reinterpreting industrial ecology

F O RU M

Reinterpreting IndustrialEcologyPaul D. Jensen, Lauren Basson, and Matthew Leach

Keywords:

diversityecological principlesindustrial ecosystemmetaphorsuccessionwaste

Address correspondence to:Paul D. JensenUniversity of Surrey CESGuildford, SurreyUnited KingdomGU2 [email protected]/ces/

c© 2011 by Yale UniversityDOI: 10.1111/j.1530-9290.2011.00377.x

Volume 15, Number 5

Summary

This article argues that industrial ecology has, to date, largelyengaged with the ecological sciences at a superficial level,which has both attracted criticism of the field and limitedits practical application for sustainable industrial development.On the basis of an analysis of the principle of succession, therole of waste, and the concept of diversity, the article high-lights some of the key misconceptions that have resulted fromthe superficial engagement with the science of ecology. It is ar-gued that industrial ecology should not be seen as a metaphorfor industrial development; industrial ecology is the ecology ofindustry and should be studied as such. There are manifoldgeneral principles of ecology that underpin our understandingof the world; however, the physical manifestation and causaleffects of these principles are particular to the system and itsconstituent elements under analysis. It is thus proposed thatcontext-specific observation and analysis of industry are re-quired before theoretical and practical advancement of thefield can be achieved.

680 Journal of Industrial Ecology www.wileyonlinelibrary.com/journal/jie

F O RU M

Introduction

The premise that the science of ecology and,more specifically, the metaphoric mimicry of nat-ural ecosystems can lead to greater conservationof resources and wider environmental protec-tion continues to be fundamental to industrialecology researchers and practitioners alike. Thestudy and, to a lesser extent, practice of indus-trial ecology have been a conceptual source ofhope for sustainable industrial development forat least two decades. The publication of the ar-ticle by Frosch and Gallopoulos (1989) entitled“Strategies for Manufacturing” arguably signaledthe start of a concerted effort to model indus-trial processes on those widely observed withinthe biosphere. The assumption is that by mod-eling industrial systems on seemingly resource-efficient and harmonious biological ecosystems,it is possible to reorganize the environmentallydamaging and resource-wasteful industrial sys-tems that humans increasingly rely on, to functionin a more environmentally benign and resource-ful manner. Despite the passing of more thantwo decades since “Strategies for Manufacturing”was first published, however, there has been slowprogress made on both the theoretical develop-ment of industrial ecology and, in particular, thepractical application of its guiding principles forcleaner production and resource efficiency. Thelack of progress within the field of industrial ecol-ogy, particularly in relation to implementation,has been attributed to several factors. Some com-mentators would assert that the lack of progressis due to the source metaphor being inappropri-ate, primarily because of fundamental differencesthat (seemingly) exist between human industryand ecosystem functioning (e.g., Ayres 2004; seealso the discussions presented by Korhonen 2005and McManus and Gibbs 2008). Others, mean-while, surmise that the active development ofso-called eco-industrial parks is unlikely to hap-pen because it is difficult, particularly at the plan-ning level, to attract and coordinate the numberand diversity of companies (and other organiza-tions) that are required to form the basis of anyself-sufficient industrial ecosystem (e.g., Ehren-feld and Gertler 1997; Gibbs and Duetz 2005,2007; Chertow 2007). In effect, it is suggestedthat human factors, such as the need for trust

and cooperation, and economic factors, such asthe self-centric need to strive for business growth,obstruct the planned development of networked,interacting, industrial ecosystems. Thus, biolog-ical ecology, other than at a highly conceptualand metaphoric level, is seemingly not a suitablescience for informing greater resource efficiency.

This article argues, however, that any mis-givings or criticisms that are held in relation toemploying ecology as a suitable guide for sustain-able industrial development are based on a lackof knowledge of the source science, or a lack ofappreciation of what ecology fundamentally per-tains to.

The Need for Reinterpretation

Philosophical Thinking and KnowledgeDeficiency

To date, industrial ecology theory has largelybeen presented on a desirability basis, and the‘cherry-picked’ principles1 of ecology are regu-larly applied (or dismissed) in a superficial man-ner, with surprisingly little or no reference toecological research. Several studies have engagedwith the wider principles of ecology and the lit-erature of the source science (e.g., Nielsen 2007;Mayer 2008; Ashton 2009). But, in many cases,arbitrary principles of ecology are routinely ap-plied in an overarching manner that is not suitedto providing the levels of complex analysis re-quired to yield sound reference points for sustain-able industrial development (discussed below andin the following section). Hence, it is readily ap-parent why tangible eco-industrial developmentremains largely elusive and why some researchersare seemingly able to dismantle much of the the-ory behind industrial ecosystem thinking (e.g.,Harte 2001; Levine 2003; Ayres 2004).

The science of ecology does not pertain to theprescription of normative ideals. It does not seekto provide a metaphoric basis for anything. And,due to inherent idiosyncrasies, many principlesof ecology do not manifest themselves in a uni-versally causal or easily predictable manner. In-deed, of the universally held principles and lawsof ecology that do exist, many derive from thestudy of physics or chemistry.2 Ecology, as oneof the natural sciences, intrinsically relates to

Jensen et al., Reinterpreting Industrial Ecology 681

F O RU M

observation, interpretation, and, one hopes, un-derstanding (and, where appropriate, prediction).On the basis, first and foremost, of observationand interpretation, ecologists seek to answer spe-cific questions or simply try to better understandthe world we live in. If applied correctly, ecol-ogy is one of the most powerful sciences availableto humans in relation to understanding the waysin which the world works. It is also, by far, oneof the most complex of scientific fields to trulyunderstand (Wilson 2001).

As a consequence of the complexity involvedin the study of ecology, and despite how thesubject is regularly portrayed in industrial ecol-ogy literature, it is still a science that containsmany uncertainties and is certainly not capa-ble of informing industrial development in anall-encompassing manner. Indeed, despite over ahundred years of studying the world from the per-spective of ecology, most ecologists will readilyadmit that they have only scratched the surfaceof the complexity involved in observing livingsystems and the ways in which they interact anddevelop (Chapman and Reiss 1999). It seems thatmany industrial ecologists are largely unaware ofor fail to acknowledge the levels of uncertaintythat exist within the wider study of ecology.

