics of the
observed system (non equivalent observations).
In social and natural sciences there is a growing awareness that socioeconomic and ecological systems share the same characteristics of complex
adaptive systems (Arthur et al., 1997; Janssen, 1998; Rammel et al., 2007). These
characteristics consist in co-evolutionary dynamics, self-organization and large
macroscopic patterns emerging out of small-scale and local interactions. Crossscale interactions and feedback loops of complex adaptive systems between
different hierarchical levels of nested hierarchies imply non-linear patterns and a
high degree of complexity. In such circumstances, the predictive power of
equilibrium models is a misleading myth (Ramos-Martín, 2003; Van den Bergh
and Gowdy, 2003).
Consequently we have to acknowledge that in complex adaptive systems
and in dealing with future scenarios and evolutionary trends we always have to
face not only simple stochastic risks, but also uncertainty and, more important,
ignorance. Risk is a condition under which the possible outcomes are given and
16
their likelihood is defined by a probability density function. We face uncertainty
when the possible results can be known but the probability of them happening is
not known (Knight 1922). Finally, we have ignorance when it is possible neither
to solve a set of probabilities nor to define a comprehensive set of outcomes
(Stirling, 1998b). In this sense our capacity to predict the behaviour of a complex
system is not only limited by the possible presence of specific statistical
variability but also by genuine ignorance and the possible presence of novelty. As
Georgescu-Roegen clearly explained “the strongest limitation of our power to
predict comes from the entropic indeterminateness and, especially, from the
emergency of novelty by combination.” (1971, p.15). This is especially relevant in
complex adaptive systems, such as socio-economic systems, where large and
increasing numbers of feedback mechanisms between elements organized at
different levels of hierarchies can give rise to unexpected novel phenomena
(Ramos-Martín, 2003).
2.3 The way out
In spite of the above epistemological predicaments, reductionist approaches
assume that uncertainty can be handled by appropriate statistical procedures, more
sophisticated analyzes, better tests and improved expertise. In addition,
reductionism assumes that what is good for society and citizens can be defined in
a substantive way. The different typologies of costs and benefits defined in
quantitative analyses can be reduced to one unique comparison rod, thus assuming
that different indicators are commensurable (Giampietro et al., 2006). Typical
reductionist analyses present one indicator (e.g. money), one scientific dimension
(e.g. economic), one scale (e.g. a region), one objective (e.g. maximize) and one
time horizon (Munda, 2004). The result is that the descriptive and normative sides
of the decision-making process are fused together. On the descriptive side, what is
calculated to be the optimum choice is supposed to be the best possible
representation of the system. This implies an uncontested agreement among all
actors on the pre-analytical choices required to describe the system. Considering
the normative side, the calculated optimum solution is assumed to be the best
possible solution to the given problem. This last assumption implies an
uncontested agreement on the set of goals, the set of options, and on the reliability
of the information from the descriptive side. All this explains why reductionist
approaches are so popular in decision making. In fact, when it is difficult to have
an agreement between all the actors representing legitimate contrasting views, it is
easier to assume that such an agreement does exist without verifying its real
existence (Giampietro et al., 2006).
17
However, tensions over the different dimensions of sustainability, and
conflicts over the use of resources are becoming so important that science can no
longer ignore them. The evident lack of agreement expressed by various
stakeholders on the choices made in the process of decision making should be
incorporated into an integrated assessment both in relation to the normative and
the descriptive sides. In this sense, rather than individuating optimal solutions in
contexts characterized by deep uncertainty, high interest and conflicting values,
scientific investigation should enhance the social resolution of sustainability
problems (Giampietro et al., 2006). Here we are in the realm of post-normal
science. This is a term introduced by Funtowicz and Ravetz (1993) to indicate a
different epistemological framework from Kuhnian normal science (Kuhn, 1962).
Post-normal science can be characterized in relation to other complementary
problem solving strategies as depicted in Fig. 2.1.
Figure 2.1: Problem solving strategies
Decision stakes
Goal: issue-driven
Post-normal
science
Goal:
client-serving
Professional
consultancy
Applied
science
Quality control:
Extended peer-review
Quality control:
client
Goal: mission-oriented
Quality control: peer-review
Systems uncertainty
Source: Adapted from Funrowicz and Ravetz (1993)
The characterization is based on two axes: system uncertainty and decision stakes.
When both are small we are in the realm of applied science and standard routines
can be safely applied for mission oriented research. Here quality control comes
18
from peer reviews. When uncertainty and stakes are in the medium range, that is,
when uncertainty cannot be managed from a technical and routine perspective
because more complex problems are relevant, we are in the domain of
professional consultancy. Here personal judgments, skills and sometimes also
courage are required. The readiness to grapple with new and unexpected situations
is often involved. Examples could be a surgeon or an engineer facing a critical
situation. In professional consultancy, clients are the main actors of quality
control. The third area is called post-normal science and is characterized as the
one where “facts are uncertain, values in dispute, stakes high and decisions
urgent” (Funtowicz and Ravetz, 1993, p.744).
In post-normal science traditional peer review is not enough for quality
assurance. On the contrary an “extended peer community” is required to involve
an ever-growing set of legitimate participants and lay persons. The establishment
of competence and the legitimacy of participants can involve broader cultural and
societal institutions. For instance, people directly affected by an environmental
problem can often have a more pressing concern with the quality of official
reassurances and a keener awareness of its symptoms than those in any other role.
A housewife, an investigative journalist, or a patient can assess the quality of
scientific results in the context of real-life situations (Funtowicz and Ravetz,
1993; 1999).
Of course, new challenges do not make traditional science irrelevant. The
point is to choose the appropriate problem-solving strategy for each particular
challenge. In this sense, post-normal science is complementary to applied science
and professional consultancy. It does not contest the findings of reliable
knowledge and certified expertise. What is questioned is the quality of that
findings in different contexts, especially with respect to environmental, societal
and ethical aspects.
This thesis regards integrated assessments for decision making. In line with
post-normal science problem-solving strategies, the integrated assessment of
sustainability problems is proposed here as a methodological approach with the
following purposes: i) keep the descriptive side (characterization of the
performance) of the decision making process separate from the normative side
(definition of the best course of action) ii) take into account incommensurable
dimensions coming from different scientific languages, different scales, different
legitimate representations of the same system iii) handle social
incommensurability and technical incommensurability consistently and
transparently.
Two integrated assessment methodologies with such properties are MultiScale Integrated Assessment of Societal and Ecosystem Metabolism
(MuSIASEM) and Social Multi-Criteria Evaluation (SMCE).
MuSIASEM deals only with the descriptive side of decision making. It is
suitable for the analysis of complex adaptive systems which show how changes in
one level of a nested hierarchical system can affect other scales/levels. It enables
19
us to assess the exchange of economic and biophysical flows between the
different levels of systems and their impacts on the eco-system embedding the
specific system of analysis. The MuSIASEM methodology entails the
simultaneous use of indicators referring to different scales and dimensions of
analysis, thus it explicitly addresses technical incommensurability problems. It is
suitable for maintaining coherence among heterogeneous variables resulting from
the use of different scales and scientific domains. As a tool for scenario analysis,
MuSIASEM measures the feasibility space of a given option. A comprehensive
description of MuSIASEM con be found in the works of Giampietro and his
colleagues (Giampietro, 2004; Giampietro and Mayumi, 2000a; 2000b;
Giampietro et al., 2001; Giampietro et al., 2009; Giampietro and Ramos-Martin,
2005).
SMCE deals with the normative side of the decision-making process. The
relevant qualities of the given problem are assessed in relation to the specific set
of goals expressed by relevant social actors. Thus the given problem is structured
in terms of options, criteria, measurement schemes and indicators that will be used
to decide the action. SMCE addresses both social and technical
incommensurability.
Arrow’s Impossibility Theorem (mentioned above) reminds us that, even in
principle, we cannot hope to find one particular sustainability strategy that is
definitely superior to another. This raises the question of how an integrated
assessment can be used to support decision making. In this regard, an integrated
assessment can facilitate a transparent and systematic exploration of different
points of view, thus contributing to informed decision making. In addition, if the
classical opposition between public deliberation and technical analysis is
abandoned, the acknowledgment of their interdependency can actually enhance
more transparent and informed decision making. For instance, the definition and
selection of technical criteria could reflect the different points of view emerging
from public participation. Different importance weights attached to the different
dimensions and criteria could be used to generate “political sensitivity maps”.
Multi-criteria methodologies, based on a comprehensive inclusion of all affected
parties, can thus provide a robust display, to different stakeholders and to decision
makers, of the ethical and political implications of alternative options, without
seeking to order them in terms of rationality (Stirling, 1999).
20
References
Aldred, J. (2006). Incommensurability and monetary valuation. Land Economics,
82 (2), 141-161.
Anderson, E. (1993). Value in ethics and economics. Harvard University Press,
Cambridge, MA
Arrow, K. (1963). Social Choice and Individual Values. Wiley, New York
Arthur, W. Durlauf, S. and Lane, D. (1997). The Economy as an Evolving,
Complex System. Addison-Wesley, Reading
Beckerman, W. and Pasek, J. (2001). Justice, Posterity and the Environment.
Oxford University Press, Oxford
Bojö, J. Mälter, K.G. and Unemo, L. (1990). Environment and Development: an
Economic Approach. Kluwer Accademic Publishers, Dordrecht
Coase, R.H. (1960). The problem of social cost. Journal of Law and Economics, 3,
1-44.
Fischhoff, B. Derby, S.L. and Lichtenstein, S. (1981). Acceptable Risk.
Cambridge University Press, Cambridge
Frey, B. and Pommerehne, W.W. (1987). International trade in art: attitudes and
behavior. Rivista Internazionale di Scienze Economiche e Commerciali, 34
(6), 465-486.
Funtowicz, S.O. Martinez-Alier, J. Munda, G. and Ravez, J.R. (1999).
Information Tools for Environmental Policy Under Conditions of Complexity.
Environmental issues series (vol 9). European Environmental Agency.
Funtowicz, S.O. and Ravetz, J.R. (1993). Science for the post-normal age. Futures,
25 (7), 739-755.
Funtowicz, S.O. and Ravetz, J.R. (1994). The worth of a songbird: ecological
economics as a post-normale age. Ecological Economics, 10 (3), 353-368.
Funtowicz, S.O. and Ravetz, J.R. (1999). Post-Normal Science. In Sardar, Z. and
Abrams, I. (Eds.), Introducing chaos. Icon Books Ltd, Cambridge, UK, Pp
457-189
Garrod, G. and Willis, K. (1999). Economic Valuation of the Environment:
Methods and Case Studies. Edward Elgar, Cheltenham
Giampietro, M. (2002). Complexity and scales: the challenge for integrated
assessment. Integrated assessment, 3 (2/3), 247-265.
Giampietro, M. (2004). Multi-scale integrated analysis of agroecosystems. CRC
Press, Boca Raton, Florida
Giampietro, M. and Mayumi, K. (2000a). Multi-scale integrated assessment of
societal metabolism: integrating biophysical and economic representations
across scales. Population and Environment: a Journal of Interdisciplinary
Studies, 22 (2), 155-210.
21
Giampietro, M. and Mayumi, K. (2000b). Multi-scale integrated assessment of
societal metabolism: introducing the approach. Population and Environment:
Giampietro, M. Mayumi, K. and Bukkens, S.G.F. (2001). Multiple-scale
integrated assessment of societal metabolism: an analytical tool to study
development and sustainability. Environment, Development and
Sustainability, 3 (4), 275-307.
Giampietro, M. Mayumi, K. and Munda, G. (2006). Integrated assessment and
energy analysis: quality assurance in multi-criteria analyses of sustainability.
Energy, 31 (1), 59-86.
Giampietro, M. Mayumi, K. and Ramos-Martin, J. (2009). Multi-scale integrated
analysis of societal ecosystem metabolism (MuSIASEM): theoretical
concepts and basic rational. Energy, 34 (313-322).
Giampietro, M. and Ramos-Martin, J. (2005). Multi-scale integrated analysis of
sustainability: a methodological tool to improve the quality of narratives.
International Journal of Global Environmental Issues, 5 (3/4), 119-141.
Holland, A. (1997). The foundations of environmental decision-making.
International Journal of Environment and Pollution, 1 (4), 484-496.
Janssen, M. (1998). Use of complex adaptive systems for the modeling of global
change. Ecosystems, 1 (5), 457–463.
Kachelmeier, S.J. and Shehata, M. (1992). Examining risk preferences under high
monetary incentives: experimental evidence from the People's Republic of
China. American Economic Review, 82 (5), 1120-1141.
Kahneman, D. Knetsch, J.L. and Thaler, R. (1990). Experimental tests of the
endowment effect and the Coase theorem. Journal of Political Economy, 98
(6), 1325-1348.
Kahneman, D. Knetsch, J.L. and Thaler, R.H. (1991). The endowment effect, loss
aversion, and status quo bias. Journal of Economics Perspectives, 5 (1), 193206.
Kahneman, D. and Tversky, A. (1979). Prospect theory: an analysis of decision
under risk. Econometrica, 47 (2), 263-291.
Knetsch, J.L. (1995). Assumptions, behavioral findings, and policy analysis.
Journal of Policy Analysis and Management, 14 (1), 68–78.
Knetsch, J.L. and Sinden, J.A. (1984). Willingness to pay and compensation
demanded: experimental evidence of an unexpected disparity in measures of
value. The Quarterly Journal of Economics, 99 (3), 507-521.
Knight , F. (1922). Risk, Uncertainty and Profit. Houghton, London
Kuhn, T.S. (1962). The Structure of Scientific Revolutions. University of Chicago
Press, Chicago
Kuttner, R. (1999). Everything for Sale: the Virtues and Limits of Markets.
Chicago University Press, Chicago
22
Lukes, S. (1996). On trade-offs between values. In Farina, F. Hahn, F. and
Vannucci, S. (Eds.), Ethics, Rationality and Economic Behaviour. Clarendon,
Oxford, U.K., Pp 36-49
Martinez-Alier, J. and Connor, M.O. (1999). Distributional issue: an overview. In
JCJM van den Berg (Ed.) Handbook of Environmental and Resource
Economics. Edward Elgar, Cheltenham, UK, Pp 380-392
Martinez-Alier, J. Munda, G. and O'Neil, J. (1998). Weak comparability of values
as a fundation for ecological economics. Ecological Economics, 26 (3), 277286.
Miller, D. and Walzer, M. (1995). Pluralism, Justice and Equality. Oxford
University Press, Oxford, U.K.
Munda, G. (2004). Social multi-criteria evaluation: methodological foundations
and operational consequences. European Journal of Operational Research,
158 (3), 662–677.
Springler, Berlin
O'Neil, J. (1993). Ecology, Policy and Politics: Human Well Being and the
Natural World. Routedge, London and New York
O'Neil, J. (1994). Ecology, Policy and Politics. Routedge, London and New York
Pearce, D. (2000). Cost-Benefit Analysis and Environmental Policy. In Helm, D.
(Ed.) Environmental Policy. Oxford University Press, Oxford, U.K.
Pearce, D.W. (1971). Cost-Benefit Analysis. MacMillan, London
Prigogine, I. (1978). From Being to Becoming. W.H. Freeman, San Francisco
Prigogine, I. and Stengers, I. (1981). Order Out of Chaos. Bantam Books, New
York
Rammel, C. Stagl, S. and Wilfing, H. (2007). Managing complex adaptive
systems -- A co-evolutionary perspective on natural resource management.
Ecological Economics, 63 (1), 9-21.
Ramos-Martín, J. (2003). Empiricism in ecological economics: a perspective from
complex systems theory. Ecological Economics, 46 (3), 387-398.
Rosen, R. (1977). Complexity as a system property. International Journal of
General Systems, 3 (4), 227-232.
Schumpeter, J.A. (1954). History of Economic Analysis. George Allen & Unwin,
London
Simon, H.A. (1976). From substatintive to procedural rationality. In Latsis, J.S.
(Ed.) Methods and Appraisals in Economics. Cambridge Univerity Press,
Cambridge, Pp 129-148
Spash, C.L. (2000). Ecosystems, contingent valuation and ethics: the case of
wetland re-creation. Ecological Economics, 34 (2), 195-215.
Stagl, S. (2009). Value articulating institutions and changing social preferences.
REFGOV Working Paper Series GPS-11. Centre for Philosophy of Law,
Université Catholique de Louvain, Louvain-La-Neuve, Belgium. [Online]
Available
from:
23
http://refgov.cpdr.ucl.ac.be/?go=publications&dc=5811d7ccaeb5d7c3e73a14
ee341c28e9984cc592 [Accessed 17 August 2011]
Stern, N. (2006). The Stern Review on the Economics of Climate Change. Cabinet
Office - HM Treasury, London. [Online] Available from:
http://www.webcitation.org/5nCeyEYJr [Accessed 18 August 2011]
Stirling, A. (1998a). Risk at a turning point? Journal of Risk Research, 1 (2), 97109.
Stirling, A. (1998b). Valuing the environmental impacts of electricity production:
a critical review of some "first generation" studies. Energy Sources, 20 (4/5),
267-300.
The Economist. (1992, Saurday 8th February). Let them eat pollution. Pp. 82
Van den Bergh, J. and Gowdy, J. (2003). The microfoundations of
macroeconomics: An evolutionary perspective. Cambridge Journal of
Economics, 27 (1), 65–84.
Vatn, A. (2004). Environmental valuation and rationality. Land Economics, 80 (1),
1-18.
Willing, R.D. (1976). Consumer's surplus without apology. American Economic
Review, 66 (4), 589-597.
Zeckhauser, R.J. and Phillips, C.V. (1989). Contingent Valuation of Damage to
Natural Resources: How Accurate? How Appropriate? Toxics Law Reporter,
4 (18), 520-529.
24
3
Multi-Scale Integrated Assessment of
Soybean Biodiesel in Brazil2
Abstract
Developing counties are often believed to have excellent conditions for biofuel
production, however studies aimed at assessing the sustainability of large scale
biofuel programmes have generally focused on a few variables related to one
scientific domain and one scale. Contrary to this approach, this paper analyzes
soybean biodiesel in Brazil using a parallel biophysical and economic assessment
at different scales. A Multi-Scale Integrated Analysis of Societal and Ecosystem
Metabolism (MuSIASEM) approach is applied as a scenario analysis tool. A
soybean biodiesel energy balance for the specific conditions of Brazil is included
and the energy ratio turns out to be 1.09. This means that that the energy delivered
is higher than the energy invested, however the net energy is very low. The
economic impacts are analyzed through input-output analysis. The results show
that soybean biodiesel increases energy consumption per hour of work without a
corresponding increase in economic labour productivity. Consequently the already
low energy efficiency of Brazilian production could get worse. Although Brazil
has large expanses of land, the substitution of 20% fossil diesel (i.e. just 3.3% of
the country's primary energy consumption) with fully renewable biodiesel might
destroy protected areas and forests and increase the GHGs emitted.
Keywords: Biofuels, MuSIASEM, Energy balance, Input–output analysis,
Integrated analysis, Societal metabolism
2
Article published in Ecological Economics (2011), Vol 70 (11), pp 2028-2038
Chapter 3 – Multi-scale integrated analysis of soybean biodiesel
3.1 Introduction
The main aim of this paper is to provide a multi-scale integrated assessment of the
introduction of soybean biodiesel in a developing country: Brazil.
Energy security concerns and environmental considerations are raised by
proponents of biofuels along with arguments regarding rural employment.
However, in the policy documents of developed countries and international
organizations, it is widely acknowledged that biofuels need to be produced above
all in developing countries if they are meant to substitute significant amounts of
fossil fuels. In this regard, developing countries3 are assumed to have considerable
expanses of unexploited land and large fractions of the labor force that are
unemployed. Consequently, land and labor would not be limiting factors.
Having displaced more than 50% of gasoline with ethanol, in 2004 Brazil
launched its biodiesel program. In only five years biodiesel substituted 5% of
diesel consumption. The fast growth of the biodiesel production program is giving
rise to an intensive debate about the possible expansion of the program, however,
to the best of my knowledge, an integrated assessment of the implications and
constraints of large-scale biodiesel production and use is still lacking.
Studies aimed at assessing the benefits and feasibility of large-scale biofuels
have generally focused on single scale analyses and individual problems (e.g. land
availability, energy balances, GHG saving, and economic impacts) with their
specific numeraires (e.g. hectares of land, joules of energy, CO2eq, US$). The
individual problems may certainly be important but they only explain one part of
the story. When several problems are addressed in the same study, it appears
impossible to reconcile them into a comprehensive and consistent framework.
This work integrates variables related to different disciplines and assesses
the feasibility of large scale biodiesel production in Brazil through a multi-scale
analysis. Consistency among the heterogeneous variables is achieved by applying
the Multi-Scale Integrated Analysis of Societal and Ecosystem Metabolism
(MuSIASEM) approach.
MuSIASEM was conceived by Giampietro and Mayumi (1997; 2000a;
2000b) as a tool for the integrated analysis of complex adaptive systems. A full
treatment can be found in Giampietro (2004). It draws on the concept of
exosomatic metabolism and societal metabolism. The term “exosomatic energy”
refers to the energy consumed outside human bodies to perform human activities4.
Societal metabolism addresses the continuous flows of materials and energy that
are absorbed by human societies from the ecosystem and transformed into goods,
3
However Asian countries are not resource extractors and depend on resource imports.
The terms exosomatic and endosomatic instruments were introduced by Alfred Lotka and taken
up by Georgescu-Roegen (1971). Endosomatic instruments are the organs which each individual is
born with, so endosomatic energy is the energy in the form of food consumed to perform
biological human activities.
4
26
services and waste (Fischer-Kowalski and Haberl, 1993). MuSIASEM
characterizes human societies as being organized into nested hierarchical systems
in the same way as ecological systems (Giampietro, 1994). These nested systems
or compartments (e.g. economic sectors) exchange flows of energy and material.
MuSIASEM draws on the Georgescu-Roegen fund-flow model (1971).
“The funds are the agents …and the flows are the elements which are used or
acted upon by the funds” (Ibid. p. 230), like a new product, energy, or a polluting
element. The flows enter into the given representation without exiting, or
disappear without entering (Ramos-Martín et al., 2007), while the funds are
essentially the converter and controller of the flows. Typical fund coordinates of
MuSIASEM are Ricardian land and human societies which act through labor time.
Whilst labor time defines the compartments in which human societies are
organized, the use of land as fund enables the analyst to characterize the
ecosystem in which human societies are embedded.
The stability of each compartment depends on the compatibility between the
pressure of the flows that are imposed on its lower level compartments and on the
capacity of the embedded compartments to deliver the required flows. In this way,
changes in one level (e.g. human time invested in a given activity or the flow of
used energy) always require adjustments in other levels and sectors. MuSIASEM
makes it possible to verify the feasibility space of a given scenario by checking
the compatibility of the whole with the parts (Giampietro et al., 2009). In this
sense, the use of a given technology is sustainable if the pace of the flows of
energy and material is compatible with the limits imposed by the funds (RamosMartín et al., 2007). In addition, MuSIASEM provides a representation of a
system both in bio-physical and economic terms. This is because models
measuring just bio-physical variables (e.g. joules, kilograms of a given material,
quantity of land, etc.) cannot address the economic impacts resulting from societal
preferences (Giampietro and Ramos-Martin, 2005).
Studies based on MuSIASEM have been applied to China (Ramos-Martín et
al., 2007), Ecuador (Falconi-Benitez, 2001), Spain (Ramos-Martín, 2001),
Catalonia (Ramos-Martín et al., 2009), the UK, (Gasparatos et al., 2009), Chile,
Brazil and Venezuela (Eisenmenger et al., 2007), and Romania, Bulgaria, Poland
and Hungary (Iorgulescu and Polimeni, 2009). All these studies are historical
analyses, while in this work MuSIASEM is used as an analysis tool for scenario
feasibility. In order to account for the changes provoked in the economic flows by
the introduction of biodiesel, input-output analysis (IOA) is applied. This does not
mean that reported results forecast what happens once biodiesel has substituted a
given amount of diesel. Rather, they are intended to support the discussion on the
constraints that the substitution of fossil fuel with biofuels implies by shedding
light on the direction of change of some key variables. Although MuSIASEM
builds on the fund-flow model (which is a close cousin of the input-out model as
Gerogescu-Roegen showed), to my knowledge this is the first time that IOA has
been coupled with the MuSIASEM methodology. Using the results of the
27
economic IOA with those of the soybean energy balance, this article analyses the
changes in energy flows and added value, together with the funds of human
activity and land use which are caused by the substitution of fossil diesel with
biodiesel. In this way, the implications and feasibility of such a substitution are
made evident.
3.2 Scenarios, general framework and methodology
3.2.1 Variables
In terms of the MuSIASEM methodology, human societies can be represented as
a nested hierarchical system made up of different compartments operating at
various levels. The size of each compartment is measured in hours of human
activity. The highest level is the whole society (WS) and is indicated as level n,
which amounts to the whole population multiplied by the number of hours in one
year (i.e. 8,760). Scaling down at level n-1 means dividing the hours of the WS
into two main compartments: the paid work sector (PW) and the household
consumption sector (HH). The PW is composed of the number of hours invested
by a population in working (and paid) activities. This sector is in charge of
producing added value, and is estimated here as the number of working persons
multiplied by an average flat value of 1,900 hours of work per year. The HH
consumes the added value generated and is made up of the time of the WS that is
not used by the PW: the dependant population, time not worked by the working
population, and non paid work. Scaling further down at the level n-2, the PW is
divided among the agricultural sector (AG), building, manufacturing, energy and
mining, and service and government. For the purpose of this research, at the n-2
level, only the distinction between the AG and the rest of the PW sector is of
interest, thus hereafter all the sectors which are not agricultural will be referred to
as 'rest of PW' (RestPW). At the level n-3, AG is further split between the soybean
sector (SOY) and the rest of the agricultural sector (RestAG). Given the scope of
this work, no further distinction is made here between the sectors that are included
in the RestAG 5 . Because of the absence of data on the hours worked in the
different sectors, it was here assumed that the worked hours per person in one
year was the same in all the sectors (i.e. 1,900). This is certainly a strong
5
The distinctions in compartments presented here are slightly different from other applications of
MuSIASEM found in literature where RestPW is clearly split into its components. However, this
