14.1 Introduction
Biomimicry may be defined as the art or science of imitating Nature. Applied initially to relatively small-scale innovations, often under the names of biomimetics and bioinspiration, in recent years the rising belief that it is through imitating Nature that we will be able to achieve the elusive goal of sustainability (Benyus, 1997) has led to increased interest in applying biomimicry at much larger scales. Permaculture (Hervé-Gruyer & Hervé-Gruyer, 2014) and certain variations of agro-ecology (Jackson, 2011) have advanced the idea that our agricultural systems should imitate Nature. Industrial ecology has sought to take the working of natural ecosystems as models for industrial systems (Erkman, 2004). And the emerging field of biomimetic urbanism has sought to take Nature as a model for cities. It seems fair to say, however, that this third major large-scale application of biomimicry is at present the least developed. Broadly speaking, there would appear to be two related reasons for this. The first is that cities are extremely complex, bringing together a wide variety of different technological objects and systems, and involving almost every major sector of technological innovation, including energy, water, the built environment, ICT, waste, food production, and transport. The second reason is that the way humans participate in cities is much more complex and introduces far more important issues relating to the human and social sciences than is the case in other artificial macro-systems, such as agricultural or industrial systems. Cities are not just places where technology and labour combine to produce things, but also economic, social, cultural, and political centres, a state of affairs that poses a major challenge to standard approaches to biomimicry operating only at the interface between the life sciences and the technical disciplines (CESE, 2015).
Given the complexity of cities, it is likely to be beyond the scope of a text of this length to address in a meaningful way the myriad issues raised by the application of biomimicry to the city. Our goal will thus be the more modest one of exploring—principally at the level of technologies and technological systems—a natural model that has occurred again and again in the recent literature on the subject: the forest. This model would appear to have been introduced first by Braungart and McDonough (2002, p. 139), who call on us to “imagine a building like a tree, a city like a forest”, but without examining this idea in a scientific or systematic way. Similarly, the contemporary biomimetic architects, Schuiten (2010) and Callebaut (2015), both take up the idea of the forest as model for the city but without subjecting it to rigorous analysis. Callebaut’s vision for the Paris of 2050, for example, is underpinned by his belief that we should “[t]ransform our cities into ecosystems, our districts into forests and our buildings into inhabited trees” (Callebaut, 2015, back cover, our translation), but when it comes to applying this model, his proposals are limited to stand-alone buildings and so can hardly be said to address the question of what it would mean for an entire city to be modelled on a forest. Other examples of architects attempting to apply the model of the forest to the city are similarly partial and unsystematic. Faced with the problem of chronic air pollution in China, Boeri (2015) has proposed the creation of what he describes as “forest cities”. In practice, however, the model of the forest serves here only as a model for tall buildings covered in vegetation in order to purify the air and regulate internal temperatures. Other ways in which the forest may serve as a model for urban design are not addressed.
If, in keeping with our definition of biomimicry as the “art or science” of imitating Nature, contemporary architects have typically tended towards the “artistic” pole of biomimetic urbanism, many academics have developed more “scientific” approaches. Newman and Jennings (2008), for example, think cities should imitate various emergent or systemic properties characteristic of natural ecosystems, most notably their status as healthy, zero-waste, self-regulating, resilient, self-renewing, and flexible. Further, they also suggest that exploring an analogy between ecological succession and urban development may help us better implement the transition towards cities characterized by such characteristics of mature ecosystems as high levels of nutrient cycling and more efficient energy use. Another notable example in this field is the work of Pedersen Zari (2015), who has argued that the systematic study and quantification of the ecosystem services provided by the pre-development ecosystem may provide not just a model, but also a standard or measure capable of helping us set targets for the reproduction of these same ecosystem services in future bio-inspired cities.
If all these approaches have the advantage of adopting scientific methodologies and focusing on the city as a whole, rather than on individual buildings or isolated problems, it is also true that they do not explicitly address the model of the forest. So, while in Pedersen Zari’s case study of the city of Wellington in New Zealand, the pre-development ecosystem was in fact a forest (low-land broadleaf podocarp), her concern is primarily with how “behavioral change” and “existing technologies” may help us reproduce the ecosystem services provided by this forest and not with how the various elements and relations within it—the organisms, abiotic elements, and the relations between them—might help us develop appropriate bio-inspired technologies and articulate them into a new urban form. The result is relative indifference to the means by which the pre-development ecosystem services are reproduced, such that, to take just one example, it matters little whether energy is produced by wind turbines or solar cells inspired by tree leaves, provided the basic ecosystem service—energy production—is reproduced to the required degree.
