APPLICATIONS AND IMPLEMENTATION

Industrial Symbiosis in
Kalundborg, Denmark
A Quantitative Assessment of Economic and
Environmental Aspects
Noel Brings Jacobsen

Keywords
Summary
by-product synergy
cascading As a subdiscipline of industrial ecology, industrial symbiosis
cogeneration is concerned with resource optimization among colocated
industrial ecosystem companies. The industrial symbiosis complex in Kalundborg,
recycling Denmark is the seminal example of industrial symbiosis in the
water reuse
industrial ecology literature. In spite of this, there has been no
in-depth quantitative analysis enabling more comprehensive
e-supplement available on the JIE understanding of economic and environmental performances
Web site connected to this case. In this article some of the central
industrial symbiotic exchanges, involving water and steam, in
Kalundborg are analyzed, using detailed economic and envi-
ronmental data. It is found that both substantial and minor
environmental benefits accrue from these industrial symbiosis
exchanges and that economic motivation often is connected
to upstream or downstream operational performance and
not directly associated with the value of the exchanged by-
product or waste itself. It is concluded that industrial symbiosis,
as viewed from a company perspective, has to be understood
both in terms of individual economic and environmental per-
formance, and as a more collective approach to industrial
sustainability.

Address correspondence to:

Noel Brings Jacobsen
Roskilde University
P.O Box 260
4000 Roskilde, Denmark
<[email protected]>

© 2006 by the Massachusetts Institute of

Technology and Yale University

Volume 10, Number 1–2

http://mitpress.mit.edu/jie Journal of Industrial Ecology 239

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Introduction change relationships between the companies in

question. The identifiable quantitative data on
Industrial ecology operates at three different the case differ in scale and scope (see table A of
levels, ranging from the global level through the the electronic supplement [e-supplement], avail-
interfirm level to the level of the individual facil- able on the JIE Web site); in particular, data on
ity (Chertow 2000). The level directly examined the economic aspects of the IS exchanges are very
here—the interfirm level—has been viewed in limited in the current literature.
terms of various models and terminologies, rang- The narrow objective of this article is to rem-
ing from eco-industrial parks (Côte and Cohen- edy these factual inadequacies by describing and
Rosenthal 1998), industrial symbiosis (Chertow analyzing some of the central quantitative eco-
2004) and industrial ecosystems (Côte and Hall nomic and environmental parameters related to
1995), to islands of sustainability (Wallner and the Kalundborg IS complex. The focus is on
Narodoslawsky 1996), industrial recycling net- water- and steam-related IS exchanges, with a
works (Schwarz and Steininger 1997), and by- view to evaluating the economic and environ-
product-synergies (Forward and Mangan 1999). mental implications of these exchanges. The
Within this framework of interfirm relationships, evaluation is based on a quantitative data record
industrial symbiosis (IS) can be categorized as spanning the period 1990–2002.
a concept of collective resource optimization The evaluation shows that some substantial
based on by-product exchanges and utility shar- and some minor environmental benefits are gen-
ing among different colocated facilities. The ex- erated by these IS relationships, and that the
plicit definition of IS is given by Chertow (2004): various water- and steam-related exchanges are
“Industrial symbiosis engages traditionally sepa- closely interconnected in a system based on
rate industries in a collective approach to com- water/energy cascading, substitution, diversifica-
petitive advantage involving physical exchanges tion, and utility-sharing.2 The question of eco-
of materials, energy, water and/or by-products. nomic viability is discussed in relation to these
The keys to industrial symbiosis are collabora- findings, and it is argued that IS relationships
tion and the synergistic possibilities offered by imply some basic direct and indirect economic
geographic proximity” (Chertow 2004, 2). benefits that make IS a reasonable approach to
Within the debate about IS, a very large num- interindustrial collaboration, with a built-in en-
ber of references have been made to one particu- vironmental effect.
lar case: the IS complex in Kalundborg, Denmark When the economic and environmental sig-
(Ehrenfeld and Gertler 1997; Côte and Cohen- nificance of the water-related IS exchanges
Rosenthal 1998; Esty and Porter 1998; Hardy is evaluated, a clearer picture appears of the
and Graedel 2002; Ehrenfeld and Chertow 2002; decision-making parameters and the reasons be-
Chertow and Portlock 2002; Brings Jacobsen and hind the operationalization of these IS ex-
Anderberg 2004). This case has been seen either changes.3 The implied broader objective of this
as a paradigmatic role model for IS or—on the article is thus to understand the operational rai-
contrary—as an isolated phenomenon where a son d’être of IS from the company perspective and
number of companies have coincidentally been thereby to contribute to the discussion of how to
locked into a web of waste, water, and energy optimize material flows at the interfirm level, as
exchanges based on contractual dependency. advocated by industrial ecology.
In spite of the fact that the Kalundborg IS
complex has played a central role in the discus-
sion of how to put the theoretical principles of
Methodology for Estimation of
IS and industrial ecology into practice (Sterr and
IS Environmental and Economic
Ott 2004), no detailed quantitative analysis of
Significance
the case has ever been conducted.1 A literature The principles of industrial ecology and indus-
review shows an immense number of references trial symbiosis predict that turning waste output
to the case over the past few years, but none from one facility into raw material for another fa-
goes much beyond mere description of the ex- cility will lead to environmental benefits caused

