Avoiding Decline: Fostering Resilience and Sustainability in Midsize Cities

Abstract

Eighty-five percent of United States citizens live in urban areas. However, research surrounding the resilience and sustainability of complex urban systems focuses largely on coastal megacities (>1 million people). Midsize cities differ from their larger counterparts due to tight urban-rural feedbacks with their immediate natural environments that result from heavy reliance and close management of local ecosystem services. They also may be less path-dependent than larger cities due to shorter average connection length among system components, contributing to higher responsiveness among social, infrastructural, and ecological feedbacks. These distinct midsize city features call for a framework that organizes information and concepts concerning the sustainability of midsize cities specifically. We argue that an integrative approach is necessary to capture properties emergent from the complex interactions of the social, infrastructural, and ecological subsystems that comprise a city system. We suggest approaches to estimate the relative resilience of midsize cities, and include an example assessment to illustrate one such estimation approach. Resilience assessments of a midsize city can be used to examine why some cities end up on sustainable paths while others diverge to unsustainable paths, and which feedbacks may be partially responsible. They also provide insight into how city planners and decision makers can use information about the resilience of midsize cities undergoing growth or shrinkage relative to their larger and smaller counterparts, to transform them into long-term, sustainable social-ecological systems.

1. Introduction

Urbanization is a hallmark of the Anthropocene. In the United States (U.S.), more than 85% of the population lives in urban areas. One in six of these urban inhabitants can be found in midsize cities, which are defined as incorporated areas in the conterminous U.S. with a population between 50,000 and 300,000 [1]. Yet, much research, outreach, and policy intervention are designed based on very large coastal cities, even though only nine U.S. cities have populations exceeding one million [1]. Here, we identify factors affecting the urban resilience of midsize cities, and describe the components of sustainability and resilience as they apply to midsize cities in both desirable and undesirable states. These descriptions inform the maintenance of desirable states into the future [2], in order to identify the system components that underpin undesirable or desirable states.

Midsize cities are unique relative to large cities because they are intimately coupled to their immediate suburban and rural surroundings through ecosystem service production and natural resources management, resulting in tight urban-to-rural feedbacks (Figure 1). Larger cities rely more on a “pipeline” delivery of food, water, and manufactured goods from places and watersheds far removed from the point of service [3], likely resulting in slower feedbacks between consumption and impacts of resource depletion at the source of consumption. We argue that relatively shorter feedbacks between the local ecosystem and the urban system are a hallmark of midsize cities compared to their larger counterparts, and provide a basis for our framework. Indeed, these feedbacks should be incorporated into any resilience-sustainability assessment of a midsize city.

2. Defining Midsize City Resilience and Sustainability

Resilience is an emergent system property that describes how much disturbance a system can withstand without losing essential structure and functioning [4]. A resilient system experiences disturbances without reconfiguring into an alternative state with new structures and functions. This definition contrasts with engineering resilience, which measures the return or “bounce back” time to a desirable state following disturbance. Here, we are primarily interested in resilience as a measure of a midsize city system’s capacity to maintain a sustainable (i.e., desirable) developmental trajectory through the preservation of desirable structures and functions across space and time.

The concepts of resilience and sustainability are related but not synonymous [5]. Resilience is often used in a normative context, because an unexpected reconfiguration into an alternative state is frequently associated with a persistent loss in desirable ecosystem services [6]. However, management intervention that seeks to intentionally transform a system from an undesirable to a desirable state may face considerable challenges if the current state is resilient to the intervention—in this context, resilience is an undesirable system property because it limits desired transformations [7]. Thus, assessing and quantifying relative, place-based resilience in social-ecological contexts requires careful definition of the resilience of what, to what, and for whom [8,9]. Here, we examine the resilience of mid-sized cities to unforeseen future disturbances for the well-being of its current and future inhabitants. We follow the broad definition of sustainability as the long-term reconciliation of society’s development goals with the planet’s environmental limits [10]. Resilience assessments capture concepts of disaster vulnerability and adaptive capacity while accounting for the complex interactions between humans and their environment [11]. Non-sustainable systems are ultimately vulnerable to both natural and human-induced disasters and hazards. A sustainable city is in a desirable state and is resilient to future intrinsic and extrinsic surprise; i.e., it can withstand disturbances coming from within and outside of the system while resisting reorganization into a new state.