In view of all this, it would seem that there is something of a lacuna in current thinking about biomimetic urbanism regarding the systematic, scientific exploration of the model of the forest. The remainder of this article will be dedicated to filling in this lacuna. In the first part, we will concentrate on justifying the model of the forest. The idea of “buildings like trees, cities like forests” clearly holds a strong poetic appeal to many biomimetic architects and urbanists, but are there rational grounds for privileging this model, such that it could potentially become the standard model for the design of all or almost all future cities, or is the forest ultimately just one model amongst others? With a view to answering this question, we will consider three potential justifications for the idea that forests possess a privileged status in biomimetic urban design: (a) the argument from fittingness, the key idea of which is that in a large proportion of cases the forest would have been the native or pre-development ecosystem destroyed to make room for the city; (b) the argument from scale, which holds that, in contrast to many other natural systems, there is a relatively close correspondence in scale between forests and cities; (c) the argument from complexity, which maintains that, as generally the most complex of terrestrial ecosystems, forests hold the most significant potential for bio-inspired innovation. In the second part of the article, we will turn instead to exploring the model of the forest. This exploration will be divided into four parts, each focusing on a specific sector of biomimetic urbanism: urban planning, urban water systems, urban energy and transport systems, and urban food and nutrient systems.
14.2 Why the Forest?
14.2.1 The Argument from “Fittingness”
Much of the literature on biomimicry and biomimetic urbanism emphasizes the importance of “fitting in” (Benyus, 1997; Braungart & McDonough, 2002). Benyus, for example, argues that whereas the mentality underlying the first industrial revolution impelled us to “learn about nature so that we might circumvent or control her”, the mentality underlying the second industrial revolution—the “biomimicry revolution”, as she calls it—involves coming to “learn from nature, so that we might fit in, at last and for good, on the Earth from which we sprang” (Benyus, 1997, p. 9). Nature, as Benyus (1997, epigraph) points out, does not just tell us what “works” and what “lasts”, but also what is “appropriate”, that is to say, what allows us—including our artefacts and our systems—to fit in in a given context. In the case of artificial macro-systems, this is typically taken to mean that we should look to native ecosystems for guidance as to how to do things in the same location. The American agro-ecologist, Wes Jackson (2011), for example, argues that it is the pre-development ecosystem—in his case the prairie—that should teach us how to farm. And Pedersen Zari (2015), as we have already seen, thinks the pre-development ecosystem should act as both model and measure for urban design.
With this focus on the pre-development ecosystem in mind, it is interesting to consider George Perkins Marsh’s claim that “[t]here is reason to believe that the surface of the habitable earth, in all climates and regions which have been the abodes of dense and civilized populations, was, with few exceptions, already covered with a forest growth when it first became the home of man” (Marsh, 1864, p. 114). While forests currently represent only around 31% of the earth’s current land surface, in pre-historic times this figure would have stood at 45% (FAO, 2012), and there can be little doubt that forests have been disproportionately chosen by humans as places to build cities (subsequent to deforestation). One single forest biome—the temperate forest—is home to many of the great cities of North America, notably those of the East Coast, to the majority of important European cities, to the cities of northeastern Asia, including the whole of Japan, North and South Korea, and Eastern China, as well as to many major cities in Australia and New Zealand. When one further factors in cities located within the tropical and sub-tropical forest biomes of South America, West Africa, India, South East Asia, and Northern Australia, as well as those found in the boreal forests of Scandinavia, Canada and Russia, it is clear that the forest would indeed have been the pre-development ecosystem for a very large proportion of the world’s existing cities. Further, the very principle that brings us to the model of the forest—the idea that Nature teaches us how to fit in in a specific place—also points towards imitating not just the features common to forests in general, but also to specific forests in specific locations.