Ecology, for the most part, is not an exactscience in which each principle or theory appliesto every organism or system under analysis ina uniform manner (that said, see the work ofJørgensen [2002] for a discussion on the existenceof universal ecosystem processes and properties).In reality, disagreements within the ecologicalsciences are rife. One brief search throughjournals such as Ecology or Oikos or through oneof the core ecology textbooks (e.g., Begon et al.2006) would highlight the levels of debate andapparent disagreement that exist within the field(the ongoing diversity-stability debate is a primeexample). Indeed, considerable debate has takenplace on the very question of whether any univer-sally applicable principles or laws3 exist in ecol-ogy (e.g., Lawton 1999; Colyvan and Ginzburg2003; Hansson 2003; Lange 2005; O’Hara 2005).Consequently, from a practical implementationstance, there is little to be gained from selectingwhat are effectively arbitrary ecological prin-ciples (e.g., Korhonen’s [2001a] widely citedfour principles for an industrial ecosystem) and

applying them to industrial development inan all-encompassing or metaphoric manner.It is important to fully engage with ecologicaltheory on a context-specific basis, otherwise theidiosyncrasies of individual principles and theirmany sound applications (and well-documentedfailings) could be missed. Any detractors of theunderlying concepts of industrial ecology are,in effect, simply highlighting deficiencies in theapplicability of many principles that have, fornumerous years, already been recognized andwidely debated by biological ecologists.

Due to widespread misuse, there have beensuggestions that the very word ecology has es-caped the controls of academia and been wronglyportrayed as a quasi-scientific philosophy for life,rather than a dedicated branch of the biologicalsciences (Westoby 1997). Surely the accusationof quasi-scientific philosophy is not a fate any re-searcher would wish to befall the study of indus-trial ecology. At a basic level, it is thus fundamen-tally wrong to reduce the field to the frail concep-tual level of a metaphor (or paradigm) for envi-ronmentally benign industrial development. Por-traying industrial ecology as a metaphor, at anyscale, only leaves the field open to criticisms ofineffectual thinking and to potentially undermin-ing statements such as: “. . . industrial systems willnever operate as nature does” (Korhonen 2001b,66). How, precisely, does ‘nature’ operate? Does‘nature’ refer to the entire biosphere, a particu-lar biome, an ecosystem, an individual habitat,or something else? What form of industry are wereferring to: all industry, an individual company,or a specific industrial sector? Industry, as a prod-uct and ‘servant’ of a living system, the humansystem, is as valid a research element of the eco-logical sciences as those studied within any otherautoecological4 study. Industrial ecology is theecology of human industry and should be stud-ied as such. It should not be trivialized and leftopen to detrimental criticisms that, as Gallopou-los (2006) suggests, can follow concepts that areseemingly underpinned on philosophical ideals.

Practice Before Prescription

The employment of philosophical metaphors(rather than ecological empiricism) to explainor dictate what occurs within the anthrosphere

682 Journal of Industrial Ecology

F O RU M

does nothing more than perpetuate the (repeat-edly calamitous) belief that we are custodians ofthe natural world rather than one small elementof the wider biosphere. Indeed, in providing le-gitimacy to metaphoric thinking, several indus-trial ecology articles are seemingly constructedaround the belief that humans do indeed tran-scend the natural environment (e.g., Boons andRoome 2001; Ehrenfeld 2003; Wells 2006). Thebelief that human industry is not part of natureand anything we do or aspire to do is not entirelysubject to the limiting factors of our one planetis fundamentally absurd and should be avoided ifindustrial systems are to be observed in an objec-tive ecological manner.

Ecological principles and ecosystem sciencemust be applied to industry in the same man-ner as they are applied to, say, a given form ofwoodland, a particular species of tree within thewoodland, or a particular community of speciesthat makes the tree its home. Essentially, ecolo-gists define specific questions about the ecologyof a given organism or system that they are inter-ested in prior to observing the subject matter un-der suitable study conditions. Observations madeare interpreted in a manner suited to answeringthe original question in hand. Fundamentally,the ecologist has an answer to a specific question;he or she rarely possesses an answer or principlethat is suitable for direct application to all queriesor observed ecological phenomena. By their verynature, all complex systems (biological or other-wise) are dynamic, and hence many principles ofthe ecological sciences must, where necessary, bedynamic and thus be molded to fit observed cir-cumstances. Indeed, if one wishes to intrinsicallyunderstand the ecology of saltmarsh, one stud-ies saltmarsh. Although they largely require thesame fundamental resources to function and aresubject to similar evolutionary forces, one wouldnot study the ecology of mangrove and expectto possess knowledge suited to predicting the po-tential future of saltmarsh. This perhaps extremeexample could be applied at any ecological levelfrom, say, genotype to organism to community tothe given example of ecosystems, but the overrid-ing message is applicable to the study and practiceof industrial ecology.

A company or, more likely, industrial sectormust be ecologically analyzed in its own right to

determine its past and, potentially, future courseof evolution (and any new feedback mechanismsresulting from its evolution). As a community,industrial ecologists need to increase researchinto the ecology of industry, rather than spec-ulating on its idealized future. Existing principlesof biological ecology that can be shown to pro-vide sound analogues between industrial systemsand biotic systems should be accepted and widelypromoted. Frivolous reinterpretation of the lan-guage or lessons of ecology should be avoided.Where, however, principles (or ecological analy-sis tools) clearly do not fit industrial developmentin general, or do not fit specific industrial sec-tors, they should be disregarded or reinterpretedto suit the given situation. This is what hap-pens within all the traditional branches of ecol-ogy; this is what should happen with industrialecology.