is not a problem since the flexibility of MuSIASEM makes it adaptable for the purposes of the
present analysis. This a why MuSIASEM is considered to be a meta model or multi-purpose
grammar.
28
assumption but it is not assumed to have a major impacts on the final results. As
previously specified, the aim of this study is not to forecast the exact values of
given variables once biodiesel has substituted the target quantity of diesel. Rather,
this work intends to shed light on the relative changes of some key variables. IOA
makes it possible to calculate such changes through the coefficients obtained by
dividing the number of hours worked in the different sectors by the output of the
same sectors.
The hours of human activity of the abovementioned compartments represent
a fund variable. Such fund variables exchange flows. In this paper two flows are
considered: added value for the economic reading and energy for the bio-physical
reading. For instance, for the soybean sector level there is a flow of added value
that is generated (AVsoy) and a flow of energy that is consumed (ETsoy). The
same can be said for AG and PW. For level n, i.e. WS, the economic flow
corresponds to the GDP (which is similar to the sum of the added value of all the
sectors of the economy) and the energy flow is the total energy throughput (TET)
which is given by the total primary energy supply consumed by the economy.
The variables just described provide quantitative information on the size of
the flows and funds and are called extensive variables. The ratio between the
flows and funds provide indicators that can be used for comparisons. These ratios
are called intensive variables. Examples are energy consumed per hour of work,
named the exosomatic metabolic rate (EMR), and added value per hour of work,
i.e. the economic labor productivity (ELP). Finally, the ratio between the EMR
and ELP for the same level (that is, using homogeneous quantities) represents the
energy efficiency, i.e. $ (or Brazilian Reais in this case) per joule of energy (see
Table 3.1 for a summary of the definitions and acronyms).
29
Acronyms
AG
AVag
AVsoy
ELPag
ELPpw
ELPsoy
EMRag
EMRpw
EMRsa
EMRsoy
ETag
ETpw
ETsoy
HH
PW
SOY
TET
THA
WS
Table 3.1: Variable definition
Definitions
Agricultural sector
value added of the agricultural sector
value added of the soybean production sector
economic labor productivity in agriculture
economic labor productivity of the paid work sector
economic labor productivity of the soybean production sector
exosomatic metabolic rate in agriculture
exosomatic metabolic rate in of the paid work sector
Exosomatic metabolic rate for the whole society
exosomatic metabolic rate of the soybean production sector
energy consumption of agricultural sector
energy consumption of the paid work sector
energy consumption of the soybean production sector
Household consumption
paid-work sector
Soybean production sector
Total energy throughput (i.e. total energy consumption) of the economy
Total human activity, i.e. the hours available in the whole society
whole society paid-work sector
3.2.2 Net delivery of biofuels
In order to analyse the feasibility of one energy source with an alternative energy
source the output/input energy ratio (ER) of the alternative energy source is
required.
This ratio measures the gross energy flow that can be yielded in a useful
form for a society out of the direct and indirect energy investment in an energy
system (Ulgiati et al., 2008). The energy investment (i.e. input) and the gross
return (i.e. output) are measured in terms of energy carriers, which are used for the
exploitation of primary energy sources (Giampietro and Mayumi, 2009). If
biofuels are intended as a primary energy source that is fully renewable, the gross
supply of biofuel (GSB) must substitute for the required energy to deliver the
inputs for the biofuel production system. After subtracting such inputs from the
GSB, a net supply of biofuel (NSB) remains available for society. The delivery of
energy carriers just described gives rise to an internal loop (see Fig. 3.1) which
amplifies the required production of biofuel according to Eq. 3.1 (Ibid.).
GSB / NSB = ER x [1 /( ER − 1)]]
(Eq. 3.1)
30
Figure 3.1: Internal loop in biodiesel production
Land & other
natural inputs
Biomass
production
conversion
Gross
supply (GSB)
Net supply
biofuel (NSB)
Input: GSB - NSB
Labour
Source: Adapted from Giampietro and Ulgiati (2005)
Failing to include this internal loop means that the biofuel production system
could cannibalize the energy matrix by consuming other energy sources with a
minimal (even zero) amount of net supply for societal use6.
In spite of the critical importance that the abovementioned internal
amplification loop has for sustainability, many well known articles and studies
empathize that biofuels have significant petroleum displacement capacity, simply
stressing that the ER is higher then one. Examples can be found in Farrel et al.
(2006), Shapouri (2004), Graboski (2002), Wang et al. (1999), Groode and
Heywood (2007), Sheehan (1998), Kim and Dale (2002; 2004), de Oliveira et al.
(2005), Lorenz and Morris (1995), among many others.
3.2.3 Scenarios
Although biodiesel in Brazil can be produced from a wide variety of available
vegetable oils and animal tallow, the vast majority of feedstock has always been
soybean oil. The official Brazilian body regulating oil, natural gas and biofuels
industries, the Agencia Nacional do Petróleo, Gás natural e Biocombustíveis
(ANP), releases a monthly bulletin on biodiesel, reporting amongst other things,
the contribution of the various feedstocks to the overall biodiesel production.
Since the first bulletin, the contribution of soybean has always been around (and
generally above) 80%. In a recent study commissioned by the industrial
association of biodiesel producers – Ubrabio- the role of soybean is not expected
to significantly decrease even with the industry’s aim of 20% fossil diesel
6
“Energy cannibalism” refers to an effect where the fast growth of an energy delivery system
provokes a need for energy that uses (or cannibalizes) the energy released by other energy sources
(Kenny et al., 2010; Pearce, 2009)
31
substitution (FGV, 2010). Thus, this paper only considers the case of soybean
biodiesel.
The scenario in which MuSIASEM is applied here envisages a substitution
of 20% fossil diesel with biodiesel (also called B20) which the Ministry of Mines
and Energy envisaged when the national program of biodiesel production was
launched. It is also considered as feasible and desirable by Ubrabio. It is worth
noting that the energy content of 20% of diesel is equivalent to just 3.3% of the
total consumption of primary energy of Brazil.
The base year for the calculation is 2005 when biodiesel was introduced in
the energy matrix. This is also the year of the most recent Brazilian input-output
tables.
Introducing biodiesel into the energy matrix at the levels envisaged here
would require several years during which many other changes that could be
included in the analysis could occur. However, rather than introducing macroeconomic and other variables (e.g. demographic trends, sector growth, changes in
land availability caused by urban growth or deforestation) for the year in which
biodiesel production would reach the target level, all figures were frozen at 2005
conditions. Thus final results should not be intended as forecasts, but as a “what
if” approach with no specific time frame.
Biodiesel is more expensive than diesel 7 , thus replacing diesel with
biodiesel increases production costs for all the sectors that use diesel for their
operations. Firms can reduce their profits if they bear the consequent rise in
production costs, alternatively they can increase the prices of products and
services. In addition, biodiesel can substitute imported diesel or domestically
produced diesel. Consequently, four scenarios were considered:
•
•
•
Scenario A - Price changes & import substitution: biodiesel substitutes all
imported diesel and part of the domestically produced diesel. The rise in
production costs that the substitution of fossil diesel with more expensive
biodiesel implies is reflected in an inflationary rise in production prices.
Scenario B - Price changes & no import substitution: biodiesel substitutes
only domestically produced diesel, thus the quantity of imported diesel
remains unchanged. As above, inflationary effects are taken into account.
Scenario C - No price changes & import substitution: biodiesel substitutes
all imported diesel (and part of domestically produced diesel). The rise in
production costs is supported by producers without increases in
production prices.
7
The most part of biodiesel cost is represented by vegetable oil costs and this leaves small room
for cost reduction. Additionally, the price of Brazilian fossil diesel is among the cheapest in the
world as a result of specific policy aimed at subsidizing the diesel price to the detriment of the
petrol price. This is because diesel is considered an important element of the whole industrial
structure and its low cost facilitates import substitution industrialization (da Silva Dias, 2007).
32
•
Scenario D - No price changes & no import substitution: biodiesel
substitutes only domestically produced diesel. There are no inflationary
effects because increases in production costs are absorbed by producers.
Obviously, the distinction between the scenarios with price effects and those
without such inflationary adjustments are relevant only for the economic analysis
and not for the energy analysis.
According to the National Energy Balance (MME, 2010), in 2005 Brazil
consumed 33.95 million tons of diesel8, of which 32.25 million were domestically
produced and 2.5 million were imported. In the same year 0.88 million tons of
diesel were exported. Of the overall diesel consumption, 80% was used for
transport. Twenty percent of the entire consumption of diesel is equivalent to
6,790,766 tons. Given the lower calorific value of biodiesel compared to diesel,
this quantity is provided by 8.24 million tons of biodiesel (see the biodiesel
energy balance section for details of the sources). This is the biodiesel level of
production and consumption in the above four scenarios.
3.2.4 Energy balance and energy analysis
Input-output analysis (IOA) can be used not only to assess economic impacts but
also for energy analyses. Since IOA is used here to represent economic flows, one
would consequently use it also to calculate energy flows.
There are two main ways to use energy IOA: direct impact coefficients of
energy intensity and hybrid models. The former method implies the construction
of a matrix of direct energy coefficients which is obtained by dividing the energy
consumed in each sector by the monetary output of the same sector. The direct
and indirect consumption of energy would be estimated by multiplying the matrix
so calculated by the Leontief’s Inverse. This method is pretty simple but it
violates the energy conservation condition (primary energy equals secondary
energy plus losses) when the energy prices are not equal across all consuming
sectors, including final demand (Herendeen, 1974; Miller and Blair, 1985). This
equal tariff condition for energy prices is certainly not verified in Brazil.
Moreover, when final demand changes are significantly different from the base
year final demand, the direct impact coefficient method introduces errors into the
estimations (Bullard and Herendeen, 1975; Casler and Blair, 1997). Thus its
results would be unreliable for this study.
In the hybrid method the purchases and sales of energy of the inter-industry
transactions are in their original unit of measurements (e.g. tons of coal, KWh of
electricity, or joules of energy) and not in monetary terms. The hybrid method
does not have the problems related to the direct impact coefficients method
mentioned above. However, not only is a complete match required between the
8
Diesel density is assumed to be 0.84 (Coppens, 2003) .
33
energy sectors where the energy data are taken from and the same sectors of the
input-output tables, but also information is required on purchases that the energy
sectors make from all the other sectors. Through the use of official Brazilian
statistics, the first of these two problems could be solved with a very high degree
of approximation, but not the second problem. Therefore, in this paper process
analysis is used for the energy balance of soybean biodiesel.
In order to make the results of the energy analysis compatible with those of
the IOA, it is assumed that all the inputs used in the biodiesel production and
delivery are produced in Brazil. Similarly, all direct and indirect energy
consumption is assumed to occur in the country. This approximation is not far
from reality because Brazil has an industry which produces all the inputs used in
biodiesel production, including equipment and machines (Ferreira and Cristo,
2006).
To the best of my knowledge, the only energy balance calculated (and
reported in a peer review journal) for the specific conditions of Brazilian soybean
production is Cavalett and Orgeta's (2010). Energy balances for the same crop
calculated in other countries are not really relevant. This is because Brazil
presents very particular conditions, such as one of the highest soybean
productivity per hectare in the world (FAO, 2011) and a very low use of fossil
fuel in electricity production.
This work presents a new energy balance for soybean biodiesel. It includes
direct and embedded energy inputs for all the steps in the biodiesel production
chain: soybean production, transport of soybeans, soybean crushing,
transesterification and biodiesel transport. These are the same phases used in IOA.
Data on input quantities for the agricultural phase are taken from Cavalett and
Ortega since they reported real Brazilian cases. The data on the input quantities of
the oil extraction and transesterification phase were collected from one Brazilian
biodiesel producer (while for this phase Cavalett and Ortega used theoretical data
found in the literature). Inferring national values from individual cases can be
risky but no other information sources were available. Additionally, the producer
in question is one the largest in Brazil9 and the technology is considered mature.
Therefore large variations among producers are not expected. Input quantities for
transport phases are again from Cavalett and Ortega.
The soybean productivity was assumed to be 2,911 Kg/ha (Estimated by
Conab for the 2009/10 cropping season and reported in March 2010 survey) and
the oil yield was estimated to be 19% (calculated from Abiove, 2010). The main
products of the soybean crushing phase are oil and cake. The energy cost of
soybean crushing was attributed proportionally to the energy content of soybean
oil. Using data from Domalsky et al. (1986)10 and assuming a cake yield of 77%
9
To accommodate the requested confidentiality the name cannot be revealed here
Although this source may appear rather old the datum can still considered valid because the
technology through which the energy content is measured has not changed.
10
34
(from Abiove), the energy content of soybean oil turns out to be 40.2% of the
combustion value of oil and cake.
The other main product of biodiesel production is glycerin. However, in this
work it was assumed that glycerin was not a source of revenue and that it did not
constitute an energy credit. In 2006, when biodiesel production was just beginning,
the glycerin produced was just 14,000 tons (ABIQUIM, 2007). Considering that 9
m3 of biodiesel entails producing 1 m3 of glycerin, the current level of biodiesel
consumption (which substitutes just 5% of fossil diesel) entails producing about
265,000 tons of glycerin per year: 7.5 times more than the domestic consumption
(Fairbanks, 2009). This production has already caused a substantial decline in the
price of glycerin from 4 R$ in 2007 to 1.8 R$ in June 2009. If biodiesel reaches
the 20% target of fossil diesel substitution, the production of glycerin is going to
become a cost rather than a revenue.
Table 3.2 reports the input required for biodiesel production with the energy
costs. The results are used in two ways: to calculate the amplification factor of Eq.
3.1, and to integrate the National Energy Balance data for the allocation of energy
consumption among the different scales of analysis.
35
Table 3.2: Soybean biodiesel process inventory
Inputs
Unit Q./ha/yr MJ/Unit Total Energy
Agricultural production
Machineries
Kg
25
92.5
2,312.5
Diesel
Lt
65
45.1
2,933.7
Phosphorus
Kg
78.8
4.6
361.3
Potassium
Kg
78.8
5.9
463.8
Lime
Kg
375
1.3
489.3
Seeds
Kg
69
4.8
331.0
Herbicides
Kg
4.8
233.8
1,122.2
Insecticides
Kg
3.2
242.7
776.7
Electricity
KWh
34
5.7
193.5
Farm buildings
M2
0.09
1.800
162
Soybean transport
Machineries
Kg
2
127.3
259.4
Diesel
Lt
5.2
45.1
234.6
Crushing
Buildings
M2
0.0005
1800
0.9
Machineries
Kg
0.3
70.5
20.5
Diesel
Lt
62
45.1
2,799.7
Electricity
KWh
202.8
5.7
1,153.7
Water
Kg
234.5
0.0026
0.6
Hexane
Kg
7
45.4
317.2
Biodiesel production
Buildings
M2
0.01
1,800
23.9
Machineries
Kg
1.2
70.5
89.7
Methanol
Kg
74.6
36.3
2,709.4
Catalyst (CH3ONa)
Kg
8.8
39.1
345.5
Electricity
KWh
24.1
5.7
137.4
Water
Kg
261.2
0.003
0.7
Wood
Kg
229.3
13.2
3,035.3
Fuel oil
Kg
0.8
51.7
41.4
Biodiesel transport
Machineries
Kg
0.4
127.3
47.5
Diesel
Lt
0.9
45.1
43.0
Total inputs
MJ
17.836.0*
Output
Kg
533.1
36.95
19.699.5
Energy ref.
1
2
3
3
4
5
3
3
6
7
1
2
7
1
2
6
8
9
7
1
10
11
6
8
12
2
1
2
11
References for energy costs 1) Scholz et al. (1998); 2) Boustead & Hancock (1979); 3) West & Marland
(2002) adjusted for the Brazilian electricity efficiency reported in Coltro et al.(2003) ; 4) Shapouri et al.
(2004); 5) Following a common procedure. the energy cost of seed production was estimated to be 150%
of the agricultural phase; 6) Coltro et al. (2003); 7) Macedo et al. (2008); 8) Haguiuda & Veneziani
(2006); 9) Ahmed et al. (1994); 10) Kamahara et al. (2010); 11) Sheehan (1998); 12) MME (2010) for
direct energy and wood density and de Oliveira & Seixas (2006) for indirect energy.
*The inputs of the crushing phase are multiplied by 40.2%
Data on the energy consumption of AG, PS, SG, HH and WS in 2005 (i.e. the
baseline year) were taken from the national energy balance. ETsoy was estimated
from the energy balance reported in Tab. 3.2 and is given by the direct energy
consumption of the agricultural phase (i.e. electricity and diesel) multiplied by the
numbers of hectares of soybean (reported by IBGE, 2005b). The energy
36
consumption of the PW is equal to the difference between the TET (of the WS
which includes all primary energy sources) and the energy consumed by HH.
Thus all energy losses are attributed to PW. In fact, PW includes the energy sector.
The changes in energy flows caused by the introduction of biodiesel in the
different levels are estimated by apportioning the energy data of the national
energy balance and from Tab. 3.2 to the specific consumption compartments.
Specifically, ETsoy was obtained by multiplying the energy consumption for
soybean production per unit of area (the agricultural phase in Tab 3.2) by the new
soybean production area after the introduction of biodiesel. The sum of ETsoy and
RestAG yielded the energy consumption of AG (i.e. ETag). For the scenarios
without substitutions of imported diesel (i.e. B and D), ETpw was calculated as
follows
ETpw = ETag + BD + ETpwbase − ETag base − Dd +ind (Eq. 3.2)
where ETag is the energy consumption of AG after the introduction of
biodiesel, BD is the additional direct and indirect energy consumption required by
biodiesel production (data are from Tab. 3.2 multiplied by total biodiesel
production) excluding the agricultural phase (which has already been included in
ETag), ETpwbase and ETagbase are respectively the energy consumption of PW and
the energy consumption of AG in the baseline case (i.e. before the introduction of
biodiesel), and Dd+ind is direct and embedded diesel energy 11 substituted by
biodiesel.
For the scenarios that envisage the substitution of imported diesel, ETpw is
calculated as above, the only difference being the last element in Eq. 3.2: the
energy of the substituted fossil diesel is equal to the direct energy of diesel12 that
is imported plus the total (i.e. direct and indirect) energy of domestically produced
diesel that is substituted.
TET in all the scenarios is calculated by summing the energy consumption
of HH, (which is not affected by the introduction of biodiesel) with ETpw.
The results section reports the outcomes of the energy analysis both
disregarding the amplification factor of fully renewable biodiesel (see Eq. 3.1)
and including this amplification. In this last case, the additional energy
consumption provoked by biodiesel is multiplied by the amplification factor.
3.2.5 Input-output analysis: data and direct coefficient matrix
IOA is used in this study to evaluate the changes in the flows of added value and
in hours of human activity that would be caused by the introduction of biodiesel.
11
12
Unit value reported in Tab. 3
It is equal to 44.84 MJ/Kg (Boustead and Hancock, 1979)
37
Through IOA it is possible to analyze the direct and indirect effects that a change
in one sector has on the others. According to the simplified characterization of
input-output tables, the economy is represented by a set of linear equations
containing technical coefficients. These represent the relationship between the
final production of each sector and its inputs. The assumptions of input-output
representations are known: no substitution effects between inputs, no supply
constraints, a fixed proportion between outputs and inputs, and constant returns to
scale.
The input-output tables used here are from IBGE (2005a) and present 55
sectors. This study considers biodiesel from soybean, consequently biodiesel
production is expected to have a strong impact on soybean production. Leaving
soybean production as part of the agricultural sector would lead to a loss of
critical information and would underestimate the overall impact. This is because
soybean has an high value added in comparison with the average of agricultural
activities. Moreover, soybean production is one of the hierarchical scales (i.e. n-3
level) of the analysis adopted here. Soybean cultivation was thus included as a
separate sector in the input-output tables. The values of this sector were taken
from the IBGE make and use tables (2005c) and were levelled to basic prices by
applying Guilhoto and Sesso Filho’s methodology (2005).
The diesel sector was disaggregated from its original sector (i.e. oil refinery)
in a pro-quota way, by multiplying all the purchases of the original sector by the
ratio between the total value of diesel and the whole value of the sector.
Unfortunately, no other information was available for a more precise calculation;
however the estimated purchase structure of the calculated diesel sector should
represent a good approximation.
As mentioned, the most recent input-output tables available were from 2005
when there was basically no biodiesel production. Consequently, biodiesel was
included as a separate sector in the tables. The sources of data on purchases
entailed by biodiesel production and oil extraction is the same biodiesel producer
whose data were used in the energy analysis. These data were collected in 2009
and adjusted to 2005 prices. The price of biodiesel was obtained through a
weighted average of prices auctioned by ANP in 2009 (the most recent and
complete year when this research was done) and was deflated to 2005 soybean oil
prices reported by Abiove (2010). In fact, the price of feedstock represents the
main cost and its high correlation with biodiesel prices can easily be shown. The
total value of biodiesel thus calculated amounted to 14,987 million Reais13 (R$).
Table 3.3 reports the technical coefficients of the biodiesel sector with all inputs
organized according to the original industrial activities they affect. The biodiesel
sector presented here integrates the crushing and transesterification phases (using
methanol as alcohol) and its technical parameters are the same as reported in the
13
The average R$/US$ exchange rate in 2005 was 2.435.
38
energy balance section. In line with the energy analysis of this study (see Section
3.2.4), all inputs are assumed not to be imported.
Table 3.3: Technical coefficients of biodiesel production
Economic activities
Direct
coefficients
Soybean cultivation
0.59566
Agricolture. forestry and logging
0.00333
Petroleum refinery and coke
0.03194
Diesel
0.03165
Chemicals
0.01858
Manufacture of machineries & equipments
0.01125
Vehicle spares and accessories
0.00045
Electricity. gas and water
0.02450
Construction
0.00048
Trade
0.03262
Transport. Storage and mail
0.03097
Information services
0.00370
Financial intermediation and insurance services
0.00348
Real estate and renting
0.00169
Maintenance and repair services
0.00044
Restaurants. cafes and hotels
0.00091
Business services
0.00280
According to the data, the biodiesel sector thus envisaged would employ around
38,450 people for the production of 8.24 millions tons required to substitute 20%
of fossil diesel.
The sale structure of biodiesel is assumed to be the same as diesel. The extra
soybean cake that the introduction of biodiesel would cause is assumed to be
completely exported at the price 428 R$/ton (reported in Abiove). Of course, this
is a strong assumption in favour of biodiesel production because the cake price is
assumed to be constant, so that biodiesel would not cause any reduction in the
price of the cake. However, given the continuous increase in global demand for
meat (cake is used as an animal feed), such an assumption may not be far from
reality. For the reasons mentioned above no earnings are envisaged for the sale of
glycerin.
The final matrix presents 58 sectors (the 55 originally present in the IBGE
inter-industry input-output table plus soybean cultivation, diesel and biodiesel). A
different technical coefficient matrix table was used for each scenario. For
scenarios C and D, the technical coefficients of all the sectors are the original ones
with the exceptions just described due to the introduction of soybean and diesel
sectors. The only difference is that coefficients reflecting the purchase of diesel
from all sectors were in this work reduced to reflect the displacement caused by
biodiesel.
39
In scenarios A and B it is envisaged that firms would charge their customers
for the increase in production costs that the introduction of biodiesel causes. The
price of biodiesel was compared with the price of diesel. It was calculated in this
work that, on an energy equivalent basis, substituting 20% of fossil diesel with
biodiesel would involve a direct expense increase of 18% for the purchase of the
fuel for all the sectors of the economy if biodiesel only substituted domestically
produced diesel (i.e. scenario B). The increase on the direct expense of diesel
would amount to 26% if biodiesel also replaced imported diesel (i.e. scenario A).
This is because imported diesel is cheaper than domestically produced diesel
(average prices of imported and domestically produced diesel were calculated by
dividing the values of diesel reported in the input-output tables by the quantities
published in the Brazilian Energy Balance).
The typical Leontief price model is:
P = A′P + v + m
(Eq. 3.3)
where P is the column vector of the price indexes, A´ is the transpose of the
technical coefficient matrix, v is the vector obtained by dividing the added value
of each sector by its output, and m is the ratio of imports to output in the i-th
sector.
The price model presented here for estimating this effect is adapted from
Valadkhani and Mitchell (2002) and is used to estimate the baseline conditions
scenario A and B . This is because, in such scenarios it is assumed that the rise in
production costs caused by the substitution of diesel with biodiesel is reflected in
an increase in the values of the sales and not in a reduction of the value added of
the sectors.
The above-mentioned equation can be partitioned into exogenous and
endogenous elements:
′   PX  v X  m X 
 PX  a XX AEX
 P  =  A′ A′  ⋅  P  + v  + m 
 E   XE EE   E   E   E 
(Eq. 3.4)
where PX represent the new price index of diesel if biodiesel were
introduced and is equivalent to 1.26 in scenario A and 1.18 in scenario B. PE is the
vector of prices in all other sectors, vX is the ratio of the added value in the diesel
sector to its output, vE is the same ratio for all the other sectors, mX is the ratio
between imports and output in the diesel sector, mE is the same ratio of imports to
outputs in all sectors excluding diesel (i.e. endogenous sectors), A´XE is the vector
representing the inputs of diesel to all other sectors, and AEE is the matrix of
technical coefficients of all endogenous sectors.
After some manipulation Eq. 3.4 becomes
40
PE = ( I − A' EE ) −1 AXE PX + ( I − A' EE ) −1 (v E + m E )
(Eq. 3.5)
In this way the vector reporting the price effects of all the endogenous
sectors is calculated. The inter-industry flow matrix Z was pre-multiplied by the
diagonal matrix obtained from the P vector. Consequently, the new inflated values
of the purchases of all the sectors ZINF were estimated. Equation 3.6 reports this
operation where the first lines z1j represent the purchase of diesel from the j-th
sector, all other zij elements of the matrix are the purchases from the j-th sector of
the i-th sectors’ production, PX is the index of the price of diesel plus biodiesel.
The elements from PEi to PEn are derived from Eq. 3.4
Z INF
 P X 0 0   z11 M z1n 