Resilience assessments capture the integrative measure of sustainability in the context of a desirable regime characterized by ecological, infrastructural (within the built environment), and socioeconomic factors. The term “regime” explicitly refers to the key structuring processes of a given system state; when a system diverges from a regime, the system state changes. An example of alternate states (i.e., regimes) in complex urban systems is a city with an economy characterized by high upward economic mobility versus a city where poor residents are caught in a poverty trap. A threshold, or tipping point, separates these alternative states (or regimes; Figure 2) and each regime has a set of reinforcing processes and structures also referred to as “basins of attraction” that maintain the system in its current state.

Some authors (e.g., [12]) have conceptualized the change from a non-sustainable to a sustainable city as a state change (regime shift), suggesting that understanding such transitions requires knowing: “(1) the triggers that have induced change; (2) situations where crisis triggers change; (3) why cities transition toward more sustainable states on their own, in the absence of crisis; (4) what we can learn from new city transitions; and (5) how resource interactions affect urban transitions [12] (p. 320)”. Given this, solutions to urban sustainability challenges may be categorized as follows: for a city in a desirable state, resilience should be fostered to ensure a long-term sustainable trajectory, and for a city in an undesirable state, change should be transformative, meaning the resilience of the system should be deliberately exceeded to foster an alternative, desirable state.

A system in a desirable state can be resilient for a short time interval but not at larger scales; that is, despite stability in terms of basic structure and function, the system is not sustainable over longer time intervals for various reasons. Examples include cities based on short-term resource extraction, such as the numerous boom-and-bust mining towns in the western U.S., and cities in the Bakken Shale region in the northern Great Plains. For cities, the scale domains of critical structuring dynamics can be categorized in many different ways. We suggest that the desirability of a resilient state can be assessed by measuring a set of key processes and structures within and across relevant domains of scale, and the tradeoffs among different components can be explicitly identified and communicated.

Midsize cities may be more responsive to sustainable transformations than their large city counterparts due to their different scales of processes, components, and feedbacks. As a result, absolute component interaction and system response to a change (e.g., policy interventions) may be more rapid or efficient than in a larger city where there are fewer and/or more deterministic connections among components. Indeed, large cities and megacities exhibit strong path-dependency that makes altering their developmental trajectory very difficult. Inertia in social systems and infrastructure is a barrier to sustainability in urban systems [12]. For these reasons, we argue that midsize cities are best suited for the study and pursuit of transitions to sustainable states, and provide a large sample size for studies in the United States.

In any urban system, the interactions of its components affects whether the city is on a sustainable trajectory. Conceptually, we categorize the resilience of urban systems as composed of three overlapping subsystems: (1) socio-economic, (2) ecological, and (3) infrastructure. Within these subsystems, we propose measurable axes that theoretically or empirically contribute to their resilience. For the ecological system, these axes are: diversity, variability, modularity, feedbacks, capital, ecosystem services, ecological footprint, cross-scale reinforcement, green space, biogeochemical cycling, and water availability [13]. For infrastructure, they are: excess capacity, capital, diversity, transportation, investment in innovation, funding, environmental impact, bottlenecks, amenities, knowledge, reuse/recycling, and age. For the social-economic and governance systems they are: trust, wealth, education, public health, economic innovation, amenities/services, diversity, public finance, culture, system age, material flows, and population density. We consider these proposed axes as hypotheses that should be explored and tested. The importance of particular variables´ feedbacks is context dependent, and this list is not exhaustive.

3. Measuring and Assessing Resilience

The tradeoffs among individual indicators of resilience in the ecological, infrastructural, and social-economic components of mid-sized cities can be visualized as “axes of resilience” in a spider web diagram for rapid comparison. This approach allows for visualization of multiple, integrated variables (each axis represents multiple variables) but does not provide a single overall measure of resilience. That type of metric can be derived using tools of Multi-Criteria Decision Analysis (see [14,15]) or Network Science techniques [16].

We suggest deconstructing complex urban systems into component subsystems and/or individual resilience metrics (

Table 1. The nine properties of a resilient system according to Walker and Salt [13], along with definitions, and possible assessment metrics for a midsize city social-ecological system.

