Dynamic heterogeneity: a framework to promote ecological integration and hypothesis generation in urban systems
- 3.8k Downloads
Urban areas are understood to be extraordinarily spatially heterogeneous. Spatial heterogeneity, and its causes, consequences, and changes, are central to ecological science. The social sciences and urban design and planning professions also include spatial heterogeneity as a key concern. However, urban ecology, as a pursuit that integrates across these disciplines, lacks a theoretical framework that synthesizes the diverse and important aspects of heterogeneity. This paper presents the concept of dynamic heterogeneity as a tool to explore how social and ecological heterogeneities interact and how they together act as both an outcome of past interactions and a driver future heterogeneity and system functions. To accomplish this goal, we relate heterogeneity to the fundamental concept of the human ecosystem. The human ecosystem concept identifies key processes that require operationalized models of dynamic heterogeneity in three process realms: the flow of materials, the assembly of urban ecosystem biota, and the locational choices humans make concerning land. We exemplify a specific dynamic model of heterogeneity in each of these realms, and indicate a range of complementary statistical approaches to integrate the drivers and outcomes of dynamic heterogeneity across the three realms. We synthesize a hierarchical framework for a theory of dynamic urban heterogeneity, noting its complementarity to other major urban theories and general model approaches. We hypothesize that human actions and structures amplify the dynamics of heterogeneity in urban systems.
KeywordsCommunity assembly Interdisciplinary Locational choice Spatial heterogeneity Theory Urban ecology Ecosystem Watershed
There are many ways to conceive of cities and urban ecosystems (Marcotullio and Solecki 2013). An important one is as spatially heterogeneous and temporally dynamic mosaics (Pickett et al. 1997; Dow 2000; Pickett and Cadenasso 2008; McGrath and Shane 2012; Qureshi et al. 2014; Grove et al. 2015). Both pattern and process in such mosaics are of interest (Liu et al. 2007; Stevenson 2011), as emphasized by the term dynamic heterogeneity. The biophysical and social heterogeneity of urban systems can take many forms. From a biological ecosystem perspective, heterogeneity can appear in cities as the differential distribution of vegetation structure, as hotspots and dead zones of nutrient and organic matter processing, as patches of high versus low soil organic matter, or as the distribution of leaf litter and coarse woody debris in streams and drainage infrastructure, for example. Heterogeneity in the physical environment appears in such conditions as cooler and warmer zones of air and surface temperatures, or as locations of high atmospheric advection and contamination. In the biota, heterogeneity exists as different levels of species richness, as differential functions of introduced species, and as sources or sinks of disease organisms and their vectors, for example. Although this list of examples is long, it does not exhaust the kinds of bioecological heterogeneity that can exist in urban systems. Such heterogeneity in the biota and the physical environment is fundamental to general ecological theory (Scheiner and Willig 2011). In fact, five of the eight fundamental principles of ecology articulated by Scheiner and Willig (2011) specifically mention heterogeneity or centrally involve heterogeneity of pattern and process.
The heterogeneity of urban areas is also of interest from the perspectives of social sciences, urban design and planning, and policy and management (Dow 2000; Forman 2008; McGrath and Shane 2012). Social-economic heterogeneity is expressed in such things as zoning regulations, neighborhood identities, housing prices, patterns of investment in green versus grey infrastructure, and in tourist guide books. It is embedded in warnings about where to go and where to avoid, and where to seek the various amenities that people desire in urban systems. Importantly, we do not endorse “spatial determinism,” in which social outcomes are explained solely by place, as exemplified by the classic Chicago School (Grove et al. 2015). Rather, we acknowledge that social and environmental relationships and the feedback between them are differentially distributed in space. Indeed, “space” in urban systems is clearly as much a social product as it is a locational description (Thierstein and Förster 2008).
There are various principles that employ urban heterogeneity in theory and in practice (Muth 1969; Zipperer et al. 2000; Band et al. 2005; Shane 2005; Jenerette et al. 2006; Pickett and Cadenasso 2008; Irwin 2010; Roy Chowdhury et al. 2011; Swan et al. 2011; Cadenasso 2013). However, in our pursuit of integrative tools for social-ecological, long-term research, we have been hampered by the lack of a clear theoretical framework of dynamic heterogeneity that can facilitate communication across disciplines and can encourage melding of social and ecological perspectives and approaches. Therefore, in this paper we articulate a theoretical framework of urban ecology that addresses dynamic heterogeneity. Our view of urban ecology implies an integration with social sciences, and our view of heterogeneity encompasses both space and time, patterns and processes.