There is, however, an obvious objection that could be put forward here. Not only are there many existing cities not located in forest biomes, but one may also wonder what is to prevent us from taking non-forest biomes as models for new bio-inspired cities. There are two important strategies one might use to respond to this objection. The first consists in exploring the idea that many existing cities located in other terrestrial biomes may either prove unsustainable in the long-term or would struggle to be viable were the native ecosystem to be taken as model. As a general rule, the two key parameters that enable forest growth are precipitation and temperature (Ghazoul, 2015). When precipitation is too low, either absolutely or relative to temperature, grassland or desert will generally result. Given that cities, like forests, require significant quantities of usable water per unit of land, the limited local precipitation characteristic of grassland and desert biomes often constitutes a significant challenge for cities, hence their frequent recourse to the exploitation of fossil aquifers or to transfers between rivers or between river basins. Both of these solutions, however, are problematic in this context. Exploiting aquifers more quickly than they are replenished—a particular problem in arid areas such as America’s Great Plains (Ogallala aquifer)—is unsustainable, and large-scale artificial transfers clearly involve “circumventing or controlling” Nature, as opposed to following or imitating it, hence philosophical critiques of this practice as “unnatural” (Rolston, 1995). Of course, it could also be objected here that biomimicry provides a perfect response to both these problems. Rather than drawing on fossil aquifers or diverting water from distant rivers, cities located in grassland or desert biomes could mimic the water-saving techniques characteristic of their native ecosystems, while at the same drawing on the parameters of these native ecosystems to set limits for sustainable water consumption. There can be little doubt that this would make cities located in these biomes more sustainable, but even then it is possible that they would still suffer shortfalls of locally available water that would put an intolerable strain on the basic functioning of the city. As for cities located in tundra biomes, the principal long-term obstacles they face to sustainability are sourcing sufficient renewable energy, particularly in winter when sunlight is much reduced, and growing enough food locally, given the obstacles to long-distance food transfers posed by a post-carbon future.
The second strategy one might use to respond to the objection that other ecosystems could also furnish important models for biomimetic cities is to note that cities typically require much greater quantities of water than is available locally in the two important biomes of grasslands and deserts, and that, were these greater quantities of water available sustainably, this surplus water may allow them to take forests as models, thus benefitting from other advantages of this model, most notably their appropriate scale and tendency to exhibit greater complexity (see below). This is particularly relevant in the case of the many cities located close to rivers in grassland or desert biomes, which could potentially use water from local rivers to realize cities based on forests. Indeed, one could even argue that exploiting local rivers in this way is simply an extension of what already happens in Nature when riparian forests form along the banks of rivers in otherwise arid areas, and in this sense the practice may be regarded as biomimetic, in contradistinction to engineering projects that redirect water flows between rivers and basins in accordance with the modern logic criticized by Benyus of “circumventing and controlling” Nature.
14.2.2 The Argument from Scale
It could perhaps be thought that we have proceeded too quickly in assuming that it is some sort of ecosystem that should constitute the basic model for the city. After all, why not use other models present in Nature? While we cannot compare the model of the forest to every other possible natural model, it will nevertheless prove instructive to consider two models frequently evoked in the literature on urban design: biological organisms and the nests of social insects. As we will see, however, both of these alternative models face significant difficulties arising from disparities in scale.
One important feature of the organismic model that should give us pause for thought is that it has for centuries already served as an important model—or at least as an important analogy or structuring metaphor—for the conception of the city. Many architects of the early Italian Renaissance, such as Giorgio Martini, Filarete, and Alberti took the human body as model for the city (Choay, 1974). Similarly, from the nineteenth century onwards, the model of the biological organism was also very common. Cerda, for example, understood the city as an “urban organism”, with the urbanist’s role being that of its “anatomist” and “doctor” (Choay, 1996). Likewise, in Urbanisme, Le Corbusier (1925) constantly describes the city in biological and physiological terms, and, in keeping with this, sees the role of the urbanist as that of either “doctor” or “surgeon”. Even today, this model is still very much present, whether in highly scientific approaches to cities (Batty, 2013) or in the context of smart cities (Picon, 2013).
One thing that proponents and theorists of the “urban organism” have typically overlooked, however, is the question of scale. For the architects of the early Italian Renaissance, the principal geometrical feature of the human body they sought to reproduce was its proportions, which, in keeping with the notion of Vitruvian man, were regarded as “perfect” (Wittkower, 1962). But taking the human or animal organism as model implies that the city will take on the ecological function of a giant consumer, incapable of carrying out analogous ecological functions to those of producers and decomposers. An unfortunate consequence will be huge and often unsustainable demands placed on external ecosystems both in terms of resource provision and waste absorption. Much the same may be said about contemporary attempts to theorize the smart city as an “urban organism” (Picon, 2013). While these have typically attempted to address the problem of unsustainability via the notion that information and communication technologies—conceived via the model of the sensory apparatus and nervous system—could make the city more efficient, thanks to the detection of leakages, rapid or pre-emptive reaction to external perturbations, and so on, they do not consider the possibility that the model of the organism may, in large part for reasons of scale, be fundamentally problematic no matter how efficient it may become.