240 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

by a reduced intake of virgin material and/or The data for this article are based on qualita-
reduced emissions (Graedel and Allenby 1995; tive key-informant interviews, internal and ex-
Chertow 2004). The environmental effects of the ternal documents from the companies involved,
selected water-related IS exchanges are there- and public statistical material. Data references are
fore analyzed in terms of their ability to re- given in tables A-F in the e-supplement.
duce the intake of high-quality water by means
of water substitution and water cascading. The
Status and Scope of the IS
theoretical principle of cascading is extensively
Complex in Kalundborg
described by Côte and Hall (1995), and Sirkin
and Houten (1994) and is discussed by Con- The development of industrial symbiosis in
nelly and Koshland (1997, 2001). It is used here Kalundborg, Denmark has been described as
in a simplified outline version as proposed by an evolutionary process in which a number of
Fraanje (1997) where cascading involves start- independent by-product exchanges have grad-
ing with high resource quality increasing lifetime ually evolved into a complex web of symbi-
per application, overall lifetime, and minimiz- otic interactions among five colocated compa-
ing quality loss per application. On the basis of nies and the local municipality (Ehrenfeld and
these principles, the selected water-related IS ex- Gertler 1997; Ehrenfeld and Chertow 2002).
changes are generally described as a number of The companies involved in industrial symbiosis
downward water-cascading sequences and the re- include a 1,300-MW (2002) power plant (As-
sulting amount of water saved is estimated and næs),5 an oil refinery (Statoil A/S), a biotech
discussed. and pharmaceutical company (Novo Group),
The steam/heat-related IS exchanges are an- a producer of plasterboard (Gyproc Nordic
alyzed in terms of the cogeneration effect4 and East), and a soil remediation company (Soilrem
the net reduction of emissions of carbon dioxide A/S).
(CO2 ), sulfur dioxide (SO2 ), and nitrogen oxides As illustrated in figure 1, the various mate-
(NOx ) are illustrated in a hypothetical compari- rial flows among the companies are based either
son with stand-alone production. The engineer- on water, solid waste, or energy exchanges. In
ing calculations of the cogeneration effects have this system, wastewater and cooling water from
not been done by the author, but are based on data the refinery are reused at the power plant: the
and calculations from the companies involved, wastewater for secondary purposes, the cooling
and as such are based on appropriate engineering water as feeder water for the boilers producing
methodology and references (see table F of the steam and electricity, and also as input water
e-supplement and acknowledgments). for the desulfurization process. The desulfuriza-
The economic aspects of the same exchange tion process in turn produces industrial gypsum
relationships are estimated and discussed as a used in the production of plasterboard at the
combination of investments at the time of ini- colocated Gyproc factory, thereby partly replac-
tiation, and direct and indirect economic savings ing the use of natural gypsum. The cogenerat-
related to upstream or downstream production- ing power plant also produces heat for the town
related issues. The direct economic savings are of Kalundborg and steam for the Novo facility
usually caused by avoided discharge fees or dis- and the Statoil refinery. The Novo facility is
posal costs or by reduced prices achieved by only supplied with steam from the power plant,
substitution. The indirect economic benefits are whereas the refinery has production-related in-
related to avoided investments, increased flexibil- house steam generation capacity, partly sup-
ity, or supply security. Thus the economic aspects plied by preheated boiler water from the power
of the IS projects are estimated and discussed as plant in a total-supply-security system. In ad-
a combination of direct cost reductions, real in- dition, heated cooling water from the conden-
vestments in relation to alternative avoided in- sation process at the power plant is piped off
vestment scenarios, and estimated payback times to a nearby fish farm, thereby increasing the
for the different exchange projects at the time of efficiency in the farm, as the heated cooling
project initiation. water ensures full-scale production of the fish

Brings Jacobsen, Industrial Symbiosis in Kalundborg 241

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 1 The industrial symbiosis, Kalundborg (in 2002). Source: The Center for Industrial Symbiosis,
Kalundborg (Hansen 2003).

throughout the year. Finally, solid by-products The Water and Steam
such as fly ash from coal combustion, sludge Exchanges in the Kalundborg IS
from public wastewater treatment, and biomass Complex
from biogenetic fermentation at the Novo facil-
ity are recycled in various ways, both locally and The Kalundborg region has a large groundwa-
nonlocally. ter deficit, and groundwater supplies have gradu-
In total, industrial symbiosis in Kalund- ally dwindled over a period of 20 years as the local
borg counts—depending on the definition— water-consuming industries have expanded in
approximately 20 different by-product exchanges size and consumption. As a result of these devel-
in operation, a number of potential projects, opments a number of public/private initiatives for
and a number of projects closed down as saving groundwater have been initiated over the
markets and technological innovations have last few decades. These public/private initiatives
developed. can be categorized into three overall strategies
Following the definition of IS as proposed by by which a number of water/steam-related
Chertow (2004), the symbiotic relationships con- industrial symbiosis projects can be identified.
cerned with water and steam have been cho- One strategy has been to replace groundwater
sen for a dedicated in-depth study, because these with surface water in the most water-consuming
exchanges illustrate a clear IS business prac- industries (1961-). A second strategy has been to
tice based on geographical proximity, by-product optimize internal water use and diversify external
reuse and business-to-business resource optimiza- water sources in the water-consuming industries
tion. (c. 1975-). A third strategy has been to upgrade

242 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Figure 2 Selected industrial symbiosis water flows in Kalundborg (m3 /yr and GJ/yr) in 2002. Source: Green
Accounts from Statoil, 2002, Novo Group, 2002, and Asnæs power plant, 2002 and information from the
companies obtained through the Kalundborg Center for Industrial Symbiosis (Hansen 2004). WT =
wastewater treatment; FGD = fluegas desulfurization; mill. = million. One gigajoule (GJ) = 109 joules
(J, SI) ≈ 2.39 × 105 kilocalories (kcal) ≈ 9.48 × 105 British thermal units (BTU); one cubic meter (m3 , SI) =
103 liters (L) ≈ 264.2 gallons (gal).

surface water to drinking water quality and to increase, stabilization, or decrease in amounts
import groundwater to the Kalundborg region exchanged among the different participating
from adjacent regions (1997-).6 companies.
These different strategies have created a di- The importance of these water-related symbi-
versified water supply system in the region, based otic flows in relation to the total water input in
on close cooperation among the various water- the different symbiotic facilities is considerable,
consuming industries. Figure 2 presents a sec- as shown by table C of the e-supplement. More
tion of this diversified system where the local than 95% of the water input at the power plant
water flows are related to the water-consuming is part of the symbiotic network, whereas 98% of
symbiosis industries. From this, in 2002 a total the water input for the refinery is symbiotic in
input of 1.9 million cubic meters (m3 ) ground- character, as is approximately 20% for the Novo
water7 and 2.7 million m3 surface water, and facility. The steam/heat supply accounts for more
a total industrial discharge of 3.6 million m3 than 50% of the energy input at the Novo facility,
of water can be observed, excluding salty sea- whereas around 4% of the energy input for the re-
water discharges. During the period 1990–2002, finery is part of the steam-related symbiotic activ-
these different flows have undergone some signif- ities (in 2002). As can be observed from figure 2,
icant quantitative changes, as is shown in table B the central water- and steam-related IS projects
of the e-supplement. The use of surface water are to a great extent centered on the power plant.
has almost doubled, whereas use of cooling wa- Thus an analysis of the water consumption and
ter, wastewater, and boiler water have shown an the historical water prognosis for this facility is a