= 0 P2E 0  ⋅  z 21 M z 2 n 
0 0 P E   z M z 
nn 
n   n1

(Eq. 3.6)
Similarly, the final demand matrix F and the output vector x were premultiplied by the same diagonalized P vector reported in Eq. 3.5 in order to obtain
the inflated expenses for the final demand FINF and the new inflated output xINF.
The first line of ZINF and the first line of FINF were then split into fossil
diesel and biodiesel.
The new matrix of domestic direct coefficients A* for scenarios A and B
was obtained by dividing each purchase of the new inter-industry inflated
transaction matrix by the inflated value of its output. Consequently, A* reflects the
fact that producers charge their customers for the increases in fuel prices that the
introduction of biodiesel causes.
The inflationary increases modeled in scenario A and B could cause a loss
of competitiveness of the country along with a small reduction in the real incomes
of families. However, including such effects goes beyond the scope of this study
and consequently it is not modeled here.
3.2.6 Input-output analysis: impact assessment
The IOA that is employed to evaluate the economic impacts of biodiesel builds on
the model of Wicke et al. (2009). The technical coefficient matrix (with the
inclusion of the biodiesel sector) was partitioned as follows:
&& =  Adb Bdb 
A
A B 


(Eq. 3.7)
41
where A is the matrix of direct coefficients for all endogenous sectors
(estimated through the price model for scenarios A and B), B describes the inputs
to diesel and biodiesel sectors from all other industries to produce a Real’s worth
of output of the two sectors, Adb represents the direct coefficients of the sale
structure from biodiesel and diesel sectors to all other sectors, and Bdb are the
inputs from biodiesel and diesel to produce a Real’s worth of their own
production.
In terms of change, the classical Leontief (I-Ä)∆X=∆Y model would result
in:
− Adb ( I − Bdb )  ∆xdb  ∆y db 
⋅
=
( I − A)
− B  ∆x  ∆y 

(Eq. 3.8)
where ∆xdb is exogenously set and represents the contraction in the diesel
sector’s output and the biodiesel sector’s output, ∆x is the change in output of all
other sectors, ∆ydb represents the change in the final demand of the diesel and
biodiesel sectors, whilst ∆y is the change in final demand for all other sectors.
The typical Leontief model is demand driven, where exogenous changes are
in final demand. In this work exogenous changes are in the output of two sectors:
biodiesel and diesel14. With the exclusion of these two sectors, the final demand
of all the sectors is assumed not to be affected by the introduction of biodiesel
(∆y=0). The solution to Eq. 3.8 is:
∆xdb

∆X = 
−1 ~ 
( I − A) B 
(Eq. 3.9)
~
∆x − Adb ( I − A) −1 B − Bdb ∆xdb 
∆Y =  db

0


(Eq. 3.10)
~
where vector B is the input cost from the sectors with endogenous output
~
(i.e. non diesel and non biodiesel sectors) to produce ∆xdb, Adb(I − A)−1 B is the
flow of sales from diesel and biodiesel sectors to all other sectors, and B∆xdb
represents the purchases of diesel and biodiesel sectors for their own production.
Using this method it is therefore possible to assess the direct and indirect
economic impacts that the introduction of biodiesel would cause. IOA is used here
to calculate the added value, the employment of each sector and the land use of
the soybean and agricultural sectors. The results were calculated including the
14
A full treatment of these models can be found in Miller and Blair (1985) and Roberts (1994).
42
amplification factor (see Eq. 3.1) in terms of the required biodiesel output and
also without such an amplification.
3.3 Discussion and results
The energy ratio (ER) of soybean biodiesel turns out to be 1.10. It is higher than
one, so the energy delivered is higher than the energy invested. However, the ER
is very small. In fact, for each joule of gross energy delivered to society, 91%
comes from other sources and only 9% constitutes the net delivery of energy from
the biodiesel production system. Since energy sources that are employed in the
delivery of gross biodiesel supply have a positive opportunity cost, this high level
of energy input means that the biodiesel delivery could lead to energy
cannibalization. This is why, in a fully renewable biodiesel system, the energy
inputs required to deliver gross energy to society must come from the same
biodiesel production system. However, the ER calculated here gives rise to an
amplification factor of 10.6. This means that 10.6 liters of biodiesel are required
to provide one net litre of biodiesel, and 9.6 liters are reinvested in the biodiesel
production and delivery system. Thus, the biodiesel quantity required to substitute
20% of fossil diesel must be multiplied by 10.6. If this amplification factor is not
taken into account, biodiesel would substitute just 0.3% of national primary
energy consumption and not 3.3% (i.e. the energy contained in 20% of diesel),
thus energy gains would essentially be nil.
Table 3.4 reports the results of the extensive variables of the economic and
energy analysis for the four levels, both for fully renewable biodiesel and non
renewable biodiesel (i.e. when the amplification factor is not taken into account).
The results reported in Tab. 3.4 refer only to scenario A, which presents the most
favorable assumptions for biodiesel while the remaining three scenarios are
reported in the appendix A3.1. The baseline conditions are included for
comparison and report the inflationary effects estimated through the price model
(as if substituted diesel were to cost the same as biodiesel).
43
Soybean
(n-3)
Baseline conditions
ETi (MJ)
7.33E+10
HAi (Hours) 6.74E+08
AVi (106 R$) 17,387
Table 3.4: Results - Extensive variables
Agriculture
Paid work
(n-2)
(n-1)
Whole society
(n)
3.50E+11
3.61E+10
105,110
7.38E+12
1.73E+11
1,841,619
9.15E+12
1.61E+12
1,841,619
Scenario A: non renewable biodiesel
ETi (MJ)
1.24E+11
4.01E+11
3.66E+10
AVi (106 R$) 29,329
117,211
7.66E+12
1.74E+11
1,862,783
9.43E+12
1.61E+12
1,862,783
Scenario A: fully renewable biodiesel
ETi (MJ)
6.08E+11
8.95E+11
4.23E+10
AVi (106 R$) 144,076
234,741
1.89E+13
1.84E+11
2,086,383
2.07E+13
1.61E+12
2.086.383
The introduction of non renewable biodiesel creates positive variations in the
extensive variables characterizing the economic and energy flows. These changes
are much higher at lower levels: AVsoy and ETsoy increase by almost 70%, AVag
and ETag by almost 15%. At higher levels, variations in these variables become
negligible. The reason for this must be found in a reduced contribution to the GDP
of the sectors whose production is stimulated by biodiesel, mainly soybean
production. Although soybean production has quite an important role for the
overall production of the agricultural sector, as an agricultural activity its
contribution to the total added value to all the sectors of the economy remains
very small. Moreover, contraction in the output of domestic fossil diesel reduces
the positive economic effects of biodiesel.
Changes in the same variables are much more evident when fully renewable
biodiesel is analyzed. The energy consumption of soybean (level n-3) and its
added value increase by more then seven times. In addition, the introduction of
fully renewable biodiesel implies such a high quantity of production of biodiesel
that changes in the values of the flows are also evident at higher scales. In fact, for
the whole society (level n), the increase in the total value added would be around
13%, but this would mean twice as much energy being consumed. This is because
biodiesel is a product with low economic value but its fully renewable production
implies enormous amounts of energy consumption. Additionally, the contribution
to the total added value of the sectors that are directly and indirectly activated by
biodiesel production is pretty small.
The above-mentioned variables indicate quantity changes in the throughput
of the flows. Changes in the pace of consumption and production of the same
flows are given by the two intensive variables (reported in Table 3.5): the
economic labor productivity (ELP) and the exosomatic metabolic rate (EMR).
44
Table 3.5: Results – Intensive variables
Soybean
Agriculture
Paid work
(n-3)
(n-2)
(n-1)
Baseline conditions
EMRi (MJ/hr)
ELPi (R$/hr)
108.8
25.8
Whole society
(n)
9.7
2.9
42.7
10.7
5.7
1.1
Scenario A: non renewable biodiesel
EMRi (MJ/hr)
108.8
ELPi (R$/hr)
25.8
11.0
3.2
44.0
10.7
5.09
1.2
Scenario A: fully renewable biodiesel
EMRi (MJ/hr)
108.8
ELPi (R$/hr)
25.8
21.2
5.6
102.6
11.3
12.8
1.3
On the assumption of constant returns to scale of IOA, changes in ELP are simply
the result of the different contributions of the value added of sectors to the value
added of the higher level in which the same sectors are aggregated. As above,
some changes can be detected for the non renewable biodiesel case as well, but
the effects of the introduction of biodiesel are much stronger if biodiesel is
intended to be fully renewable. In fact, the value added that is produced per hour
of work in agriculture (i.e. ELPag) would increase by 90% as a result of the
additional soybean production required to produce biodiesel. This is because
soybean farming is generally highly mechanized and just one individual farmer
can easily control 200 hectares (Roessing and Lazzarotto, 2004). However, in
comparison with the scenario with no biodiesel in the energy matrix, the
biophysical capitalization that the soybean production requires would raise the
energy consumed per hour of work by 118% for the whole agricultural sector
(level n-2). Scaling up to the paid work level (n-1), would imply an increase in
ELP of just 6% and a rise in EMR of more than 140%.
This result contrasts with the historical trend that ELP and EMR generally
have. In fact, a high correlation between ELPpw and EMRpw has been found for
countries such as Spain (Ramos-Martín, 2001), Ecuador (Falconi-Benitez, 2001)
and the USA (Cleveland et al., 1984). Regarding Brazil, I found a Pearson
correlation coefficient of 0.85 between EMRpw and ELPpw using a time series
from 1990 to 2007. An in-depth historical analysis on the reasons why these two
variables result to be correlated for the case of Brazil would go beyond the scope
of this scenario assessment study. However, a tentative explanation is that higher
consumption of energy per hour of work in the PW sector means the use of bigger
and more machineries and equipments, which, in turn, implies the consumption of
more direct energy for running the same machines. The final result is an increase
of the productivity of labor (Cleveland et al., 1984; Hall et al., 1986; RamosMartín et al., 2009). Contrary to these historical results, the substitution of fossil
diesel with soybean biodiesel would strongly increase EMRpw with a much lower
45
increase in ELPpw. The consequence would be a worsening in the energy
intensity (TET / total value added) of the country which would increase by 100%.
The two funds of the MuSIASEM methodology are human time and land.
Of these, land is certainly the most limiting factor. In 2005, the area dedicated to
soybean amounted to 23.3 million hectares and with fully renewable biodiesel,
this area would have to expand to cover 85% of the land that the last agricultural
census (IBGE, 2006) reported as being used for agricultural purposes. Such an
expansion of soybean cultivation would entail clearing forest land and using the
area currently protected by law.
If biofuels are intended to reduce GHG emissions the carbon generated on
land by the bioenergy crops must be higher than the carbon captured by
sequestration and storage on the cultivated field (Searchinger et al., 2008) .
However, this is hardly the case if energy crops cause forest destruction. In this
regard Fargione et al. (2008) calculate that the emissions per hectare caused by
deforestation in the Amazon region would require 320 years to repay the carbon
debt with soybean biodiesel and from 17 to 37 years if forest clearing occurs in
the Cerrado biome. The “Soybean Moratorium”15 introduced in 2006 under the
promotion of Greenpeace has proven to be an effective way for preventing forest
substitution with soybean planting in the Amazon (Lapola et al., 2010). However,
this initiative cannot prevent indirect land use changes. If soybean is planted on
rangelands and if new pastures are consequently obtained by forest clearing, direct
land use change is avoided but the overall GHG balance turns out be extremely
negative. The land that can be purchased by ranchers in the agricultural frontier
after having sold their pastures to soybean planters in the South can be 10 times as
large the original farm because of the differentials in land prices (Sawyer, 2008).
Nowadays the new agricultural expansion frontier is the Centre-West and the
North region. However, the great part of soybean expansion is talking place in the
Cerrado (Girardi, 2008; Sawyer, 2008). This is a large tropical savannah biome
much more vulnerable then the Amazon because it has less protection (there is no
Soybean Moratorium and forest clearing up to 80% of the farm is perfectly legal).
The Cerrado has a vast range of animal and plant biodiversity (Klink and
Machado, 2005; Ratter et al., 1997) and is considered one of the world
“biodiversity hotspot” (Myers et al., 2000), i.e. “areas featuring exceptional
concentrations of endemic species and experiencing exceptional loss of habitat”
(Ibid. p.853). Due to the more famous deforestation process of the Amazon forest,
the deforestation of the Cerrado is less acknowledged at world level (Janssen and
Rutz, 2011; Mazzetto Silva, 2009). But the deforestation pace in the Cerrado is
much higher than in the Amazon (Mazzetto Silva, 2009; Sawyer, 2008). Forest
destruction in the Cerrado is considered as an alternative to the forest clearing in
the Amazon (Sawyer, 2008). It is in the Cerrado where the typical single mono15
It implies that soybean cannot be traded based on private arrangements among trading
companies if soybean was planted in recently deforested areas.
46
crop plantations take place mainly for soybean cultivation. This area is also
experiencing pressure for sugarcane expansion in addition to its traditional use for
cattle breading. Biodiesel production in Brazil started only in 2005 but some
worrying signals can be detected. According to the Conab crop survey of May
2011, soybean areas has been expanding with an annual average rate of 3,95%
from 2006/07 to 2010/1116.
The results of the other scenarios (reported in Table A3.1 of the appendix)
are not very different, especially for the case of fully renewable biodiesel. This is
because the amplification factor is so high as to cancel any minor differences.
3.4 Conclusions
It is often argued that developing countries have perfect conditions for developing
biofuels, especially Brazil and more in general Latin America. This area is
deemed to have a very high potential for biomass production (Smeets et al., 2007)
due to its surplus agricultural land, favorable climate and soil conditions, a good
infrastructure, and an abundance of labor force.
It is becoming increasingly acknowledged that diversification away from oil
is more and more urgent. This work has attempted to show how the use of a
parallel biophysical and economic reading at different scales can shed light on the
consequences and sustainability of alternative options to oil. Specifically, this
paper has illustrated how MuSIASEM can be applied to enrich the discussion on
the implications of fossil fuel substitution with biodiesel in Brazil.
Of course, the development of a new industry has positive effects on the
overall value added of a country, but merely looking at the GDP will not indicate
if something is actually positive or negative. The energy delivered by soybean
biodiesel is higher than the energy invested. However, the resulting net energy is
not much compared with the energy source it is intended to substitute. Net energy
has been essential for the evolution of present day organisms (Lotka, 1922; Odum
et al., 1995). In the same way, the industrialization of human societies implies the
use of high ER energy sources17.
This all means that replacing fossil diesel with fully renewable biodiesel
would lead to dramatic changes in the energy flows . The amount of energy per
hour of work tends to increase without a corresponding growth in the ELP. This is
because biodiesel mainly stimulates the production of sectors with low value
added while upstream linkages with high value added sectors are rather limited.
Biofuel proponents could argue that solutions should be searched for in order to
16
17
The soybean area in 2010/11 is estimated in the Conab survey.
See Cleveland (2005) for an estimation of oil and gas ER (or EROI to use his words).
47
reduce the energy consumed. However, this would entail a reduction in ELP and
consequently a loss of competitiveness for the whole economy. By the same token,
if the ELP were increased through economies of scale this would end up
increasing the level of mechanization, and thus the consumption of even more
energy. Moreover, further mechanization would reduce rural employment, and
this would go against one of the main purposes of the Brazilian National Program
of Biodiesel Production.
Eisenmenger et al. (2007) showed how Brazil has had a faster growth in the
energy consumed than in the GDP. Compared with the other Latin America
countries in their study, they highlighted that Brazil is very inefficient in the use
of energy. The use of fully renewable biodiesel would aggravate this problem. In
addition, the increase in energy consumption would occur without any direct
growth in the energy consumed by households, but only of the producing sectors.
The consequences of the introduction of non renewable biodiesel would be
much less dramatic and the constraints much less stringent. However, gains in net
energy would essentially be nil. In fact, the substitution of 20% of fossil diesel
would be equivalent to just 0,3% of the country energy consumption.
Brazil has a surface area of more than 850 million hectares. In purely
theoretical terms (i.e. without distinguishing among land cover and land use
classes) this area is so large that there would be enough land to reach the target of
20% of fossil diesel substitution even with fully renewable biodiesel. However,
soybean would cover the vast majority of the land currently used for agricultural
purposes, so direct and indirect land use shifts would be inevitable. Such a way to
de-carbonize energy consumption does not seem to be very wise. Unlike
developed countries, the majority of Brazilian GHG emissions do not come from
fossil fuel use, but from deforestation and changes in land use, which accounted
for 58% of CO2eq emissions in 2005 (MCT, 2009). So rather than contributing to
GHG displacement, soybean biodiesel might actually increase CO2emissions.
The upshot is that large-scale soybean biodiesel production and use in Brazil
may be feasible. But its desirability is highly questionable.
Further research could explore the implications of biofuel use and
production obtained by using different feedstocks.
48
References
Abiove. (2010). Associação Brasileira das Indústrias de Óleos Vegetais: Dados
do
complexo
soja.
[Online]
Available
from:
http://www.abiove.com.br/menu_br.html [Accessed 14 November 2010]
Ahmed, I. Decker, J. and Morris, D. (1994). How much energy does it take to
make a gallon of soybean biodiesel. Report prepared for the National
SoyDiesel Development Board. Institute for Local Self Reliance. [Online]
Available from: http://www.carbohydrateeconomy.org/ [Accessed 12 March
2010]
Associação Brasileira da Indústria Química - ABIQUIM. (2007). Anuário da
indústria química brasileira. São Paulo.
Boustead, I. and Hancock, G.F. (1979). Handbook of industrial energy analysis.
Horwood Limited, Ellis, UK
Casler, S.D. and Blair, P.D. (1997). Economic structure, fuel combustion, and
pollution emissions. Ecological Economics, 22 (1), 19-27.
Cavalett, O. and Ortega, E. (2010). Integrated environmental assessment of
biodiesel production from soybean in Brazil. Journal of Cleaner Production,
18 (1), 55-70.
Cleveland, C.J. (2005). Net energy from the extraction of oil and gas in the United
States. Energy, 30 (5), 769–782.
Cleveland, C.J. Costanza, R. Hall, C.A.S. and Kaufmann, R. (1984). Energy and
the US economy: a biophysical perspective. Science, 225 (4665), 890-897.
Coltro, L. Garcia, E.E.C. and Queiroz, G. (2003). Life cycle inventory for electric
energy system in Brazil. International Journal of Life Cycle Assessment, 8 (5),
290-296.
Companhia Nacional de Abastecimento (Conab). (2010). Safras. [Online]
Available from: http://www.conab.gov.br/ [Accessed 24 November 2010]
Coppens, G. (2003). Sojadossier. Soja voor mens en dier. Werkgroep voor een
Rechtvaardige en Verantwoorde Landbouw (WERVEL).
da Silva Dias, G.L. (2007). Um desafio novo: o biodiesel [A new challenge:
biodiesel]. Estudos Avançados, 21 (59).
de Oliveira, E.D. and Seixas, F. (2006). Análise energética de dois sistemas
mecanizados na colheita do eucalipto [Energy analysis of Eucalyptus
harvesting operations in two mechanized systems]. Scientia Florestalis, 70
(49-57).
de Oliveira, E.D. Vaughan, B.E. and Rykeil, E.J. (2005). Ethanol as fuel: energy,
carbon dioxide balances, and ecological footprint. BioScience, 55 (7), 593602.
Domalsky, E.S. Jobe, T.L. and Milne, T.A. (1986). Thermodynamic data for
biomass conversion and waste inceniration. Solar Technical Information
49
Program of the Solar Energy Research Institute. [Online] Available from:
http://www1.eere.energy.gov/biomass/ [Accessed 22 February 2010]
Eisenmenger, N. Ramos-Martín, J. and Schandl, H. (2007). Análisis del
metabolismo energético y de materiales de Brasil, Chile y Venezuela. Revista
Iberoamericana de Economía Ecológica, 6, 17-39.
Fairbanks, M. (2009). Glicerina. Revista Química e Derivados. No. 487. [Online]
Available from: http://www.dequi.eel.usp.br/ [Accessed 30 November 2010]
Falconi-Benitez, F. (2001). Integrated assessment of the recent economic history
of Ecuador. Population and Environment: a Journal of Interdisciplinary
Studies, 22 (3), 257-280.
FAO. (2011). FAOSTAT database. [Online] Available from: http://faostat.fao.org/
Fargione, J. Hill, J. David Tilman Polasky, S. and Hawthorne, P. (2008). Land
clearing and the biofuel carbon debt. Science, 29 (February), 1235-1237.
Farrell, A.E. Plevin, R.J. Turner, B.T. Jones, A.D. O’Hare, M. and Kammen, D.M.
(2006). Ethanol can contribute to energy and environmental goals. Science,
311, 506-508.
Ferreira, J.R. and Cristo, C.M.P.N. (2006). O Futuro da Indústria: biodiesel. Série
Política Industrial, Tecnológica e de Comércio Exterior. Ministério do
Desenvolvimento, Indústria e Comércio Exterior - MDIC, Brasília.
FGV. (2010). Biodiesel: sua importância e perspectivas para o Brasil. Fundação
Getulio Vargas & União Brasileira do Biodiesel - UBRABIO. [Online]
Available from: http://www.ubrabio.com.br/ [Accessed 4 November 2010]
Fischer-Kowalski, M. and Haberl, H. (1993). Metabolism and colonization.
Modes of production and the physical exchange between societies and nature.
Innovation in Social Sciences Research, 6 (4), 415-442.
Gaboski, M.S. (2002). Fossil energy use in the manufacture of corn ethanol.
Prepared for the National Corn Growers Association. [Accessed 18
November 2010]
Gasparatos, A. El-Haram, M. and Horner, M. (2009). Assessing the sustainability
of the UK society using thermodynamic concepts: Part 1. Renewable and
Sustainable Energy Reviews, 13 (5), 1074-1081.
Giampietro, M. (1994). Using hierarchy theory to explore the concept of
sustainable development. Futures, 26 (6), 616-625.
Giampietro, M. (2004). Multi-scale integrated analysis of agroecosystems. CRC
Press, Boca Raton, Florida
Giampietro, M. and Mayumi, K. (1997). A dynamic model of socioeconomic
systems based on hierarchy theory and its application to sustainability.
Structural Change and Economic Dynamics, 8 (4), 453-469.
Giampietro, M. and Mayumi, K. (2000a). Multi-scale integrated assessment of
societal metabolism: integrating biophysical and economic representations
50
across scales. Population and Environment: a Journal of Interdisciplinary
Studies, 22 (2), 155-210.
Giampietro, M. and Mayumi, K. (2000b). Multi-scale integrated assessment of
societal metabolism: introducing the approach. Population and Environment:
Giampietro, M. and Mayumi, K. (2009). The biofuel desillusion: the fallacy of
large scale agro-biofuel production. Forthcoming
Giampietro, M. Mayumi, K. and Ramos-Martin, J. (2009). Multi-scale integrated
analysis of societal ecosystem metabolism (MuSIASEM): theoretical