A similar issue arises regarding the nests of social insects. Lovelock (2014), for example, thinks that the nests of termites provide a good overall model for the city, in part because they could allow us to solve the problem of excessive urban temperatures resulting from global warming. More specifically, he thinks it may be possible to build entire cities largely closed off from the external environment, as buildings generally are, and thereafter regulate their internal temperature using bio-inspired passive cooling techniques pioneered by termites. Again, however, one obvious danger with this model is that cities are orders of magnitude larger than insects’ nests and that, like living organisms, these nests also assume the ecological role of consumers and are thus heavily dependent on their external environment both for the provision of resources and the absorption of wastes.
If the above arguments do indeed suggest that the forest may be a more appropriate model for the city than either biological organisms or the nests of social insects, it may nevertheless be objected that there is no reason why one single model should be taken as fundamental. Indeed, it could even be argued that privileging the model of the forest amounts to a sort of fundamentalism and that a more pragmatic and pluralist approach would have the advantage of enabling the articulation of multiple models for the city, without any one of them being accorded priority. The principal weakness with this objection is its assumption that highly diverse features of natural systems operating at radically different scales may be abstracted from their natural context and thereafter blended, modified, and manipulated at will. One important issue this assumption overlooks is the way that within living systems the presence of certain elements or properties will often strongly favour or disadvantage others. For example, it may well turn out that taking trees as models for energy generation at building-level will also strongly favour the adoption of a comparable mechanism to that found in the forest for temperature regulation, namely evapotranspiration; conversely, it may prove very difficult to combine tree-inspired energy generation systems with a gigantic city-wide air-conditioning system that requires the city to be largely sealed off from the external environment, as is the case in termites nests. To say this does not entail commitment to a strong holism, according to which specific parts of living systems cannot be taken as models and thereafter translated into other contexts, for that would by definition render all biomimicry impossible. Seeing the forest as the fundamental model for the city, understood as that model that gives the biomimetic city its fundamental characteristics, but without determining its every last detail, would still allow us to draw on biological models not found in forest ecosystems. To take a simple example, sea sponges have aroused the interest of biomimicry researchers thanks to the advanced fibre optics they possess (Kulchin et al., 2008). But the simple fact they are found in the sea is no reason not to import this technology into urban telecommunications networks. Indeed, the “appropriateness” of this natural technology has little to do with the specifically marine environment in which it first developed and more to do with its advantageous mechanical parameters and low-temperature production process. This in turn reminds us of the importance of not interpreting “appropriateness” or “fittingness” in too narrow or dogmatic a way, as if the sole relevant criterion were whether the model to be imitated already exists in the local ecosystem.
14.2.3 The Argument from Complexity
A third argument for taking forests as models for cities, and one which we have already alluded to in our earlier comparison of the earth’s major biomes, concerns the fact that forests are by and large the most complex land-based ecosystems. This is of course a broad generalization. Barthlott, Erdelen, and Rafiqpoor (2014) note that biodiversity varies in large part as a function of what they call “geodiversity”, understood as diversity of abiotic factors, such as climate, geomorphology, and geology. Biodiversity is interpreted here in a rather narrow way as species diversity, which is clearly not the same thing as complexity, but it does nevertheless suggest that some non-forest ecosystems may exhibit greater complexity than some forest ecosystems. This proviso aside, the basic reason for the tendency of forests to exhibit greater complexity is the availability of significant quantities of water, minerals, and organic compounds, which, coupled with appropriate temperatures, generally allow forests to contain more diverse elements and more diverse relationships between these elements than other land-based ecosystems. These elements and relations in turn make possible such important system-wide properties as relatively high rates of nutrient recycling, effective self-regulation, and resilience to perturbations. Further, it is also true that those forests that benefit from the most favourable climatic conditions, namely tropical rainforests, are not only generally the most diverse of forest ecosystems, but also the most stable and resilient (Ghazoul, 2015).
The high complexity often characteristic of forests constitutes a significant advantage for biomimetic urbanism, for it allows them to offer a greater variety of natural models to imitate, whether at the level of individual elements, relationships, or systemic properties. Further, given the great complexity of cities, it is also debatable whether ecosystems working at relatively low levels of complexity offer viable models in this context. Grasslands may well offer important models for biomimetic agricultural systems (Jackson, 2011), but, as we have already seen, the limited water resources and the reduced complexity that result may pose a significant obstacle to their viability as models for cities. Deserts may contain clusters of cacti, which could perhaps be taken as models for small villages that emulate the water-saving and energy generating capacities of these succulent plants, but it is doubtful whether they exhibit sufficient complexity to act as a model for an entire city. So, while it is beyond doubt that the fauna and flora of non-forest ecosystems constitute interesting models for biomimetic design, whether they can function as models for entire cities remains an open question.