Property	Definition	Assessment Metrics
Biological Diversity	The number and evenness of species, functional groups, and response to disturbance in the ecosystem	Diversity of native land covers, and rate land use change within city limits and around the city
Ecological Variability	Natural variability and fluctuations in ecological processes, structures, and populations	Presence of wetlands, which may indicate flood repression
Modularity	System components are connected to one another so that information is transferred effectively, but not so overly connected that shocks cause disproportionate damage	The degree to which the damage/loss of a single institution affects the city
Acknowledging Slow Variables	Incorporation of information about long-term outcomes that result from near-term decision making	Use of long-term time horizons in city planning; acknowledgement of system-specific slow variables in these plans
Tight Feedbacks	Feedbacks among critical system components respond quickly, allowing practitioners to avoid dangerous thresholds	Infrastructure and planning that ensures essential services, like water quality, is continuously maintained
Social Capital	The net sum of benefits generated from relationships among components in a system’s social network	Unemployment rate, poverty levels, voter participation, public servant approval ratings
Innovation	Degree of learning, experimentation, education, and locally developed rules in order to embrace change and creatively improve conditions	What proportion of the economy is due to new technology; ethnic, racial and gender diversity in business ownership and governance
Overlap in Governance	Institutions have redundancy in their roles and responsibilities	Multiple institutions are authorized to provide essential services in the case of natural and human disasters
Ecosystem Services	Essential and nonessential benefits people obtain from ecosystems	Accessibility to local agricultural lands, hunting and fishing, green spaces, wetlands

[13]) because assessing resilience in the overall system requires quantifying resilience along multiple axes, and assessing the tradeoffs among those axes. To integrate these subsystems, a “system of systems” approach should be utilized. Quantification first occurs for each subsystem, where a collective evaluation of all subsystems within a city provides the opportunity for an explicit assessment of the resilience of the system as a whole while incorporating tradeoffs that occur amongst the subsystems (Figure 3). As an example, natural disasters such as fire may have devastating ecological and infrastructural impacts, but resilience to natural disasters may be enhanced by wealth, through more options in response and a faster recovery.

In estimating the resilience of midsize cities and other urban systems, we propose a suite of measureable factors across the three component subsystems (ecological, infrastructural, and social-economic), following an approach similar to that of the planetary boundaries. The key properties, or structuring variables, which describe the performance of the ecological, infrastructural, and social-economic subsystems are not commensurate in that they cannot be weighed against one another equally, but are useful for relative comparisons. The nine properties are also not an exhaustive list, but a starting point (Table 1). In estimating such variables within an urban system, one can draw conclusions regarding the relative resilience of midsize cities of similar geographic, demographic, and cultural extents and across the individual subsystems.

3.1. Ecological Subsystem

Human communities rely upon the goods and services provided by ecosystems [17]. Inhabitants of urban systems in the U.S. rely upon a suite of ecosystem services such as waste removal, water filtration, food provisioning, and recreation. Maintaining these goods and services for long periods of time, over the course of successive urban planning cycles, is critical. Resilient ecosystems can withstand some degree of disturbance and may continue to produce the desired services. If, however, the level of disturbance exceeds its resilience, the system will reorganize into an alternative state [18]. Often this reorganization results in significantly fewer and/or less desirable goods and services, which may be unanticipated by humans who depend upon those services to maintain quality of life standards [19]. The reorganization is also often marked by a loss of biodiversity and overall ecological integrity. This alternative state is often considered undesirable.

There are several ecological questions that relate to the sustainability and resilience of urban systems. Basic questions related to sustainability and resilience focus on the relationship between ecological systems and anthropogenic built environments, the generation of ecosystem services in urban settings, the benefits of green infrastructure and vulnerability to natural disasters. Because midsize cities are so tightly coupled to their surrounding landscape (Figure 1), changes to the urban environment can have direct, immediate consequences for the surrounding ecosystems. If an urban economy experiences a drastic change, the effect on the local ecosystem could disrupt activities that maintain essential ecosystem services. An example of the disruptive influence of economic changes in ecosystem services is the case of intense mining or agriculture extraction. While such disruptive, extractive management is likely unsustainable over long time horizons; it may be requisite for supporting human populations over a shorter period at certain quality of life standards.