concepts and basic rational. Energy, 34 (313-322).
Giampietro, M. and Ramos-Martin, J. (2005). Multi-scale integrated analysis of
sustainability: a methodological tool to improve the quality of narratives.
International Journal of Global Environmental Issues, 5 (3/4), 119-141.
Giampietro, M. and Ulgiati, S. (2005). Integrated assessment of large-scale
biofuel production. Critical Review in Plant Science, 24 (5/6), 365-384.
Girardi, G. (2008). Aquecimento global e a nova geografia da produção agrícola
no Brasil. Embrapa - Unicamp. [Online] Available from:
http://www.embrapa.br/publicacoes/ [Accessed 10 October 2010]
Groode, T.A. and Heywood, J.B. (2007). Ethanol: a look ahead. MIT publication,
Cambridge, MA
Guilhoto, J.J.M. and Sesso Filho, U.A. (2005). Estimação da matriz insumoproduto a partir de dados preliminares das contas nacionais. Economia
Aplicada, 9 (1), 1-23.
Haguiuda, C. and Veneziani, R. (2006) Eficiência energética no saneamento
básico [Energy efficiency in basic sanitation]. Paper presented at the 3º
Congresso Brasileiro de Eficiência Energética e Cogeração de Energia da
ABESCO,
São
Paulo,
Brazil.
[Online]
Available
from:
http://www.metodoeventos.com.br/3eficienciaenergetica/downloads/2905/tar
de/celso_haguiuda.pdf [Accessed 13 October 2009]
Hall, C.A.S. Cleveland, C.J. and Kaufmann, R. (1986). Energy and Resource
Quality. John Wiley & Sons, New York
Herendeen, R.A. (1974). Affluence and energy demand. Mechanical Engineering,
96 (10), 18-22.
IBGE. (2005a). Matriz de Insumo-Produto. Instituto Brasileiro de Geografia e
Estatística.
[Online]
Available
from:
http://www.ibge.gov.br/home/estatistica/economia/matrizinsumo_produto/def
ault.shtm [Accessed 14 May 2010]
IBGE. (2005b). Produção Agrícola Municipal. Instituto Brasileiro de Geografia e
Estatística. [Online] Available from: http://www.sidra.ibge.gov.br/ [Accessed
16 May 2010]
IBGE. (2005c). Tabelas de Recursos e Usos de Bens e Serviços. Instituto
Brasileiro de Geografia e Estatística, Brasilia.
51
IBGE. (2006). Censo Agropecuário. Instituto Brasileiro de Geografia e Estatística.
[Online] Available from: http://www.sidra.ibge.gov.br/ [Accessed 3 March
2010]
Iorgulescu, R.I. and Polimeni, J.M. (2009). A multi-scale integrated analysis of
the energy use in Romania, Bulgaria, Poland and Hungary. Energy, 34 (3),
341-347.
Janssen, R. and Rutz, D.D. (2011). Sustainability of biofuels in Latin America:
risks and opportunities. Enery Policy, in press.
Kamahara, H. Hasanudin, U. Widiyanto, A. Tachibana, R. Atsuta, Y. Goto, N.
Daimon, H. and Fujie, K. (2010). Improvement potential for net energy
balance of biodiesel derived from palm oil: A case study from Indonesian
practice. Biomass & Bioenergy, 34 (12), 1818-1824.
Kenny, R. Law, C. and Pearce, J.M. (2010). Towards real energy economics:
Energy policy driven by life-cycle carbon emission. Energy Policy, 38 (4),
1969-1978.
Kim, S. and Dale, B.E. (2002). Allocation procedure in ethanol production system
from corn grain. International Journal of Life Cycle Assessment, 7 (4), 237243.
Kim, S. and Dale, B.E. (2004). Cumulative energy and global warming impact
from the production of biomass for biobased products. Journa of Industrial
Ecology, 7 (3-4), 147-162.
Klink, C. and Machado, R.B. (2005). Conservation of the Brazilian Cerrado.
Conservation Biology, 19 (3), 707-713.
Lapola, D.M. Schaldacha, R. Alcamoa, J. Bondeaud, A. Kocha, J. Koelkinga, C.
and Priess, J.A. (2010). Indirect land-use changes can overcome carbon
savings from biofuels in Brazil. Proceedings of the National Academy of
Sciences, 107 (8), 3388-3393
Lorenz, D. and Morris, D. (1995). How much energy does it take to make a gallon
of ethanol? . Institute for Local-Self Reliance. [Online] Available from:
http://www.carbohydrateeconomy.org/ [Accessed 12 November 2010]
Lotka, A.J. (1922). Natural selection as a physical principle. Proceedings of the
National Academy of Sciences, 8, 151-154.
Macedo, I.C. Seabra, J.E.A. and Silva, J.E.A.R. (2008). Green house gases
emissions in the production and use of ethanol from sugarcane in Brazil: The
2005/2006 averages and a prediction for 2020. Biomass & Bioenergy, 32 (7).
Mazzetto Silva, C.E. (2009). Ordenamento Territorial no Cerrado brasileiro: da
fronteira monocultora a modelos baseados na sociobiodiversidade.
Desenvolvimento e Meio Ambiente, 19, 89-109.
Miller, R.E. and Blair, P.D. (1985). Input-Output Analysis: Fundations and
Extenstions. Prentice-Hall, Englewood Cliffs
Ministério da Ciência e Tecnologia - MCT. (2009). Inventario brasileiro das
emissões e remoções antrópicas de gases de efeito estufa. Informações gerais
52
e valores preliminares. [Online] Available from: http://www.oc.org.br/
[Accessed 30 November 2010]
Ministério de Minas e Energia - MME. (2010). Balanço energético nacional.
Ministério de Minas e Energia. [Online] Available from:
https://ben.epe.gov.br/ [Accessed 24 November 2010]
MME. (2010). Balanço energético nacional. Ministério de Minas e Energia.
[Online] Available from: https://ben.epe.gov.br/ [Accessed 24 November
2010]
Myers, N. Mittermeier, R.A. Mittermeier, C.G. da Fonseca, G.A.B. and Kent, J.
(2000). Biodiversity hotspots for conservation priorities. Nature, 403, 853858.
Odum, W.E. Odum, E.P. and Odum, H.T. (1995). Nature's pulsing paradigm.
Estuaries, 18, 547-555.
Pearce, J.M. (2009). Optimizing greenhouse gas mitigation strategies to suppress
energy cannibalism. 2nd Climate Change Technology Conference. Hamilton,
Ontario.
[Online]
Available
from:
http://me.queensu.ca/people/pearce/publications/documents/asp3.pdf
[Accessed 12 December 2010]
Ramos-Martín, J. (2001). Historical analysis of energy intensity of Spain: from a
conventional view to an integrated assessment. Population and Environment:
Ramos-Martín, J. Cañellas-Boltà, S. Giampietro, M. and Gamboa, G. (2009).
Catalonia’s energy metabolism: Using the MuSIASEM approach at different
scales. Energy Policy, 37 (11), 4658-4671.
Ramos-Martín, J. Giampietro, M. and Mayumi, K. (2007). On China's exosomatic
energy metabolism: an application of multi-scale integrated analysis of
societal metabolism (MSIASM). Ecological Economics, 63 (1), 174-191.
Ratter, J.A.J. Ribeiro, J.F. and Bridgewater, S. (1997). The Brazilian Cerrado
vegetation and threats to its biodiversity. Annals of Botany, 80, 223-230.
Roberts, D. (1994). A modified Leontief model for analysing the impact of milk
quotas on the wider economy. Journal of Agricultural Economics, 45 (1), 90101.
Roessing, A.C. and Lazzarotto, J.J. (2004). Criação de Empregos pelo Complexo
Agroindustrial da Soja. ISSN 1516-781X - 233. Embrapa. [Online] Available
from: www.cnpso.embrapa.br [Accessed 27 February 2009]
Sawyer, D. (2008). Climate change, biofuelsand eco-social impacts in the
Brazilian Amazon and Cerrado. Philosophical Transaction of the Royal
Society, 363 (1498), 1747-1752.
Scholz, V. Berg, W. and Kafulß, P. (1998). Energy balance of solid biofuels.
Journal of Agricultural Engineering Research, 71 (3), 263-272.
Searchinger, T. Heimlich, R. Houghton, R.A. Dong, F. Elobeid, A. Fabiosa, J.
Tokgoz, S. Hayes, D. and Yu, T.-H. (2008). Use of U.S. croplands for
53
biofuels increases greenhouse gases through emissions from land-use change.
Science, 340, 1238-1240.
Shapouri, H. Duffield, J. McAloon, A. and Wang, M. (2004). The 2001 Net
Energy Balance of Corn Ethanol. [Online] Available from: http://ncga.com/
[Accessed 10th May, 2009]
Sheehan, J. Camobreco, V. Duffield, J. Graboski, M. and Shapouri, H. (1998).
Life cycle inventory of biodiesel and petroleum diesel for use in an urban bus.
No. SR-580-24089. National Renewable Energy Laboratory. [Online]
Available from: http://www.nrel.gov/ [Accessed 13 March 2009]
Smeets, E.M.W. Faaij, A.P.C. Lewandowski, I.M. and Turkenburg, W.C. (2007).
A bottom-up assessment and review of global bio-energy potentials to 2050.
Progress in Energy and Combustion Science, 33 (1), 56-106.
Ulgiati, S. Russi, D. and Raugei, M. (2008). Biofuel production in Italy and
Europe: benefits and costs, in the light of the present European Union biofuel
policy. In Pimentel, D. (Ed.) Biofuels, Solar and Wind as Renewable Energy
Systems: Benefits and Risks. Springer, New York, Pp 465-491
Valadkhani, A. and Mitchell, W.F. (2002). Assessing the impact of changes in
petroleum prices on inflation and household expenditures in Australia. The
Australian Economic Review, 35 (2), 122-132.
Wang, M. Saricks, C. and Santini, D. (1999). Effects of Fuel Ethanol Use on FuelCycle Energy and Greenhouse Gas Emissions. Center for Transportation and
Research, Argonne National Laboratory. [Online] Available from:
www.transportation.anl.gov/pdfs/TA/58.pdf [Accessed 11th March, 2007]
West, T.O. and Marland, G. (2002). A synthesis of carbon sequestration, carbon
emissions, and net carbon flux in agriculture: comparing tillage practices in
the United States. Agriculture Ecosystems & Environment, 91 (1-3), 217-232.
Wicke, B. Smeets, E. Tabeau, A. Hilbert, J. and Faaij, A. (2009). Macroeconomic
impacts of bioenergy production on surplus agricultural land — A case study
of Argentina. Renewable and Sustainable Energy Reviews, 13 (9), 2463–2473.
54
Appendix A3.1 – Additional results
Table A.3.1: Results
Soybean
Agriculture
Paid work
(n-3)
(n-2)
(n-1)
Scenario B: Baseline conditions
ETi (MJ)
7.33E+10
3.50E+11
7.38E+12
HAi (Hours)
6.74E+08
3.61E+10
1.73E+11
AVi (106 R$)
17,384
105,069
1,841,282
Scenario B: Non renewable biodiesel
ETi (MJ)
1.24E+11
4.01E+11
7.66E+12
HAi (Hours)
1.14E+09
3.66E+10
1.74E+11
AVi (106 R$)
29,239
117,211
1,862,783
Scenario B: Fully renewable biodiesel
ETi (MJ)
6.09E+11
8.95E+11
1.89E+13
HAi (Hours)
5.59E+09
4.22E+10
1.84E+11
AVi (106 R$)
144,051
234,646
2,084,187
Scenario C & D: Baseline conditions
ETi (MJ)
7.33E+10
3.50E+11
7.38E+12
HAi (Hours)
6.74E+08
3.61E+10
1.73E+11
AVi (106 R$)
17,398
105,163
1,852,199
Scenario C: Non renewable biodiesel
ETi (MJ)
1.24E+11
4.01E+11
7.66E+12
HAi (Hours)
1.14E+09
3.66E+10
1.74E+11
AVi (106 R$)
29,230
116,383
1,855,370
Scenario C: Fully renewable biodiesel
ETi (MJ)
6.09E+11
8.95E+11
1.89E+13
HAi (Hours)
5.61E+09
4.23E+10
18.84E+11
AVi (106 R$)
143,857
233,712
2,079,291
Scenario D: Non renewable biodiesel
ETi (MJ)
1.24E+11
4.01E+11
7.66E+12
HAi (Hours)
1.14E+09
3.66E+10
1.74E+11
AVi (106 R$)
29,226
116,331
1,853,678
Scenario D: Fully renewable biodiesel
ETi (MJ)
6.09E+11
8.95E+11
1.89E+13
HAi (Hours)
5.61E+09
4.23E+10
1.84E+11
AVi (106 R$)
143,846
233,652
2,077,592
Whole society
(n)
9.15E+12
1.61E+12
1,841,282
9.43E+12
1.61E+12
1,862,783
2.07E+13
1.61E+12
2,084,187
9.15E+12
1.61E+12
1,852,199
9.43E+12
1.61E+12
1,855,370
2.07E+13
1.61E+12
2,079,291
9.43E+12
1.61E+12
1,853,678
2.07E+13
1.61E+12
2,077,592
55
4
Social-Multi Criteria Evaluation of
Alternative Geothermal Power
Scenarios: The case of Mt. Amiata in
Italy
Abstract
Italy was the first country in the world to exploit geothermal resources for the
production of electricity. In Europe it is still the first country in terms of installed
capacity. Currently, the only region in Italy with geothermal power plants is
Tuscany. This study focuses on Mt. Amiata, one of the two geothermal areas in
Tuscany, where there is strong opposition to the exploitation of geothermal
resources. The context is characterized by contested scientific results regarding
crucial issues such as the impact of geothermal exploitation, the conservation of
water resources and human health. A social multi-criteria evaluation is proposed
to explore the different legitimate perspectives of the actors involved. Scenarios
are distinguished in terms of their installed capacity, technology and plant site. A
Condorcet consistent aggregation algorithm is applied and results are analyzed
using a sensitivity analysis. The alternative scenarios are evaluated in a multidimensional way by attaching different weights to the criteria reflecting divergent
points of view.
Keywords: geothermal power, multi-criteria analysis, integrated assessment,
conflict analysis, sensitivity analysis, Italy
Chapter 4 – Social-multi criteria evaluation of alternative geothermal power
scenarios: The case of Mt. Amiata in Italy
4.1 Introduction
This paper intends to show the potential use of a social multi-criteria evaluation
(SMCE) in managing problems related with conflicts arising around geothermal
power. Specifically, it explores the case of Mt. Amiata, in the region where
geothermal power originated: Tuscany.
The first experiments to use geothermal energy to produce electricity took
place in Tuscany in 1904 in Larderello. Since then Italy has remained the first
producer of electricity from geothermal sources in Europe and is the fifth
internationally after the USA, Philippines, Indonesia, and Mexico (Bertani and
Fridleifsson, 2010). At the moment of writing all the geothermal power plants in
Italy are located in Tuscany. Here geothermal power made up 24% in 2009 of
electricity consumption (and 32% of net production), while nationally the
contribution of geothermal power to electricity consumption is just 1.6% (Terna,
2010). Currently there are 35 power plants with 882,5 MW of installed capacity18
(ARPAT, 2010; ENEL, 2010). In Tuscany the geothermal power plants are
located in two areas: the so-called traditional area around Larderello where 30
plants (and 794,5 MW of installed capacity) are located, and Mt. Amiata area (in
the south of Tuscany) where five plants (with 88MW) have been installed (see Fig.
4.1). It is in the Mt. Amiata area where geothermal energy has been facing strong
opposition during the last few years.
Fig. 4.1: Geothermal power plants in Italy
18
Corresponding to 770 MW of net capacity.
57
Opposition to renewable energies is not uncommon and it is often considered as a
NIMBY (not in my back yard) attitude. The geothermal power industry therefore
tends to classify such behavior as a social acceptability problem (Cataldi, 2001;
De Jesus, 1997). However more than simple social acceptability, opposition
should be considered as being part of a more general environmental management
problem which presents elements of energy policy, economic considerations, local
pollution, water conservation concerns, employment effects, quality of life and
aesthetical aspects. This kind of environmental management problem reflects
conflicts of interests and values. In the presence of plural values,
incommensurability is the norm and not the exception. Incommensurability refers
to the absence of a common unit of measurement to evaluate alternatives
(Martinez-Alier et al., 1998). This is because simply deciding what to measure
implies value conflicts. However incommensurability does not imply that rational
comparability is impossible. On the contrary, with value-pluralism, alternatives
can be “weakly comparable”, without resorting to a single value (and to a single
unit of measurement). Simon (1976) distinguishes between substantial rationality
and procedural rationality. The former refers to the rationality of the result
irrespectively of the way in which decisions are taken, while the latter refers to the
rationality of the decision-making process itself. In deciding between weakly
comparable alternatives, procedural rationality must substitute for substantial
rationality (Martinez-Alier et al., 1998).
Where environmental management is characterized by conflicts in values
and interests, it is very difficult to arrive at a straightforward and unambiguous
solution. This implies that planning processes should be characterized by the
search for acceptable compromise solutions through an adequate evaluation
methodology (Munda et al., 1994).
Multi-criteria decision aid has proven to be a powerful tool to deal with
complex environmental and energy management problems. Several examples can
be found in Gamboa and Munda (2007), Diakoulaki et al. (2005), Barda et al.
(1990), Georgopoulou et al. (1997), Cavallaro and Ciraolo (2005), Afgan and
Carvalho (2002), Goumas and Lygerou (2000), Beccali et al. (2003),
Haralambopoulos and Polatidis (2003), Kowalski et al. (2009), Paruccini (1994),
and Janssen (1992). The objective of multi-criteria aid is not to discover some
particular truth or an optimizing solution, but rather the final result should be seen
as a creation (and not a discovery) aimed at facilitating “an actor taking part in a
decision process to shape, and/or argue and/or transform his preferences, or to
make decision in conformity with his goals” (Roy, 1990 p. 328).
From a practical point of view, one of the main advantages of multi-criteria
decision aid is that it makes it possible to handle great amounts of data in a multidimensional way. It is a very transparent method because different valuations are
not translated into a single numeraire (e.g. US$ or energy or exergy). Using data
from scientific dimensions in their original units of measurement, it is also
suitable for interdisciplinary approaches (Munda, 2008).
58
Of course, multi-criteria analysis cannot solve all conflicts but it can help
decision making by shedding light on the nature of the conflict and on the way to
find compromises, thus increasing the transparency of the decision-making
process (Martinez-Alier et al., 1998). The most common use of multi-criteria
analysis is in providing a final ranking of alternatives based on different criteria.
In order to address possible quality problems with data, a sensitivity analysis is
often added. This work proposes a very different approach. A sensitivity analysis
is included here mainly to give political weights to the different criteria reflecting
diverging perspectives. The final rankings thus represent “politically sensitivity
maps”, to use Stirling’s (1999) words.
The next section describes the methodological framework. Section 4.3
provides a historical-institutional analysis of the context of this study and includes
a brief summary of the main social actors involved. Section 4.4 introduces the
chosen alternatives and explains which criteria were used and how they were
estimated. The results are included in section 4.5. The last section presents some
final remarks on the overall process and on the specific results.
4.2 Methodological framework
A multi-criteria problem can typically be described by a finite set A of feasible
alternatives a1, a2, … an (later called scenarios) and a family G of criteria g1, g2,
….gm (representing the different points of view), by which alternatives are
evaluated. Alternative a1 is considered better than alternative a2 if, according to
the gi criterion g( a1 ) > g( a2 ).
Given the set A of alternatives and the set of criteria G, it is possible to build a n x
m matrix whose elements report the performance of each alternative according to
each criterion. Depending on the methodology used, the matrix can include
quantitative, qualitative and also both types of evaluations (Munda, 1995; Munda
et al., 1994).
The multi-criteria exercise can be summarized as follows (adapted from
Gamboa, 2006; Gamboa and Munda, 2007):
•
•
•
•
•
Problem structuring
• Historical-institutional analysis
• Identification of social actors
• Definitions of preferences and aspirations
Identification of alternatives
Identification and estimation of criteria
Selection and application of a ranking algorithm
Analysis of results and sensitivity analysis
59
These phases are not intended to follow a chronological order. Rather, they
influence each other dynamically. Once the results analysis has been performed, a
new cycle can begin because the knowledge acquired may enable the social actors
and analysts to change their perspectives and structure the initial problem in a
different way.
The historical-institutional analysis is mainly aimed at defining the given
problem by identifying social actors and eliciting their preferences and aspirations.
An analysis of the actors cannot be a simple enumeration of the agents
involved. Important aspects to be included are the actors’ main interests and
stakes, the perception of the problem, the degree of influence, and access to
technical knowledge (Funtowicz et al., 1998). This phase of the research
facilitates the generation of alternatives.
The institutional analysis in this research involved a review of various
documents such as laws, policy documents, press releases and newspapers. This
phase made it possible to identify the main actors. Subsequently semi-structured
interviews (SSI) were held with exponents and representatives of the identified
social actors (Appendix A4.1 reports a list of the interviews held). A question
guide was previously prepared based on the information collected during the
secondary data review. The main objective of the interviews was to gain
knowledge on the perceptions, needs and aspirations of the social actors identified.
In addition, following a snowball methodology, the interviews made it possible to
identify other social actors.
As Roy (1985) specifies, the preference model used to evaluate the
alternatives is not based on the alternatives themselves but on their consequences,
which result from the alternatives and from the subjective evaluations of the social
actors. The consequences are evaluated using certain criteria.
The choice of criteria is a technical translation of the social actors’ desires
and needs (Gamboa and Munda, 2007). Essentially, criteria represent the different
points of view of the social actors, i.e. the axes along which the social actors argue,
transform and justify their preferences. The comparisons obtained through these
criteria should be considered as partial preferences because they are limited to the
aspects taken into account by the point of view represented by the definition of
each criterion (Bouyssou, 1990).
According to the multi-criteria problem reported above, in order to state that
j is preferred to k (with j and k belonging to the set of N feasible alternatives), it is
sufficient that gi ( j ) > gi ( k ). This preference description represents a “true
criterion”. In this case, any difference between gi ( j ) and gi ( k ) implies a strict
preference relation. However, even when the decision maker is a real person, their
preferences are seldom clearly stated. Among areas of firm conviction may lie
nebulous zones of uncertainty. Moreover, the data used to evaluate the
performance of each alternative may be imprecise (Roy, 1990). This is why the
introduction of discrimination thresholds is advisable. Here an indifference
threshold is used, as depicted in Eq. 4.1 and 4.2.
60
j P k ⇔ g m ( j ) > g m (k ) + q
j I k ⇔ | g m ( j ) − g m (k ) | ≤ q
(Eq. 4.1)
(Eq. 4.2)
where P represents a preference relation, I an indifference relation and q is
the indifference threshold, i.e. the greatest value of the difference between the
alternatives j and k which is not large enough to differentiate j from k on criterion
gm (Roy et al., 1986). The type of model in Eq. 4.1 and 4.2 is called “quasicriterion”.
From the chosen preference relations, a multi-criteria algorithm must be
applied in order to derive the aggregate result. Given the context of this study, one
important characteristic is that the result should not be an isolated alternative but a
ranking 19 . Thus, if the first alternative cannot be chosen because of political
reasons (e.g. it gives rise to a strong conflict), other alternatives can be considered
in their ranked order. In addition, it is important that the algorithm be noncompensatory so that a very good performance in one criterion cannot compensate
for a bad one in an environmental criterion or vice versa. It is also advisable that
the intensity of the preference information is not accounted for in order to avoid
compensability. Weights should reflect importance coefficients and not tradeoffs 20 (Munda, 2004; Vincke, 1992). Finally, it is essential that algorithm be
simple and transparent. The Condorcet consistent rule developed by Munda
(2005; 2009) has such properties. This is based on the maximum likelihood
concept, that is, the maximum likelihood ranking supported by the maximum
number of criteria for each pair-wise comparison, summed over pairs of
alternatives. An N x N outranking matrix E can be built respecting the axioms of
diversity (a complete order of alternatives can be obtained for each criterion),
symmetry (only ordinal pair-wise information is accounted for, so intensity of
preference is disregarded), and positive responsiveness (the degree of preference
between alternative j and k is a strictly increasing function of the number of
criteria and weights, which ranks j before k).
Any element of E:ej,k (j ≠ k) is obtained by a pair-wise comparison between
alternative j and k according to all M criteria. This pair-wise comparison is
obtained by applying Eq. 4.3.
1