Ecosystem science alone provides a myriad ofenlightening system genesis, development, andmaintenance principles that industrial ecologistscan apply on a testable basis if it is accepted thatthere is no contradiction in the literal analysisof industrial systems on an ecological basis. In-deed, it has been recently argued that “. . . realindustrial ecology . . .” can only occur once a fullunderstanding of the current conditions of in-dustrial systems, in relation to our knowledge ofecosystem functioning, is acquired (see Nielsenand Muller 2009, p. 1914). The potential needto reinterpret existing principles of ecology to fitwhat is witnessed within industrial systems doesnot translate as the rejection of ecology as a suit-able platform for promoting sustainable industrialdevelopment; it translates as the evolution of in-dustrial ecosystem science.

All fledging scientific fields develop in a hap-hazard, opportunistic manner (Wilson 2004). Af-ter at least two decades of development, how-ever, it is time for industrial ecology theoriststo concentrate on specificity. The age of oppor-tunism that saw the promotion of the ecologicalmetaphor has arguably reached its natural (andvery limited) conclusion. Only after appropriateprinciples of ecosystem development are categor-ically shown to be applicable to industry (anddemonstrated to policy makers) can industrialecology be taken seriously as a science capable ofdelivering tangible eco-industrial development.


F O RU M

To consolidate what has been stated here, in-dustry is as valid an ecological study subject asany other product of a given organism’s continu-ing adaption to its environment and its compul-sion to fulfill its daily needs. As a consequence,industrial ecosystems do not require creation;they already exist, albeit in a (largely) resource-inefficient form. It will be argued that at the ini-tial points of system genesis and succession, theydo develop in accordance with the same ‘princi-ples’ and ‘laws’ as all other ecosystems. At thecommunity development and functioning lev-els, however, the sheer magnitude of complexityand innumerable system feedback mechanismsmean that many existing ecological principlesare unlikely to be directly transferable in theirraw form. It is thus argued that discovering whenand where existing principles of ecology are ap-plicable, when and where they require adaption,and when alternative principles require creation,constitutes the study of industrial ecology and,ultimately, the development of the principles ofindustrial ecosystem science.

To continue the core argument of this article,some of the regularly proposed core principles ofindustrial ecology that generate criticism and de-bate from both detractors of industrial ecologyand natural ecologists alike will be highlighted.By way of conclusion, the article endeavors tohighlight the necessity to perform and consultecology in a literal and case-specific manner and,in the process, attempts to also provide an in-sight into the intrinsically complex nature of thescience and study of ecology.

Misconception, Contradiction,and Debate

One potential criticism of using ecosystemsas a source model for industrial development isthe belief that the two systems seemingly oper-ate according to fundamentally different prin-ciples of evolution and maintenance. Due tointernal recycling and system inputs primarilyderiving from renewable sources, it is widely be-lieved that ecosystems are inherently resource ef-ficient and thus sustainable. In contrast, it is sug-gested that industry is inherently unsustainabledue to the linear nature of resource movement

and energy inputs primarily deriving from non-renewable sources (e.g., Frosch and Gallopoulos1989; Jelinski et al. 1992; Graedel 1996; Korho-nen 2001a; Korhonen et al. 2004; Gibbs 2008).Superficially, these differences between the twosystems seem undeniable; however, as will be ar-gued, the apparent differences only exist due tomisconception or inappropriate spatial or tempo-ral comparisons being made.

System Genesis and Succession

The laws of thermodynamics are the biggestconstraint to production within industrial andnatural ecosystems (Nielsen 2007)—this positionis irrefutable. Exactly how the laws of thermo-dynamics manifest themselves within the evo-lution of biotic systems is, however, a point ofdebate. Without an exposition of the widely dis-cussed relationship between thermodynamic effi-ciency and system development, it can be safelystated that at the point of genesis, the global in-dustrial system developed, and continues to de-velop, along the same macroecological principlesand laws as any other ecosystem (for chronolog-ical discussion on thermodynamics and ecosys-tem development, see Lotka 1922a, 1922b; Odumand Pinkerton 1955; Schneider and Kay 1994;Jørgensen 2002).

At the point of colonization, the primary (andsecondary) genesis of habitats occurs with the ap-pearance of select pioneer species that, via theirfundamental niches,5 are adapted to sequesteringand assimilating underutilized in-situ resources(Chapman and Reiss 1999; Begon et al. 2006).As pioneer species take hold within a habitat,they systematically, by their very appearance, cre-ate or facilitate further fundamental niches thatprovide and provoke, through system feedback,multispatial habitat change and the developmentof new resources or the release of existing re-sources. As later sere6 organisms fill and exploitthe newly evolved fundamental niches, the pro-cess of niche opening and realization occurs ona continual basis in continued response to in-ternal system changes. As this process of auto-genic succession, as it is termed, takes place, andan ecosystem continues to develop, communitystructure typically evolves toward diversificationuntil the ecosystem reaches a point of relative


F O RU M

temporal stability (Chapman and Reiss 1999). Inmany cases, the primary colonizers that set sys-tem evolution and diversification in motion, withtheir unlocking of previously difficult-to-utilizeresources, become victims of their own successas they promote the development of ecosystemsthat are rich in competition or simply no longersuited to their maintenance or, more specifically,their fecundity (Chapman and Reiss 1999; Be-gon et al. 2006). The unintentional but self-propagated changes to a given habitat, such asthe modification of soil chemistry or structurethrough an accumulation and decomposition ofleaf litter, which are disastrous for many incum-bent species, are essential for the appearance ofothers. Neither organic nor industrial pioneersintentionally make their ‘homes’ less suited totheir proliferation; stochastic and potentially un-desirable environmental change is an inevitableconsequence of their existence and their neces-sary resource (in the widest sense) exploitation(for a discussion on the general trends of systemsuccession, see Odum 1969).