14.3 Exploring the Model of the Forest
14.3.1 City Planning as “Biome-Mimicry”
In the previous sections, we referred to both forest ecosystems and forest biomes. Biomes differ from ecosystems in two respects: they are large-scale entities, whereas ecosystems may be defined at varying scales; and whereas ecosystems are defined by interactions between biotic and abiotic components, biomes are defined by their principal form of vegetation. This is not to say, however, that within a given biome various ecosystems not defined by their principal vegetation cannot also be present at smaller scales. The temperate forest biome, for example, is defined by its principal form of vegetation, though any given portion of this biome may also contain such ecosystems as rivers, lakes, wetlands, treeless floodplains, and steep slopes or cliffs. In what follows, we will draw on the distinction between ecosystems and biomes to propose a vision of city planning as the imitation of the native biome, or “biome-mimicry”, an approach that has thus far been applied mainly to regional planning (Biomimicry Guild and HOK, 2013).
The traditional way that environmentally conscious urban planners have taken the different ecosystems of the native biome into account, as opposed to just smothering them by clear-cutting forests, burying rivers, draining wetlands, building on flood plains, and so, was through the method of “ecological determinism” advanced and practised most notably by McHarg (2006). Ecological determinism consists in mapping the various natural features—geological, hydrological, biological, ecological…—of the local site, categorizing them in terms of their relative ecological value as established via considerations of productivity, biodiversity, role in risk mitigation, and so on, and then privileging the development of those sites whose ecological value is lowest.
An important aspect of McHarg’s ecological determinism is that, as Corner (2006) has noted, it continues to suppose a dichotomy between nature and the city. In doing so, it also remains within the “dualist” mind-set characteristic of what environmental philosopher, Mathews (2011, p. 364), calls “the traditional project of environmentalism”, the basic focus of which is Nature preservation and conservation. Where Mathews thinks biomimicry differs from this dualist mind-set is in its belief that humans are not necessarily a destructive ecological force, since, through following Nature’s basic rules and principles, such as running on solar energy, recycling nutrients, or using energy efficiently, we may come to participate constructively in the life of the biosphere, a change in perspective that Mathews thinks represents a “turning point in Western thinking” (Mathews, 2011, p. 368). Applied to the city, this would mean that any natural ecosystem within the forest biome could at least in principle be the subject of biomimetic development or, in those cases where it had already been smothered by traditional urban development, regenerated in such a way that it could participate effectively in a city working in analogous ways to a forest ecosystem. As an example of the former, undeveloped wetland areas could be taken as sources of inspiration for new wastewater treatment systems, whether through adapting existing wetlands to that purpose, building specialized sewage treatment plants based on wetland ecosystems (Todd & Todd, 1993), or by some combination of the two. An example of the latter would be the daylighting and regeneration of urban streams and rivers such that they could henceforth carry out various important ecological services—e.g., flood protection and water purification—important to the functioning of “cities like forests”.
In keeping with the status of the forest as the defining ecosystem of forest biomes, it is also true, however, that it is this ecosystem that would constitute the principal model for the built urban environment. One important planning strategy this model suggests is increasing density through verticality, whether through upward or downward growth, both of which could draw inspiration from a key feature of tree growth: modularity, as seen in the growth of trunks, leaves, branches, and roots. This does not imply, however, that skyscrapers would become the norm. Within forests, upward growth is subject to limiting factors and the same would also be true of biomimetic cities; cities composed primarily of skyscrapers, for example, would no doubt struggle to generate sufficient solar energy, for increasing volumes would lead energy demand to increase disproportionately with respect to supply. As for downward growth, it is interesting to note that the celebrated French architect, Dominique Perrault, often frames his key concept for the densification of cities—the “groundscape”—in biomimetic terms (Degioanni, 2015), a topic to which we will return when we come to examine various sector-specific biomimetic innovations.