Midsize cities that are truly sustainable balance the tradeoff between disruptive extractive activities and ecological resilience by fostering healthy ecosystems in desirable states with long-term sustainable economic practices. Ecological resilience offers a buffer, lest the tradeoffs between ecological integrity and economic growth occur in direct opposition. When this buffer is lost, the city becomes vulnerable to decline.

3.2. Infrastructure Subsystem

Infrastructure is a critical foundation shaping the manifestation of cities within their ecological and social contexts. Infrastructure operates at the nexus of engineering systems, the natural environment, and socioeconomic systems. Because resilience and sustainability are emergent properties of the interactions of all of these sub-systems, sustaining the combined social-engineering capacity requires understanding and managing feedbacks and interrelations among the subsystems (Figure 4) across temporal and spatial scales. Critical infrastructures underlying water supply, transportation, energy production, etc. are interdependent, especially in urban areas. This interdependence affects the overall performance of the system, as well as the sub-systems it comprises. For example, recent biofuel and shale gas/oil development (energy production) in the U.S. will create unique challenges for critical lifeline infrastructural systems including food production, energy supply, transportation, water supply and others [20], as well as for the environment and the communities. Infrastructural components related to emerging energy development are interdependent, primarily by virtue of physical proximity, operational interactions, and competition for natural and human resources at both local and regional scales [21]. Interdependencies may stabilize the overall system operation; however, damage to one infrastructure can cascade and damage connected components with system-wide consequences, impacting large geographic regions and sending ripples through the regional and national economy [22].

Infrastructure expansion and maintenance is critical for the sustainable development of midsize cities. Infrastructure development in midsize cities is closely tied to rural development in some cases (e.g., biofuel development in the Midwest can drive the growth of many small and midsize cities in the region); and existing infrastructures and future expansion can have synergistic or competitive interactions with nearby large cities (e.g., water supply expansion can compete with that existing in a nearby large city; highway and railway expansion in those cities can enlarge and/or complicate the region transportation network).

The principles of green infrastructure [23] are especially suitable for infrastructure development in midsize cities given the dependence of the urban area to the surrounding environment. However, past infrastructural system design focused on efficiency and optimization of limited key services to some human societies and ignored the needs of natural systems, and/or affected the needs of other societies (e.g., water supply), which left these systems vulnerable to change. The implementation of green infrastructures needs proper policy guidance based on both economic and environmental performance, and societal acceptance of the infrastructure development that would accompany city development purposes. The choice among various technology options that differ in their environmental costs and infrastructure requirements will be influenced by the policies adopted. Infrastructure development that prioritizes the maximization of one outcome (i.e., economic growth) at the expense of others can produce an array of effects at the community level [24], including beneficial effects (such as increased employment opportunities, landowner royalties, and new sources of tax revenue) and deleterious effects (such as overburdened municipal services, road network congestion and pavement deterioration, reduced water availability for conventional uses, economic volatility, a disruption of social and cultural patterns, and the stigma associated with environmental health risk and industrialization). It is reasonable to assume that at the scale of midsize cities, the impact of infrastructure development is more significant and more diversified [24,25].

3.3. Social-Economic Subsystem

There are several questions related to social-economic subsystems that focus on the relationship between environmental factors, measures of resilience, and the cohesion, wealth and transformability of society. Here we focus on the relation between environmental factors and social network cohesion, the relation between results of measures of resilience and property value, and the means to identify undesirable states and foster transformation.

Environmental governance serves as a bridge between nature and society, and reflects the values society places on nature. Governance structures and processes in midsize cities are the nexus for landscape-provided ecosystem services, the infrastructure that delivers the services, and the urban populace that depends on them. Thus, we propose governance as a nexus within the social-economic subsystem. In the context of midsize cities, governance refers to the means through which political actors choose the goals of urban planning, management, development and protection; the means through which they take action to achieve those goals; and the means through which planning is integrated with regional management to affect ecosystem services. Thus, governance includes the formal laws, regulations, policies, and processes of government, as well as the informal institutional frameworks in which social and political actors influence policy decisions [26,27,28,29,30,31].