Other inclusive theories of urban systems exist, though none has been embraced as predominant or comprehensive of all factors of interest (Box 1; Batty 2013; Brenner 2014b; McHale et al. 2015). We do not propose heterogeneity theory as a replacement for those theories. All can contribute to a complete understanding of urban systems. However, those additional theories are beyond the scope of this paper.
• Local-regional urban gradients and global change (Grimm et al. 2008).
• The continuum of urbanity (Boone et al. 2014).
Plan of this paper
This paper develops the concept of dynamic heterogeneity using the following scheme. First, we acknowledge that concern with heterogeneity in the urban realm is relevant because of the growing importance of city, suburban, and exurban (CSE) systems, along with their regional and global connections (Boone et al. 2014; Brenner 2014a). We then focus on the key insight that dynamic heterogeneity is important both as a driver and an outcome of biophysical and social urban processes. Heterogeneity in one feature of a CSE can act as a driver by interacting with temporal changes, pulsed events, or other patterns of heterogeneity to generate new spatial templates of heterogeneity -- that is, heterogeneity as outcome. As driver and outcome, dynamic heterogeneity mediates between the social and the biological in CSE systems. We then apply the concept of dynamic heterogeneity to urban systems and present a hierarchical structure of a general theory of urban heterogeneity. The hierarchy uses an inclusive theory that has a general principle – that urban systems are organized as spatially differentiated, functionally significant mosaics. This general theory encompasses more focused, middle-level theories that are intended to represent the full range of processes and patterns that contribute to the higher level, general theory. Middle level theories themselves encompass a variety of more specific and mechanistic models of urban pattern and process that can be tested or applied in specific situations. The articulation of expectations and hypotheses representing middle-level theories and the integration among middle- and lower-level theories is an important frontier for urban ecology.
Defining dynamic urban heterogeneity
The spatial relevance of dynamic heterogeneity is broad because urban systems consist of CSEs and because they are parts of regionally and globally connected systems (Schmid 2014). Heterogeneity can appear at all spatial scales that occur within CSEs, as well as across urban agglomerations and megaregional networks (Harrison and Hoyler 2014). However, urban heterogeneity is not stationary in time (McGrath and Shane 2012). Rather, human and natural processes and the interaction between them are dynamic (Alberti 2008). Therefore, spatial heterogeneity is as much about time and processes as it is about spatial structure, a feature emphasized by the phrase “dynamic heterogeneity.” This sketch of an inclusive theory of dynamic heterogeneity is part of a long and broad tradition concerning the spatial structure of urban systems (Burgess 1925; Gottdiener and Hutchison 2011). It reflects one of the classic definitions of urban systems as contrasted with socially more homogeneous villages (Wirth 1945). Urban heterogeneity has motivated those interested in immigration (Taubenbock et al. 2014), social justice (Bryant 1995), urban planning (Sukopp and Weiler 1988), political power (Raleigh 2014), lifestyle identity (Grove et al. 2003), and indeed, many other topics. The social production of space (Lefebvre 1991), the patchy reinvention of cities and their enclaves over time (Shane 2005; McGrath and Shane 2012), and the recognition of fine scale social dynamics (Jacobs 1961) are examples of concern with dynamic heterogeneity beyond ecology.
Ecology has likewise had an important research focus on the dynamics of spatial heterogeneity. Population ecology, where the evolutionary significance of spatial difference within populations is clear (Levins 1968), perhaps began modern ecology’s concern with heterogeneity. This continued through spatialization of community ecology via dispersal and metacommunity concerns (Chesson and Chase 1986), and on to the invention of landscape ecology as a disciplinary home for studies of heterogeneity (Forman 1981). Similarly, ecosystem ecology has come to acknowledge that both the heterogeneity represented by species and by spatial patchiness are significant for ecosystem functioning (Jones and Lawton 1995; Leibold et al. 2004; Band et al. 2005; Pickett and Cadenasso 2013).