e jk = ∑  w m ( Pjk ) + wm ( I jk ) 
2

m =1 
M
(Eq. 4.3)
where wm is the weight for criterion m.
19
Also called γ problem (Roy, 1990)
Weights as trade-offs indicate how much a good performance in one criterion can compensate
for a bad one in another (the analogy in economic jargon is the marginal rate of substitution).
Weights as importance coefficients indicate how important a criterion is, but no compensation is
implied.
20
61
Let T be the set of all N! possible complete rankings of alternatives, and τs
each individual ranking belonging to T. The score φs of each τs is obtained by the
N 
summation of ejk over all the   pairs j, k of alternatives (i.e. ϕ s = ∑ e jk ,
2 
where j ≠ k, s = 1,2,…N! and ejk ∈ τs).
The final ranking τ* is the one which maximizes φs (see Eq. 4.4):
τ ∗ ⇔ ϕ * = max ∑ e jk
(Eq. 4.4)
where ejk ∈ T.
4.3 Historical-institutional analysis
4.3.1 Historical context
Until the beginning of 1900 Mt. Amiata was a typical mountain area of volcanic
origin where the main activities included agriculture, forestry and animal
production, after which the mining for cinnabar radically changed the economic
profile of the area. The mining sector grew so much that in 1965 it satisfied 35%
of the world’s mercury demand. Subsequently, a fast decline took place until 1976
when mines were closed down with hundreds of redundancies (Serafini and Sani,
2007).
Geothermal explorations started at the end of the 1950s with the installation
of the first small plants in the municipality of Piancastagnaio to the east of the
mountain and, in Bagnore (belonging to the municipality of Santa Fiora) in the
west. Geothermal activity was soon perceived as an opportunity to counteract the
economic depression caused by the end of the cinnabar mining. Government
policies were set up to create new jobs. These included ornamental plant
production in greenhouses benefiting from geothermal heat. These greenhouse
areas were set up near Piancastagnaio, in an area named Casa del Corto.
ENEL, the once state-owned company operating and installing geothermal
plants, was privatized at the beginning of the 1990s. During the same period a
new plan called “Geotermia 2000” was launched to install 200 additional MW of
capacity in Mt. Amiata (Bertini et al., 1995). It was around this plan that
opposition to geothermal exploitation originated. At that time, the visual impact of
the new plants was the main concern. In any case, three new plants were installed
in Piancastagnaio and one in Bagnore (20MW each). Compared to the plants
previously installed these new plants entailed the decoupling of installed capacity
from on-site employment. This is because all plants are remotely controlled at a
62
center a long distance away. As a consequence, the reduced employment effects
of geothermal power plants became another reason for discontent.
The municipality of Piancastagnaio asked various experts to contribute to
the publication of a new book (1994) on the effects of geothermal exploitation.
Some of the articles gave cause for concern among a small minority of the
population. In the meantime the first reports on air quality released by the
authority in charge of environmental control (called ARPAT) revealed that the
emissions of the individual plants were much higher in Mt. Amiata than in the socalled traditional geothermal areas (further north, around Larderello).
In September 2000 two explosions occurred near Piancastagnaio because of
geothermal fluid escaping from the soil. The inhabitants were evacuated and all
the farm animals in the area died. These events gave rise to a new surge of
opposition which included local rallies.
Another important element which caused concern among the population was
the arsenic concentration in drinking water. In 2001 a decree set the limit of
arsenic concentration at 10 µg/l. Since then, authorities have permitted
expectation to the law regarding drinkable water. Given the high quantity of
arsenic in the water, limits were often raised to 20 or 30 µg/l. The problem was
solved in 2009-2010 with the installation of arsenic abatement plants for drinkable
water. However opponents to geothermal exploitation suspect that the high
arsenic concentration may in some way be linked to the presence of the power
plants.
At a regional level (i.e. within Tuscany in general), the growth of
geothermal power plants was almost nil during the 2000s. In designing its new
energy policy the regional government decided that the abundance of geothermal
resources was an opportunity not to be wasted for the development of the
renewable energies sector. In the energy plan approved in 2008 the installation of
200 MW21 were planned. The regional government of Tuscany spearheaded an
important negotiation with ENEL to set up a new compensation fund for the
geographical areas where the geothermal plants are located. This gave rise to the
so-called “general agreement on the exploitation of geothermal resources” signed
in 2007 by all the municipalities of geothermal areas in Tuscany apart from the
one in Mt. Amiata (i.e. Abbadia San Salvatore).
In addition to the final compensation, the general agreement included the
acquisition of EMAS certification for all power plants, the commitment of ENEL
to endorse new agreements with unions and industrial associations to enhance
local employment and entrepreneurship, and the promotion of scientific studies
and research on the impact of geothermal exploitation.
21
The plan was for the whole Tuscan area (and does not specify where 200 MW are to be
installed). So the 200MW target is not a target just for the Mt. Amiata area.
63
4.3.2 The scientific debate
The scientific debate is mainly around two crucial issues: the effects of
geothermal exploitation on water conservation and the effects on human health.
The main question on water regards the effects of geothermal exploitation
on the quantitative and qualitative conservation of the potable aquifer. In order to
introduce the reader into the geological context of this case study it is necessary to
briefly describe the geothermal field of the Amiata volcanic complex (Southern
Tuscany). There are two distinct water-dominated reservoirs: a so called shallow
reservoir and a deep reservoir. The shallow reservoir is sited in the Mesozoic
carbonatic formations at 500-1000 m depth. The deep one is hosted in the
Paleozoic metamorphic basement at 2500-4000 m depth. These two reservoirs are
separated by a low permeable layer and are considered part of a unique
geothermal system 22 (Barelli et al., 2010). The shallow geothermal reservoir is
overlain by cap rocks namely “Liguridi”. Above them is located a layer of
volcanic rocks or “Volcanites” which host the potable aquifer.
The effect of the geothermal exploitation on the conservation of the potable
aquifer depends on various elements which, due to their complexity, are highly
debated among the scientific community. It is worth remembering that the springs
in Mt. Amiata are characterized by water shortages.
The first cartographic reconstruction of the potable aquifer can be found in a
study by Calamai et al. (1970) who identified its piezometric level at 950 m.a.s.l.
A geophysical survey carried out in 2003-2006 by the Italian National Research
Council (CNR) identified an important depression in the phreatic aquifer 23
(Manzella, 2006). Finally the recent piezometer installed by the regional
government revealed that the water table is at 780 m.a.s.l (thus suggesting a
reduction of 170 m compared with the level identified in Calamai et al.).
This debate basically has two main positions. One position claims that the
potable aquifer and the geothermal system are not connected; water shortages are
mainly due to a reduction in rainfall, to the continuous drawings of water for
drinking purposes from wells (many of which are illegal) and tunnels connected to
waterworks, to the general crumbling conditions of the local waterworks and to
the presence of the tunnels of the old mine. According to this position, the original
reconstruction of the piezometric level of the phreatic aquifer of Calamai et al. is
probably subject to errors due to the techniques used, to the few measurements
taken and to the interpretations of the results. The depressions identified by the
CNR study are also subject to the approximation typical of the technique used. In
22
Other characteristics of the geothermal reservoirs are: the shallow reservoir presents temperature
ranging from 150 (in Bagnore) to 200°C while the temperature of the deep reservoir are more
homogeneous and generally greater than 300°C (Bertini et al., 1995). In the deep reservoir the non
condensable gas content is between 4% and 15% (Bertini et al., 1995) while in the shallow one is
much higher: around 40% in Piancastagnaio and 85% in Bagnore (Barelli et al., 2010).
23
The magnetotelluric method was applied for carrying out the geophysical survey.
64
addition, the presence of contaminants in the water could be due to the natural
presence of the same substances in the area and to the now closed mining activity.
Different aspects of this view can be found in the EIA reports submitted by ENEL
(2005; 2008; 2009a; 2009b; 2009c), in scientific articles by ENEL personnel, and
in the study commissioned at the University of Siena (2008) by the Tuscany
regional government. These studies find confirmation of their arguments in the
prior works of Focacci et al. (1993), Barazzuoli et al. (2004) and Papalini (1989),
among others.
The other position argues that the exploitation of geothermal power has
provoked a depression in the geothermal reservoir and this depression has drawn
water from the phreatic aquifer thus reducing the water table. The depression
identified by the CNR study consequently indicates a recharge of the geothermal
reservoir by the potable aquifer. Since the phreatic aquifer reduces its weight, the
pressure that the water table causes on gasses coming from below also diminishes,
consequently the ascent of contaminants from the geothermal reservoir is
facilitated. In addition, the reduction in springs causes an increase in the
concentration of poisoning contaminants. In summary geothermal exploitation can
negatively impact the conservation of the potable aquifer. Different aspects of this
view are held by Borgia (2007), by a study commissioned at EDRA by the
regional government (EDRA, 2006a; 2006b), and by some geologists from the
offices in charge of land protection and the prevention of hydraulic and
hydrogeological risks of the regional government. The upholders of this position
find confirmation of their arguments in older studies conducted by ENEL
personnel with ENEL data (Burgassi et al., 1965; Calamai et al., 1970; Cataldi,
1965).
Just to give an idea of how the scientific debate is polarized, Borgia (2007)
found a clear correlation between the vapor extracted for geothermal use and a
reduction in the Mt. Amiata spring flows. However this correlation is completely
negated in the study by the University of Siena24. Moreover, the legitimacy of the
University of Siena study is contested by residence committees opposing
geothermal exploitation because the study was conducted by a research team
which included a member who was appointed by ENEL as an expert in previous
civil suits.
The other main issue is the effect of geothermal exploitation on human
health. A specific statistical-epidemiologic study (ARS, 2010) was conducted by
comparing mortality and hospitalization statistics of the population in the
geothermal areas with that of nearby and similar areas25. The results showed that
considering the whole set of geothermal areas (i.e. including also the so-called
traditional geothermal area), there was a small excess of mortality among males
(+6%) with respect to the expected value and no excess of hospitalization.
24
The input data in the two studies was different.
The analysis covered 2000-2006 for mortality statistics and 2002-2004 for hospitalization
statistics.
25
65
However, considering only the Mt. Amiata area, among males there was a
significant excess of mortality (+13%), an excess of cancer (+19%), and an excess
of mortality for breathing apparatus illnesses. While females presented an excess
of mortality for acute breathing illnesses. Regarding hospitalization there were
some excesses due to stomach cancer, breathing illnesses (only for females) and
kidney failure. However, the study concluded that in all likelihood the excess of
mortality and hospitalization was not due to the presence of geothermal plants
because the most worrying indicators referred only to males26 (and not to females
who are exposed to the geothermal presence in the same way as males).
According to the study, the excesses revealed were probably due to lifestyle and
past employment, mainly mining. In spite of the reassuring conclusions, the
results of the epidemiologic study remain a cause of concern and the regional
government has recently agreed to finance further investigations.
4.3.3 Current status
Two mining concessions have been awarded to ENEL: one is to the east of Mt.
Amiata (where Piancastagnaio and Abbadia S. Salvatore are located) and one is to
the west (where Santa Fiora and Arcidosso are located)27.
Four plants are currently operating in the east, all in the Piancastagnaio area. The
characteristics of each plant are reported below with their official name (data are
from the Environmental Impact Assessment submitted by ENEL, 2008; 2009b;
2009c):
•
•
PC2. This is the oldest plant and currently has 8 MW of installed capacity.
It was installed in 1969 and it is fuelled only by the shallow reservoir
(which presents a very high quantity of non condensable gases with all
their harmful elements). It is a dry steam power plant with no re-injection
of fluid and no filters for air emission. This plant is responsible for the vast
majority of the geothermal emissions due to geothermal exploitation of the
area. The plant provides heat to the nursery activities of a nearby area
called Casa del Corto, where greenhouses are located. These nurseries
employ around 250 people.
PC3. It has 20MW of installed capacity and was set up in 1990. It is
fuelled only by the deep reservoir. It has a flesh steam technology28, which
26
According to the study, the excess in breathing illnesses among females were consistent with
regional trends.
27
The mining concession of the West known as Bagnore, has an extension of 45.87 Km2 and all
the municipalities involved here belong to the Province of Grosseto. The mining concession of the
east side is called Piancastagnaio, it extends over 47.91 Km2 and all the municipalities belong to
the province of Siena.
28
A description of the technologies available for geothermal power plans can be found in Kagel et
al. (2007), DiPippo (1991; 2005), and Bacci (1998) among others.
66
•
partially re-injects the extracted geothermal fluid and is endowed with
filters for the abatement of hydrogen sulphide (H2S) and mercury (Hg)
emissions (the filter is called AMIS). It is located in the south of
Piancastagnaio.
PC4 and PC5. These two plants are located near to each other in the north
of Piancastagnaio. Each of the two plants has 20MW of installed capacity.
PC4 was set up in 1991, and PC5 in 1996. The two plants are fuelled only
by the deep reservoir. Their operation capacity is slightly lower than the
theoretical capacity because of a lack of geothermal fluid. In order to
operate at full capacity new wells need to be drilled. Without new wells
the two plants reduce their power capacity every year. The two plants
exploit flash steam technology with partial re-injection of the geothermal
fluid and both are now equipped with the H2S and Hg abatement filters.
A so-called “re-organization plan” for Piancasatgnaio mining license was
submitted by ENEL and authorized by the regional government. This plan
involves interventions only in Piancastagnaio. The main characteristics are: PC2
would be closed down, the heat that PC2 was providing to Casa del Corto would
be provided by a new heat pipe connected to PC3, another heat pipe would be
installed to provide the citizens of Piancastagnaio with heat, various wells would
be drilled29 to increase the production of existing power plants, and more than
16.3 km of various steam pipelines would be installed in order to connect the new
wells with the power plants and to make the three plants part of a single system of
steam distribution (details are from ENEL, 2008).
In the west of Mt. Amiata there is only one operating plant named Bagnore
3 (BG3 for short) from the name of the concession and the locality where the
plant is located. It has 20MW of installed capacity with flash steam technology,
partial re-injection of the extracted fluid and H2S and Hg abatement filters. As
mentioned the plant is located within the municipality of Santa Fiora.
A new project for the construction of a new 40 MW power plant (named
BG4) in the west of Mt. Amiata was submitted by ENEL. Currently, the regional
government has not yet authorized this new project.
4.3.4 Social actors
There are many stakeholders involved in the policy arena of this case and
deciding which ones to include inevitably presents some degree of arbitrariness.
The total list could include research organizations (the University of Siena, the
29
Five new production wells would be drilled, two old wells would be reactivated, one old well
would be deepened (all of them would be about 3,500 meters deep, thus reaching the deep
reservoir). In addition, one new re-injection well of about 1,000 meters (up to the shallow
reservoir) would be drilled (ENEL, 2005; 2009a).
67
University of Florence, a technical expert committee set up by the Regional
government, and the National Research Council), the local association of hotels
and the local association of service providers, environmental NGOs with a minor
presence (e.g. Legambiente or Amici della Terra), a lawyers’ NGO which assisted
the resident associations during various legal actions, and also a Buddhist
organization which attracts several followers in the west of Mt. Amiata. However,
the social actors believed to have been the most active in recent years and/or
which present a clear stake in the geothermal exploitation of Mt. Amiata are
reported in Table 4.1.
68
Table 4.1: social actors
Social actor
Tuscany
Regional
government
Type
Regional
government
ENEL
Private
company
Piancastagnaio
municipality
Local
authority
Abbadia S.
Salvatore
municipality
Local
authority
Santa Fiora
municipality
Local
authority
Arcidosso
municipality
Local
authority
Description - Position
The regional government has taken over the 20-20-20 EU objectives.
According to the most recent energy plan, the Region should cover
39% of the electricity consumption (and 10% of thermal energy) with
renewable energy sources by 2020 (Tuscany regional government,
2008). The additional amount of electricity that will have to be
produced by all renewable energy sources is planned to be 3,542
GWh, of which 1,600 GWh by geothermal power.
These objectives show the essential role that geothermal power is
expected to have in order to achieve the desired targets. In addition,
the regional government is in charge of authorizing the construction
and operation of geothermal power plants.
This is the sole company currently operating geothermal power plants
in Italy (including on Mt. Amiata). Depending on the expected costs
and revenues, it is interested in expanding the geothermal exploitation
to produce more electricity and to be entitled to more green certificates
(or to new incentive schemes).
Four plants are located within its area with a total of 68MW of
installed capacity. The municipality supports the re-organization
proposed by ENEL in Piancastagnaio for several reasons: it involves
the closing of PC2 which is a plant emitting high levels of air
pollution; it entails the construction of a heat pipeline allowing
inhabitants and companies to access low heat costs; it guarantees
maximum capacity of electricity production and consequently the
maximum level of royalties (which, to some extent, depend on the
quantity of electricity produced).
Part of the area is included in the mining concession awarded to ENEL
for the exploitation to the east of Mt. Amiata. The municipality has
never considered geothermal power as a driver of development and it
opposes the construction of any new plants that would exploit high
and medium enthalpy resources. Geothermal exploitation is perceived
to be at odds with the development of the already important tourist
sector. It is the only municipality in the geothermal area which did not
sign the general agreement with ENEL and the Regional government,
thus turning down the funds that this would have involved. In any
case, it supports the re-organization plan because it means closing
down PC2, so less air emissions would affect the municipality.
This is the local authority is on the west side of Mt. Amiata. A 20MW
plant (called BG3) is located in its district. The new 40MW plant
(called BG4) would also be located in the area, if installation is finally
authorized. The municipality supports the presence of BG3 and the
new construction of BG4. The main perceived benefit for the new
plant is the possible development of small companies that could access
low cost sources of heat. Royalties are also considered important. In
fact, the vast majority of royalties are allocated to the municipality
where the new plant is physically located.
Part of the district is included in the mining concession awarded to
ENEL for the exploration of the west side of Mt. Amiata however no
plant is located in its area. Nevertheless, given the prevalent wind
direction, the majority of air emissions from the BG3 (and BG4 if it is
finally constructed) are deposited in its district (and not in the Santa
Fiora area where the plant is located). The municipality is worried that
the construction of BG4 would imply further emissions. It would
tolerate its presence if the technical authorities guaranteed that the
emissions will be maintained within acceptable levels and that the new
plant would not interfere with the aquifer conservation.
69
Social actor
Prospettiva
Comune di
Piancastagnaio
Type
Residents
association
with elected
representatives
in the city
council
WWF
Environmental
NGO
Rete Comitati di
Difesa del
Territorio
Regional
network of
associations
Comitato
per la Tutela
dell’Ambiente
Abbadia S.S
Residents’
association
with elected
representatives
in city council
Rifondazione
Comunista Santa
Fiora
Local branch
of a left-wing
party
Description - Position
They are worried that exploitation of high enthalpy resources may
provoke a geothermal fluid discharge (as has already occurred) and
interfere with the conservation of the aquifer. They oppose the reorganization plan because it involves new wells and new pipelines,
thus more exploitation of high enthalpy resources and a negative
visual impact. They do not consider the closing of PC2 to be a positive
element of the re-organization plan because dismissing PC2 should
have been agreed independently from the plan. They ask for a
moratorium of additional exploitations of the high enthalpy resources.
In the past they submitted a request for further integrations of the
environmental impact assessment of BG4 regarding the effects of the
planned plant on the ecological stability and on the food chains. They
also submitted a formal claim to the European Union regarding the
fact that the mining concession for BG3 was extended without an
environmental impact assessment. They are worried about the
additional emissions that BG4 would provoke and about the possible
detrimental effects on the aquifer.
This is an network of associations committed to the natural and
preservation of the area. It operates on a regional scale. They are
worried that geothermal exploitation may deplete the water table,
contaminate water resources with heavy metals, provoke superficial
discharges of geothermal fluids, and cause dangerous emissions. They
oppose the exploitation of high enthalpy resources. They also oppose
the re-organization plan and the construction of the new plant in
Bagnore.
This is a citizens association from the town of Abbadia San Salvatore.
They fear that geothermal power plant emissions may affect human
health. They are worried about the conservation of the aquifer and
they believe the presence of geothermal power plants does not
stimulate the economic development for the area. They oppose the
exploitation of medium and high enthalpy resources.
It is the local branch of a political leftist party. It represents the
opposition to the development of geothermal power in the small town
of Santa Fiora. Members are mainly worried about the environmental
impact that the construction of the new plant in Bagnore (BG4) would
involve. They oppose the construction of BG4 because it implies a
three times larger capacity (in Santa Fiora area) with the same
technology of BG3.
70
4.4 The multi-criteria matrix
The multi-criteria methodology entails identifying a set of alternatives and a set of
criteria to compare such alternatives.
4.4.1 Generation of alternatives
The scenarios taken into consideration are seven with four overall origins: 1) the
preservation of the status quo 2) the projects planned by ENEL, 3) scenarios
generated after in-depth discussions with technical experts and scientists from the
geothermal sector in order to address (at least partially) the worries of some of the
social actors 4) the formulation in “scenario terms” of the requests of the
opponents to the ENEL projects.
The scenarios considered are:
A. BaU (Business as Usual). This scenario means maintaining the current
conditions. All five plants remain operating as they are. At the same time,
two plants in Piancastagnaio (PC4 and PC5) experience a reduction in
their production capacity because of a lack of geothermal fluid.
B. Reorg (reorganization). This plan is proposed by ENEL and has already
been authorized. The details are reported in section 4.3.3.
C. ClosingPC2. This scenario envisages that PC2 would be closed down and
a new heat pipe would be installed from PC3 in order to provide the Casa
del Corto area with heat (as in the previous case). No other interventions
are envisaged, so the annual electricity production of PC4 and PC5 would
decrease.
D. Reorg+BG4. This scenario joins the two projects proposed by ENEL. In
Piancastagnaio a re-organization is planned exactly as explained in B. To
the west of the mountain a new plant of 40MW capacity (BG4 for short)
would be installed with flash steam technology beside the existing plant
(the total installed capacity in Mt. Amiata would be increased from the
current 88MW to 120 MW). In addition to the installation of a power
system (which in this case involves cooling towers with six cells and the
H2S and Hg abatement filters), the construction of the new plant entails
drilling new wells30 and the installation of about 12 km of steam pipelines.
E. Reorg+40CC. As in the previous case this scenario involves the reorganization plan proposed by ENEL in Piancastagnaio. The new power
plant to be constructed in Bagnore would have a closed cycle technology.
30
Six new wells would be drilled and two old wells would be re-activated (all of which would
reach the deep reservoir and would be used for production purposes). In addition, two new wells
reaching the shallow reservoir would be drilled for re-injection purposes.
71
This means that the fluid extracted from the geothermal reservoir would be
totally re-injected (and not partially as normally happens with traditional
flash steam technology). The only technology presenting this characteristic
and already on the market is binary cycles 31, which is the technology
envisaged here 32 . The construction of this new plant with 40MW of
installed capacity would mean using a wider area and a higher cooling
towers than in the traditional flash stem technology (because of the
different cooling systems), along with a higher number of wells to be
drilled in order to totally re-inject the extracted fluid.
F. ClosingPC2+20CC. This scenario envisages that the PC2 plant would
close down and that a new heat pipe would be installed to provide Casa del
Corto area with heat. In Piancastagnaio no further interventions would be
made. In addition, a new plant with 20MW of installed capacity and closed
cycle would be built in Bagnore. As in the previous case the technology
would be binary cycles. Obviously, the area occupied by this new plant
would be smaller than in the previous case (but much larger than
traditional plants based on flash steam technology).
G. Reorg+20CC. In Piancastagnaio the re-organization plan would take place
exactly as in B. In addition, a new 20MW plant with a binary cycle
technology would be installed in Bagnore.
Each scenario is assumed to have a 30-year duration period.
4.4.2 Choice and estimation of criteria
Eleven criteria were taken into consideration representing the results of the
institutional analysis described in Section 4.3: 1) electricity produced, 2)
profitability of the plants, 3) municipality revenues, 4) direct heat use, 5)
greenhouse gas (GHGs) emissions avoided, 6) H2S emissions, 7) Hg emissions 8)
ammonia (NH3) emissions, 9) arsenic (As) emissions, 10) possible impact on the
phreatic aquifer 11) visual impact. Initially it was considered also direct
employment among the set of criteria. However, it was excluded because the local
31
With binary cycles, the geothermal water heats another liquid. The two liquids are kept
completely separate through the use of a heat exchanger used to transfer the heat energy from the
geothermal water to the working fluid. The secondary fluid vaporizes into gaseous vapor and turns
the turbines that power the generators. With air cooling the geothermal fluids never make contact