Brings Jacobsen, Industrial Symbiosis in Kalundborg 243

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

first step toward evaluating the operational raison These IS arrangements have made it possible to
d’être of the water and steam-related IS projects replace and diversify the intake of water to the
and their present economic and environmental power plant and to optimize the energy efficiency
significance. of the facility. By means of these IS relationships,
Table 1 illustrates how over the last 12 years the power plant has managed to replace ground-
the power plant has gradually managed to reduce water with surface water, surface water with cool-
the intake of groundwater by drawing on many ing water, and cooling water with wastewater.
different sources, including groundwater, surface Parallel with this gradual substitution/cascading,
water, cooling water, and wastewater, in cooper- the energy content in the water is used at different
ation with the colocated companies. levels ranging from high-energy steam exchanges
The initial cause of this change at the power to low-energy exchanges of salty cooling water.
plant was the installation of a number of water- In conjunction with these arrangements, utility
consuming processes. It was estimated that the in- sharing is a third factor used to further optimize
stallation of a wet flue gas desulfurization plant in economic and environmental performance (see
1992 would require approximately 450,000 m3 /yr the next section).
of water, and the potential installation of another The forecast need for 3.0 million m3 of
desulfurization plant by 1995/1996 would require high-quality water at the power plant by the
another 400,000 m3 /yr. Furthermore, the export year 2000 has been turned into consumption
of steam to the refinery and the pharmaceutical of approximately 1.2 million m3 of low-quality
plant was expected to increase the run-through water (in 2002) as a consequence of these
of process water by 222,000 m3 /yr (in 2000), IS exchanges among the companies, combined
thereby creating an equivalent need for raw water with reduced production capacity at the power
input. plant.8
Table 2 shows the complete forecast for water Thus the various IS exchanges in current op-
supply and increased run-through of process water eration help to optimize the water/energy flows
at the power plant (1988–2000). It was estimated and economic performance at the power plant,
that the water intake would increase from the and the operational raison d’être of the IS ex-
mid-1980s over the next 20 years by a factor of 3 changes seems to be primarily based on this fact.
from approximately 1.0 million m3 /yr to approx. To acquire a clear quantitative picture of these IS
3.0 million m3 /yr. Thus, to secure a steady and optimization processes, each water/steam IS ex-
cheaper supply of water for the production pro- change is analyzed further below in terms of eco-
cess, a radical shift was needed from high-quality nomic and environmental significance and per-
water to low-quality water where possible. This formance.
was further prompted by the increasing intake of
water by the colocated companies, especially the
pharmaceutical plant, Novo (see table 1).
Analysis of Water/Steam IS
Five different sources, all of which included a
Exchanges Involving the Asnæs
symbiotic element, were to form the basis of the
Power Plant
water supply to the power plant. A first source
was increased use of surface water jointly with As indicated by figure 2, the core water/steam
the refinery and the Novo facility. A second IS exchanges associated with the power plant
source was use of cooling water from the refin- involve exchanges of surface water, cooling
ery. A third source was use of wastewater from water, wastewater, steam/heat, pretreated boiler
the refinery. The last two proposed sources were water, and salty cooling water. Table D of the
reused wastewater from the pharmaceutical com- e-supplement summarizes the environmental and
pany Novo and from the public wastewater treat- economic significance of these exchanges, each
ment plant. of which is evaluated in the following sections.
Today, the first three of these original IS ex- Table D also includes the performance of a num-
changes are in operation, in close conjunction ber of other IS exchanges in the Kalundborg IS
with the downstream steam-related IS exchanges. complex.

244 Journal of Industrial Ecology

 Table 1 Water consumption at Asnæs power plant, Novo plant, and Statoil plant, 1990–2002, in thousand cubic meters (1,000 m3 )
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Asnæs Power Plant in Kalundborg

Groundwater 335 412 356 313 119 73 61 51 39 36 52 37 51
Brings Jacobsen, Industrial Symbiosis in Kalundborg

Surface water 105 87 167 407 791 733 522 433 926 827 623 578 686
Cooling/waste water 715 705 802 847 849 820 753 634 580 518 315 647 492
Novo Group in Kalundborg
Total water intake 1,400 1,600 1,700 1,900 1,900 2,200 2,500 2,500 2,200 2,300 2,300 2,600 2,700
Statoil Plant in Kalundborg

A P P L I C AT I O N S A N D I M P L E M E N TAT I O N
Total water intake (*) (*) (*) 1,200 1,300 1,300 1,300 1,300 1,300 1,300 1,300 1,600 1,600

Source: Green Accounts from Statoil refinery (covering 1997–2002), Novo Group (covering 1994–2002), Asnæs power plant (covering 1996–2002) and internal publication from Asnæs
power plant: “Asnæs power plant—Recycling of wastewater (1992/93).” Novo Group water intake includes water imported as steam.
Notes: (*) = Not available. One cubic meter (m3 , SI) = 103 liters (L) ≈ 264.2 gallons (gal).
245
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Table 2 Water consumption (1988) and water prognoses (2000) for Asnæs power plant
1988 (consumption) 2000 (prognosis)
m3 m3

Boiler water Unit 1–5 — 390,000

Boiler water Novo 192,000 400,000
Boiler water Statoil 136,000 150,000
Boiler water district heating 5,000 10,000
Sluice water unit 1–5 — 340,000
Granulation/dust — 830,000
Desulfurization unit 3–5 — 850,000
Sanitary water — 30,000
Total internal use 736,000 2,400,000
TOTAL consumption including water loss to 1,069,000 3,000,000
IS activities

Source: Future water supply for Asnæs power plant, internal document, Asnæs power plant, June 1989.
Notes: (—) = Consumption not explicated in this category, but accumulated as part of the total consumption.

Replacement of Fresh Groundwater with whereas the delivery of cooling water from the
Surface Water refinery to the power plant is based on water sub-
The power plant takes in surface water jointly stitution and cascading. In this symbiotic rela-
with the other symbiosis industries from a nearby tionship, the power plant replaces surface water
lake. The surface water is upgraded and thereby with cooling water piped from the refinery. Thus
made viable as boiler water for steam production. the refinery takes in surface water (which replaces
Thus, the intake of surface water at the power groundwater) and uses the surface water as cool-
plant has had a direct effect on the intake of fresh ing water; then the cooling water is piped to
groundwater, amounting to an annual saving of the power plant. The power plant treats and up-
686,000 m3 in 2002, or more than 6.9 million grades the cooling water to boiler-water quality in
m3 for the period 1990–2002 (table 1). In the an annual amount of approximately 500,000 m3
same period all of the symbiosis industries saved and uses the water in steam production. This
more than 30 million m3 of groundwater, as seen IS exchange has meant a saving of more than
in table B of the e-supplement, as a result of the 7.6 million m3 of surface water for the selected
replacement of groundwater with surface water. period 1990–2002 in terms of reduced power
Table E of the e-supplement shows the publicly plant surface water intake. The cooling water ex-
available price difference between surface water change required an investment of approximately
and groundwater in the period 1990–2002. 2.0 million DKK for technical installations and
Based on this, the replacement of groundwater piping, shared by the partners at the time of
with surface water led to an annual saving of project initiation. An additional investment of
approximately 7.6 million Danish Kroner (DKK) 40 million DKK was required for a pretreatment
or $960,000 U.S. for 20029 by the power plant, facility to treat both the surface water and the
or more than 35 million DKK for the selected cooling water to boiler-water quality at the power
period. A joint private investment of 32 million plant. The price of cooling water from the refin-
DKK in utility sharing in the present pipeline ery is linked to the price of surface water by a 50%
system has meant a reasonable payback time for discount. This means that the refinery gained a
the individual symbiosis industries involved. direct saving of approximately 1.8 million DKK
or approximately $228,000 U.S. (in 2002) by
trading the cooling water with the power plant,
Replacement of Surface Water with
whereas the power plant achieved an equivalent
Cooling Water
saving caused by a corresponding reduced intake
The intake of surface water is based on utility of surface water. Furthermore, the power plant
sharing among the various symbiosis industries, achieves an indirect economic savings due to an