14.3.2 Urban Water Systems
In most forest ecosystems, the predominant source of water is precipitation and only to a minor extent streams, rivers and groundwater. This model implies the generalization of decentralized rainwater harvesting, whether for non-potable uses (irrigation and gardening, laundry, industry, fire-fighting reserves), to reduce surface runoff and flooding, to refill aquifers (Han & Mun, 2011), or even—especially in places with water scarcity or uneven precipitation regimes—for drinking water production. The transition to decentralized rainwater harvesting would require improved water quality and thus also a reduction of anthropogenic pollutant emissions and the development of appropriate rainwater treatment technologies. The forest biome offers appropriate models for both of these, for it not only runs on non-toxic biological nutrients, but also filters water in ways we might profitably imitate, including physical filtration, phytodepuration, and mycofiltration (Stammets, 2005). In addition, decentralized rainwater harvesting would significantly increase resilience in the face of technological failures, terrorist or military attacks, natural disasters, and other disruptions.
Another important feature of the forest biome is the use of rainwater for cleaning purposes. The leaves of many plants, including most famously the lotus, have hydrophobic surfaces which enable rainwater droplets to collect and remove impurities. This property has already been imitated in the form of self-cleaning paints and surfaces (Forbes, 2005), but its natural deployment in leaves suggests a particularly important application: self-cleaning solar panels, which could prevent efficiency reductions of up to 30% caused by surface impurities (Crawford, 2012). Further, forest rainwater that has previously been used for cleaning purposes will later reach the ground where it can be re-used. Drawing on this model, the local recycling and reuse of greywater—generally used initially for cleaning purposes (washing buildings and vehicles, showers, laundry…)—could be developed for, amongst others uses, irrigation or gardening (Li, Wichmann, & Otterpohl, 2009).
The model of the forest could also underlie a paradigm shift in stormwater management. In existing densely urbanised environments, more than 50% of precipitation is transformed into surface runoff and evacuated by means of artificial systems (drains, sewers) into surface waters or aquifers, approximately 15% is infiltrated into the soil through the remaining pervious areas, and the last 30–35% either evaporates directly or evapotranspires. In forest ecosystems, by contrast, surface runoff is typically only around 10%, infiltration 50% and evapotranspiration 40% or above (Barraud et al., 2009). Taking the forest ecosystem as model would thus imply a significant reduction in the imperviousness of urban ground surfaces, by means of porous pavements and roads, infiltration ponds and swales, raingardens, and bio-filters. Further, drawing on the model of the tree, architects could design buildings in such a way that they mimic tree roots in facilitating the infiltration of water into the soil, thus reducing the need to dedicate large surface areas to artificial infiltration basins. Increasing infiltration in these ways would offer many advantages, in particular groundwater recharge, contribution to river base flow, and, with adequate storage similar to water retention in forest soils, the generation of additional water resources for use in dry periods, as advocated by proponents of the “sponge city” (Tu & Tian, 2015). As for evapotranspiration, it could be imitated via a diversity of techniques, including vegetated roofs and walls (which also reduce runoff), growing creepers up buildings, raingardens, natural and constructed wetlands, and tree planting (Dover, 2015). The principal advantages of these techniques include a reduction of the urban heat island effect, aesthetic and psychological benefits for humans, and an increase in urban biodiversity.
Wastewater treatment is another sector the forest biome allows us to rethink in a more ecological manner. The model of the forest biome implies both the greater use of decentralized wastewater treatment and the development of treatment techniques that approximate more closely to the complex ecological treatment of wastes characteristic of natural wetlands. Regarding decentralized treatment, dry toilets, faeces composting, and separate urine treatment systems could not only reduce water consumption and the pollution of surface waters (especially in combined sewer systems), but also make possible efficient treatment of pharmaceutical wastes and nutrient recovery for re-use as fertilizers. As for dedicated wastewater treatment systems, it would be possible further to develop and generalize the use of wetland-inspired water-treatment plants, which not only facilitate the recovery of nutrients (biogas, nitrogen, phosphorous) but also reduce energy use for oxygenation (Todd & Todd, 1993). Further, as noted above, the restoration of natural wetlands or the construction of artificial wetlands could also play an important role in wastewater treatment, though the viability of this method would likely depend on the adoption of other biomimetic practices, notably a reduction of the quantity (via less wasteful consumption and greywater reuse) and an improvement of the quality (via avoidance or treatment of anthropogenic pollutants) of the wastewater entering these ecosystems (Kadlec et al., 2000).