In direct comparison, it can be stated that theIndustrial Revolution blossomed on the back ofthe exploitation of vastly underutilized resourcesin the same way pioneer primary producers ex-ploit seemingly bare earth. The evolved abilityto exploit underutilized sources of energy andthus develop greater system production capac-ity generated unprecedented industrial develop-ment and, consequently, significant changes inthe local and wider environment. As the tech-nosphere has developed and diversified on theback of continued resource innovation and fun-damental (techno) niche realization, increasedresource competition and environmental changehave been generated. Over the last two centuries,the accumulation of environmental ‘bads’ (froma human perspective) has promoted successionalforces within the technosphere that have, in turn,promoted changes to the way we allow the in-dustrial ecosystem to proliferate. For example,the systematically evolved physical and politi-cal tools of cleaner production can be seen asa clear response to the autogenic poisoning anddegradation of the human habitat.

Due to our higher cognitive skills, we are ef-fectively able to recognize impending phases ofautogenic succession and thus adapt or evolve far

quicker than most colonizer species. Thus, thelevels of industrial adaption and diversificationthat are capable of facilitating greater resourceefficiency do not necessarily evolve due to a per-ceived lack of immediate necessity. Arguably theforces of autogenic (or possibly allogenic7) suc-cession can, however, only be avoided for so longbefore industry, in its current form, degrades thehuman habitat to a point where it succumbs tosuccession and perishes or adapts in a way bestsuited to reaching relative environmental stabil-ity. In the same manner as many other colonizersbecome victims of their own success, resource-inefficient and environmentally degrading indus-try will, no doubt, eventually become a victim ofits own considerable success.

Whilst fundamental niches continue to ex-ist within immature ‘habitats,’ geographic disper-sion will, however, take place—industrial prac-tices that have succumbed within industriallyevolved countries invariably continue to colonizeand prevail within newly industrialized countries.The components of all systems will, if allowed byadaption or lack of competition, take the leastresistance or ‘easy’ existence and developmentoption until such time as environmental influ-ences dictate that adaption (or extinction) is anecessity. The ‘easy’ option and the feedbackcontrols required to promote the necessity (orwill) to evolve will differ vastly between systemsand the specific components that constitute agiven system’s community because, in the processof realizing a fundamental niche, a given com-pany or industry must, by definition, be locallyadapted.8

The idea that local factors may be importantto the understanding of industrial ecology is notnew. It has been widely acknowledged that thedevelopment (or non-development) character-istics of individual industrial ecosystems are af-fected by local factors (e.g., Korhonen, 2001a;Baas and Boons 2004; Roberts 2004; Deutz andGibbs 2008). Like all systems, industry is sub-ject to the physical and temporal availability ofthe resources that allow it to function. Thus,on one level or another, there must be locallypresent influences and instances of operationalfeedback that allow a given industrial ecosystemto proliferate. Although it is difficult to ascer-tain what is controlling what within an ecosystem


F O RU M

(Nielsen 2007), ascertaining what system devel-opment and proliferation influences exist withinindividual industrial ecosystems, at a specific ge-ographic scale or on a sectoral basis, will allow amore complete, but admittedly complex, pictureof potential industrial evolution to be painted.Ultimately, however, industrial ecologists needto remember that everything in the present is aproduct of the past; arguably, nothing in natureis adapted to the future.

Nature: The Perfect Recycler?

Biological ecosystems are driven by circu-lar recycling of resources; in contrast, indus-trial systems are outwardly characterized by linearmovement of resources and the accumulation ofwaste (e.g., Graedel 1996; Ehrenfeld and Gertler1997; Korhonen 2001a; Korhonen et al. 2004).It is widely believed that, as “. . . masters of re-cycling . . .” (Korhonen 2001b, 57), ecosystemsgenerate “. . . little or no waste” (Roberts 2004,998). In many ways, these statements are per-fectly sound, and, thus, as a core tenet of in-dustrial ecology theory, the goal of emulating thestatus of ‘master’ recycler within industrial ecosys-tems is an admirable one. These statements arealso technically incorrect, however, as at any onetime, vast amounts of waste exist within indi-vidual ecosystems and the wider biosphere (asalso acknowledged and demonstrated by Levine[2003] and Ayres [2004]).

Fossil fuels, our primary energy source and, ar-guably, the source of contemporary industry, are awaste product of nature’s ‘incomplete’ biologicalassimilation of ancient carbon-based life. Manyof the materials we employ in construction activi-ties are formed from waste products that have notbeen fully assimilated by the environment. Oneof many possible examples is limestone, which isprimarily composed of the calcite remains of long-dead marine organisms. The current compositionof our atmosphere is the product of waste gener-ation: prior to the appearance of life on Earth(and, by definition, ecosystems), the balance ofelements that were found in the planet’s atmo-sphere differed vastly from its present composi-tion. Indeed, without the oxygenic ‘pollution’ ofthe atmosphere, humans (and, arguably, all eu-karyote life forms) would not exist! Numerous

further apparent instances of waste in the en-vironment could be presented to highlight thefact that waste, as we perceive it, does indeedexist within the biosphere. Furthermore, the no-tion of closed cyclical systems can be automati-cally debunked, at least at the local scale, withthe knowledge that ecosystems regularly lose andgain resources through instances of wet-fall, dry-fall, gaseous atmospheric dispersion, and in-situconsumption or collection by organisms prior toex-situ deposition (see Begon et al. 2006).

If ecosystems are not as resourceful (and lo-cally closed) as we are led to believe they are,the question thus arises: is internal or externallydeposited waste actually a bad thing, or is it sim-ply a human construct that has no meaning inecology? In some cases, ‘waste’ is arguably good.The accumulation of waste and perceived pol-lution or degradation of a given ecosystem isoften the precursor to evolution and the emer-gence of diversity and, consequently, increasedresource optimization—this, in part, is the basicbut much debated principle of (autogenic) suc-cession. Waste is only seen as bad when it im-pinges on or threatens the long-term existence ofa given organism. Waste, in effect, does not ex-ist; it is a force of evolution within both systemsand should be studied as such within industrialecology. Indeed, the perception that industrialecosystems are wasteful due to the linear move-ment of materials only exists if one puts a spatialor temporal boundary around a study site. Withthe knowledge that waste does exist in natureand that ecosystems are not completely cyclical,it could be easily argued (and proved) that ecosys-tems are also linear systems if, again, one doesnot extend the study boundary (both in time andspace).