246 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

energy-cascading effect of the 30◦ C heated cool- As can be observed from table B of the
ing water, which is approximately the tempera- e-supplement, the amount of exchanged
ture required for the boiler-water pretreatment wastewater has decreased during the selected pe-
facility. The power plant has thereby avoided an riod 1992–2002, which is explained by a parallel
investment of approximately 100,000 DKK10 for internal recycling program at the power plant,
preheating installations for the pretreatment fa- where a 200,000 m3 buffer basin has made it pos-
cility. In the same way the refinery gains an ex- sible to pool and recycle wastewater from the
traordinary indirect saving, because the piping of power plant, wastewater from the refinery, and
used cooling water to the power plant has meant drain water from the surrounding fields in order to
a postponed 8–10 million DKK investment in an extend the time frame between water intake and
extended wastewater treatment facility at the re- final discharge. This allows the reuse of the same
finery, which would otherwise have had to treat secondary water several times, thereby reducing
the cooling water before discharge. Additionally, the intake of water of differing levels of quality.
the cooling water exchange has postponed a col- As mentioned above, one of the triggers for op-
lective upstream investment by extending the life erationalizing the three symbiotic water projects
of the current-capacity joint pipeline system for was the installation of a water-consuming desul-
surface water intake and at the same time keep- furization plant at the power plant. The adjust-
ing the intake of surface water below the carrying ment of these three symbiotic exchange projects
capacity of the surface water donor lake, which is and the internal buffer capacity at the power
5.0 million/m3 /yr (in 2004). plant in relation to one another made it pos-
sible to run the desulfurization plant on low-
quality water alone (see figure 2), thereby re-
ducing the intake of an equivalent amount of
Replacement of Surface Water and high-quality water. Reducing the intake of high-
Cooling Water with Wastewater quality water for the desulfurization plant made
an equivalent amount of high-quality water avail-
Wastewater from the refinery is piped to the
able as feeder water for the boilers producing
power plant, where it partly replaces the intake of
steam for the refinery and the pharmaceutical
surface water and cooling water. Approximately
facility.
9,000 m3 of wastewater was delivered from the
refinery to the power plant in 2002, and more
than 1.1 million m3 in the period 1992–2002,
Steam and Heat Cogeneration at the
thereby replacing the same amount of recycled
Asnæs Power Plant
cooling water or surface water for various sec-
ondary purposes at the power plant. The waste- The symbiotic projects concerned with sur-
water exchange between the refinery and the face water, cooling water, and wastewater are to
power plant required a total investment of ap- a great extent interconnected in an integrated
proximately 2 million DKK at the time of initia- circuit. Further down the supply chain the water
tion. The wastewater is a giveaway and a contrac- types are converted into steam and delivered as a
tual evergreen, in the sense that the wastewater high-energy (by-)product to the neighboring fa-
is not priced by the refinery and the contract has cilities or as heat for the district heating system.
no time limitation. The economic benefit is, for Following this process flow, it can be noted that
the refinery, primarily related to the observance the contractual delivery of steam pushed the ini-
of discharge permissions and improved upstream tiation or expansion of the three water-related
availability of surface water due to a reduced in- IS exchanges as a result of the increased run-
take by the power plant. The direct economic though of process water, whereas the steam sup-
gains for the power plant are achieved by the re- ply itself arose for a number of different reasons.
placement of surface water with wastewater and Among these, the economic and environmen-
amount to approximately 58,000 DKK for 2002 tal arguments for cogeneration of steam, heat,
and about 4.5 million DKK for the period 1992– and electricity compared with stand-alone pro-
2002. duction turned out to be the most important, in

Brings Jacobsen, Industrial Symbiosis in Kalundborg 247

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

combination with the right timing for boiler ren- (DKK/ton).15 This economic profile can by
ovation and increased steam consumption at the illustrated in more detail by examining the
colocated facilities. present-day steam exchange between the power
A detailed engineering calculation of the full- plant and the Novo facility. In 1982 the first
scale cogeneration effect of the IS relationships steam pipeline was established between the power
is beyond the scope of this article, partly be- plant and the Novo facility (maximum capacity
cause conditions (e.g., fuels and boilers) have 55 tons/hr), but in the late 1990s further capac-
changed over the years, and partly because access ity was needed due to production expansion at
to relevant technical data and decision scenar- the Novo facility. Thus, the steam prognosis for
ios (e.g., alternatives to IS solutions) is difficult the period 1999–2005 showed drastic increases
in the longer historical perspective. More cur- in steam demand, as documented in figure A of
rent data have therefore been chosen to illustrate the e-supplement.
the emission savings from the IS steam/heat ex- On the basis of these prognoses, the Novo
changes, but still based on a number of hypotheti- Group investigated three different cost scenarios
cal assumptions. Table 3 illustrates a hypothetical for producing steam at a level above 55 tons/hr.
emission-reduction scenario achieved by deliver- The first scenario outlined an extension of the ex-
ing steam and heat from the Asnæs power plant isting pipeline system to the power plant; a sec-
compared with production of the same number ond scenario projected a company-based stand-
of gigajoules per year (GJ/yr)11 from a hypothet- alone facility operated by an independent utility
ical stand-alone facility fueled with natural gas supplier; and the third scenario was based on a
(see table F of the e-supplement for data refer- company-based stand-alone facility operated by
ence and assumptions).12 Thus, for the selected the Novo Group. Table 4 indicates comparative
period (1997–2002), an emission reduction of cost profiles for the three different scenarios.
154,000 tons13 of CO2 and 389 tons of NOx has The most cost-effective scenario was found
been achieved by the delivery of steam and heat to be the one based on an extended pipeline
from the power plant compared with the produc- system between the power plant and the Novo
tion of the same number of GJ/yr from a hypo- facility—the IS solution. The investment/price
thetical stand-alone facility fueled with natural ratio of the IS solution showed reasonable advan-
gas. tages compared with stand-alone alternatives, as
The calculations in table 3 can only be seen documented by table 4. A new steam pipeline was
as a partial indication of the environmental ben- therefore constructed and put into operation in
efits of the steam/heat IS relationships because 2001 with a considerable excess capacity for fur-
the calculation is based only on production from ther production expansion at the Novo facility.
Asnæs unit 5, fueled with Orimulsion.14 There- The present steam price is thus based on three in-
fore, the steam/heat production from Asnæs units terrelated factors: an investment contribution for
2–4 (fueled by coal in 2002) is excluded, and the new pipeline construction, an energy price,
so are the effect of fuel switching among the and a price for water pretreatment. The price of
units and the adjustment to the grid in gen- steam is constantly fluctuating, but as a general
eral. A more detailed engineered calculation of tendency it can observed that the price of steam
the full cogeneration effect is beyond the scope has changed from the initial project start (1982)
of the article, but table 3 gives a qualified pic- to the present date, reflecting the fact that steam
ture based on a comparison with a theoretical is regarded today as a pure commercial product,
alternative. and no longer as a by-product. Nonetheless, ex-
Turning to the economic performance of the panded steam capacity between the power plant
steam/heat exchanges, the same complexity can and the Novo facility in 2001 remains economi-
be observed. Following the hypothetical setup cally feasible for both companies when compared
above, the central question narrows down to with alternative scenarios, especially when re-
the price of steam/heat produced by a stand- lated factors such as supply security, technologi-
alone facility fueled with natural gas compared cal insights, and operational expertise are taken
to the price of imported power plant steam/heat into consideration.