An example of emerging dynamic heterogeneity
Feedbacks between and among ecological processes flow through filters of spatial heterogeneity (Wiens 2000). These feedbacks also construct new patterns of heterogeneity that can affect subsequent interactions (Pickett et al. 2000). An example might be the existence of a spatially heterogeneous pattern of forest trees, including ash trees (Fraxinus spp.; Fig. 1b). As the introduced emerald ash borer (Agrilus planipennis) invades and kills ash trees in North American cities (Jones and McDermott 2015), a new pattern of heterogeneity is generated -- the distribution of forest canopy gaps created by ash mortality. Of course the actual mortality of ash trees in time and space is conditioned by such things as preemptive cutting of ash trees by people, or by decisions to protect particularly valued ash trees through extraordinary actions (Fig. 1b; Jennings et al. 2015; Jones and McDermott 2015). Such conservative or adaptive actions are patchy and differentially distributed across a CSE mosaic. The subsequent heterogeneity of the forest is affected by natural succession or by compensatory planting of different tree species by forest managers. This emerging “ash spiral” of interacting heterogeneities with natural and socially driven events is now underway in many urban areas in the eastern U.S. Thus, the interactions between different elements of heterogeneity are key mechanisms of urban dynamics. Of course, urban dynamics are also affected by social and biophysical structures and fluxes from outside their boundaries, or which operate on very long time frames in which the urban system is embedded (Olson 1982; Seto et al. 2012).
The human ecosystem: A foundation for dynamic heterogeneity
All four components of human ecosystems (Fig. 2) can be heterogeneous. The role of heterogeneity in the functioning of biologically focused models of ecosystems is widely recognized (Hutchings et al. 2000; Lovett et al. 2005; Scheiner and Willig 2011). The human ecosystem concept suggests that heterogeneity also involves human decision-making processes and the buildings and infrastructure that result from these decisions (Fig. 2). The heterogeneous control of ecosystem structure and function (Shaver et al. 2007; Pickett and Cadenasso 2013) is an important general reserach question in ecology. To understand how heterogeneity influences the interaction among ecosystem components, we employ three broad process realms that structure ecosystems: 1) critical ecosystem fluxes, including their controls, pools, and transformations (Likens 1992); 2) the assembly of biological potential, that is, what organisms are present, their traits, community dynamics and spatial arrangement, and what influences their distribution (Chapin et al. 2011); and 3) the structures people build, the technologies they employ, their cultural contexts, and the decisions they make (Spirn 1984; Berry 1990; Smith and Stirling 2010). The flows of matter and energy, the assembly and biogeochemical potential of the biota, and human choices all interact reciprocally. The three process realms focus on entities and phenomena that hybridize the four components of the human ecosystem. All likewise interact with spatial heterogeneity through time so that pattern and processes are addressed. Material flows, biotic differentials, and human decisions are broad processes that all landscapes share (McDonnell and Pickett 1993). Therefore, these three realms are fundamental to understanding any ecosystem that includes people, their artifacts, or their effects. These three realms integrate across the components of the human ecosystem, and allow for examination of dynamic heterogeneity. While the process realms can be viewed through the familiar lens of pattern and process, the dynamic heterogeneity approach emphasizes that pattern is itself dynamic and is an active contributor to ecological change.
A nested hierarchy of heterogeneity theory
These three process realms are extraordinarily broad. In order to put them to work in generating clear explanations and hypotheses about dynamic heterogeneity in urban systems, more specific models and research approaches are required. Seeking such intermediates is akin to the use of “middle-level theories” as recommended in social science research (Merton 1968; Cadwallader 1988), and the use of hierarchical theories as employed in the natural sciences (Wu and Loucks 1995; Pickett et al. 2007; Scheiner and Willig 2011). The general theory for the entire discipline of ecology is supported by the key principles that organisms are variable, that environments are heterogeneous and historically contingent, and that organisms are mortal and are resource-limited. This broad and comprehensive theory is operationalized by more narrowly focused, middle-level theories. Such constituent theories are the familiar chapter headings of ecology textbooks, such as community assembly, ecosystem processes, life history evolution, negative and positive species interactions, environmental gradients, and so on (Scheiner and Willig 2011).
Although there are many model types that might operationalize the three process realms, our choices are guided by 1) the ability to connect with broader disciplinary areas and 2) relevance to long-term urban processes. For the broad area of flow of materials, represented by water and contaminants, we employ an urban version of the watershed continuum concept (Groffman et al. 2004; Walsh et al. 2005; Kaushal and Belt 2012). For the broad area of differential spatial distribution and performance of organisms, we choose metacommunity theory as the representative translational concept (Swan et al. 2011). Finally, for the human decision making arena of theory, we apply locational choice theory that draws together concepts from economics, geography and urban planning (Irwin et al. 2009). We acknowledge that other modeling strategies are possible (e.g., Box 1); however, our roster is not determined by judgments against the ones not included but rather by its relevance to evolving urban CSE systems. We do not focus on such specific approaches as urban metabolism, the power law approaches to urban prediction, theories of the social production of space, or political ecology, for example. However, our explanations, specific models, and relevant data streams do address components of these other model approaches, and there is necessarily some mechanistic overlap with the set of theories outside our scope.