14.3.3 Urban Energy and Transport Systems
The principal energy source of forests is sunlight. Taking trees as models for buildings not only suggests the possibility of fitting each individual building with an array of solar panels, but could potentially also lead to developments in biomimetic solar technologies, including water-splitting technologies that reproduce many of the functional properties of leaves (Chong, Colón, Ziesack, Silver, & Nocera, 2016; Reece et al., 2011). Similarly, the study of leaf arrangements on stems, phyllotaxis, may help us arrange these solar technologies in optimal ways, and the imitation of heliotropism could help maximise solar exposition. The French biomimicry specialist, Gauthier Chapelle (2015), even goes so far as to imagine translucent solar panels positioned at different levels and capable of capturing different wavelengths of light, as is the case in forests. Beyond plant photosynthesis, various other living organisms could offer useful models either for the improvement or development of other renewable energy sources (Harman, 2013) or for temperature regulation (Gruber, 2011).
A well-known limitation of solar energy is its intermittency. In forests, this issue is resolved via a number of adaptations, including seasonal changes (lower growth rates, leaf loss) and energy storage in the form of sugars, lipids, or other types of biomass. Regarding the former, it is quite possible that in strongly seasonal climates urban energy use would need to decline significantly in winter months, though drawing on geothermal energy—as some animals do for warmth—could partially offset this problem. Regarding the latter, while energy storage is a widely recognized necessity for the transition to renewables, the important question it raises is whether it would be possible to store energy in simple chemical form—hydrogen and hydrogen compounds being the obvious examples—rather than in conventional battery technologies. Indeed, were water-splitting artificial photosynthesis technologies to become viable at a large scale, then this could potentially signal a move away from the current focus on electricity storage and towards a hydrogen-based energy storage system closer to that found in forest ecosystems.
Another important feature of forests is the rhizosphere, which could provide a response to the problem of uneven energy access and demand. In forests, trees exchange energy for nutrients with an extensive network of mycelium, a process which also makes excess energy available to other trees. The net result is the transfer of energy from trees that have an excess of energy to those that, for a variety of reasons—especially differences in species, age and height—are energy deficient (Simard et al., 1997). This system may provide a model for a distributed underground energy network enabling buildings that are at any given moment “energy positive” to transfer their excess energy to those that are “energy negative”. Further, the information flows that manage these transfers and exchanges in what has been referred to as “nature’s internet” (Stammets, 2005) and the “underground information superhighway” (Bais, Park, Weir, Callaway, & Vivanco, 2004) could also provide a natural model for the convergence of distributed energy and communication technologies advocated by Rifkin (2011).
A further way in which energy use in forests could provide a model for cities concerns the transport sector. The lightweight bone structures and morphological adaptations of animals have long been used as a source of inspiration for vehicle design, often with a view to improving energy efficiency (Harman, 2013). As far as the model of the forest is concerned, however, perhaps the most important source of inspiration is the interactions between producers (especially trees) and consumers (insects and animals). Just as the insects and animals of the forest obtain their energy from the trees and other plants, so the vehicles of the city could obtain their energy directly from the buildings, as has also been advocated by Rifkin (2011). As for urban vehicle parks, it would again be possible to draw on the model of the forest to design multiple tree-shaped structures—posts on which an artificial canopy of solar technologies would be mounted—, which could not only recharge electric or hydrogen-powered vehicles, but also provide shade, just as the trees in the forest do for animals and other organisms. Lastly, nature could even provide models for innovative urban transport systems. Schuiten (2010), for example, has drawn on the model of pine processionary caterpillars to propose an automated transport system that can work in both “private” (separate) and “public” (processionary) ways, thus reducing both fuel consumption and congestion.
14.3.4 Urban Food and Nutrient Systems
The idea of “buildings like trees, cities like forests” was put forward in Braungart and McDonough’s Cradle to Cradle (Braungart & McDonough, 2002), a book better known for the idea that there are two basic types of nutrients—biological (biodegradable) and technical (non-biodegradable), both of which may be infinitely recycled. In combination with the idea of “cities like forests”, this implies a transition towards much greater recycling of both types of nutrients within the urban environment.