Effectively unused by-products emanatingfrom processes within industrial (and biologi-cal) systems have not become system waste; theyhave simply entered resource sinks similar tothose that exist within all ecosystem compart-ments. Any resource sink will perpetuate withinan ecosystem until such time as an organism orprocess enters the system or evolves that is ca-pable of directly using or assimilating the saidwaste product. Given sufficient time, there is al-ways the possibility that something will evolvethat can economically, thermodynamically, or


F O RU M

chemically assimilate or make best use of a re-source, regardless of how unlikely that may seemat a given point in time. For example, if overseveral millions of years humans had not evolvedinto their specific niche and had not continued torealize many fundamental niches, resource sinkssuch as crude oil would, from the perspective ofmany other elements of nature, remain a uselessand potentially harmful waste product. This ex-ample and general line of thought reiterates acore question within industrial ecology theory:where is the boundary, if one exists, betweenindustrial and biological systems? More specif-ically, when does the human system start, orstop, being part of wider biosphere evolution andfunctioning?

It is accepted that undesirable and potentiallydisastrous side effects are associated with the un-locking of some resource sinks that are deemedto be examples of waste within nature. For an ob-vious example, the release of carbon dioxide is aseemingly undesirable side effect of the use of fos-sil fuels. If anthropogenic cognition is taken outof the theoretical equation, however, there are noundesirable side effects emanating from our useof ecosystem by-products. We have simply cre-ated new environmental influences. Waste, or,in effect, the existence of a new resource or suc-cessional influence within the environment is aprecursor and force behind evolution and invari-ably the appearance of further optimized systems.It is undeniable, however, that our biggest prob-lem, as a species, is the rate at which environ-mental ‘problems’ are multiplying, persisting andthus potentially threatening our current way oflife. But, without suitable consideration of (andswift reaction to) the consequences of our actions,this is seemingly an inevitable and typically un-predictable consequence of niche realization andconsequent resource exploitation.

Diversity and Ecosystem Functioning

It has been argued that industrial diversityis essential to the functioning and stable de-velopment of industrial ecosystems (e.g., Coteand Smolenaars 1997; Korhonen 2001a, 2001b;Korhonen and Snakin 2005; Ashton 2009;Liwarska-Bizukojc et al. 2009). It is widely be-lieved that increased diversity and, consequently,

collective adaptability are able to protect againstharmful perturbations in the local and wider en-vironment. It is also believed that the increasednumber of potential system linkages that derivefrom system diversity also promote the possibil-ity of localized by-product reuse and, thus, in-creased productivity (Korhonen 2001a, 2001b;Hardy and Graedel 2002; Sterr and Ott 2004;Korhonen and Snakin 2005; Liwarska-Bizukojcet al. 2009). At the highest point of applica-tion, these statements are both sensible and intu-itive within the development and understandingof both industrial and biological systems theory.When parallels are drawn between observationsmade within biological and industrial ecology,however, there is a belief that industrial systemstypically develop toward homogenization of pro-duction and, in some cases, monopolies, which isin contrast to the individuality and diversity thatexists within ecosystems (Nielsen 2007; Ashton2009). Arguably, however, this divergence be-tween the two systems does not strictly exist. In-dividual species do dominate at various succes-sional seres and would, no doubt, evolve towarda full monopoly of resources and habitat ‘markets’if suitable environmental conditions existed andtheir fundamental niche allowed such domina-tion. For instance, many mature ecological com-munities can be almost entirely dominated by asingular species of plant life. Indeed, the commonreed (P. australis) has been known to develop‘natural monocultures’ that cover vast areas ofwetland.

Although the likes of the common reed canentirely dominate the plant life within a givenarea (on many scales), it is debatable to what ex-tent it exists in biotic isolation (i.e., nutrient up-take is largely facilitated by symbiotic mycorrhizalfungi, and the stands form a refuge for many an-imals). Furthermore, their continued dominanceat a given point in time is continually subject tostochastic and potentially destructive forces (e.g.,storms and disease) that could lead to the reopen-ing of a given area’s resources to a wider group ofspecies. In parallel to this example, it is debatableto what extent a given industry can or does op-erate in isolation or how long any company cancontinue to monopolize the use of a resource orthe share of a market. As evidenced by the re-cent world economic downturn, even the most


F O RU M

dominant of companies are subject to, and canultimately be victims of, wider market forces be-yond their control.

Discussion on the potential benefits of diver-sity and its effects on overall system stability andlevels of production have to be conducted in acontext-specific manner. Diversity, in the senseof the richness and relative abundance of systemelements, simply provides options for adaptabil-ity. In theory, diversity is excellent for promotingsystem productivity and stability; however, thesesystem traits could, arguably, be achieved by asmall number of companies if they are sufficientlyadaptable and prepared for change or prepared toexploit an opportunity to evolve toward a ‘fitter’state of existence. Neither internal nor systemdiversity, however, guarantees an increase in theproductive efficiency of the individual or the sys-tem as a whole. Indeed, as our current economicsystems are largely driven by finite resources, theefficiency of our productions systems is arguablya far more important aspect of ecosystem func-tioning to focus on than, for example, a dogmaticconsideration of productivity or the increased cy-cling of materials.

Ultimately, system efficiency and stability (inthe form of resilience and, initially, resistance)cannot be boiled down to a generalized beliefthat more diversity equates to desirable systemcharacteristics. Such a statement needs empiricaltesting at various scales within defined industrialecosystems, not least because several researchershave conducted subject-specific research thatconcludes that a positive relationship betweenbiotic diversity, per se, and overall system stabil-ity is not as universal as one would intuitively be-lieve it to be (see McCann 2000; Cameron 2002;Pfister and Schmid 2002). Indeed, although in-creased productivity, in the form of primary pro-duction, has been largely shown to be a productof increased species richness (e.g., Tilman et al.2001; Hooper et al. 2005; Flombaum and Sala2008), the effects (or at least our understanding)of diversity on system stability and productiveefficiency are less conclusive.