To illustrate the theoretical structure of each of these areas, we outline them using “if-then” statements that expose their conditional structure, which is an important form for biological theories (Pickett et al. 2007). If a stated condition or relationship holds, then a certain outcome is predicted. Such predictions can be tested in various ways: experiments, modeling, comparison, and long-term studies (Carpenter 1998). This structure is common to general theories in the physical and biological sciences, and it suggests the key hypotheses of those theories (Gould 1977; Pickett et al. 2007).
The urban engineered watershed: A specific model for integrating flows
Urban engineered watersheds combine biological, physical, social, and built patterns and processes, making them an important tool for studying human ecosystems comprising cities, suburbs, and exurbs. In particular, applying the watershed concept to river networks that run through CSE systems requires its own set of specific assumptions and boundary conditions (Walsh et al. 2005). For example, the substantial re-plumbing of watersheds in urban ecosystems introduces new connections and blurs ecological boundaries between the terrestrial and riverine zones in the forms of pipes, drainage infrastructure, and altered hydrological flow paths (Fig. 4; Kaushal and Belt 2012). This leads to cascading impacts of heterogeneity on the water chemistry and ecosystem functions of urban watersheds across longitudinal, lateral, and vertical dimensions as well as through time (Kaushal and Belt 2012; Kaushal et al. 2014a): 1) Longitudinally, there can be significantly enhanced “hot spots” of gross primary production along the urban watershed continuum up to five times greater than forest streams due to light availability, which can be heterogeneous due to patchy riparian cover (Kaushal et al. 2014a). 2) Laterally, downcutting of urban streams due to the flashy delivery of rainwater from impervious surfaces typically disconnects the stream from its former floodplain. 3) Vertically, hydrological connections exist between ground water, drinking water supply pipes, storm drains, sanitary sewers, and surface streams. Longitudinal, lateral, and vertical patterns of disconnection and connection among pathways can shift biogeochemical functions from sinks to sources of contaminants, such as reactive nitrogen, which can impact eutrophication in streams and receiving waters (Groffman et al. 2002). Finally, 4) there is evolution in heterogeneity of ecosystem structure, function, and services of urban watersheds across the temporal dimension based on response to degradation, management, and restoration activities (Kaushal et al. 2014b).
The urban watershed continuum concept requires a set of detailed “if clauses” appropriate to urban systems. If re-plumbing in CSE watersheds enhances the hydrologic connections between streams and the urbanized uplands, then we expect that organic inputs, temperatures, and nutrient pollution will be increased in all orders of streams affected. Temperature- and pollution-sensitive aquatic organisms will be disproportionally impacted. If streams of any size are disconnected from their former floodplains, then nitrogen removal functions of urban riparian zones will be impaired. Species composition will also shift with the drying out of stranded floodplains. We expect that the various interventions, connections, and effects in urban watersheds can all be spatially and dynamically heterogeneous.
The metacommunity: A specific model for biotic assembly and differential performance
The metacommunity concept focuses on the differential distribution of organisms across heterogeneous space, especially those in which patches are isolated from one another by inhospitable or suboptimal matrix (Leibold 2011; Alexander et al. 2012). Patch configuration can be crucial in urban metacommunity dynamics. For example, some analyses find that patch adjacencies are strongly associated with the diversity and abundance of bees (Hinners et al. 2012) and migratory birds (Carlson 2006). Because both aquatic and terrestrial habitats in urban areas are heterogeneous, we expect the metacommunity approach to apply to both those kinds of ecosystem patches. These additions to metacommunity theory, which account for human actions and decisions, are needed to apply metacommunity theory to urban situations (e.g., Swan et al. 2011).
The metacommunity approach makes predictions based on the following logic, in which several if statements identify the causal factors, and together those factors can result in several outcomes or then statements: If (1) sites are discrete, (2) sites are separated by contrasting and less- or un-suitable habitat, (3) species have differential abilities to disperse across a heterogeneous spatial mosaic, and (4) different species are differentially prone to extirpation in different habitat types, then (A) the local diversity of biotic communities will be determined by their spatial relationship to other patches, (B) the nature of the intervening mosaic will affect community composition, and (C) these factors mediate the development of biodiversity at multiple spatial scales. A layer of complexity is added by human actions requiring an if statement unique to CSE systems: If (5) the role of human-mediated dispersal of species, either as propagules or active adults, predominates then (D) human mediation can override the effects of unassisted dispersal and the unmodified rate of local extinctions of sessile organisms.