with the atmosphere before they are pumped back into the underground geothermal reservoir
(Kagel et al., 2007). ENEL itself installed two binary cycles plants in Nevada amounting to 65Mw
of total capacity and has acquired rights to install 150 MW of additional capacity in different USA
states (Roxborough, 2010)
32
Theoretically a closed cycle can also be obtained with total re-injection and flash steam
technology. In Iceland there are plans to install this type of prototypal plant, but there are no
operating and commercial cases at the moment of writing. Consequently, it was decided not to
consider this possibility in this work.
72
actors never actually mentioned it in the interviews. The plants are controlled
remotely at a control centre a long distance away. In Mt. Amiata there are few
locals employed in maintenance. In addition, the number of local employees
would not be significantly different in the considered scenarios. Of course, during
a plant’s construction, the employment effect can be important. However, this
effect would be limited just to a few years and the majority of employees and
companies contracted for the construction of the plant would not be from Mt.
Amiata. Moreover, the employment effect of the construction phase could only be
estimated with a very high degree of approximation. The potential of new
companies accessing low cost heat sources may have some positive employment
effect. However, such an effect is already reflected in criterion 4. Some studies
also include social acceptability among the criteria (Beccali et al., 2003; Cavallaro
and Ciraolo, 2005; Chatzimouratidis and Pilavachi, 2008; Liposcak et al., 2006).
However social acceptability is probably a consequence of the evaluation of other
criteria.
The criteria taken into consideration are reported below along with the way
they were estimated.
Criterion 1: Electricity produced
This criterion reflects the point of view of the regional government. In fact,
Tuscany is required to reach specific electricity targets produced from renewable
resources. Of course, this criterion is also of interest for the plant operator because
the electricity produced is sold on the market.
The amount of electricity produced by each plant was extracted from the a
regional government database (2011).
In the scenarios that do not include the reorganization plan, the electricity
production of PC4 and PC5 diminishes over time. This is clearly evident from the
time series extracted from the abovementioned database. The average annual
change rates of electricity production were calculated for each plant. These rates
were used to estimate the annual amount of electricity produced by each plant for
the duration of the scenarios. It was also assumed that once a power plant
produces just 40% of its net capacity the plant is closed down (DiPippo, 2005).
This is the case of PC4 in BaU, in Closing PC2 and in Closing PC2+20CC. It was
also assumed that once PC4 closes down, all the geothermal fluid which was
originally used by PC4 is directed towards PC5, which returns to full capacity
(this is because the wells connected to PC4 and PC5 are part of the same pipeline
system). As previously mentioned all the scenarios excluding BaU involve PC2
closing down, thus no electricity would be produced by this plant.
73
For the scenarios that include the re-organization plan and for all new plants,
the annual electricity production is estimated by multiplying the net capacity of
each plant by 8,000 hours33.
The net capacity of a traditional flash steam power plant is 95% of the gross
capacity. With binary cycle technology, the thermodynamic losses are much
higher and on average the net capacity is 77% of the gross capacity34.
As previously mentioned, in all scenarios where the reorganization plan is
not included, some power plants would slightly reduce their electricity production.
When a criterion varies over time (or in space) a “point-reduction” is needed to
sum up a given distribution by a single value (Roy, 1985). In this work the median
value of the annual electricity production was used. The results for each scenario
are reported in Table 4.2.
Table 4.2: Electricity production (MWh)
BaU
531,670
Reorg
620,800
ClosingPC2
504,670
Reorg+BG4
924,800
Reorg+40CC
867,200
ClosingPC2+20CC
577,350
Reorg+20CC
744,000
Criterion 2: Company profitability
This reflects the point of view of the company operating and installing the power
plants. The profitability is measured by the net present value (NPV) of each
scenario.
The main sources of revenue are the electricity produced and the incentive
scheme. The price of electricity was obtained by means of a weighted average of
the price of electricity exchanged in the electricity market managed by the
company in charge (GME, 2011) from 2005 to 2010. The current incentive
scheme for geothermal power plants is the green certificate (GC), that is, a
market-based mechanism. The GC market in Italy is characterized by an excess of
supply (GSE, 2011) so the withdrawal price set by law was chosen as the
reference price. During the period when this research was carried out, a new law
was introduced, radically changing the incentive system. Basically, from 2011 to
2015 the GC system is maintained as is and the withdrawal price is set at 78% of
the price at which the GCs are placed35 by the company in charge of allocating
incentives for renewable energies (i.e. GSE). Thus the assumed prices are 72.32
33
This is the average yearly duration of working hours of each plant as indicated in the EIA report
for BG4.
34
Personal communication from ENEL Ricerche
35
Such a price is set by law as the difference between 180 €/MWh and the reference price of the
previous year for renewable energies set by the relevant government authority. In 2011 this price
was 113.1 €/MWh.
74
€/MWh for the electricity and 88.22 €/MWh36 for the GCs until 2015. The price
of the GCs is certainly not the price that would be revealed if the withdrawal
mechanisms were not in place. In fact, the rationale for a withdrawal price system
is to avoid a too low price because of the excess supply of GCs. Consequently,
once the withdrawal system is not in place, the price of the GCs (or of their
substitute) is expected to be much lower. The recently introduced law establishes
that after 2015, the GC system will be substituted by an auction system. The price
resulting from the auction system was assumed to be 45€/MWh. This is the price
simulated by REF (2011) through the GreeCe model in the absence of a
withdrawal price for the GCs. Of course such an estimation may easily be wrong.
Therefore a robustness analysis is needed.
Investment, maintenance and operational costs were taken from various
sources (Bertani, 2009; Entingh and McVeigh, 2003; Hance, 2005; Petty, 2005;
Sanyal, 2004), updated and adapted to the Italian case under the supervision of a
geothermal plant expert. The cost structure of each scenario required for the
estimation of the discounted cash flow is reported in Appendix A4.2.
In order to take into account the entrepreneurial risk in choosing the
discount rate, I decided to double the interest rate37 earned by the government
bonds expiring in 30 years (i.e. the entire duration of each scenario). The resulting
discount rate is 10%. Table 4.3 reports the NPV of the seven scenarios and
includes the effects of different GCs values.
Table 4.3: Profitability (NPV in thousands €)
GC(€/MWh)
45.00
66.61
88.22
BaU
153,148
153,148
153,148
Reorg
196,155
202,793
209,432
ClosingPC2
148,517
148,517
148,517
Reorg+BG4
232,372
256,898
281,424
Reorg+40CC
184,249
205,260
226,271
ClosingPC2+20CC
130,164
137,682
145,201
Reorg+20CC
183,462
197,619
211,776
Criterion 3: Municipality revenues
This reflects the point of view of the town councils. For each municipality the
revenues generated by geothermal activities consist of the following:
a. 0.13 cents per KWh produced. At least 60% of this sum is for the
municipality where the plant is located and the remaining part is
proportionally distributed to the municipalities according to the mining
license area of each municipality.
b. The compensation fund in the general agreement on the exploitation of
geothermal resources (see Appendix A4.2).
36
This value must be multiplied by a given coefficient, which depends on the type of renewable
source from which the electricity is produced. The coefficient for the electricity produced by
geothermal energy is 0.9
37
Auction held on 14 February 2011
75
c. The real property tax. According to the interviews with the mayors, in
Piancastagnaio this amounted to about €50,000 and in Santa Fiora to
€3,000.
Abbadia S. Salvatore is the only municipality which did not sign the general
agreement on the exploitation of geothermal resources. Consequently this
municipality benefits only from the revenues in point a.
The annual flow of revenues is discounted though its NPV. Since this flow
depends on the power plants and production, it was decided to use the same
discount rate proposed for the NPV of the profitability criterion: 10%. The results
are reported in Table 4.4.
Table 4.4: NPV of the municipality revenues (NPV in thousands of Euros)
Santa Fiora
Arcidosso
Piancastagnai
Abbadia S.S
Total
BaU
5,762
3,403
9,748
1,143
20,056
Reorg
5,762
3,403
16,045
551
25,761
ClosingPC2
1,205
3,710
10,438
2,396
17,749
Reorg+BG4
21,179
11,083
16,062
3,481
51,806
Reorg+40CC
19,532
10,678
13,752
34,81
47,445
ClosingPC2+20CC
12,987
7,036
10,134
2,396
32,554
Reorg+20CC
12,994
7,043
13,724
3,481
37,242
Criterion 4: Direct heat uses
The possibility to access a low cost heat source arose several times during the
interviews. Direct heat use is considered important both for house heating and for
small industrial activities. In Tuscany the main energy source for house heating is
natural gas which is distributed though pipelines. However, one of the four
villages - Piancastagnaio - is not connected to a natural gas network, so houses are
heated using GPL and diesel boilers or through electric systems. Consequently
heating is more expensive than in the rest of the region. In addition, even in the
areas that are connected to a natural gas network, it is believed that access to low
cost heating would make local companies more competitive and would encourage
new companies to be set up. This is believed to be very important to limit the
emigration flow due to the few employment opportunities available in the area.
Geothermal power plants can provide a low cost source of heat by selling
the excess heat which is not used in the plant (e.g. after the steam resulting from
the geothermal fluid has fuelled the turbine).
The availability of heat from geothermal power plants is evaluated in
linguistic terms. Following the approach used by Roy and Silhol (1986), the
qualitative evaluation was translated into a quantitative scale, which is reported
below in Table 4.5. Since the scale reports increases for worse performances, the
desired direction is decrease.
76
Table 4.5: Qualitative evaluation of direct heat uses
Evaluation
Scale
Perfect
1
Very
good
2
Good
3
More or
less good
4
Moderate
5
More or
less bad
6
Bad
7
Very
bad
8
Extremely
bad
9
Heat availability essentially depends on the size of the new plants (the larger the
size, the more excess heat is available), on the technology used (binary cycle
plants are less efficient in producing electricity than flash steam power plants, so
they present a higher quantity of excess heat) and on the specific arrangements
offered by the plant operators. In this regard, the aforementioned reorganization
plan involves the construction of a new heat pipeline to provide Piancastagnaio
with heat.
The small town of Santa Fiora is already provided with heat from BG4.
Thus a direct heat use is already an option for a very small part of the whole
Amiata area. The BaU and ClosingPC2 scenarios envisage that direct heat use is
maintained at the current level (which benefits only Santa Fiora), so the
Piancastagnaio area would still need the high cost heating systems that it is using
now. The evaluation is considered “more or less bad”. Closing PC2+20CC means
that more excess heat is available for the west side of the mountain (where
heating from the geothermal plant is already available) in comparison with the
current level. An evaluation of this scenario is therefore obtained by a one step
increase in the scale to the level of “Moderate”. As already mentioned, Reorg
entails the construction of a new pipeline for heating Piancastagnaio (which is not
connected to the natural gas network), which means that this scenario is
considered “more or less good”. In addition to the new pipeline in Piancastagnaio,
Reorg+BG4 and Reorg+20CC envisage the construction of a new plant in the west,
thus the evaluation for these two scenarios is a step further: “good”. As the above
scenarios Reorg+40CC entails installing a pipeline in Piancastagnaio and also
envisages the construction of the largest plant with the highest excess supply, the
evaluation is “very good”.
The evaluation of each scenario is reported in Table 4.6.
Table 4.6: Direct heat use
BaU
More or
less bad
Reorg
More or
less good
ClosingPC2
More or less
bad
Reorg+BG4
Good
Reorg+40CC
Very good
ClosingPC2+20CC
Moderate
Reorg+20CC
Good
77
Criterion 5: GHG emissions avoided
This criterion is of interest for the regional government. In fact the regional
government took over the EU 20-20-20 target,38 and the production of electricity
from renewable energy is part of the GHG abatement strategy. Geothermal power
plants can emit a large amount of GHGs in the form of CO2 and CH4 and their
exact value depends on the specific composition of geothermal fluid. However
their emissions are not included in the quotas allocated to EU countries. Therefore
the amount of GHGs emitted from geothermal power plants is not part of the
amount of GHGs that Italy and Tuscany need to reduce 39 . Thus, the GHG
emissions caused by geothermal power plants are not accounted for in this study.
The amount of electricity produced in Tuscany from each fossil fuel source
was derived from Terna (2010) and from the Tuscany regional government (2009).
In 2008 the electricity production obtained from fuel oil was 13%, while the rest
was obtained from natural gas. An amount of 557.1 Kg of CO2eq is avoided for
geothermal MWh. This value was calculated using data from the regional
government ’s database (providing data from individual power plants) which
shows that the average emissions of CO2eq per MWh produced by fuel oil is 763.2
Kg and 526.2 Kg by natural gas. It was then assumed that the electricity obtained
by the geothermal power plants replaces the electricity produced by burning fuel
oil and natural gas in the same proportion as such plants contribute to the total
quantity of electricity produced by fossil fuels.
The median value of the annual GHG emissions avoided for each scenario is
shown in Table 4.7.
Table 4.7: Tons of CO2eq emissions avoided
BaU
296,187
Reorg
345,840
ClosingPC2
281,145
Reorg+BG4
515,194
Reorg+40CC
483,106
ClosingPC2+20CC
281,145
Reorg+20CC
414,473
38
The 20-20-20 are two main targets to be achieved by the EU by 2020: at least 20% of GHGs
reduction in comparison to the 1990 emissions and at least 20% of energy consumptions must be
obtained by renewable energy. On 22 June 2011 the European commission also proposed a new
directive to achieve an increase of at least 20% in energy saving compared to the PRIMES 2007
baseline.
39
This is because it is generally assumed that the GHG emissions from geothermal power plants
would naturally occur in a diffused way, so geothermal power plants would be simply
concentrating emissions they cannot be held responsible for.
78
Criteria 6, 7, 8, 9: H2S, Hg, NH3 and As emissions
The emissions can be a cause for concern for the various social actors involved in
the study (municipalities, the regional government, ENEL, etc) however, they
represent the greatest worry for residents.
These criteria include the emissions that are discharged in the highest
amounts, and which are considered most dangerous, namely H2S, Hg, NH3, and
As40.
H2S produces an unpleasant smell at low levels of concentration (but its
perception diminishes with prolonged exposition), and beyond certain levels it
represents a serious hazard for human health. Hg, NH3 and As can also represent a
health problem beyond certain concentration levels41. The last ARPAT report on
the emissions of geothermal power plants shows that although the concentration
value of the WHO guidelines for health protection in the 1997-2009 measurement
period was occasionally exceeded, the concentration of H2S and Hg in Mt. Amiata
is much higher than in the traditional geothermal area.
It is worth mentioning that H2S and NH3 contribute to the formation of
inorganic secondary particulate matter (PM) whose effects are on a regional scale.
In this regard the regional government has specific objectives for PM reduction.
From a comparison of the geothermal areas in Tuscany, the total Hg
emissions flow in Mt. Amiata is much higher than in other areas (Tuscany
regional government, 2010). In addition, ARPAT (2010) reports a frequent
overflow of the maximum Hg and NH3 flow allowed by law among plants (but the
regulation is still respected because the maximum concentration limits are not
exceeded)42.
NH3 emissions from geothermal power plants are especially important in
Tuscany because they represent the second source of NH3 emission after
agriculture, amounting to 30-40% of the total emissions of this substance
(Tuscany regional government, 2010).
Many different variables should be taken into consideration to estimate the
concentrations of emissions in the air (such as wind speed and direction,
temperature and rainfall) and a specific model should be used. This is certainly
very important but goes beyond the scope of this study. Consequently, only the
annual quantities of air emissions are calculated and not air concentrations.
40
Others could have been included such as antimony, methane, and boric acid, however according
to the literature consulted, given their emissions levels, they are not thought to represent a problem.
41
The maximum concentration of the polluting elements of the WHO guidelines and other
authorities for health protection are reported in ARPAT (2010), Tuscany regional government
(2010) and Bacci (1998).
42
The regulation on geothermal power emissions set a first maximum limit on the flow and a
second limit on the maximum concentration of the polluting substance. Only when the first limit is
not respected, does the second take place. Thus, when the maximum flow limit is exceeded, the
regulation is still respected if pollutant concentration does not exceed the level indicated by the
second limit.
79
The emission factors indicating the amount of emissions per MWh were
calculated by averaging the individual ARPAT (2010)43 emission measurements.
The ARPAT database reports the emissions both in the presence and absence of
AMIS. The environmental impact assessment (EIA) report for BG4 specifies that
on average the abatement filters for H2S and Hg (called AMIS) work 90% of the
time. Consequently the emission factors were calculated as a weighted average
emission in the presence and absence of AMIS. In order to calculate the annual
emissions, the resulting emission factors were multiplied by 8,000, which is the
number of hours a power plant normally works (ENEL, 2005). For the remaining
760 hours the plant is assumed not to work due to maintenance. When the power
plant is not working, the flow of the wells is reduced to about 50% of its working
flow, and wells discharge directly into open air, that is, without AMIS and without
re-injection of the fluid (at plant level). Thus the emissions during the maintenance
period were estimated as the emissions that would occur without AMIS, with 50%
of flow and increased by the quantity normally re-injected. The quantity normally
re-injected was assumed to be 25% of the flow that reaches the plant44 (ENEL,
2005).
Tables 4.8, 4.9, 4.10 and 4.11 report the median annual values of H2S, Hg,
NH3 and As emissions for each scenario.
BaU
1,825
Reorg
1,070
ClosingPC2
1,015
Table 4.8: H2S emissions (Tons/yr)
Reorg+BG4 Reorg+40CC ClosingPC2+20CC
1,727
1,119
966
Reorg+20CC
1,021
BaU
605
Reorg
309
ClosingPC2
251
Table 4.9: Hg emissions (Kg/yr)
391
317
244
Reorg+20CC
302
BaU
3.088
Reorg
3.392
ClosingPC2
2.929
Table 4.10: NH3 emissions (Tons/yr)
7.827
3.530
2.792
Reorg+20CC
3.255
BaU
16
Reorg
19
ClosingPC2
15
Table 4.11: As emissions (kg/yr)
26
19
15
Reorg+20CC
18
43
In the PC4 plant the AMIS system was only installed recently and no measurements were
available. The abatement efficiency was thus estimated by averaging the efficiency of the same
filters on all the other plants in Mt. Amiata.
44
Even though the quantity to be re-injected was taken from an ENEL source, it should be noted
that it represents an approximation and the actual level could change according to different levels
of condensation.
80
Criterion 10: Impact on the aquifer
The evaluation of this criterion is unavoidably subject to strong uncertainty. The
heated scientific debate mentioned in Section 4.3.2 is also a result of this
uncertainty.
Given the uncertainty underlying the effects of geothermal exploitation on
the conservation of the water aquifer and the critical importance of this issue (the
aquifer provides water to more than 700,000 people), a a precautionary principle
is here proposed. Consequently it is assumed that geothermal exploitation may
affect the conservation of the aquifer45. Therefore if the extraction of vapor from a
geothermal reservoir can cause a depression, which draws water from the potable
aquifer, the consequence is that the less vapor is extracted, the better it is for the
conservation of the potable aquifer.
The quantities of extracted vapor in the different scenarios was estimated
from the EIA data (ENEL, 2005; 2009c) and are reported in Table 4.12. With
binary cycle plants all extracted fluid is assumed to be re-injected.
BaU
284
Reorg
194
Table 4.12: net quantity of extracted fluid (T/h)
ClosingPC2 Reorg+BG4 Reorg+40CC ClosingPC2+20CC
164
280
194
164
Reorg+20CC
194
Criterion 11: Visual impact
There are many tools for assessing the visual impact of a project, however given
the scope of this study no sophisticated techniques were used. Similarly to the
approach proposed in Munda et al. (2006), a matrix aimed at facilitating the
evaluation of the visual impact was built with two axes: distance of the additional
work from the main villages of the area (Piancastagnaio in the east and Santa
Fiore in the west) and volume of the work (see Fig 4.2). Thus the higher the
distance and the smaller the volume, the better the visual impact. The visual
impact would naturally be evaluated though a qualitative judgment. As for
criterion 4, the qualitative evaluation was translated into a quantitative scale,
which is reported in Figure 4.2. The result is that the higher values of the scale
mean a worse visual impact, so lower values are preferred to higher values.
The visual impact of the BaU scenario is considered as being “moderate”.
So the additional work of the other scenarios involves changes in the visual impact
evaluation with respect to the “moderate” level of the BaU scenario.
45
A similar view is assumed in the advice on the re-organization plan of Piancastagnaio provided
by the three watershed authorities (Tevere, Ombrone and Fiora), the office in charge of the water
resources protection and management and by the office in charge of the prevention of hydraulic
and hydro-geologic risks of the regional government. The document concludes that it is not
possible to rule out that the vapor extraction cannot provoke an important impact on the phreatic
aquifer.
81
Volume of the works
Figure 4.2: visual impact matrix evaluation
Extremely Bad:
8
Very bad:
7
Very bad:
7
Bad:
6
Bad:
6
More or less bad:
5
More or less bad:
5
Moderate:
4
Moderate:
4
More or less good:
3
More or less good:
3
Good:
2
Good:
2
Very good:
1
Distance
On the basis of the above considerations, the visual impact of the different
scenarios is reported in Table 4.13.
Table 4.13: visual impact
BaU
Moderate
Reorg
More or
less bad
ClosingPC2
Moderate
Reorg+BG4
Bad
Reorg+40CC
Very bad
ClosingPC2+20CC
More or less bad
Reorg+20CC
Bad
82
4.5 Ranking alternatives
In this work the decision maker is not a real person whose preferences can be
elicited in some way. Consequently the model only represents a system of
preferences aimed at answering certain questions (Roy, 1991).
A choice must be made about weights. These reflect the importance of a
given criterion with respect to the others. Different techniques can be used46 but in
the context of this work with no individual decision maker, it is impossible to
establish a set of weights to satisfy all the social actors. Some models like
ELECTRE IV (Roy and Hugonnard, 1982) and NAIADE (Munda, 1995) simply
avoid assigning weights to criteria. However these models do assign weights in an
implicit way. In fact, if no weights are assigned, the result is that all criteria have
the same weight. As previously stated, one of the main advantages of a multicriteria analysis is its inherent transparency. If criteria are assumed to have the
same importance, it is advisable that all criteria are assigned an equal weight in an
explicit way.
Another approach suggested by Munda (2008) consists in assigning each
criterion to one of the three dimensions of the sustainability concept (economic,
social and environmental). The weights are allocated to criteria proportionally so
that each dimension has an equal weight. Such an approach is certainly defensible
from a theoretical point of view. However, its main problem is that often criteria
can be assigned to the three different dimensions only with a very high degree of
arbitrariness. For instance, considering the criteria used in this study, the
profitability of the plant would certainly be considered as being ‘economic’, but
what about electricity production? Is it economic (because it is sold on the
market), social (because it is used by humans), or environmental (because it
comes from a renewable energy source)? The same would apply to direct heat use.
And what about polluting substances? Are they environmental because they affect
the environment, or social because they can also affect human health?
This work does not claim to provide a complete answer to the conflict
described, but rather to explore the problem from different points of view. A
sensitivity analysis applied to relative weights is thus an extremely powerful
technique.
Here a final ranking is presented assuming equal weights of all the criteria,
and further results are explored by changing the relative weights of criteria.
Table 4.14 represents the multi-criteria impact matrix derived by joining the
evaluation vectors of the previous section. Table 4.15 reports the outranking
46
See Edwards (1977) for SMART, Edwards and Barron (1994) for SMARTER, Jia et al (1998)
for SWING, Simos (1990) and its amendments (Figueira and Roy, 2002) and Wang et al. (2008)
for pair-wise comparison techniques
83
matrix by applying Eq. 4.3 with equal weights and the indifference threshold
indicated in Table 4.14.
The choice of threshold value is very often based on common sense. In
addition, it nearly always contains a certain amount of arbitrariness (Roy et al.,
1986). Yet, in many situations, any reasonable value of the indifference thresholds
other than zero, leads to a model of preference that seems more convincing than
equating the indifference threshold to zero (Bouyssou, 1990).
In this research project, indifference thresholds were set using two common
sense approaches. When an external benchmark was available, the indifference
thresholds were set as a minimum percentage of achievement of the objectives
reflected by the selected criteria. This was the case for electricity produced and
GHGs avoided. These two criteria are mainly of interest to the regional
government. In fact, the regional government has specific objectives for electricity
production from geothermal power and GHG reduction. Thus, the threshold
values reflect minimum percentages of achievements of the regional government’s
stated objectives. When an external benchmark was not available, the thresholds
were set as the minimum percentage of current levels. This is the case for all
criteria except electricity produced and GHGs avoided. In any case, a robustness
analysis is included to verify that arbitrariness does not significantly affect the
final results.
Table 4.14: multi-criteria impact matrix
Criteria
Electricity
prod.
Profitability
Municipalities
rev.
Direct heat
uses
Avoided
GHGs em,
H2S emissions
Hg emissions
NH3
emissions
As emissions
Impact on
aquifer
Visual impact
Dir.
BaU
Reorg
ClosingPC2
Reorg
+BG4
Reorg
+40CC
ClosingPC2
+20CC
Reorg
+20CC
Threshold
value
↑
531,670
620,800
504,670
924,800
867,200
577,350
744,000
100,000
↑
153,148
196,155
148,517
232,372
184,249
130,164
183,463
15,000
↑
20,056
25,761
17,749
51,806
47,445
32,554
37,242
5,000
↓
6
4
6
3
2
5
3
-
↑
296,187
345,840
281,145
515,194
483,106
281,145
414,473
150,000
↓
↓
1,825
605
1,070
309
1,015
251
1,727
391
1,119
317
966
244
1,021
302
250
50
↓
3,088
3,392
2,929
7,827
3,530
2,792
3,255
500
↓
16
19
15
26
19
15
18
3
↓
284
194
164
280
194
164
194
50
↓
4
5
4
6
7
5
6
-
84
Table 4.15: outranking matrix
BaU
Reorg
ClosingPC2
Reorg+BG4
Reorg+40CC
ClosingPC2+20CC
Reorg+20CC
BaU
Reorg
ClosingPC2
Reorg+BG4
Reorg+40CC
0
0.7273
0.7273
0.6364
0.8182
0.6364
0.7727
0.2727
0
0.4545
0.4545
0.5909
0.6364
0.5909
0.4545
0.5455
0
0.4545
0.5455
0.5000
0.5455
0.3636
0.5455
0.5455
0
0.6818
0.5455
0.6818
0.1818
0.4091
0.4545
0.3182
0
0.4545
0.4091
ClosingPC2
+20CC
0.3636
0.4545
0.5000
0.4545
0.5455
0
0.5000
Reorg+20CC
0.2273
0.4091
0.4545
0.5000
0.5909