The increases in productivity that are seen tobe a product of species richness are, in general,intuitively believed to be a result of increases inboth spatial and temporal resource complemen-tarity (i.e., niche differentiation and facilitation),

which follow the maturation and diversificationof a system9 (Cardinale et al. 2007). Apart fromproductivity, it is probable that complementar-ity and functional traits, at and between multi-ple trophic levels, are important to the develop-ment of all ostensibly positive ecosystem prop-erties. Acquiring a greater understanding of thedevelopment of genotypic and phenotypic traitsis almost certainly the best way to derive a fullerknowledge of system diversity and, in particular,its context-specific causal properties and ‘bene-fits.’ Unfortunately, research into which and howmany species act in a complementary way hasnot been widely conducted (Hooper et al. 2005).Moreover, in specific relation to the long-runningdiversity-stability debate, it has been suggestedthat research has (perhaps wrongly) been over-taken by theoretical thinking—long-term fieldstudies and experimentation are required if weare to fully understand what facilitates systemstability (Hooper et al. 2005).

Although both systems (biological and indus-trial) are clearly diverse, it is difficult to determinein which system diversity plays the more impor-tant role (Nielsen and Muller 2009). It is onlywith context-specific research that the roles ofdiversity within industrial systems can even be-gin to be understood. With the level and breadthof industrial data that are available—in manycases to an extent biological ecologists can onlydream of—it should be possible to conduct theresearch needed on the evolution of functionaltraits and the many aspects of resource comple-mentarity. As Mayer (2008) stated, use of suchdata offers great potential for advancing bothecologies.

With the number of industrial data that existwithin many developed countries, it should notbe difficult to go beyond concept-driven specu-lation and start confronting the wider diversitydebate with empirical data and analysis. Indeed,the authors of this article have, in partnershipwith the United Kingdom’s National IndustrialSymbiosis Programme, started the process of map-ping the industrial diversity of England (in termsof both industrial sectors and resource flows) inthe pursuit of discovering how it relates to knowninstances of eco-industrial development and op-portunities for industrial symbiosis. The work (tobe published) is still at the data analysis stage,


F O RU M

but preliminary findings have been made thatnot only confirm some of the most basic assump-tions made within the literature of both ecologiesbut also add to some of the more complex ques-tions on the relationship between diversity andecosystem functioning.

It is perhaps sensible to increase diversityamong power producers to ensure future energysecurity, but, other than possibly increasing oppor-tunities for localized industrial symbiosis, whatbenefits can be drawn from the generalist pre-scription to promote diversity within industrialecosystems? As can be surmised from the abovediscussion, individual ecosystem functioning andstability will be as dependent on the specificcharacteristics of potential immigrant species asthey are on the collective functional (and re-dundant) traits of existing species. Because eachecosystem possesses a non-uniform number of re-source flow compartments and internal linkages,there is arguably no way of knowing, withoutcontext-specific research (both at and betweenmultiple trophic levels), how diversity affects agiven system. For example, there could be sig-nificant levels of diversity within a system, butthe loss of just one keystone or functional species(e.g., mycorrhizal fungi within plant communi-ties) could be disastrous in terms of maintain-ing what are deemed to be anthropogenicallydesirable conditions or, simply, levels of systemproduction. In the same vein, without sufficientcontext-specific information, the artificial intro-duction of a new species to promote diversitycould, in fact, lead to a reduction in diversity if thespecific resource requirements of the immigrantlead to extinctions or promote competitive ex-clusion (e.g., the extinction of many animals onislands where predatory alien species have beenintroduced, or, within England, the wide-scalemarginalization of the red squirrel [S. vulgaris] bythe nonnative grays [S. carolinsensis]). The spe-cific goal of maintaining or increasing the diver-sity of an industrial ecosystem must be clear andwell thought out before prescription can be made.Diversity, in short, is a highly relative conceptwhose scale is largely dictated by levels of nichedifferentiation and realized niche overlap. Diver-sity, in whatever form it is observed, tends to existwhere the prevailing system conditions allow it toexist.

Debate Summary

In summary, it has been shown that, at thehighest level, many principles of ecology can beapplied to industrial systems. For instance, themany aspects of niche theory and its role in sys-tem succession are invaluable to understandinghow industrial development has taken place andwill theoretically take place. Nevertheless, somewidely promoted ideological principles of indus-trial ecology, such as the idea that ecosystemsare masters of recycling, waste is bad, and diver-sity is essential to the development and stabilityof ecosystems, are not applicable in a universalmanner. Ascertaining the effects and existenceof waste within a given system and determin-ing the contribution diversity makes to promot-ing stable and efficient industrial ecosystems are,among other, important questions within indus-trial ecology research. But, due to the complexityinvolved in ecological research at the commu-nity and systems levels, studies relating to theirapplication must be conducted widely and theirresults presented on a case-by-case basis if anyuseful patterns of eco-industrial development areto be determined and made (prescriptively) trans-ferable to similar systems.

Conclusion

A note of caution is required before one stud-ies the dynamics of industrial ecosystems: “[In-dustrial] ecologists, like the [companies] organ-isms they study, cannot make [industry] natureconform to their perfect liking” (adapted from:Wilson 2001, 155). Thus, industrial ecosystemscannot be studied in relation to biological sys-tems with any form of ideological preconceptionin mind, otherwise it will only lead to widespreadfalse hope or misconceived disappointment.

Industrial ecology is exactly that, the ecologyof industry—it is not a metaphor or paradigmfor anything. Industrial ecology needs to learnand borrow from the many natural sciences,but it must not be confined by their observa-tions of the world if theoretical and practicalprogress in the field is to be made. Due to theidiosyncrasies inherent within all levels of eco-logical structure, the principles of ecology do notmanifest themselves in a universally predictablemanner. Consequently, there exists a theoretical


F O RU M

industrial ecology vacuum that needs filling withcontext-specific eco-industrial analysis.