Adaptive governance (and transformative governance [32]) is a key strategy for maintaining the sustainability and resilience of cities. Adaptive governance is a term used to describe system-level approaches to collective choice and action within a variety of environmental governance settings that is sufficiently flexible to allow adjustment, adaptation, and response to change in a manner that fosters the resilience of desirable regimes for social-ecological systems [33,34,35,36,37,38]. Adaptive governance warrants our attention in the context of problems that arise at the interface of nature and society, including solutions to the challenges of water scarcity and quality, the increasing loss of agricultural land to urban development in landscapes surrounding midsize urban areas.

Resilience of the social-economic system is non-normative. It describes a system property rather than how society values that property. But when we apply terms such as resilience and sustainability to social systems, we necessarily overlay normative concepts. The literature on adaptive governance thus recognizes the need for attention to legitimacy, transparency accountability and other attributes of good governance in any effort to sustain its implementation [34,38,39], and we propose that no system that ignores these issues is likely to be sustained in a developed, Western, democratic nation. Thus, our concept of the attributes of governance that will enhance the resilience and sustainability of a desired state must include legitimacy, equity, and justice as well as flexibility and the capacity to adapt.

3.4. Applied Example: A Resilience Assessment of Two Midsize Cities

Several frameworks have been proposed for conducting resilience assessments [16,40], yet the applicability of these techniques for midsize cities is underdeveloped [41,42]. Here we build on existing approaches [16,40,41,42], using established social-ecological properties [13] to estimate the relative resilience of two midsize cities in the Midwestern region of the United States. These nine metrics (Table 1.) together describe the performance of the ecological, infrastructural, and social-economic subsystems. The individual properties are not commensurate with one another, but can be examined individually for weaknesses in the whole system, and across multiple systems.

The nine properties are also not an exhaustive list for each specific city, but serve as a general starting point to cover the three subsystems. They are: high (1) biological (species, response, and trait) diversity; (2) ecological variability and (3) modularity; decision makers that (4) acknowledge slow variables, but work to (5) tighten feedbacks among components that may trigger system reconfigurations; high (6) social capital; (7) innovation, and (8) overlap in critical roles, especially regarding emergency response; and finally, a reliable flow of (9) ecosystem services to society. In order to assess the relative resilience of two mid-sized cities in the Midwestern U.S., we assessed these nine properties of a resilient system in Lincoln, Nebraska and Springfield, Illinois (for detailed methods and results, see the Appendix A).

Lincoln, Nebraska (NE) and Springfield, Illinois (IL) are midsize, capital cities with similar ecological, social-economic, and demographic features located in the Midwestern region of the U.S. (US Census 2014). Lincoln, NE is experiencing economic and population growth relative to Springfield, IL (

Table 2. Demographic and economic data for Lincoln, Nebraska and Springfield, Illinois, USA, according to 2014 U.S. Census data.

Variable	Lincoln, NE, USA	Springfield, IL, USA	National
Population	272,996	116,809	318,900,000
Population Annual % Change	5.60%	0.40%	3.30%
Unemployment Rate	2.10%	4.60%	6.2%
Gross Domestic Product (millions of USD)	15,482	8714	17.914 × 1012
Gross Domestic Product Annual % Change	2.60%	−0.80%	2.40%

). This provides an opportunity to probe whether the nine social-ecological resilience indicators provide insight into this divergence, and whether the predicted ability to respond to a disturbance or stressor is linked to GDP and population growth, both of which carry high social and political importance.

Response diversity describes the multitude responses to a disturbance from a community of species that contribute to the same ecological function [43]. An example is the continuity of primary production in the face of an extreme drought event followed by a flood event in a relatively short period. Land cover/land use diversity is an indicator of the larger landscape’s response diversity. If one type of land cover/land use is particularly vulnerable to a social-ecological disturbance in that system (e.g., drought, economic fluctuations), monocultures of that cover type in and around a midsize city could reduce its resilience. As a result, we assume that high diversity of land covers confers a measure of resilience to the system. Our analysis of land cover diversity in Lincoln, NE, and Springfield, IL indicate slightly higher diversity in and around Lincoln, NE relative to Springfield, IL (Figure 5, panels a and b, respectively; for detailed methods see Appendix A). Our approach included buffers surrounding the cities, based on the assumption that midsize cities are tightly coupled with their immediately adjacent ecosystems. The highest land cover diversity, normalized for area, was found in the buffers surrounding Lincoln’s city center, and indicates a greater capacity for the system to absorb shocks while continuing to provide a range of ecosystem services.