In other cases, where the assumptions of metacommunity theory do not apply, local conditions may better explain heterogeneity and suggest new model complexity. In such cases, other specific hypotheses may address the responses to land cover types present in a mosaic landscape, the relative roles of the regional versus local species pools, people’s lifestyle contrasts that affect biotic diversity and composition, the resources available for maintenance of plants and animals in different locations, the differential vagility of various taxa, and so on (Matteson and Langellotto 2010; LaDeau et al. 2013; Beauchamp et al. 2015; Johnson et al. 2015; Swan et al. 2016). For example, 94 species of birds were observed at 132 bird sampling points in Baltimore during the breeding season. The BirdLife International Range Maps suggest that 161 bird species could live in Baltimore (Aronson et al. 2014). The difference indicates that heterogeneity in such features as land use and land cover act as filters for some species during the breeding season. In other cases, local conditions explain heterogeneity in bird and mobile insect species (Croci et al. 2008; Matteson and Langellotto 2010). Metacommunity theory provides a tool for diagnosing the mechanisms yielding these different associations between spatial heterogeneity and diversity and abundance of biota. Urban-specific if statements generate new models.
Locational choice: A specific model for human decisions and the built environment
Locational choice theory focuses on the differential distribution of households and firms (Weber 1929; Hoover 1948; Lösch 1954; Isard 1956; Tiebout 1956; Myrdal 1957; Alonso 1964; Christaller 1966; von Thünen 1966; Krugman 1991) based upon three fundamental economic forces that influence location: (1) natural advantages that attract households and firms; (2) economies of concentration that enhance the productive efficiency of firms that cluster; and (3) transportation and communication costs that spatially differentiate markets and their geographical extent (Edgar and Giarratani 1999). Key extensions of locational choice theory (Graves 1980; Haurin 1980; Roback 1982) have demonstrated the importance of amenities and disamenities to explain the locational choices of households and jobs. These include urban amenities such as culture (Glaeser et al. 2001; Florida 2005) and natural amenities such as climate (Cragg and Kahn 1997) and coastlines (Rappaport and Sachs 2003; Oliva 2006); and disamenities such as crime (Cullen and Levitt 1999).
Locational choice theory can make predictions based upon “if-then” statements: if (1) households make locational choices to improve their well-being, and (2) households are separated by locations providing greater or lesser well-being (i.e. locations are heterogeneous), then a household’s location will reflect its optimal combination of preferences for amenities and disamenities for a given set of constraints, such as income, that limit that choice. A range of economic, social, and institutional drivers may influence locational choices (Alonso 1964; Muth 1969; Harr et al. 1975; Alperovich 1982; Logan and Molotch 1987; Mieszkowski and Mills 1993; Cho 2001; Bayoh et al. 2006).
How do the choices people make about where they live and how they use and manage their land reciprocally interact with dynamic heterogeneous ecological conditions? The decisions people make appear at multiple scales and can influence ecological features of an urban region, and these ecological conditions in turn feed back to influence households, neighborhoods, municipalities, and governance networks. Ecological conditions can be perceived as amenities or disamenities that influence locational, land use or land management choices. While some ecosystem structures and functions may influence individual locational and management choices directly, many other feedbacks involve higher levels such as governance networks, zoning, and regulations to protect habitats and ecosystem functions.
Integrating the three middle level theories: Flow, biotic potential, and human choice
The three realms and their middle-level theory should not be treated as disconnected models and empirical pursuits. As important as each of the three conceptual areas are in orienting long-term data collection and in motivating hypotheses, they are equally important as stimuli for integration. Urban ecology is continually challenged to integrate over spatial scales, over temporal scales, and across the many disciplines that are concerned with the structure, functioning, and shaping of urban systems (Pickett et al. 2001; Grimm and Redman 2004; Alberti 2008; Hahs et al. 2009). The three middle-level theories focusing on flux, biotic assembly and potential, and human choices, help integration by narrowing the universe over which integration must be accomplished. Integration among the middle-level theories is accomplished by specific models that identify types and magnitudes of connections. In essence, integration can focus on how each of the major process areas relates to the other two and what feedbacks exist among them. How does the nature of locational choices explain community assembly and biodiversity patterns in the metacommunity; how does it help us understand urban watershed structure? Likewise, how do watershed processes and patterns explain patterns in locational choices and metacommunity structure? Finally, how does the structure and change in the metacommunity mediate the structure and function of the watershed and people’s locational choices? The feedbacks embedded in this set of questions are particularly important for ferreting out the integrative interactions that occur in the urban social-ecological system.