The study and characterization of specific in-dustrial ecosystems will provide the ‘principles ofindustrial ecosystem science’. The principles willno doubt be unique to some forms of company, ag-glomerations, or industrial sectors; some will ap-ply on a cross-sector basis, and some, maybe, willbe applicable to all industrial ecosystems and thusbe deemed law. The primary question for indus-trial ecology is, however, not what principles arepotentially important to promoting environmen-tally benign manufacturing, but in what contextis one principle more important than another.

Lawton (1999) stated that ecological patternsemerge most clearly at the single species level, be-cause their study contingencies are manageable,and at larger scales, because general statistical or-der emerges from the “scrum.” In contrast: “Themiddle ground is a mess. It is fascinating to study,and rich in wonderful biology. But by studying it,do not expect universal rules, even simple contin-gent general rules, to emerge. If and when they do,treasure them” (Lawton 1999, 188). It is by ex-ploring the richness and divergence of industrialecosystems that industrial ecology will become arespectable and defensible field in its own right,rather than through simplification and general-ization of what is an inherently complex subject.

Acknowledgements

The presented article forms one element ofresearch being conducted into regional resourceplanning by Paul D. Jensen in accordance withhis enrollment on the University of Surrey’s En-gineering Doctorate (EngD) Programme in Envi-ronmental Technology. The Regional ResourcePlanning project is financially supported by theUnited Kingdom’s Engineering and Physical Sci-ences Research Council and International Syner-gies Ltd. The authors would like to thank the fouranonymous reviewers for their insightful com-ments and helpful suggestions for improving thearticle.

Notes

1. Principle: an observed or logically proven statementof truth.

2. For example, the laws of thermodynamics (widelyemployed in ecosystem development research) de-rive from the study of physics.

3. Law: a principle (or collection of related principles)that is true without exception.

4. “Autoecological” pertains to the ecological study ofa single species of organism.

5. Fundamental niche: the complete range of nichesan organism can fill in the absence of competition.

6. A “sere” refers to the sequential change in the com-position of an ecological community during the pro-cess of succession.

7. Allogenic (succession): a change to communitycomposition as a result of external environmentalinfluences.

8. Although the realization of a fundamental nicheis described here as being achieved as a result oflocal adaption, parallel and convergent evolutionshow that the development of a given ‘phenotype’ iscapable of replication wherever suitable conditionsexist.

9. Although, as a caveat to this statement, it shouldbe noted that diverse plant communities can takeseveral generations of growing seasons before com-plementarity facilitates system production on scalesgreater than that of highly productive monocultures(see Cardinale et al. 2007).

References

Ashton, W. S. 2009. The structure, function, and evo-lution of a regional industrial ecosystem. Journalof Industrial Ecology 13(2): 228–246.

Ayres, R. U. 2004. On the life cycle metaphor: Whereecology and economics diverge. Ecological Eco-nomics 48(4): 425–438.

Baas, L. W. and F. A. Boons. 2004. An industrial ecol-ogy project in practice: Exploring the boundariesof decision-making levels in regional industrialsystems. Journal of Cleaner Production 12(8–10):1073–1085.

Begon, M., C. R. Townsend, and J. L. Harper. 2006.Ecology: From individuals to ecosystems. Fourth edi-tion. Oxford, UK: Blackwell.

Boons, F. and N. Roome. 2001. Industrial ecology asa cultural phenomenon: On objectivity as a nor-mative position. Journal of Industrial Ecology 4(2):49–54.

Cameron, T. 2002. 2002: The year of the “diversity-ecosystem function” debate. Trends in Ecology andEvolution 17(11): 495–496.

Cardinale, B. J., J. P. Wright, M. W. Cadote, I. T.Carroll, A. Hector, D. S. Srivastava, M. Loreau,


F O RU M

and J. J. Weis. 2007. Impacts of plant diversityon biomass production increase through time be-cause of species complementarity. Proceedings ofthe National Academy of Sciences, USA 104(46):18123–18128.

Chapman, J. L. and M. J. Reiss. 1999. Ecology:Principles and applications. Second edition. Cam-bridge, UK: Press Syndicate of the University ofCambridge.

Chertow, M. R. 2007. “Uncovering” industrial symbio-sis. Journal of Industrial Ecology 11(1): 11–30.

Colyvan, M. and L. R. Ginzburg. 2003. Laws of natureand laws of ecology. Oikos 101(3): 649–653.

Cote, R. P. and T. Smolenaars. 1997. Supporting struc-tures for industrial ecosystems. Journal of CleanerProduction 5(1–2): 67–74.

Deutz, P. and D. Gibbs. 2008. Industrial ecology and re-gional development: Eco-industrial developmentas cluster policy. Regional Studies 42(10): 1313–1328.

Ehrenfeld, J. 2003. Putting a spotlight on metaphorsand analogies in industrial ecology. Journal of In-dustrial Ecology 7(1): 1–4.

Ehrenfeld, J. and N. Gertler. 1997. Industrial ecologyin practice: The evolution of interdependence atKalundborg. Journal of Industrial Ecology 1(1): 67–80.

Flombaum, P. and O. E. Sala. 2008. Higher effect ofplant species diversity on productivity in natu-ral than artificial systems. Proceedings of the Na-tional Academy of Sciences, USA 105(16): 6087–6090.

Frosch, R. A. and N. E. Gallopoulos. 1989. Strate-gies for manufacturing. Scientific American 261(3):144–152.

Gallopoulos, N. E. 2006. Industrial ecology: Anoverview. Progress in Industrial Ecology—an Inter-national Journal 3(1–2): 10–27.

Gibbs, D. 2008. Industrial symbiosis and eco-industrialdevelopment: An introduction. Geography Com-pass 2(4): 1138–1154.

Gibbs, D. and P. Deutz. 2005. Implementing industrialecology? Planning for eco-industrial parks in theUSA. Geoforum 36(4): 452–464.

Gibbs, D. and P. Deutz. 2007. Reflections on imple-menting industrial ecology through eco-industrialpark development. Journal of Cleaner Production15(17): 1683–1695.

Graedel, T. E. 1996. On the concept of industrial ecol-ogy. Annual Review of Energy and the Environment21(69–98): 69–98.