248 Journal of Industrial Ecology

 Table 3 Hypothetical emission savings from steam and heat generation at Asnæs, Unit 5
Emissions
Heat Production CO2 SO2 NOx
Steam Novo+Statoil District Heating Asnæs Stand-alone N-gas Asnæs Stand-alone N-gas Asnæs Stand-alone N-gas
Year [GJ/yr] [GJ/yr] [tons/yr] [tons/yr] [tons/yr] [tons/yr] [tons/yr] [tons/yr]
Brings Jacobsen, Industrial Symbiosis in Kalundborg

1997 346,579 723,099 51,025 74,200 41 0 68 124

1998 384,440 1,088,667 67,789 101,411 54 0 90 169
1999 510,928 731,179 62,148 87,065 49 0 82 145
2000 568,302 730,127 65,880 91,298 52 0 87 152
2001 588,881 730,467 67,247 92,864 54 0 89 155

A P P L I C AT I O N S A N D I M P L E M E N TAT I O N
2002 682,083 569,784 67,102 89,142 53 0 89 149
Total 3,081,213 4,573,323 381,191 535,979 304 0 505 893
Total emissions avoided (tons) 154,788 −304 389

Notes: N = natural gas; CO2 = carbon dioxide; SO2 = sulfur dioxide; NOx = nitrogen oxides. Tons refers to metric tons; one metric ton (t) = 1 megagram (Mg, SI) = 103 kilograms
(kg) ≈ 1.102 short tons.
249
 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

Table 4 Comparative index cost profile for steam production, Novo (Scenario 1 = Index 100)
Scenario 1 (IS Scenario 2 (Stand-alone Scenario 3 (Stand-
solution with facility operated by alone facility
Asnæs power plant) external operator) operated by Novo)

Price per ton 100 108 100

Total investment 100 174 120

Source: Tang (2004).

Pretreated Boiler Water from Power Plant the sharing of boiler water, underlining an ongo-
to Oil Refinery ing adjustment to the economic conditions guid-
ing the IS exchanges.
The economic complexity of the cogener-
ated steam exchanges is underlined by the fact
that steam has developed gradually into a pure Salty Cooling Water from Power Plant to
commercial product with a strong conventional Fish Farm
market-determined price. Unlike the Novo facil-
The final step in the total water- and energy-
ity, the refinery had a feasible means of produc-
cascading sequence among the companies is a re-
ing its own process-related steam, which is why
cycling project where salty cooling water from the
market price increases meant a decrease in the
power plant is used in a colocated fish farm. After
delivery of steam to the refinery, but an increase
producing electricity, steam, and district heating,
in in-house production of process-related steam
the boiler water at the power plant is condensed
(see table B of the e-supplement, Row 7). This
with salty cooling water, causing the temperature
development, however, has also meant that the
of the cooling water to increase by 7–8◦ C, and
cooperation between the refinery and the power
thereby rendering the salty cooling water usable
plant has evolved into a supplementary cooper-
by the fish farm. Approximately 23 million m3 of
ative strategy based on exchanges of pretreated
heated salty cooling water from the power plant
boiler water. The increasing in-house production
was recycled in the fish farm in 2002, whereas
of process-related steam has meant that the refin-
the rest of the salty cooling water was discharged
ery has temporary shortages of pretreated boiler
directly into the Kalundborg Fjord. The total
water for its own process-related steam facility.
amount of discharged salty cooling water from
On the other hand, the power plant has a perma-
Asnæs Power plant is equivalent to 16 million
nent excess capacity in the pretreatment facility
GJ of thermal pollution (2002), so the intake
for boiler water. Piping treated boiler water from
in the fish farm utilizes approximately 2.5% or
the power plant to the refinery has enabled the
39,000 GJ (2002) of this residual heat stream.
refinery to shave its peaks and thereby avoid an
The exchange is a giveaway and a contractual ev-
investment of 10 million DKK in an extension of
ergreen requiring a smaller investment of 75,000
its own pretreatment facility, whereas the power
DKK and a payback time of less than 2 years,
plant has achieved corresponding increased ef-
thanks to the 15 percent increase in production
ficiency of its pretreatment facility. The direct
capacity in the fish.
investment in this boiler-water IS project has
been small (1.5 million DKK), because existing
pipelines and infrastructure have been used to
Discussion of Economic and
facilitate the IS exchange, which is intended to
Environmental Aspects of IS in
deliver about 50,000 m3 of pretreated boiler water
Operation
per year. Thus the quantitative decline in the
high-energy steam exchange between the refin- Evaluation of the selected IS exchanges has
ery and the power plant has been counterbal- shown that some significant and some smaller
anced by a high-quality water exchange based on environmental benefits have been achieved as a