If a process is well-understood, then mechanistic or simulation models can be important tools for generating hypotheses about what types of environmental influences drive deviations from predicted outcomes and for evaluating how processes will be affected by changes in common urban variables such as land cover, management, and climate.
When there is uncertainty about the relative importance of predictive variables or process structure, then it is important to let the data drive understanding through statistical models. In such cases, hierarchically structured state-space models can be employed. A hierarchical or multilevel model structure defines how information is shared across sampling units and processes, allowing researchers to accommodate important differences (e.g., among groups, individuals, or regions) while still allowing for shared characteristics (e.g., Clark and Bjornstad 2004; Clark et al. 2005). Bayesian or likelihood-based analysis (LaDeau and Clark 2006; Luo et al. 2011), or state-space formulations for understanding a process or state that is not directly observed can be used.
When the goal is to understand how ecological processes will be affected by some real or hypothetical scenario, e.g., change in land cover, management, climate, and uncertainty in process or data are a concern, then research can employ Bayesian Networks to evaluate likely outcomes (Fig. 8). Bayesian Network models are important tools in decision science and have gained attention as important tools for probabilistic evaluation of management outcomes in ecological systems (McCann et al. 2006; Leigh et al. 2011; Maxwell et al. 2015).
Conclusion: Back to inclusive, general theory
The theory of urban heterogeneity is a nested hierarchical conceptual structure that puts the following general principle into practice – urban systems are organized as spatially differentiated, functionally significant mosaics (Fig. 3). Specifically, it first assumes that urban systems are preeminently heterogeneous. Second, it assumes that there are different layers of heterogeneity, representing both biophysical and social features. Third, the layers of heterogeneity are assumed to interact at various scales. Fourth, the interactions of different layers of heterogeneity generate new heterogeneities. This leads to a fifth statement, the general prediction that heterogeneity can act as both driver and outcome in urban systems. Finally, this train of logic and empirical predictions is significant because heterogeneity is hypothesized to affect ecosystem functioning. This sixth proposition is consonant with a test of general theory of ecology with its core focus of the role of spatial and heterogeneity and its temporal changes. A novel hypothesis especially relevant to urban systems now emerges: dynamic heterogeneity is amplified by the intensity and speed of human-biophysical feedbacks in urban systems.
The general principle of the inclusive theory we have adopted can be tested, operationalized, and explained by three main sorts of process: the flow of materials, the differential distribution and performance of the biota that are key elements in ecosystem processes, and the choices that people, represented by individuals, households, organizations, and many other kinds of institutions, make about where to locate and what activities to conduct in specific places.
Each of these three constituent middle-level theories of urban heterogeneity is itself still a large topic. As in any theory, more specific models and research approaches are required to further translate the broad concepts and principles into operational models and specific hypotheses. We have explored a dynamic heterogeneity spiral as a tool to represent the interaction of different kinds of heterogeneity and urban processes and events. The spiral of dynamic heterogeneity can be considered a template to guide development of specific models of the structure, change, and importance of urban heterogeneity as driver and outcome. Our conceptual hierarchy is capped by an inclusive theory of urban spatial heterogeneity (Fig. 3). The theoretical framework has a second level that explores three broad subtheories, which are derived from the components of the human ecosystem. These subtheories supply mechanisms for the general theory, and they require still more detailed specific models and hypotheses to motivate data collection, empirical evaluations, and statistical tests.
This research was supported by funding from the NSF Long-term Ecological Research (LTER) Program under Grant No. DEB-1027188 and the Urban Sustainability RCN Grant No. RCN 1140070.