Despommier (2011, p. 3) has recently applied this principle to urban agriculture, arguing that in order to transform cities into the “functional urban equivalent of a natural ecosystem”, we must both produce and recycle the food consumed in the city within the city itself, hence his advocacy of vertical farming. On our view, this argument overlooks two key issues. The first concerns the fact that, while biomimetic cities would come to resemble forests in certain important respects, they would nevertheless remain cities, a state of affairs which raises the important question of what a city is. Without advancing a full response to this question, it is important to note that ever since their initial development in Natufian and Neolithic cultures, cities have depended on surrounding agricultural systems capable of generating high quantities of calories in the form of cereals, meat, and dairy products (Mithen, 2004), in which case it seems plausible to say that dependency on extra-urban agricultural production has always been part of what makes a city a city. The second concerns the fact that, unlike forests, cities contain a very high concentration of one species, homo sapiens (and their domestic companions). Deploying sufficient numbers of vertical farms to feed all the human residents of the city would amount to a radical departure from the workings of the forest, for a forest could never produce enough food to feed current urban populations of humans. Further, it also seems likely that the attempt to increase urban food production to anything like 100% of local consumption would in many cases lead to problematic competition with other uses of available space. In view of all this, it is very hard to see how the calories currently provided by cereals and dairy animals—both of which inhabit non-forest biomes (grasslands)—could be produced in “cities like forests” using nothing other than the renewable resources directly available in the urban environment, a state of affairs that raises the important question of the models of both the prairie (Jackson, 2011) and the forest in biomimetic agriculture (Hervé-Gruyer & Hervé-Gruyer, 2014). To be sceptical of the idea that all food production could potentially take place in the urban environment does not imply, however, that urban farming is to be excluded. On the contrary, the forest offers many natural models one could draw on in this context, including fruit-producing trees, the production of herbs, vegetables and berry-producing plants in clearings, the cultivation of fungi by certain species of ants, honey production from bees, and raising forest-dwelling scavenging animals like pigs and chickens.
As far as the technical nutrients currently dominant in the manufacturing and construction sectors are concerned, while it is true that more or less closed-loop recycling of these nutrients implies a move away from the linear flows of the “extractive economy” (Benyus, 1997) and the imitation of the high-levels of nutrient cycling characteristic of forest ecosystems, it is also true that biomimicry has been critical of the “heat, beat, and treat” techniques used in the production and recycling of technical nutrients (Benyus, 1997). With this in mind, the generalization of the model of the forest would seem to point towards a reduction of technical nutrients and the transition to a bio-based, solar economy, as advocated by Scheer (2002). Of particular interest in the context of architecture and urbanism would be the generalization of infinitely recyclable, biological construction materials, the most obvious one—given the model of the forest—being wood. As for manufacturing, it is interesting to note that 3D printers and other decentralized manufacturing techniques, which could potentially be designed to run on biological nutrients (Kennedy, Fecheyr-Lippens, Hsiung, Niewiarowski, & Kolodziej, 2015), could enable us to approximate to the hyper-local production and recycling systems characteristic of forests, thus reducing ecologically disruptive and energy-intensive flows of materials at national, continental, and global scales.
14.4 Conclusion
This chapter had two principal objectives. The first was to justify the idea that the forest is at least the most important model, and, more controversially, perhaps even the only viable model, for biomimetic cities. If the three arguments put forward—from fittingness, scale, and complexity—were no doubt insufficient to prove the stronger of these two positions, such that it clearly remains an open question as to whether tundra, grassland, or desert biomes may function as models for cities in corresponding locations, they do nevertheless strongly suggest that the forest is likely to prove the most important model for biomimetic urbanism. Moreover, even if not presented as such, the exploration of this model with respect to urban planning and design carried out in the second part of the article is perhaps the model’s most important justification, for it is through exploring the concrete possibilities it opens up that one becomes aware of the awe-inspiring potential of forests as models for sustainable cities.
It is also true, however, that our exploration of the model of the forest was highly selective, both in that it focused on certain features of the forest at the expense of others and in the sense that it adopted a sectorial approach, when other approaches would also no doubt have been possible. This in turn suggests that much further research is needed into this model, including: a more detailed study of a wider range of key sectors of forest-based urban planning and design, as well as explicit reflexion on their articulation and integration; analysis of the pertinence of different types of forest for biomimetic urban design; consideration of how the temporality of forest ecosystems, including such issues as seasonality, succession, evolution, resilience, and adaptation to climate change may serve as models for urban design; exploration of the spatiality of forest ecosystems qua models for urban design, an obvious starting point for which would be the common division of the forest into different levels (the rhizosphere, the forest floor, the understory, the canopy, and the emergent level), but which could also extend to a consideration of the embeddedness of forests at regional, continental and global scales; and, perhaps most difficult of all, critical reflection on the relation between forest-based cities and the human and social sciences, whether in the form of such relatively practical considerations as the institutional obstacles to realizing biomimetic cities, or deeper, anthropological and philosophical questions about what taking forests as models for cities would mean for our understanding of what it means to be human and of our overall place in Nature.