250 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

result of direct substitution, utility-sharing, and of IS exchanges. On the one hand, it can be
water/energy cascading. For example, substantial observed that the IS exchanges involve environ-
benefits accrue from exchange of cooling water mental savings in terms of substitution or energy
between the power plant and the refinery, and cascading. On the other hand, it can be ques-
some minor benefits are achieved by wastewater tioned whether these savings are substantial, if we
exchange between the same two companies. In compare them with the potential for further IS ar-
total, the different IS arrangements contributed rangements and the total flows of waste material,
more than 95% of the total water supply to the energy, and water. We may further ask whether
power plant in 2002, compared with 70% in 1990, the downstream IS solutions have become an
thus demonstrating a gradual development to- obstacle to radical upstream environmental im-
ward a more comprehensive IS supply system and provements (e.g., reduction at source or the appli-
a “system ability” to save groundwater. cation of cleaner technologies) (Oldenburg and
A more critical view of the system also reveals Geiser 1997). In the end, these points raise the
some potential for further optimization, because question of whether IS can be viewed as a com-
more than 1.2 million m3 of wastewater is dis- prehensive strategy for environmental improve-
charged from the refinery, with only 9,000 m3 ments.
reused at the power plant (in 2002) (table C of It seems reasonable to say that the inter-
the e-supplement). Similarly, there seems to be a pretation of the success of IS exchanges de-
possibility of reusing some of the 2.3 million m3 pends greatly on the context and the perspec-
of wastewater from the Novo group beyond the tive in which these intercompany arrangements
present co-treatment arrangement with the pub- are viewed. The findings from the present anal-
lic wastewater facility (see figure 1). Such expan- ysis indicate that the IS exchanges only form a
sion was investigated in the 1980s and 1990s, but single element in an overall process of improv-
was rejected for economic and technical reasons. ing the environmental performance of a number
The amounts of wastewater discharged from the of companies, each with a significant environ-
system, though, still seem to imply possibilities mental impact factor. A large number of the IS
for further IS activity, perhaps primarily in con- exchanges are closely connected with the exis-
nection with further co-treatment arrangements tence of air pollution controls or water purifi-
or utility sharing. cation plants, and the use of these installations
The same picture can be observed in the case has paved the way for the trading of reusable by-
of the steam/heat projects, where some significant products (e.g., industrial gypsum, fly ash, waste,
cumulative CO2 reduction effects are achieved and cooling water). Thus IS exchange is not re-
when one compares the system with hypothet- garded as an isolated, ultimate environmental so-
ical stand-alone production. Still, the total fuel lution, but rather as part of a process of improv-
efficiency of the Asnæs power plant is—among ing the total environmental performance of the
the other factors—dependent on local heat and individual company. As in most other cases, in-
steam demand. In general, Kalundborg (with only ternal environmental improvements specified by
20,000 inhabitants) is far too small compared to annual improvement targets are given first prior-
the total capacity of Asnæs (approx. 1,300 MW), ity by each company in the present case.
again reflecting the fact that Asnæs was located The potential for further IS improvements of
in Kalundborg in a time (1959) before cogen- the total regional material flows thus has to be
eration was a consideration. A complete picture evaluated in this context, and has to be consid-
of the environmental effect of cogeneration is ered in combination with a number of other in-
beyond the scope of this article, and it is recog- house environmental initiatives, each of which
nized that such a full-scale analysis would nat- seems to be central to the understanding of the
urally include the life-cycle assessment (LCA) rationale behind a given IS exchange. The analy-
and a system analysis of Danish power gene- sis of the water-related IS exchanges has demon-
ration.16 strated the importance of this point, and this con-
These clarifications raise a number of ques- textual understanding of the IS exchanges seems
tions in relation to the environmental benefits to be important if IS is to be judged as part of

Brings Jacobsen, Industrial Symbiosis in Kalundborg 251

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

a comprehensive strategy for environmental im- appears, stressing the need to understand the
provements. context in which the IS exchanges find their
The same kind of contextual interpretation relevance and motivation. In general it can be
is important in the discussion of the economic observed that low-value by-product exchanges
aspects of the IS exchanges. The present evalu- are often motivated by indirect economic bene-
ation has shown that the pure water-related ex- fits, whereas high-value by-product exchanges are
changes lead to economic benefits because of the often motivated more by direct economic bene-
scarcity and costliness of groundwater resources. fits related to the value of the by-product itself.
As shown by table E of the e-supplement, direct Between these two extremes there are a number
economic benefits can be achieved by replacing of intermediate stages, which often move from
groundwater with other water sources. Never- low-value status to high-value status as a result
theless, from a corporate perspective, the direct of upgrading, or the gradual creation of a mar-
economic savings from the use of other water ket and thus more direct economic benefits. This
sources are minor, which is why some indirect sequence is followed by increasing contractual
economic arguments are required to understand complexity and a tendency toward a redefini-
the economic motivation of the water-related ex- tion or reconception of the exchanged materi-
changes. als from waste/by-product status to commercial
These indirect economic arguments are, to a product status. If we take IS as an expression
great extent, associated with long-term strategic of a business-to-business economic transaction
planning based on the desire for increased supply with a built-in environmental effect, it follows
security, operational capability, and the expan- that the basic economy of these arrangements
sion of production without the obstacle of water is driven by both long-term and short-term di-
shortages in the longer term. The direct economic rect and indirect economic considerations de-
returns related to the value of the exchanged by- pending on commercial market status and the
product are thus balanced by indirect economic production-related importance of the exchanged
returns related to the operation in general. The by-product.
latter represent the real economic argument for In sum, these findings raise the question of
a number of the water-related IS exchanges, and whether IS relationships can be viewed as solely
the same tendency can be seen with a number of market-driven arrangements that evolve spon-
other IS exchanges in the case studied here. taneously, or whether the initiation of IS rela-
This picture of indirect economic arguments tionships requires something beyond pure market
changes, however, when one considers by- forces. The primary objective of this article has
products with a higher initial value due, for exam- been to investigate the quantitative aspects and
ple, to higher energy content or the market value operational raison d’être of a selected number of
of substances. The indirect economic arguments IS exchanges, and these quantitative aspects do
are of less importance to the steam exchanges to a certain extent explain the immediate logic
presented in this study; it can be observed here of the selected IS exchanges. Nevertheless, these
that it is the direct value of the steam itself that quantitative findings also point to some addi-
is the fundamental driving force and focal point tional perspectives relevant to a full explanation
for the exchange relationships. Furthermore, the of the IS exchanges.
recipient can focus on its core business instead of On one hand, the economic and environmen-
running utilities. Thus, the steam exchanges dif- tal parameters listed in table D (e-supplement)
fer radically from the water exchanges in terms seem to support a market-based view that ex-
of direct economic motivation and the commer- plains the IS exchanges as a combination of op-
cial setup. This difference is clearly reflected by erational, environmental, and economic benefits.
complex contractual agreements in the case of This explanation seems to be important in the
steam, whereas the water exchanges are based on case of high-value by-products such as steam and
simpler contracts. heat, a picture that also seems to be supported
A highly differentiated picture of the eco- by other studies of high-value IS exchanges
nomic stimuli for the selected IS exchanges thus (Chertow and Lombardi 2005).