0.5000
0
One disadvantage of the aggregation procedure applied here is that there can be
more than one ranking with the same maximum likelihood ranking τ*. This is why
the results presented in the following tables include more than one ranking. The
rankings presenting the highest score when equal weights are applied are reported
in Table 4.16. It is worth noting that equal weight methods are the most common
approach in renewable energy analyses (Wang et al., 2009).
1°
Reorg+Bin40
Reorg+Bin40
Table 4.16: ranking for equal weights among all criteria
2°
3°
4°
5°
6°
Bin20
Reorg+Bin20 Reorg ClosingPC2 Reorg+BG4
Reorg+Bin20
Bin20
Reorg ClosingPC2 Reorg+BG4
7°
BaU
BaU
Some interesting results can be observed. The current scenario is the worst. In this
sense, the discontent that geothermal power has generated can be justified. Also,
with equal weights, the scenario joining the two ENEL proposals (i.e. Reor+BG4)
is the second worst.
Scenarios including binary cycles technologies score between best positions.
In fact, Reorg+Bin40 ranks first. However, as explained in the institutional
analysis section, the reorganization plan (included in Reor+Bin40 and in
Reorg+Bin20) would be strongly opposed by the Prospettiva Comune di
Piancastagnaio and Comitati di Difesa del Territorio. Bin20 does not score as well
as Reor+Bin40 but might receive less social opposition.
A sensitivity analysis was applied to evaluate how rankings change by
varying the relative weights of criteria. A robustness analysis was also applied to
the indifference thresholds. Of course, an extremely high number of sensitivity
analyses are possible by combining all possible weights of each criterion with the
other weights of all the other criteria and with all possible values of the
indifference thresholds. Limits need to be set. It was decided to limit the possible
number of sensitivity analyses to the following possible combinations: an increase
in the weight of each criterion by one and maintaining all other weights at their
original value of one (all weights are normalized to make a total of one), increase
the threshold value of the same criterion by 50%, reduce the threshold value of the
85
same criterion by 50%, and increase the weights of two criteria that reflect a
specific point of view. The most significant changes that were observed by
increasing or reducing the indifference thresholds are included in the tables. When
the different values of the indifference thresholds do not cause significant changes
in the rankings, the robustness analysis of the indifference threshold is not
reported. Only the most interesting results obtained by the sensitivity analysis are
reported here.
The profitability criterion is mainly of concern for ENEL. The results
obtained by changing the value of its weight are in Fig. 4.3 (for reasons of space,
just the three best positions are included).
N of times higher then other weights
Fig. 4.3: Sensitivity analysis of Profitability
20
Reorg+BG4
Reorg+Bin40
Reorg+Bin20
Reorg+BG4
Reorg
Reorg+Bin40
6
Reorg+BG4
Reorg+Bin40
Reorg+Bin20
Reorg+BG4
Reorg
Reorg+Bin40
5
Reorg+BG4
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Reorg
Reorg+Bin40
4
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+BG4
Reorg
Reorg+Bin40
3
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Reorg
Reorg+Bin40
Reorg
Reorg+Bin40
Reorg+BG4
Reorg+Bin40
Reorg+BG4
Reorg
Reorg+Bin20
2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg+BG4
Bin20
Reorg+Bin40
Reorg
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin40
Reorg
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Ranking
Reorg+Bin20
Bin20
3°
1
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin40
Reorg+Bin20
Bin20
1°
2°
3°
Ranking - threshold reduced by 50%
The position of Reorg+Bg4, i.e. the projects proposed by ENEL, improves by
increasing the weight of the profitability criterion. However, Reog+Bin40 keeps
scoring very well. When an indifference threshold is reduced and the profitability
weight increased, Reorg reaches a very high position. Figure 4.3 does not report
the tails of the ranking. These would show that if the profitability weight is five,
Bin20 is in last position. This results suggest that with increasing importance for
this criterion, Bin20 would probably be rejected by ENEL unless it is heavily
subsidized.
Figure 4.4 shows different rankings obtained by increasing the weight of the
Electricity Production.
Reorg+Bin40 remains in first position even with a high weight. Only if the
indifference threshold is strongly reduced and a weight of five is applied, would
86
Reorg+Bin40 be surpassed by Reorg+BG4. Again the tails are not included but
they show that BaU would stay in last position.
Fig. 4.4: sensitivity analysis of Electricity Production
20
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+BG4
Reorg+Bin40
Reorg+Bin20
5
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+BG4
Reorg+Bin40
Reorg+Bin20
4
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+BG4
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
3
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
1
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Ranking
Reorg+Bin20
Bin20
3°
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin40
Reorg+Bin20
Bin20
1°
2°
3°
Figure 4.5 reports the same typology of analysis for the H2S Emission criterion.
Changes can be detected only by reducing the indifference threshold. In so doing,
Bin20 would be the first option if the weight were doubled.
Fig. 4.5: Sensitivity analysis of H2S emissions
20
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
Bin20
Reorg+Bin40
Reorg+Bin20
3
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
Bin20
Reorg+Bin40
Reorg+Bin20
2
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
Bin20
Reorg+Bin40
Reorg+Bin20
1
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
1°
2°
Ranking
3°
Reorg+Bin40 Bin20
Reorg+Bin20
Reorg+Bin40 Reorg+Bin20 Bin20
Bin20
Reorg+Bin40 Reorg+Bin20
1°
2°
3°
The sensitivity analysis for the Hg emission criterion is depicted in Fig. 4.6. By
increasing the weight of this criterion by three, ClosingPC2 and Bin20 reach the
first position. So, when the emissions of Hg are actually considered as a major
concern (e.g. because of further investigations announced by the regional
government following the results of the epidemiological study) these alternatives
could be justified. If the threshold value is increased by 50%, Reorg+Bin40 rank
first, Reorg+Bin20 second, and Bin20 third.
87
N of times higher then other
weights
Fig. 4.6: Sensitivity analysis of Hg emissions
20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Bin20
3
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Bin20
2
ClosingPC2
Reorg+Bin40
Bin20
ClosingPC2
Reorg+Bin40
Bin20
Reorg+Bin40
Reorg+Bin20
Bin20
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Ranking
Reorg+Bin20
Bin20
3°
Reorg+Bin40
Reorg+Bin20
Bin20
1
1°
2°
3°
Ranking - threshold increased by 50%
The sensitivity analysis for the impact on aquifer is reported in Fig. 4.7. The use
of binary cycles improves the position of the scenario. However, if the threshold
value is reduced, ClosingPC2 and Bin20 rank better than the alternatives which
include the reorganization plan.
Fig. 4.7: sensitivity analysis of Impact on aquifer
20
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
2
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
1
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin20
Bin20
Reorg+Bin20
Bin20
Reorg+Bin20
1°
2°
Ranking
3°
Bin20
Reorg+Bin40 Reorg+Bin20
Bin20
Reorg+Bin40 ClosingPC2
Reorg+Bin40
Bin20
ClosingPC2
ClosingPC2
Reorg+Bin40 Bin20
Reorg+Bin40
Bin20
Reorg+Bin20
ClosingPC2
Bin20
Reorg+Bin40
Reorg+Bin40
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
1°
2°
3°
Figure 4.8 reports the sensitivity analysis obtained by changing the weights of the
two criteria at the same time. The criteria are Electricity production and GHGs
avoided. This type of analysis would reflect the importance of regional energy
policies.
88
Fig. 4.8: Sensitivity analysis of Electricity Production (E) and GHGs (G)
E: 20
G: 20
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Bin20
Reorg
ClosingPC2
BaU
E: 4
G: 4
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Bin20
Reorg
ClosingPC2
BaU
E: 3
G: 4
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Bin20
Reorg
Reorg
ClosingPC2
ClosingPC2
BaU
BaU
E: 3
G: 3
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Bin20
Reorg
Reorg
ClosingPC2
ClosingPC2
BaU
BaU
E: 3
G: 2
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Bin20
Reorg
Reorg
ClosingPC2
ClosingPC2
BaU
BaU
E: 2
G: 3
Reorg+Bin40
Reorg+Bin20
Reorg+BG4
Bin20
Reorg
ClosingPC2
BaU
E: 2
G: 2
Reorg+Bin40
Reorg+Bin20
Reorg+BG4
Bin20
Reorg
ClosingPC2
BaU
1°
2°
3°
Ranking
4°
5°
6°
7°
The sensitivity analysis of the other criteria is included in Appendix A4.3.
Results were also calculated for the considered values of green certificates but the
changes observed are minimal and are related only to the tails of the rankings.
89
4.6 Conclusions
The context of this work is characterized by strong uncertainty concerning crucial
issues such as the impact of a given economic activity on human health and the
conservation of an extremely precious resource: water. It is my contention that the
problem presented here is a typical post-normal science problem, where “facts are
uncertain, values in disputes, takes high and decisions urgent” (Funtowicz and
Ravetz, 1993: p744 ). In post-normal science, the classical dichotomies of facts
and values, and ignorance and knowledge are transcended. Incomplete control and
a plurality of legitimate perspectives should be openly acknowledged. The social
actors included in this study do have different legitimate perspectives and
conflicting values. This paper has attempted to show how a social-multi-criteria
evaluation can be applied in such a post-normal science case.
Decision making cannot accommodate all the legitimate claims from
different social actors. Some people will benefit and others will be negatively be
affected. If decision making is based on optimizing mono-disciplinary models,
best alternatives could certainly (and easily) be identified. However, these
optimizing models tend to make the problems that have not been captured by the
selected variables, reappear in a stronger form in other models. For example,
profit-maximizing models, which cause ecological stress, and models that
optimize ecological conservation variables, which imply profit compression and
the absence of employment opportunities. In addition, by boosting the expected
benefits of the selected mono-disciplinary variables in conditions of diverging
perspectives, social and environmental conflicts can easily be aggravated. In fact,
the social actors whose interests are not reflected by the selected variables will be
negatively affected. This is why decision support tools should facilitate decisionmaking processes based on an interdisciplinary selection of variables, aimed at
identifying compromise solutions rather than providing optimizing results.
The main objective of this work was not to indicate a definitive solution for
the geothermal development scenarios in Mt. Amiata, but rather to explore
possible alternatives in the light of different concerns and different points of view.
The results do not intent do relieve policy makers of their responsibilities to take
very difficult decisions but are aimed at shedding light on the consequences of
specific options by assigning more or less importance to certain criteria and
certain points of view. In this way, the paper contributes to the decision-making
process by modeling preferences through weights and criteria. The ultimate hope
is to have contributed to making the decision-making process more transparent.
With this caveat, some tentative conclusions for this specific case study are
reported. Current scenarios become the worst of all considered alternatives when
90
criteria have an equal weight. In addition, current scenarios never get beyond the
penultimate position by changing relative weights. The two projects proposed by
ENEL become the first option when the profitability criterion weighs at least five
or six times more than all the others. Between these two extremes lie various
alternatives, and their rank depends on the weights of the criteria. Therefore
depending on the relative weights, this work provides some answers for the
decision-making process. Binary cycles tend to move the given alternatives
between the highest positions. Regarding social reactions, the alternatives which
include the so-called reorganization plan would be vetoed by three social actors
based in the east of the mountain. The scenarios reflecting the views of the
residents committees (i.e. ClosingPC2) rank in first position when air emissions or
impact on aquifer acquire more importance. Specifically, it is in the first or second
position when the weight of NH3 emissions criterion is a least three times higher
than the others, when Hg emissions are at least twice as high as the others, and
when the weight of As emissions is two or three times more then the others. It
would obtain a first or second position when the weight on the impact on aquifer
criterion is doubled along with an halving of the indifference threshold. One
social compromise alternative could be the installation of binary cycles on the
west side. However, the position of the different social actors is not determined
once and for all, and opposition may become stronger when the feasibility of a
given project becomes a concrete option. In addition, the installation cost of a
20MW binary cycle plant should probably be subsidized in addition to the
envisaged green certificate price.
It is worth recalling that a specific criterion for employment effects was not
included. The reasons for this were explained in Section 4.4.2 and include the lack
of data and the fact that employment was never indicated as being important by
the interviewees. This is because the number of permanent employees in the
geothermal industry in Mt. Amiata is small and is not expected to grow
significantly in the expansion scenarios. However, inclusion of employment
effects for the limited period of the construction phase of the scenarios,
comprising new investments would probably have provided different results.
Moreover with a larger scale analysis, the effects on ancillary industries could
also be included. These employment effects could be evaluated through inputoutput analyses along with other economic and environmental impacts. In fact,
depending on the objectives of the analysis, multi-criteria techniques could easily
be used with other impact assessment tools in order to evaluate the effects of the
selected scenarios on different socio-economic and environmental variables.
91
References
Afgan, N.H. and Carvalho, M.G. (2002). Multi-criteria assessment of new and
renewable energy power plants. Energy, 27 (8), 739–755.
ARPAT. (2010). Monitoraggio delle aree geotermiche: controllo delle emissioni qualità dell'aria. Tuscany Region - Agenzia Regionale di Protezione
dell'Ambiente della Toscana. Regione Toscana. [Online] Available from:
http://www.arpat.toscana.it/temi-ambientali/aria/areegeotermiche/ar_areegeotermiche.html [Accessed 22 February 2010]
ARS. (2010). Progetto di ricerca epidemiologica sulle popolazioni residenti
nell'intero bacino geotermico toscano - Progetto geotermia. Agenzia
Regionale di Sanità della Toscana e Fondazione Toscana Gabriele
Monasterio
[Online]
Available
from:
http://www.regione.toscana.it/regione/export/RT/sitoRT/Contenuti/minisiti/pier/visualizza_asset.html_378191505.html [Accessed
7 March 2011]
Bacci, E. (1998). Energia geotermica: impieghi, implicazioni ambientali e
minimizzazione dell'impatto. ARPAT - Litografia I.P., Florence, Italy
Barazzuoli, P. Bianchi, S. B, B.M. Nocchi, M. Pratesi, T. Rigati, R. and Solleoni,
M. (2004). Studio Idrogeologico per la Valutazione della vulnerabilità
all’Inquinamento dell’Acquifero Vulcanico del M. Amiata. Commissionato
dall'Amministrazione Provinciale di Grosseto. C&P Adver effigi, Arcidosso.
Barda, O.H. Dupluis, J. and Lencoini, P. (1990). Multicriteria location of thermal
power plants. European Journal of Operations Research, 45 (2/3), 32–346.
Barelli, A. Ceccarelli, A. Dini, I. Fiordelisi, A. Giorgi, N. Lovari, F. and
Romagnoli, P. (2010). A review of the Mt. Amiata geothermal system (Italy).
Paper presented at the World Geothermal Congress. Bali, Indonesia.
http://b-dig.iie.org.mx/BibDig/P10[Online]
Available
from:
0464/pdf/0613.pdf [Accessed 19 March 2011]
Beccali, M. Cellura, M. and Mistretta, M. (2003). Decision-making in energy
planning. Application of the Electre method at regional level for the diffusion
of renewable energy technology. Renewable Energy, 28 (13), 2063–2087.
Bertani, R. (2009). Geothermal energy in the world: current status and future
scenarios. Presentation held at the International Workshop "Geotheramal
energy development:opportunities and challenges", Pomarance, Italy.
[Online] Available from: http://cegl.it [Accessed 14 March 2010]
Bertani, R. and Fridleifsson, I.B. (2010). The contribution of geothermal energy to
the reduction of CO2 emissions and to the mitigation of the climate changes.
Notiziario dell'Unione Geotermia Italiana. [Online] Available from:
www.unionegeotermica.it [Accessed 19 March 2011]
Bertini, G. Cappetti, G. Dini, I. and Lovari, F. (1995). Deep drilling results and
updating of geothermal knowledge on the Monte Amiata area. Paper
92
presented at the World Geothermal Congress. Florence, Italy. [Online]
Available from: http://www.geothermal-energy.org/ [Accessed 20 March
2011]
Borgia, A. (2007). Lettera inviata alla Regione Toscana. Oggetto: vostre note del
16 gennaio 200. Prot. A00GRT/13761/124.02.02 e dle 22 gennaio 2007, Prot.
A00GRT/19166/124/02/02.
[Online]
Available
from:
18 January 2011]
Bouyssou, D. (1990). Building Criteria: a prerequiste for MCDA. In Costa,
C.A.B.e. (Ed.) Readings in multiple criteria decision aid. Springer, Berlin,
Heidelberg, New York, Pp 58-80
Burgassi, R. Cataldi, R. Mounton, J. and Scadenllari, F. (1965). Prospezioni delle
anomalie geotermiche e sua applicazione alla regione Amiatina. Industria
Mineraria, 16 (5), 231-243.
Calamai, A. Cataldi, R. Squarci, P. and Taffi, L. (1970). Geology geophysics and
hydrogeology of Monte Amiata geothermal field. Geothermics, Special issue
1.
Cataldi, R. (1965). Remarks on the geothermal reserach in the region of Monte
Amiata (Tuscany - Italy). Paper presented at the IAV International
symposium of volcanology. New Zeland.
Cataldi, R. (2001). Social acceptance of geothermal projects: problems and costs.
International
Summer
School.
[Online]
Available
from:
http://www.geothermal-energy.org [Accessed 5 May 2011]
Cavallaro, F. and Ciraolo, L. (2005). A multicriteria approach to evaluate wind
energy plants on an Italian island. Energy Policy, 33 (2), 235–244.
Chatzimouratidis, A.l. and Pilavachi, P.A. (2008). Multicriteria evaluation of
power plants impact on the living standard using the analytic hierarchy
process. Energy Policy, 36 (3), 1074–1089.
Comune di Piancastagnaio. (1994). Geotermia: possibili effetti sulla salute e
risanamento delle aree compromesse. Piancastagnaio
De Jesus, A.C. (1997). Environmental sustainability of geothermal development.
Energy Sources, 19 (1), 35-47.
Diakoulaki, D. Henggler Antunes, C. and Gomes Martins, A. (2005). MCDA and
energy planning. In Figueira, S.G. and Ehrgott, M. (Eds.), Multiple-criteria
Decision Analysis. State of the Art Surveys. Springer International Series in
Operations Research and Management Science, New York, Pp 859–898
DiPippo, R. (1991). Geothermal energy Electricity generation and environmental
impact. Energy Policy, 19 (8), 798-807.
DiPippo, R. (2005). Geothermal power plants: principles, applications and case
studies. Butterworth-Heinemann, Amsterdam, the Netherlands
EDRA. (2006a). Inquadramento generale dei processi gravitativi profondi
(volcanic
spreading)
sui
vulcani.
[Online]
Available
from:
93
18 January 2011]
EDRA. (2006b). Rilievo geostrutturale preliminare dell’apparato vulcanico del
Monte
Amiata
[Online]
Available
from:
18 January 2011]
Edwards, W. (1977). How to use multiattribute utility measurement for social
decision making. IEEE Transactions on Systems Man and Cybernetics, 7 (5),
326–340.
Edwards, W. and Barron, F.H. (1994). Smarts and smarter: improved simple
methods for multiattribute utility measurement. Organizational Behavior and
Human Decision Processes, 60 (3), 306–325.
ENEL. (2005). Concessione di coltivazione Bagnore. Studio di impatto
ambientale: Centrale geotermoelettrica Bagnore 4 [Online] Available from:
http://www.regione.toscana.it/via [Accessed 25 February 2011]
ENEL. (2008). Concessione di coltivazione Piancastagnaio. Riassetto dell'area
geotermia di Piancastagnaio: Studio di Impatto Ambientale. [Online]
Available from: http://www.regione.toscana.it/via [Accessed 25 February
2011]
ENEL. (2009a). Concessione di coltivazione Bagnore. Centrale geotermica di
Bagnore 4: Integrazioni allo studio di impatto ambientale. [Online] Available
from: http://www.regione.toscana.it/via [Accessed 25 February 2011]
ENEL. (2009b). Concessione di coltivazione Piancastagnaio. Riassetto dell'area
geotermia di Piancastagnaio: Integrazioni allo studio di Impatto Ambientale.
[Online] Available from: http://www.regione.toscana.it/via [Accessed 25
February 2011]
ENEL. (2009c). Riassetto dell'area geotermica di Piancastagnaio: Integrazioni
volontarie. [Online] Available from: http://www.regione.toscana.it/via
[Accessed 25 February 2011]
ENEL. (2010). Bilancio di sostenibilità. [Online] Available from:
http://www.enel.com/ [Accessed 12 June 2011]
Entingh, D.J. and McVeigh, J.F. (2003). Historical improvements in geothermal
power system costs. Geothermal Resources Council Transactions, 27.
Figueira, J. and Roy, B. (2002). Determining the weights of criteria in the
ELECTRE type methods with a revised Simos’ procedure. European Journal
of Operational Research, 139 (2), 317–326.
Focacci, G.A. Madrucci, W. and Nenci, D. (1993). Studio e pianificazione delle
risorse idriche nel territorio della Comunità Montana del Monte Amiata
Comunità Montana del Monte Amiata.
Funtowicz, S.O. De Marchi, B. Lo Cascio, S. and Munda, G. (1998). The Troina
water valuation case Study. Social processes for environmental valuation:
94
Procedures and institutions for social valuations of natural capitals in
environmental conservation and sustainability policy (VALSE). Research
project ENV4–CT96–0226.
Funtowicz, S.O. and Ravetz, J.R. (1993). Science for the post-normal age. Futures,
25 (7), 739-755.
Gamboa, G. (2006). Social multi-criteria evaluation of different development
scenarios of the Aysén region, Chile. Ecological Economics, 59 (1), 157-170.
Gamboa, G. and Munda, G. (2007). The problem of windfarm location: a aocial
multi-criteria evaluation framework. Energy Policy, 35 (3), 1564-1583.
Georgopoulou, E. Lalas, D. and Papagiannakis, L.A. (1997). Multicriteria
decision aid approach for energy planning problems: the case of renewable
energy option. European Journal of Operational Research, 103 (1), 38–54.
Gestore Mercati Energetici - GME. (2011). Statistiche e Monitoraggio. [Online]
Available from: http://www.mercatoelettrico.org/ [Accessed 14 March 2011]
Gestore Servizi Energetici - GSE. (2011). Incentivazione delle fonti rinnovabili
con i Certificati Verdi e le Tariffe Onnicomprensive. [Online] Available
from: http://www.gse.it/ [Accessed 12 February 2011]
Goumas, M. and Lygerou, V. (2000). An extension of the PROMETHEE method
for decision making in fuzzy environment: ranking of alternative energy
exploitation projects. European Journal of Operational Research, 123 (3),
606-613.
Hance, C.N. (2005). Factors affecting costs of geothermal power development
Geothermal Energy Association. [Online] Available from: http://www.geoenergy.org/ [Accessed 28 January 2011]
Haralambopoulos, D.A. and Polatidis, H. (2003). Renewable energy projects:
structuring a multicriteria group decision-making framework. Renewable
Energy, 28 (6), 961–973.
Janssen, R. (1992). MultiobjectiveDecision Support for Environmental Problems.
Kluwer, Dordrecht
Jia, J.M. Fisher, G.M. and Dyer, J.S. (1998). Attribute weighting methods and
decision quality in the presence of response error: a simulation study. Journal
of Behavioral Decision Making, 11 (2), 85–105.
Kagel, A. Bates, D. and Gawell, K. (2007). A guide to geothermal energy and the
environment. Geothermal Energy Association. [Online] Available from:
http://www.geo-energy.org [Accessed 28 January 2011]
Kowalski, K. Stagl, S. Madlener, R. and Omann, I. (2009). Sustainable energy
futures: Methodological challenges in combining scenarios and participatory
multi-criteria analysis. European Journal of Operational Research, 197,
1063–1074.
Liposcak, M. Afgan, N.H. Duic, N. and da Graca Carvalho, M. (2006).
Sustainability assessment of cogeneration sector development in Croatia.
Energy, 31 (13), 2276–2284.
95
Manzella, A. (2006). Convenzione tra la Regione Toscana e l’Istituto di
Geoscienze e Georisorse del CNR per la realizzazione della campagna
geofisica triennale tramite elettromagnetismo relativa all' acquifero dell'
edificio vulcanico del Monte Amiata.
Martinez-Alier, J. Munda, G. and O'Neil, J. (1998). Weak comparability of values
as a fundation for ecological economics. Ecological Economics, 26 (3), 277286.
Munda, G. (1995). Multi-criteria Evalution in a Fuzzy Environment: Theory and
Applications in Ecological Economics. Physica-Verlag, Heiderlberg
Munda, G. (2004). Social multi-criteria evaluation: methodological foundations
and operational consequences. European Journal of Operational Research,
158 (3), 662–677.
Munda, G. (2005). Multi-criteria decision analysis and sustainable development.
In Figueira, J. Greco, S. and Ehrgott, M. (Eds.), Multiple-Criteria Decision
Analysis. State of the Art Surveys. . Springer International Series in
Operations Research and Management Science, Springer, New York, Pp 953986
Springler, Berlin
Munda, G. (2009). A conflict analysis approach for illuminating distributional
issues in sustainability policy. European Journal of Operational Research,
194 (1), 307-322.
Munda, G. Gamboa, G. Russi, D. and Garmendia, E. (2006). Social-Multi Criteria
Evaluation of Renewable Energy Sources: Two Real World Catalan
Examples. Report for the EU research project “Development and Application
of a Multicriteria Decision Analysis software tool for Renewable Energy
sources (MCDA-RES)”. Contract NNE5-2001-273. Universitat Autonoma de
Barcelona.
Munda, G. Nijkamp, P. and Rietveld, B. (1994). Qualitative multicriteria
evaluation for environmental management. Ecological Economics, 10 (2), 97112.
Papalini, C. (1989). La risorsa acqua nell’Amiata. Alcuni dati sulle sorgenti e sul
loro sfruttamento. Amiata Storia e Territorio, 5, 51-54.
Paruccini, M. (1994). Applying multiple criteria aid for decision to environmental
management. Kluwer, Dordrecht
Petty, S. (2005). Cost and Supply of Geothermal Power. Presentation held at the
"Utah Geothermal Power Generation Workshop" in [Online] Available from:
http://geology.utah.gov/emp/geothermal/ugwg/workshop0805/index.htm
[Accessed 14 April 2011]
Ricerche e Consulenze per l'Economia e la Finanza - REF. (2011). Generazione
Rinnovable e Mercato CV: scenari di breve e lungo termine. Osservatorio
Energia.
96
Roxborough, S. (2010). ViaLogy taps into renewable sector. Energyboom.
[Online] Available from: http://www.energyboom.com [Accessed 5 July
2011]
Roy, B. (1985). Méthodologie Multicritère d’aide à la Décision. Economica, Paris
Roy, B. (1990). Decision-aid and decision-making. European Journal of
Operational Research, 45 (2/3), 324-331.
Roy, B. (1991). The outranking approach and the fundations of Electre methods.
Theory and Decision, 31 (1), 49-73.
Roy, B. and Hugonnard, J.C. (1982). Ranking of suburban line extension projects
on the Paris metro system by a multicriteria method. Transportation
Reserach, 16A (4), 301-312.
Roy, B. Présent, M. and Silhol, D. (1986). A programming method for
determining which Paris metro stations should be renovated. European
Journal of Operational Research, 24 (2), 318-334.
Sanyal, S.K. (2004). Cost of geothermal power and factors that affect it. Paper
presented at the Twenty-Ninth Workshop on Geothermal Reservoir
Engineering.
Stanford,
California.
[Online]
Available
from:
http://www.geothermex.com [Accessed 19 January 2011]
Serafini, G. and Sani, G. (2007). Monte Amiata. Frammenti di storia di miniere,
minatori e lotte sociali. Quei pezzi di cinabro lungo il fosso del Siele. C&P
Adver Effigi, Arcidosso
Simon, H.A. (1976). From substatintive to procedural rationality. In Latsis, J.S.
(Ed.) Methods and Appraisals in Economics. Cambridge Univerity Press,
Cambridge, Pp 129-148
Simos, J. (1990). Evaluer l’impact sur l’environnement: Une approche originale
par l'analyse multicritè re et la négociation. Presses Polytechniques et
Universitaires Romandes, Lausanne
Terna. (2010). Dati statistici. [Online] Available from: http://www.terna.it/
[Accessed 20 April 2011]
Tuscany regional government. (2008). Piano di Indirizzo Energetico Regionale.
[Online]
Available
from:
http://www.regione.toscana.it/regione/multimedia/RT/documents/119971453
6712_Piano_indirizzo_energetico_regione_toscana.pdf
[Accessed
13
November 2009]
Tuscany regional government. (2009). Documento di monitoraggio del P.I.E.R.
[Online] Available from: http://www.regione.toscana.it/ [Accessed 18 March
2011]
Tuscany regional government. (2010). Criteri direttivi per il contenimento delle
emissioni in atmosfera delle centrali geotermoelettriche. Settore Qualità
dell'Aria, Rischi Industriali, Prevenzione e Riduzione Integrata
97
dell'Inquinamento.
[Online]
Available
from:
http://www.nextville.it/repository/normativa/954_dgr344allegato.pdf