- Band L, Cadenasso M, Grimmond CS, et al. (2005) Heterogeneity in urban ecosystems: patterns and process. In: Turner M, Jones C, Weathers K (eds) Lovett G. Ecosystem Function in Heterogeneous Landscapes, Springer New York, pp. 257–278Google Scholar
- Battaglia M, Buckley G, Galvin M, Grove M (2014) It’s not easy going green: obstacles to tree-planting programs in East Baltimore.Google Scholar
- Batty M (2013) The new science of cities. MIT Press, CambridgeGoogle Scholar
- Berry BJL (1990) Urbanization. In: Turner II B, Clark WC, Kates RW, et al. (eds) The earth as transformed by human action: global and regional changes in the biosphere over the past 300 years. Cambridge University Press, New York, pp. 103–120Google Scholar
- Boone CG (2013) Social dynamics and sustainable urban design. In: Pickett STA, Cadenasso ML, McGrath B (eds) Resilience in ecology and urban design: linking theory and practice for sustainable cities. Springer, New York, pp 47–61Google Scholar
- Boone C, Cadenasso ML, Grove JM, et al (2009) Landscape, vegetation characteristics, and group identity in an urban and suburban watershed: why the 60s matter. Urban Ecosystems 13:255–271. doi: 10.1007/s11252-009-0118-7
- Boone CG, Redman CL, Blanco H, et al. (2014) Reconceptualizing land for sustainable urbanity. In: Seto KC, Reenberg A (eds) Rethinking urban land use in a global era. MIT Press, Cambridge, pp. 313–330Google Scholar
- Bormann FH, Likens GE (1969) The watershed-ecosystem concept and studies of nutrient cycles. In: van Dyne GM (ed) The ecosystem concept in natural resource management. Academic Press, New YorkGoogle Scholar
- Brenner N (2014a) Introduction: urban theory without an outside. In: Brenner N (ed) Implosions/explosions: towards a study of planetary urbanization. jovis Verlag, Berlin, pp. 14–30Google Scholar
- Brenner N (2014b) Implosions/explosions: towards a study of planetary urbanization. jovis Verlag, BerlinGoogle Scholar
- Bryant JP (ed) (1995) Environmental justice: issues, policies, and solutions. Island Press, Washington DCGoogle Scholar
- Burgess EW (1925) The growth of the city: an introduction to a research project. In: The city. University of Chicago Press, Chicago, pp. 47–62Google Scholar
- Cadenasso ML (2013) Designing ecological heterogeneity. In: Urban design ecologies: AD reader. John Wiley & Sons, Hoboken, pp. 271–281Google Scholar
- Cadenasso ML, Pickett STA (2008) Urban principles for ecological landscape design and management: scientific fundamentals. Cities Environ 1:Article 4.Google Scholar
- Carlson CE (2006) The relationship between breeding bird community structure in urban forest patches and the human-mediated resources in the surrounding residential matrix. University of Georgia, PhDGoogle Scholar
- Chesson PL, Chase TJ (1986) Overview: nonequilibrium community theories: chance, variability, history, and coexistence. In: Community ecology. Harper and Row, New York, pp. 229–239Google Scholar
- Christaller W (1966) Central places in southern Germany. Prentice-Hall, New YorkGoogle Scholar
- Edgar HM, Giarratani F (1999) An introduction to regional economics. West Virginia University, West VirginiaGoogle Scholar
- Fisher SG (1992) Pattern, process and scale in freshwater systems: some unifying thoughts. In: Aquatic ecology. Oxford University Press, London, pp. 575–591Google Scholar
- Florida RL (2005) The flight of the creative class. Harper Business, New YorkGoogle Scholar
- Forman RTT (1981) Interaction among landscape elements: a core of landscape ecology. In: Proceedings of the international congress of landscape ecology. Pudoc, Wageningen, pp. 35–48Google Scholar
- Gottdiener M, Hutchison R (2011) The new urban sociology, 4th edn. Westview Press, PhiladelphiaGoogle Scholar
- Gould SJ (1977) Ever since Darwin: reflections in natural history. Norton, New YorkGoogle Scholar
- Harrison J, Hoyler M (2014) Megaregions. Globalization’s new urban form? Edward Elgar, LondonGoogle Scholar
- Heynen N, Kaika M, Swyngedouw E (eds) (2006) In the nature of cities: urban political ecology and the politics of urban metabolism. Routledge, New YorkGoogle Scholar
- Hoover EM (1948) The location of economic activity. McGraw Hill, New YorkGoogle Scholar
- Hutchings M, John EA, Stewart AJA (eds) (2000) The ecological consequences of environmental heterogeneity. Cambridge University Press, New YorkGoogle Scholar
- Isard W (1956) Location and space-economy. John Wiley & Sons, HobokenGoogle Scholar
- Jacobs J (1961) The death and life of great American cities. Random House, New YorkGoogle Scholar
- Jones CG, Lawton JH (eds) (1995) Linkng species and ecosystems. Chapman and Hall, New YorkGoogle Scholar
- Krugman PR (1991) Geography and trade. MIT press, CambridgeGoogle Scholar