252 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

On the other hand, a number of the selected tal considerations. The former argument has a
IS exchanges in the present case have to be un- narrow individual focus on economic viability in
derstood in a wider perspective where the direct a wider operational context. The second implies a
economic benefits and market logic are more dif- more collectivist approach to resource efficiency,
ficult to grasp. The wastewater exchange between driven by individual benefits and contexts. The
the refinery and the power plant provide one ex- water- and steam-related IS projects thus have
ample of this. When the direct short-term eco- to be explained at a variety of levels and from
nomic benefit is minor and the risk is high it often different perspectives. First, the low groundwater
seems to be the case that companies refrain from availability in the municipality created pressure
engaging in IS exchanges unless something else for a diversified water supply strategy to secure
drives the establishment of the IS exchange. a resource basis for production expansion at the
It has often been pointed out that social rela- different facilities in the region. Second, the in-
tions and individual agency at the interfirm level stallation of specific water-consuming processes
(Cohen-Rosenthal 2000; Korhonen 2004) might at the power plant triggered the cooling water and
be important factors when by-products are pushed wastewater exchanges between the power plant
“across the factory fence.” These factors might and refinery. Third, the contractual obligations
also explain why certain IS exchanges are ini- for the export of steam and heat from the power
tiated despite only minor short-term economic plant to the neighboring industries and the city of
benefits. The present article does not investigate Kalundborg created a high flow-through of pro-
whether these “soft” social factors have had any cess water at the power plant, in turn triggering
influence on the initiation of the selected IS ex- the cooling water and wastewater exchanges and
changes, but it can be observed that certain of the increased intake of surface water. Fourth, di-
the IS exchanges in the Kalundborg IS complex rect and indirect economic arguments based on
are such that these factors might provide further increased flexibility, capacity extension, technol-
important explanations. To judge whether the ogy access, supply security, and water substitu-
social factors actually are important for IS initia- tion created an economic motivation for both
tion and operation would require further research the water and steam exchanges.
(Brings Jacobsen 2005), and the question seems On the basis of these clarifications it can be
to be fruitful, because it touches on the discus- concluded that the water/steam IS exchanges in
sion of the drivers of and barriers to IS exchanges Kalundborg are driven to a great extent by the
in particular (Ehrenfeld and Gertler 1997) and search for a diversified water supply and the pos-
industrial ecology in general (Van Berkel et al. sibility of optimizing energy efficiency and supply
1997). security for economic and operational reasons.
Nevertheless, this ongoing search for solutions
to the problem of water supply and the extent
Conclusion
to which some of the IS exchanges involve only
This study has presented some central eco- minor and long-term benefits presumably indi-
nomic and environmental facts related to se- cate that a purely market-based explanation of
lected IS exchanges in Kalundborg, and in so the IS exchanges has to be supplemented by
doing differs in detail and contextual scope from an explanation that takes individual agency and
other analyses (see table A of the e-supplement). social factors during the project development
It has shown that some significant and some phases into consideration. Especially in the early
smaller environmental savings are related to the project development phase of an IS exchange,
individual IS exchanges based on water substitu- when exact benefits are unclear and uncertainty
tion and cascading. Likewise, it has been shown is high, individual agency can be assumed to be
that the economy of the IS exchanges differs vital to project development; but how far this as-
in scale and scope depending on the specific sumption is correct must be verified by further
production-related context. It seems clear that research.
each exchange project is justified by an individ- The clarification of this latter point is impor-
ual economic argument and related environmen- tant, because it is relevant to any full explanation

Brings Jacobsen, Industrial Symbiosis in Kalundborg 253

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

of the IS complex in Kalundborg and the degree 7. One cubic meter (m3 , SI) = 103 liters (L) ≈
to which this example can be used as a model for 264.2 gallons (gal).
IS initiatives at other locations. The economic, 8. The calculation of proportion of the reduction in
environmental, and operational facts pointed out consumption attributable to IS exchanges versus
in this article are all related to a specific con- reduced production capacity is complex and be-
yond the scope of this article if the result is to
text, partly explain why a number of IS exchanges
be exact. As one example, however, IS exchanges
were initiated in the case studied here, and high-
concerned with cooling and waste water reduced
light the obvious fact that IS exchanges have to de facto the intake of surface water to Asnæs by
contribute in some way to the bottom line of a approximately 55% during 1990–2002.
given company. Nevertheless, the findings have 9. Based on an average currency exchange rate of 788
also indicated that this obvious truth about IS DKK = $100 U.S. in 2002. Obtained from Danish
requires a more sophisticated view of the bottom national bank statistics. The same exchange rate
line that considers both short-term economic and and year are used when DKK are converted to
long-term environmental and operational aspects U.S. $ in other sections of the article.
that might lower initial entry barriers to engage- 10. An approximation referring to the time of the
ment in exchanges. Taken together, these ele- project’s initiation (Christensen 2003).
11. One gigajoule (GJ) = 109 joules (J, SI) ≈ 2.39 ×
ments involve a more universal logic that might
105 kilocalories (kcal) ≈ 9.48 × 105 British ther-
be transferable to other agglomerations of com-
mal units (BTU).
panies searching for IS solutions and point to IS 12. At the time of the initiation of the steam ex-
as an element in a rational business perspective changes (1982), the fuel for stand-alone produc-
that takes sustainability into account. tion would probably have been fuel oil. In this
analysis, natural gas is chosen as a hypothetical fuel
Acknowledgments for a stand-alone facility because this fuel would
probably have been relevant to stand-alone pro-
The author wants to thank the symbiosis in- duction in a present-day picture, according to the
dustries in Kalundborg for providing time and companies.
resources for this study. Special thanks to Lars 13. Tons refers to metric tons; one metric ton (t) =
Munkø for doing the engineering calculations 1 megagram (Mg, SI) = 103 kilograms (kg) ≈
in table 3 and table F (of the e-supplement), to 1.102 short tons.
Weslynne Ashton of Yale University, and to the 14. Orimulsion is a fuel composed of 70% bitumen
four anonymous reviewers. and 30% water. The fly ash from Orimulsion con-
tains 12.2% vanadium and 2.7% nickel on aver-
Notes age, which is extracted for recirculation. Orimul-
sion was used at Asnæs in the period 1994/1995–
1. This assertion is based on a literature review and 2002/2003.
an evaluation of incoming research requests on 15. The hypothetical stand-alone natural gas facility
IS to the Symbiosis Institute and the Symbiosis used in Table 3 is—technically—not identical to
Industries, Kalundborg, 1995–2004. either of the two stand-alone facilities used in Sce-
2. See the methodological section for terminological nario 2 or Scenario 3 in Table 4.
explanations. 16. See work by Energy E2 (2000).
3. The question of environmental regulation is not
touched upon in this article, but it is recognized
that environmental regulations often work as a
References
central precondition for IS implementation.
4. In this context cogeneration refers to power plants Brings Jacobsen, N. 2005. Industrial symbiosis in the
that utilize residual heat from electricity produc- making: Bridging social and technical explana-
tion for, for example, industrial purposes or district tions. Paper presented at the 11th Annual In-
heating systems. ternational Sustainable Development Research
5. One megawatt (mW) = 106 watts (W, SI) ≈ Conference, June 6–8, Helsinki.
56.91 × 103 British thermal units (BTU)/min. Brings Jacobsen, N. and S. Anderberg, 2004. Under-
6. Editor’s note: For an optimatization model of water standing the evolution of industrial symbiotic
reuse in industrial parks, see the article in this networks—The case of Kalundborg. In Economics
journal by Keckler and Allen (1998). of industrial ecology: Materials, structural change,