Hansson, L. 2003. Why ecology fails at application:Should we consider variability more than regular-ity? Oikos 100(3): 624–627.

Hardy, C. and T. E. Graedel. 2002. Industrial ecosys-tems as food webs. Journal of Industrial Ecology6(1): 29–38.

Harte, J. 2001. Business as a living system: The valueof industrial ecology (a roundtable discussion).California Management Review 43(3): 16–25.

Hooper, D. U., F. S. Chapin III, J. J. Ewel, A. Hector, P.Inchausti, S. Lavorel, J. H. Lawton, et al. 2005. Ef-fects of biodiversity on ecosystem functioning: Aconsensus of current knowledge. Ecological Mono-graphs 75(1): 3–35.

Jelinski, L. W., T. E. Graedel, R. A. Laudise, D. W.McCall, and C. K. N. Patel. 1992. Industrial ecol-ogy: Concepts and approaches. Proceedings of theNational Academy of Sciences, USA 89(3): 793–797.

Jørgensen, S. E. 2002. Integration of ecosystem theories:A pattern. Third edition. Dordrecht, the Nether-lands: Kluwer Academic.

Korhonen, J. 2001a. Four ecosystem principles for anindustrial ecosystem. Journal of Cleaner Production9(3): 253–259.

Korhonen, J. 2001b. Some suggestions for regional in-dustrial ecosystems: Extended industrial ecology.Eco-management and Auditing 8(1): 57–69.

Korhonen, J. 2005. Do we really need the debate on thenatural ecosystem metaphor in technology man-agement and sustainable development literature?Clean Technologies and Environmental Policy 7(1):33–41.

Korhonen, J. and J.-P. Snakin. 2005. Analysing theevolution of industrial ecosystems: Concepts andapplication. Ecological Economics 52(2): 169–186.

Korhonen, J., F. von Malmborg, P. A. Strachan, andJ. R. Ehrenfeld. 2004. Management and policyaspects of industrial ecology: An emerging re-search agenda. Business Strategy and the Environ-ment 13(5): 289–305.

Lange, M. 2005. Ecological laws: What should they beand why would they matter? Oikos 110(2): 394–403.

Lawton, J. H. 1999. Are there general laws in ecology?Oikos 84(2): 177–192.

Levine, S. H. 2003. Comparing products and produc-tion in ecological and industrial systems. Journalof Industrial Ecology 7(2): 33–42.

Liwarska-Bizukojc, E., M. Bizukojc, A. Marcinkowski,and A. Doniec. 2009. The conceptual model ofan eco-industrial park based upon ecological re-lationships. Journal of Cleaner Production 17(8):732–741.

Lotka, A. J. 1922a. A contribution to the energetics ofevolution. Proceedings of the National Academy ofSciences, USA 8(6): 147–151.


F O RU M

Lotka, A. J. 1922b. Natural selection as a physical prin-ciple. Proceedings of the National Academy of Sci-ences, USA 8(6): 151–154.

Mayer, A. L. 2008. Ecologically-based approaches toevaluate the sustainability of industrial systems.International Journal of Sustainable Society 1(2):117–133.

McCann, K. S. 2000. The diversity-stability debate.Nature 405(6783): 228–233.

McManus, P. and D. Gibbs. 2008. Industrial ecosys-tems? The use of tropes in the literature of indus-trial ecology and eco-industrial parks. Progress inHuman Geography 32(4): 525–540.

Nielsen, S. N. 2007. What has modern ecosystem the-ory to offer cleaner production, industrial ecologyand society? The views of an ecologist. Journal ofCleaner Production 15(17): 1639–1653.

Nielsen S. N. and F. Muller. 2009. Understandingthe functional principles of nature: Proposing an-other type of ecosystem services. Ecological Mod-eling 220(16): 1913–1925.

Odum, E. P. 1969. The strategy of ecosystem develop-ment. Science 164(3877): 262–270.

Odum, H. T. and R. C. Pinkerton. 1955. Time’s speedregulator: The optimum efficiency for maximumpower output in physical and biological systems.American Scientist 43(2): 331–343.

O’Hara, R. B. 2005. The anarchist’s guide to ecologicaltheory. Or, we don’t need no stinkin’ laws. Oikos110(2): 390–393.

Pfister, A. B. and B. Schmid. 2002. Diversity-dependent production can decrease the stabilityof an ecosystem. Nature 416(6876): 84–86.

Roberts, B. H. 2004. The application of industrial ecol-ogy principles and planning guidelines for the de-velopment of eco-industrial parks: An Australian

case study. Journal of Cleaner Production 12(8–10):997–1010.

Schneider, E. D. and J. J. Kay. 1994. Life as a man-ifestation of the second law of thermodynam-ics. Mathematical and Computer Modelling 19(6–8):25–48.

Sterr, T and T. Ott. 2004. The industrial region asa promising unit for eco-industrial development:Reflections, practical experience and establish-ment of innovative instruments to support indus-trial ecology. Journal of Cleaner Production 12(8–10): 947–965.

Tilman, D., P. B. Reich, J. Knops, D. Wedin, T. Mielke,and C. Lehman. 2001. Diversity and productiv-ity in a long-term grassland experiment. Science294(5543): 843–845.

Wells, P. E. 2006. Re-writing the ecological metaphor:Part 1. Progress in Industrial Ecology—an Interna-tional Journal 3(1/2): 114–128.

Westoby, M. 1997. What does “ecology” mean? Trendsin Ecology and Evolution 12(4): 166.

Wilson, E. O. 2001. The diversity of life. Reprint edition.London: Penguin.

Wilson, E. O. 2004. On human nature. Second edition.Cambridge, MA, USA: Harvard University Press.

About the Authors

Paul D. Jensen is a research engineer at theNational Industrial Symbiosis Programme (de-livered by International Synergies Ltd) and theUniversity of Surrey’s Centre for EnvironmentalStrategy (CES) in Guildford, Surrey, UK. LaurenBasson is a lecturer at CES, and Matthew Leachis a professor and director of CES.


reinterpreting industrial ecology

Documents