Tuscany regional government. (2011). Inventario Regionale delle Sorgenti di
Emissione. APEX - Emissioni puntuali. Direzione PTA.
Università di Siena. (2008). Studio geostrutturale, idrogeologico e geochimico
ambientale
dell’area
amiatina.
[Online]
Available
from:
20 December 2010]
Vincke, P. (1992). Multicriteria Decision-aid. Wiley, New York
Wang, J.-J. Jing, Y.-Y. Zhang, C.-F. Shi, G.-H. and Zhang, X.-T. (2008). A fuzzy
multi-criteria decision-making model for trigeneration system. Energy Policy,
36 (10), 3823–3832.
Wang, J.-J. Jing, Y.-Y. Zhang, C.-F. and Zhao, J.-H. (2009). Review on multicriteria decision analysis aid in sustainable energy decision-making.
Renewable and Sustainable Energy Reviews, 13 (9), 2263–2278.
98
Appendix A4.1 – Summary of the interviews
Social actor
Piancastagnaio municipality
Santa Fiora branch Communist
Party
Arpat
Prospettiva Comune Piancastagnaio
WWF
Comitato per la Tutela
dell’Ambiente dell’Amiata Abbadia San Salvatore
Arcidosso municipality
Rete Comitati per la Difesa del
Territorio
Enel Green Power Ricerche
Residents’ association of Arcidosso
(no more active)
Santa Fiora Municipality
Abbadia San Salvatore
Municipality
Table A4.1: Interviews
Participants
Place
Mayor
Mountain authority office,
Arcidosso
1
Santa Fiora
Date
09/03/2011
09/03/2011
1
3
1
3
Arpat office, Siena
Piancastagnaio
Monte Labbro
11/03/2011
17/03/2011
17/03/2011
18/03/2011
Mayor
1
Town hall, Arcidosso
18/03/2011
18/03/2011
2
22/03/2011
1
Enel Green Power office,
Pisa
Arcidosso
Mayor
Mayor’s deputy
Mayor
Mountain authority office,
Arcidosso
Florence
26/03/2011
25/03/2011
05/04/2011
Appendix A4.2 – Cost structure
The operational and maintenance (O&M) costs are assumed to be 2.5% of the
capital investment for a new plant. This means that the O&M costs of a 20MW
plant amounts to €2M.
The main taxes include a corporate tax, a regional business tax and a
property tax. The first amounts to 27% of the annual profits, the second is equal to
3.9% of the annual gains (minus a €5,000 tax allowance per employee), and the
third was asked to the mayors during the interviews (currently €50,000 in
Piancastagnaio and €3,000 in Bagnore).
In addition, the company has to pay a compensation fund (including the
mining license) to the municipalities and to the regional government. This
compensation consists of the following: i) 0.13 cents Euros per KWh produced (al
least 60% of the sum is paid to the municipality where the plant is located and the
rest is proportionally distributed according to the mining license area of each
municipality), ii) 0.195 cents€ per KWh produced (to the regional government),
iii) 650€ per Km2 of the mining concession (to the regional government), iv)
6.7M€ annually (to the regional government) for all the geothermal activity that
ENEL is carrying out in Tuscany, v) 65,000€ for ten years for each new MW of
installed capacity (to the Regional government’s fund) vi) 1,25M€ for each MW
of newly installed capacity for research and innovation activities on renewable
energies 47 and for other interventions specifically indicated in the general
agreement on the exploitation of geothermal resources signed by ENEL, the
regional government and the 15 municipalities with geothermal areas in Tuscany48.
The 6.7M€ of point iv goes to the compensation fund without reference to the
individual mining license or plant, so some means must be introduced to allocate
its cost to the individual plant/area/enterprise. In this work it was decided to
proportionally allocate the total sum in relation to the license areas of the 15
municipalities in Tuscany that benefit from this fund.
For reasons of space, the complete cash flow has not been included but only
the main cost components. The time schedule of the different works is derived
from EIA reports submitted by ENEL.
Below the investments costs for each scenario are reported. The
decommission costs are assumed to be €3.6M for each plant.
47
This last component is not directly allocated to the municipalities.
For points i, ii and iii the reference is article 16 of decree 22 of 11/2/2010 (substituting law
867/86), for the others the legal reference is the aforementioned general agreement on the
exploitation of geothermal resources.
48
100
BaU
No investments are included in this scenario
Reorg
Investments are reported in Table A4.2.1 (details are derived from the EIA
reports)
Table A4.2.1
Reorganization plant investments costs (Thousands €)
Well Details
Unit
Wells drilling (3,500 m.)
5
Wells reactivation (3,500 m.)
3
Wells deepening (up to 3,500m)
1
Well drilling (1,000 m) – reinjection
1
Pipelines details49
Length (m.)
North-South pipe
3,600
PC38-PC29 pipe
1,100
PC29-PC36/3 pipe
1,300
PC36-North/South pipe
500
PC35-PC3-PC25 pipe
2,400
PC25-PC3-C pipe
2,700
PC3-PC8 heat pipe
1,900
PC3- Casa del Corto area heat pipe
2,800
Unit cost
5,500
2,750
3,500
1,500
Cost/Km
300
250
300
300
250
200
200
250
ClosingPC2
In this scenario the only investment is the heat pipe to connect PC3 to the Casa del
Corto area. Its value is taken from Table A4.2.1 and is estimated to be 700,000 €.
Reorg+BG4
This scenario involves investments for the construction of a new 40MW plant in
the west plus the same investments reported in Table A4.2.2 of the Reorg scenario.
The details of the investments for the new 40MW plant are derived from the
EIA reports submitted by ENEL.
49
Pipes have different unit costs because have different diameters
101
Table A4.2.2
BG4 Investment costs (Thousands)
Wells details
Well drilling (4,000m)
Well drilling (3,000m)
Well reactivation (4000m)
Well drilling (1000m) - reinjection
Pipelines
Two-phase steam pipe
Other investment costs
Permitting
Acquisition well site
Surface exploration for well field dev.
Acquisition plant site
Plant design
Turboexpander & generator
Power station
Cooling towers
Other civil works
Amis
Connection to grid
Others
Unit
5
1
2
2
Length (m)
6,200
Unit cost
6,000
5,000
3,000
1,500
Cost/Km
350
1,000
750
5,100
1,700
1,848
13,600
16,898
2,890
4,012
5,100
1,700
7,456
Reorg+40CC
This scenario involves the same costs expected in Reorg plus the investment costs
for the new binary cycle plant to be installed in Bagnore (reported in Table
A4.2.3)
102
Table A4.2.3
Investment costs for 40 Mw binary cycle plant (Thousands €)
Wells details
Unit
Unit cost
Production wells
6
5,000
Injection wells (1,000m)
5
1,500
Injection wells (3,000m)
3
5,000
Reactivation of production wells
2
2,500
Pipelines
Length (m.) Cost/Km
Pipes for production wells
6,000
350
Pipes for injection wells
9,000
350
Permitting
1,000
Surface exploration (exploratory drilling)
3,000
Acquisition well sites
1,250
Surface exploration (well field dev).
5,000
1,700
Plant design
2,469
17,000
Power station
30,685
Cooling system
7,803
Other civil works
4,012
Connection to grid
1,700
Others
18,638
Closing PC2+20CC
This scenario does not envisage the reorganization plan (whose costs are reported
Table A4.2.1). In Piancastagnaio the only investment would be the heat pipe to
connect PC3 to Casa del Corto.
In Bagnore a new 20MW binary cycle power plant would be installed. Its
investment costs are reported in Table A4.2.4
103
Table A4.2.4
Investment costs for 20 Mw binary cycle plant (Thousands €)
Wells details
Unit
Production wells
2
Reactivation of production wells
2
Re-injection wells (shallow reservoir)
3
Re-injection wells (deep reservoir)
1
Pipelines
Length (m.)
Pipes for production wells
4,150
Pipes for injection wells
6,000
Permitting
Surface exploration (exploratory drilling)
Acquisition wells site
Surface exploration (well field dev.)
Plant design
Power station
Cooling system
Other civil works
Connection to grid
Others
Unit cost
5,000
2,500
1,500
5,000
Cost/Km
350
350
1,000
2,000
750
3,000
1,000
1,452
10,000
18,050
4,590
2,360
1,000
10,173
Reorg+20CC
Investment costs of this scenario are the sum of the costs of the reorganization
plan (Table A4.2.1) and of the 20MW binary cycle (Table A4.2.4).
104
Appendix A4.3 – Additional results: sensitivity analysis
Table A4.3.1: Sensitivity analysis of NH3 emissions
20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
BaU
BaU
Reorg+BG4
Reorg+BG4
5
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
BaU
BaU
Reorg+BG4
Reorg+BG4
4
ClosingPC2
Bin20
Bin20
Bin20
Bin20
ClosingPC2
Bin20
ClosingPC2
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Bin20
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
Reorg+Bin20
Reorg
Reorg
Reorg
ClosingPC2
ClosingPC2
Reorg
Reorg+BG4
BaU
Reorg+BG4
BaU
Reorg+BG4
BaU
BaU
Reorg+BG4
BaU
Reorg+BG4
BaU
Reorg+BG4
3
Bin20
Reorg+Bin40
Reorg+Bin20
Reorg
ClosingPC2
Reorg+BG4
BaU
2
Reorg+Bin40
Reorg+Bin40
Bin20
Bin20
Reorg+Bin20
Reorg+Bin40
Reorg+Bin20
Bin20
Reorg+Bin20
Reorg
Reorg
Reorg
ClosingPC2
ClosingPC2
ClosingPC2
Reorg+BG4
Reorg+BG4
Reorg+BG4
BaU
BaU
BaU
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Reorg+Bin20
Bin20
3°
Reorg
Reorg
4°
Ranking
ClosingPC2
ClosingPC2
5°
Reorg+BG4
Reorg+BG4
6°
BaU
BaU
1
7°
Table A4.3.2: Sensitivity analysis of As emissions
20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
BaU
BaU
Reorg+BG4
Reorg+BG4
5
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
BaU
BaU
Reorg+BG4
Reorg+BG4
4
ClosingPC2
ClosingPC2
Bin20
Bin20
Bin20
Bin20
ClosingPC2
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
Reorg
Reorg
BaU
Reorg+BG4
BaU
Reorg+BG4
Reorg+BG4
BaU
Reorg+BG4
BaU
3
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
Reorg+BG4
Reorg+BG4
BaU
BaU
2
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
Bin20
Bin20
ClosingPC2
Bin20
Reorg+Bin40
ClosingPC2
Bin20
Bin20
Bin20
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin20
Reorg+Bin20
ClosingPC2
Reorg+Bin20
Reorg+Bin20
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
ClosingPC2
Reorg+Bin20
Reorg
ClosingPC2
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
Reorg
ClosingPC2
Reorg
Reorg
ClosingPC2
Reorg
Reorg
Reorg
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Reorg+Bin20
Bin20
3°
Reorg
Reorg
4°
Ranking
ClosingPC2
ClosingPC2
5°
Reorg+BG4
Reorg+BG4
6°
BaU
BaU
1
7°
Table A4.3.3: Sensitivity analysis of municipality revenues
105
20
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+BG4
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
Reorg
Reorg
ClosingPC2
ClosingPC2
BaU
BaU
4
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+BG4
Bin20
Reorg+Bin20
Reorg+Bin20
Bin20
Reorg
Reorg
ClosingPC2
ClosingPC2
BaU
BaU
3
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+BG4
Reorg+Bin20
Bin20
Reorg+Bin20
Reorg+BG4
Reorg+Bin20
Bin20
Bin20
Reorg
Reorg
Reorg
ClosingPC2
ClosingPC2
ClosingPC2
BaU
BaU
BaU
2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
Bin20
Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Bin20
Bin20
Reorg
Reorg
Reorg+BG4
Bin20
Reorg
Reorg
Reorg+BG4
ClosingPC2
Reorg+BG4
Reorg
Reorg
ClosingPC2
Reorg+BG4
Reorg
Reorg+BG4
ClosingPC2
ClosingPC2
ClosingPC2
Reorg+BG4
ClosingPC2
ClosingPC2
BaU
BaU
BaU
BaU
BaU
BaU
BaU
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Reorg+Bin20
Bin20
3°
ClosingPC2
ClosingPC2
5°
Reorg+BG4
Reorg+BG4
6°
BaU
BaU
7°
1
Reorg
Reorg
4°
Ranking
Table A4.3.4: Sensitivity analysis of GHGs avoided
20
Reorg+Bin40
Reorg+Bin20
Reorg+BG4
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
3
Reorg+Bin40
Reorg+Bin20
Reorg+BG4
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
Bin20
Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Bin20
Bin20
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Reorg+BG4
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+BG4
Bin20
Bin20
Bin20
Reorg+Bin40
Reorg+Bin40
1°
Bin20
Reorg+Bin20
2°
Ranking
Reorg+Bin20
Bin20
3°
Reorg+Bin40
Reorg+Bin20
Bin20
1
1°
2°
3°
106
Table A4.3.5: Sensitivity analysis of H2S (s) and Hg (g) emissions
H2S:20
Hg: 20
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
Reorg+BG4
Reorg+BG4
BaU
BaU
H2S: 2
Hg: 3
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin20
Reorg+Bin20
Reorg
Reorg
Reorg+BG4
Reorg+BG4
BaU
BaU
H2S: 3
Hg: 2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
ClosingPC2
ClosingPC2
Reorg+Bin40
Bin20
Bin20
Bin20
Bin20
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
ClosingPC2
Reorg+Bin20
Reorg+Bin20
Reorg+Bin40
Bin20
Reorg+Bin40
Bin20
ClosingPC2
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
ClosingPC2
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
ClosingPC2
Reorg
ClosingPC2
Reorg
Reorg
Reorg
Reorg
Reorg
Reorg
ClosingPC2
Reorg
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
ClosingPC2
Reorg+Bin40
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
Bin20
Bin20
Bin20
Bin20
1°
Bin20
Bin20
Reorg+Bin40
ClosingPC2
Bin20
Bin20
ClosingPC2
Reorg+Bin40
Reorg+Bin40
Reorg+Bin40
2°
Reorg+Bin40
ClosingPC2
Bin20
Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin40
ClosingPC2
Reorg+Bin20
Reorg+Bin20
3°
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg+Bin20
Reorg
ClosingPC2
Reorg+Bin20
Reorg+Bin20
Reorg
ClosingPC2
4°
Ranking
Reorg
Reorg
Reorg
Reorg
ClosingPC2
Reorg
Reorg
Reorg
ClosingPC2
Reorg
5°
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
Reorg+BG4
6°
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
BaU
7°
H2S: 2
Hg: 2
107
5
Conclusions
When we analyze the replacement of fossil fuels with alternative energies, we
necessarily face trade-offs and conflicting arguments with regard to energy quality,
conservation of natural resources, environmental health and societal preferences.
Assessments which comprehensively take into account the multi-dimensional
aspects and multi-scale implications must be applied.
This thesis has attempted to show how these integrated assessments can be
applied in two cases. In Chapter 2 the inadequacy of the reductionist approaches
for the assessment of sustainability in complex systems was illustrated along with
the desired characteristics of proper sustainability assessment strategies. In the
following chapters two cases were analyzed. The Brazilian biodiesel case was
chosen because biodiesel production and use is a fairly recent enterprise in Brazil
(while the Brazilian ethanol experience is much older). An intensive debate is
ongoing on whether to expand biodiesel production and use. However, to the best
of my knowledge, no integrated assessment of the constraints and implications of
biodiesel expansion currently exists. Such an integrated assessment is the subject
of Chapter 3. A Multi-Scale Integrated Analysis of Societal and Ecosystem
Metabolism (MuSIASEM) was applied to assess the feasibility space of a given
scenario. The geothermal case was presented in Chapter 4 where a different kind
of assessment – a social multi-criteria evaluation (SMCE) – was proposed to
explore the possible alternatives of geothermal development in a specific area.
Both cases are characterized by the presence of contrasting values and
interests, uncertainties on the effects of the given renewable energy expansion and
by a real urgency of the concerned decisions. These are cases where post-normal
science solving strategies should definitely be used.
This thesis has shown how MuSIASEM can be used to shed light on the
possible constraints affecting a given choice. A given policy option is analyzed
using the parallel use of non equivalent descriptions related to different scales of
analysis and different scientific readings. The definition of the feasibility and
desirability is obtained by analyzing the reciprocals effects of the chosen
indicators across hierarchical scales, and economic and biophysical constraints.
The proposed methodology is thus used as a tool to support discussions on the
(un)sustainability of the specific policy choice.
Chapter 5 - Conclusions
The second method, i.e. SMCE, was applied in a context where the
legitimacy of scientific research was often contested by some of the parties
involved. In this context, the applied method explores socially constructed
solutions using a multi-criteria approach. SMCE is thus used as a tool to support
decision-making between various alternatives.
The proposed methods have shown how various relevant aspects can be
simultaneously taken into account. Geothermal power and biodiesel were thus
assessed with the following results: i) an analysis of the pros and cons of the
specific alternatives was proposed with their implications and constraints ii)
relevant indicators representing the performance of the alternative(s) were
identified for different scales and different non reducible dimensions iii) different
trade-offs between environmental, economic and social dimensions were
pinpointed iv) the discussion regarding the present situation and the desirability
and feasibility of specific future scenarios was contributed to.
The scale issue results to be critical in both cases and methods. It is
explicitly addressed in MuSIASEM and it implicitly affects the results of SMCE.
In the geothermal case, the relevant scale was arbitrarily chosen as regional.
However, other scales could have been considered. And if scales are changed, the
social actors (whose preferences represent the main input for the definition of
criteria) are also different. Consequently, the characterization of the chosen
criteria would also be different. For instance, where a global scale was chosen, a
life cycle analysis would be more appropriate for the characterization of the
geothermal MWh emissions. Thus all emissions caused by the construction of the
different equipment involved in a power plant could have been included along
with all the materials consumed. A European or a national scale could also have
been chosen. In fact, all the electricity that is not produced in Italy will have to be
purchased in Europe. In any case, the social actors whose preferences are
translated using relevant criteria would certainly have been different.
In both cases one main issue is how the given problem is structured. This
research phase helps the analyst to make Allen’s key questions on sustainability
(mentioned in the introduction chapter) transparent: sustainability of what,
sustainability for whom, sustainability for how long and sustainability at what
cost? In any case, some methodological and epistemological concerns will always
arise. Who decides what the relevant aspects are? Whose concerns should be
considered? How important are they? How can they be quantified or described?
The way these questions are answered, unavoidably involves some degree of
subjectivism. However, this is not necessarily negative. In fact models that claim
to represent an objective reality of a given complex system involve two crucial
problems: epistemological cheating and practical rigidity. Concerning the first
point, an objective representation of a complex system does not exist almost by
definition because all modeling exercises are a focusing and narrowing device,
whereby some variables, some scales, and a specific time frame are kept (the ones
of interest for the specific point of view) and others are excluded (which may
109
however be of interest for other points of view). Concerning the practical aspect,
models devised to avoid any personal subjectivity are often so rigid that they
barely adhere to the situation which the analyst intends to model. Most likely the
way subjectivism affects the usefulness and reliability of a model depends on the
ethical behavior and capacity of the analyst to structure the problem.
In the Brazilian biodiesel case, the problem was structured in order to show
the consequences of the given policy: i) at four different hierarchical scales ii), in
economic terms iii), in energy consumption terms iv), and emphasizing land use
implications. Other perspectives could also have been included. For instance, the
effects at a farm level would have been of interest for the individual farmer.
Alternatively, the analysis could have focused on specific geographical scales (e.g.
a given Brazilian region), or on the effects on other natural resources, such as
water.
In the geothermal case the proposed criteria were chosen to reflect the
points of view of specific stakeholders. Other variables could probably have been
included if the number of stakeholders was increased or reduced. Or more simply,
the translation of stakeholder preferences in the specific criteria used could have
been done in a different way.
These examples show how the analyst’s subjectivism could have provided
other representations and other conclusions. This is not a proclamation for
anarchy in research methodology. Rather, the examples highlight that: i) the way a
decision is achieved is as important as the specific aspects of the chosen decision,
ii) transparency should not simply be advocated but put into practice in every step
of the research process, iii) the decision process should include negotiation and
dialogue with those who have a stake in the specific problem.
Socio-economic and ecological systems are complex. Our capacity to
foresee their evolution is extremely limited, especially when ecological systems
and socio-economic systems interact with each other. Their interaction creates
reciprocal feedback, non-linear dynamics, legacy effects, time lags, heterogeneity,
and surprise (Liu et al., 2007). Moreover, when facts are uncertain, values are in
dispute, stakes are high and decisions urgent (to recall the post-normal science
definition), decision making cannot be limited to a one-shot activity. Instead, it
becomes a continuous learning (and adaptive) process with a cyclical nature: the
problem is structured, a given policy option is characterized by relevant indicators
and criteria, an evaluation is obtained. Based on the computed results the problem
can again be structured differently, the alternative (or more than one alternative) is
again characterized and evaluated, and so on. The alternatives considered, their
perceived impact and the way the problem is structured may suddenly be judged
in a completely different way. In this sense, the application of highly flexible
evaluation procedures such as MuSIASEM and SMCE should be welcomed.
The cases applied in this thesis demonstrate that MuSIASEM and SMCE
can contribute to enhancing these cyclical endeavors by making how the given
problem was characterized transparent, by contributing to the characterization of
110
the given alternative(s), and by providing a consistent framework for evaluating
heterogeneous variables.
Given the underlying uncertainly in sustainability problems and the
presence of unavoidable subjectivity in the analyst’s work, the only way to gain
legitimacy in public decision making is through consistent stakeholder
engagement and transparency. The geothermal case clearly shows that facts
concerning crucial issues such as health impacts and water conservation remain
contested in spite of the ever-growing scientific research on these topics. The
apparent paradox of this case is that the more research is performed on the
impacts of geothermal exploitation, the more doubts are raised. In a kind of
infinite loop, policy makers keep promoting more research.
The analysis included in this thesis does not provide a definite answer so
that decision makers can avoid assuming responsibility for their choices. On the
contrary, this thesis stresses that science cannot either legitimize policies in
conditions of uncertainty and different values, nor can it relieve policy makers
from taking very difficult decisions. In conditions of high uncertainty and
conflicting interests, the question is how to base decision making on the best
honest information regarding consequences, uncertainty and conflicting values.
In this sense, the proposed approaches can contribute to fostering a shared
knowledge regarding the specific sustainability of energy problems by keeping
actors informed so that compromise solutions can hopefully be reached.
The two alternative energies proposed need some protection if they are to be
expanded. Both biodiesel in Brazil and geothermal power in Italy are incentivized
through a quota system (a blending system in the former case and green
certificates in the latter). In any case, the burden of such protection mechanisms is
on consumers. Very often, alternative energies are not economically competitive
with fossil fuels. If their use is to be expanded they need to be protected in some
way. In this period of growing financial constraints, integrated assessments are of
paramount importance in order to make the best use of direct and indirect
subsidies.
Areas of improvement could focus on how to make the applied
methodological tools more stakeholder friendly. Graphic techniques could be used
in order to make results easier to communicate.
It is also important for further research to work on how MuSIASEM and
SMCE can be integrated into a more unified framework and in individual case
studies. Once alternatives are thus developed as in SMCE, their implications (and
feasibility space) could be assessed by MuSIASEM. Some alternatives could be
ruled out and the indicators generated through MuSIASEM could be plugged into
the multi-criteria impact matrix.
One innovative methodological approach introduced in this thesis is the
coupled use of input-output analysis (IOA) and MuSIASEM. Using IOA,
economic flows were generated to feed the MuSIASEM framework. It would also
be interesting to explore the combined use of multi-criteria evaluation and IOA.
111
The increasing availability of data sets for physical flows of energy, materials and
emissions can be complemented with monetary transactions of input-output tables
at regional and national levels. This kind of physical IOA was applied by Leontief
(1970), Leontief and Ford (1972), Bullard and Herendeen (1975), Griffin (1976),
Carter (1976), Polenkse and Lin (1993), Duchin and Steenge (1999), Nakamura
and Kondo (2002), Gutmanis (1975), Schäffer and Stahmer (1989), Lenzen et al.
(2004), among many others.
Environmental adjusted forward and backward linkages could be developed
(as done in Lenzen, 2003) for different environmental effects such as GHG
emissions, emissions at regional and local levels (e.g. NOx or SO2), water and
energy consumption, along with traditional forward and backward linkages
measured in monetary terms. The impact of investing or stimulating final
demands in different sectors could then be evaluated using a multi-criteria
analysis. Economic sectors would therefore become the set of alternatives of
multi-criteria analyses and the different economic and environmental adjusted
backward and forward linkages would represent the elements of the impact matrix.
MuSIASEM and SMCE are certainly not a panacea, but they can help in
providing a multi-perspective assessment of the proposed alternatives. The two
methods compete for a niche in the market place of sustainability appraisal along
with many other methodological approaches (cost benefit analysis, life cycle
analysis, material flow accounting, etc). No method can be considered the best in
conditions of high uncertainty and conflicting stakes. Moreover, both in science
and in decision making, value judgments cannot be avoided.
In a world constrained by energy limitations, environmental problems and
economic concerns, the implementation of more pluralistic and multi-criteria
approaches should be essential when assessing alternatives. In this way, science
can contribute to ensuring that decision-making is more informed and more
transparent. As scientists, this is probably all we can hope for.
112
References
Carter, A.P. (1976). Energy and the environment. a structural analysis. Brandeis
University Press, Hanover, N.H.
Duchin, F. and Steenge, A.E. (1999). Input-output analysis technology and the
environment. In Berg, J.C.J.M.v.d. (Ed.) Handbook of Environmental and
Resource Economics. Edward Elgard, Cheltenam, UK, Pp 1037-1059
Griffin, J. (1976). Energy: input-output modelling. Electric Power Research
Institute, Palo Alto, CA
Gutmanis, I. (1975). Input-output models in economic and environmental policy
analyses. Proceedings of the IEEE, 63 (3), 431-437.
Lenzen, M. (2003). Environmentally important paths, linkages and key sectors in
the Australian economy. Structural Change and Economic Dynamics, 14 (1),
1-34.
Lenzen, M. Pade, L.-L. and Munksgaard, J. (2004). CO2 multipliers in multiregional input-output models. Economic Systems Reasearch, 16, 391-417.
Leontief, W. (1970). Environmental repercussions and the economic structure: an
input-output approach. The Review of Economics and Statistics, 52 (3), 262271.
Leontief, W.W. and Ford, D. (1972). Air pollution and the economic structure:
empirical results of input-output computations. In Brody, A. and Carter, A.P.
(Eds.), Proceedings of the Fifth International Conference on Input-Output
Techniques, Geneva. Elsevier, Geneva, Pp 9-30
Liu, J. Dietz, T. Carpenter, S.R. Alberti, M. Folke, C. Moran, E. Pel, A.N.
Deadman, P. Kratz, T. Lubchenco, J. Ostrom, E. Ouyang, Z. Provencher, W.
Redman, C.L. Schneider, S.H. and Taylor, W.W. (2007). Complexity of
Coupled Human and Natural Systems. Science, 317 (5844), 1513-1516.
Nakamura, S. and Kondo, Y. (2002). Recycling, landfill consumption, and CO2
emission: analysis by waste input–output model. Journal of Material Cycles
and Waste Management, 4 (1), 2-11.
Polenske, K.R. and Lin, X. (1993). Conserving energy to reduce carbon dioxide
emissions in China. Structural Change and Economic Dynamics, 4 (2), 249265.
Schäfer, D. and Stahmer, C. (1989). Input-output models for the analysis of
environmental protection activities. Economic Systems Reasearch, 1 (2), 203228.
113

Tesi defintiva10 - international ph.d. in management

Transcrição

Documentos relacionados

determination of thermal and oxidation stability of sunflower

STUDIE OF THE SOLUBILIZATION OF THE SOYBEAN LECITHIN

Mozambique and the Feasible Development of the Geothermics

TUMEO 3.1.p65

Untitled - Repórter Brasil

sergio schlesinger - Portada

Human Rights violations and environmental destruction through

The Impacts of Soybean Cultivation on Brazilian Ecosystems

Eduardo Chagas Ferreira da Silva