- LaDeau SL, Leisnham PT, Biehler D, Bodner D (2013) Higher mosquito production in low-income neighborhoods of Baltimore and Washington, DC: understanding ecological drivers and mosquito-borne disease risk in temperate cities. Int J Environ Res Public Health 10:1505–1526. doi: 10.3390/ijerph 10041505 PubMedPubMedCentralCrossRefGoogle Scholar
- Lefebvre H (1991) The production of space. Blackwell, OxfordGoogle Scholar
- Leibold MA (2011) The metacommunity concept and its theoretical underpinnings. In: Scheiner SM, Willig MR (eds) The theory of ecology. University of Chicago Press, Chicago, pp. 163–183Google Scholar
- Levins R (1966) The strategy of model building in population biology. Am Sci 54:421–431Google Scholar
- Levins R (1968) Evolution in changing environments. Princeton University Press, PrincetonGoogle Scholar
- Likens GE (1992) The ecosystem approach: its use and abuse. Ecology Institute, Oldendorf/LuheGoogle Scholar
- Logan JR, Molotch HL (1987) Urban fortunes: the political economy of place. University of California PressGoogle Scholar
- Lord C, Norquist K (2010) Cities as emergent systems: race as a rule in organized complexity. Environ Law 40:551–596Google Scholar
- Lösch A (1954) The economics of location: translated from the second rev. Yale University Press, New Haven, German ed. by William H. Woglom with the Assistance of Wolfgang F. StolperGoogle Scholar
- Marcotullio PJ, Solecki W (2013) What is a city? an essential definition for sustainability. In: Boone CG, Fragkias M (eds) Urbanization and sustainability: linking urban ecology, environmental justice, and environmental change. Springer, New York, pp 11–25Google Scholar
- McDonnell MJ, Pickett STA (eds) (1993) Humans as components of ecosystems. Springer New York, New York, NYGoogle Scholar
- McGrath B, Shane G (2012) Introduction: metropolis, megalopolis, and metacity. In: Crysler CG, Cairns S, Heynen H (eds) The SAGE handbook of architectural theory. SAGE, Washington, DCGoogle Scholar
- Merton RK (1968) Social theory and social structure. The Free PRess, New YorkGoogle Scholar
- Mills ES (1967) An aggregative model of resource allocation in a metropolitan area.Google Scholar
- Muth RF (1969) Cities and housing: the spatial pattern of urban residential land use. University of Chicago Press, ChicagoGoogle Scholar
- Myrdal G (1957) Rich lands and poor: the road to world prosperity. Harper & Row, New YorkGoogle Scholar
- Oliva S (2006) The effects of watershed development on housing prices: the case of eastern Baltimore. University of Maryland, MAGoogle Scholar
- Pickett STA, Cadenasso ML, Jones CG (2000) Generation of heterogeneity by organisms: creation, maintenance, and transformation. In: Hutchings ML, John EA, Stewart AJA (eds) Ecological consequences of habitat heterogeneity. Blackwell, Malden, MAGoogle Scholar
- Pickett STA, Kolasa J, Jones CG (2007) Ecological understanding. Academic Press, San DiegoGoogle Scholar
- Reader J (2004) Cities. Atlantic Monthly Press, New YorkGoogle Scholar
- Robbins P (2004) Political ecology: a critical introduction. Blackwell, Malden, MAGoogle Scholar
- Roy Chowdhury R, Larson K, Grove JM, et al (2011) A multi-scalar approach to theorizing socio-ecological dynamics of urban residential landscapes. Cities Environ 4:Article 6.Google Scholar
- Scheiner SM, Willig MR (eds) (2011) The theory of ecology. University of Chicago Press, ChicagoGoogle Scholar
- Schmid C (2014) Networks, borders, differences: towards a theory of the urban. In: Brenner N (ed) Implosions/explosions: towards a study of planetary urbanization. jovis Verlag, Berlin, pp. 67–80Google Scholar
- Shane DG (2005) Recombinant urbanism: conceptual modeling in architecture, urban design and city theory. John Wiley & Sons, Hoboken NJGoogle Scholar
- Spirn AW (1984) The granite garden: urban nature and human design. Basic Books, New YorkGoogle Scholar
- Swan CM, Johnson AE, Nowak D (2016) Differential organization of phylogenetic, taxonomic and functional diversity in an urban woody plant metacommunity.Google Scholar
- Thierstein A, Förster A (eds) (2008) The image and the region: making mega-city regions visible! Lars Müller. Baden, SwitzerlandGoogle Scholar
- von Thünen JH (1966) Isolated state: an English edition of der isolierte Staat. Pergamon PressGoogle Scholar
- Weber A (1929) Theory of the location of industries [translated by CJ Friedrich from Weber’s 1909 book]. The University of Chicago Press, ChicagoGoogle Scholar
- Wiens J (2000) Ecological heterogeneity: an ontogeny of concepts and approaches. In: Hutchins MJ, Stewart AJA (eds) The ecological consequences of environmental heterogeneity. Blackwell, Malden, MAGoogle Scholar
- Wirth L (1945) Human ecology. Am J Sociol 50:483–488Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.