254 Journal of Industrial Ecology

 A P P L I C AT I O N S A N D I M P L E M E N TAT I O N

and spatial scales, edited by J. Van den Bergh, and Fraanje, P. 1997. Cascading of pine wood. Journal of
M. Janssen. Cambridge, MA: MIT Press. Resources, Conservation and Recycling 19: 21–28.
Chertow, M. 2000. Industrial symbiosis: Literature and Graedel, T. and B. Allenby. 1995. Industrial ecology.
taxonomy. Annual Review of Energy and Environ- Englewood Cliffs, NJ: Prentice Hall.
ment 25: 313–337. Hansen, J. 2003. Personal communication with
Chertow, M. 2004. Industrial symbiosis. In Encyclope- J. Hansen, Kalundborg Center for Industrial
dia of energy, edited by C. J. Cleveland. Oxford: Symbiosis, November, 2003.
Elsevier. Hansen, J. 2004. Personal communication with
Chertow, M. and R. Lombardi. 2005. Quantifying eco- J. Hansen, Kalundborg Center for Industrial
nomic benefits of colocated firms. Environmental Symbiosis, January, 2004.
Science and Technology 39(17): 6535–6541. Hardy, C. and T. Graedel. 2002. Industrial ecosystems
Chertow, M. and M. Portlock, eds. 2002. Develop- as food webs. Journal of Industrial Ecology 6(1):
ing industrial ecosystems. Approaches, cases, and 29–38.
tools. Bulletin Series, No. 106. New Haven, Keckler, S. E. and D. T. Allen. 1998. Material reuse
CT: Yale School of Forestry & Environmental modeling: A case study of water reuse in an indus-
Studies. trial park. Journal of Industrial Ecology 2(4): 79–92.
Christensen, V., former vice-director, Asnæs power Korhonen, J. 2004. Industrial ecology in the strategic
plant, Kalundborg, Denmark, 2003. Personal sustainable development model: Strategic appli-
communication, December 2003. cations of industrial ecology. Journal of Cleaner
Cohen-Rosenthal, E. 2000. A walk on the human side Production 12(8–10): 809–823.
of industrial ecology. American Behavorial Scientist Oldenburg, K. and K. Geiser. 1997. Pollution pre-
44(2): 245–264. vention and . . . or industrial ecology? Journal of
Connelly, L. and C. Koshland. 1997. Two aspects of Cleaner Production 5(1–2): 103–108.
consumption: Using an exergy-based measure of Schwartz, E. and K. Steininger. 1997. Implementing
degradation to advance the theory and implemen- nature’s lesson: The industrial recycling network
tation of industrial ecology. Journal of Resources, enhancing regional development. Journal of
Conservation and Recycling 19:199–217. Cleaner Production 5(1–2): 47–56.
Connelly, L. and C. Koshland. 2001. Exergy and indus- Sirkin, T. and T. Houten. 1994. The cascade chain.
trial ecology, Part 1: An exergy-based definition Journal of Resources, Conservation and Recycling
of consumption and thermodynamic interpreta- 11: 215–277.
tion of ecosystem evolution. Exergy International Sterr, T. and T. Ott. 2004. The industrial re-
Journal 1(3): 145–65. gion as a promising unit for eco-industrial
Côte, R. and E. Cohen-Rosenthal. 1998. Designing development—Reflections, practical experience
eco-industrial parks: A synthesis of some ex- and establishment of innovative instruments
periences. Journal of Cleaner Production 6(3–4): to support industrial ecology. Journal of Cleaner
181–188. Production 12(8–10): 947–965.
Côte, R. and J. Hall. 1995. Industrial parks as ecosys- Tang, P. 2004. Personal communication with P. Tang,
tems. Journal of Cleaner Production 3(1–2): 41–46. Novo Group Energy Supply Unit, Bagsværd,
Ehrenfeld, J. and M. Chertow. 2002. Industrial Denmark, 18 May, 2004.
symbiosis: The legacy of Kalundborg. In A Van Berkel, R., E. Willems, and M. Lafleur. 1997. De-
handbook of industrial ecology, edited by R. Ayres, velopment of an industrial ecology toolbox for the
Northampton, UK: Edward Elgar. introduction of industrial ecology in entreprises,
Ehrenfeld, J. and N. Gertler. 1997. Industrial ecology Part I. Journal of Cleaner Production 5(1–2): 11–25.
in practice: The evolution of interdependence Wallner, H. P. and M. Narodoslawsky. 1996. Evolution
at Kalundborg. Journal of Industrial Ecology 1(1): of regional socio-economic systems towards “is-
67–79. lands of sustainability.” Journal of Environmental
Energy E2. 2000. Livscyklusvurdering af dansk el og Systems 24(3): 221–240.
kraftvarme [Life cycle assessment of Danish Heat
and Power Generation]. <www.e2.dk/upload/
About the Author
files/presse/LCA.pdf>. Accessed January 2004.
Esty, D. and M. Porter 1998. Industrial ecology and Noel Brings Jacobsen is a postgraduate fellow and
competitiveness: Strategic implications for the Ph.D. candidate at Roskilde University in Roskilde,
firm. Journal of Industrial Ecology 2(1): 35–43. Denmark. He also served as the project manager at the
Forward, G. and A. Mangan. 1999. By-products Industrial Symbiosis Institute in Kalundborg from 1998
synergy. The Bridge 29(1): 12–16. to 2001.

Brings Jacobsen, Industrial Symbiosis in Kalundborg 255