Ecological variability in this assessment captures the degree to which social drivers mitigate ecological variability, which is a key structuring driver in many natural systems. Variability could include seasonal fluctuations in stream flow that maintain riverine wetlands, or small wildfires that mitigate woody encroachment into grasslands and thus prevent more dangerous forest canopy wildfires. However, ecological variability that contributes to system resilience is often sacrificed for more immediate ecosystem service production. Therefore, a resilient social-ecological system is one that balances long-term ecological variability with ecosystem services production. In this resilience assessment, we include metrics for both ecological variability and ecosystem services production.

To measure ecological variability for Lincoln, NE and Springfield, IL, we choose to measure the cover and total number of individual riverine wetlands along the two major streams running through the cities. Riverine wetlands are likely good indicators of ecological variability because periodic, minor flooding events recharge and maintain wetlands adjacent to streams [44]. As a result, flood repression and stream channelization common in midsize cities result in decreased wetland cover. Wetlands directly contribute to ecological resilience in the system, absorbing energy, water, and materials in larger flood events. They are also zones of high reactive nutrient removal, reducing the nutrient concentrations in the larger watershed and mitigating hypoxic zones downstream. While the two rivers were comparable in length through the city and total number of riverine wetlands, Springfield had nearly double the total riverine wetland area than Lincoln, NE, indicating greater potential to absorb flooding events (Figure 5, panels c and d).

Modularity of the social-economic system is a measure of how components of the system are arranged in relation to one another. A modular system is arranged tightly enough so that an input (e.g., a shock, information inputs, wealth) is communicated throughout the system evenly, but not so tightly or overly connected in one area of the system so that inputs reverberate, resulting in disproportionate effects on one area of the system. There are multiple ways to approach a modularity assessment of a midsize city. Given that in both cities the top employer is the state, we decided to determine how impactful losing this employer might be in terms of the relative amount of jobs lost. We found that the role of Lincoln’s top ten employers presents a sharp contrast to Springfield’s (

Table 3. The top ten employers in Lincoln, NE, USA, and Springfield, IL, USA (“Top Ten Employer”), the number of workers they employ (“# Employees”), and the percent of the workers employed by a top ten employer that the company employs (“% of Workers Employed”. Data presented is only based on workers employed by a top ten employer.

Lincoln, NE, USA			Springfield, IL, USA
Top Ten Employer	# Employees	% of Workers Employed	Top Ten Employer	# Employees	% of Workers Employed
State of Nebraska	8988	22.54	State of Illinois	17,500	48.77
Lincoln Public Schools	7975	20.00	Memorial Hospital	4433	12.35
University of Nebraska-Lincoln	6179	15.50	St. John’s Hospital	3073	8.56
Bryan Health	3796	9.52	Springfield Public Schools	2189	6.10
US Government	3206	8.04	Springfield Clinic	1953	5.44
City of Lincoln	2589	6.49	City of Springfield	1510	4.21
St. Elizabeth Regional Medical Center	2350	5.89	Southern Illinois University School of Medicine	1485	4.14
Burlington Northern Railroad	2000	5.02	University of Illinois at Springfield	1435	4.00
Madonna Rehabilitation Hospital	1400	3.51	BlueCross BlueShield	1256	3.50
B&R Stores	1391	3.49	Horace Mann	1050	2.93
Total # Employed by a Top 10 Employer	39,874	-	Total # Employed by a Top 10 Employer	35,884	--

). For instance, in Lincoln, NE the top employer (the state) employs only five percent of the active workforce, compared to Springfield, IL, where the top employer (the state) employs almost a quarter of the active workforce. If the top employer in each city were to lay off the same proportion (or all) of its employees, the impacts would be felt more strongly in Springfield, IL, versus Lincoln, NE, indicating a lower social-ecological resilience to this particular type of stressor.

Slow variables in social-ecological systems are the long-term outcomes that result from short-term alterations of key social-ecological feedbacks. Recognizing these slow variables in decision making is essential to protecting the long-term resilience of a midsize city. We compared the land-use and emergency preparedness plans (required by law) of both Springfield and Lincoln to assess their acknowledgment of slow variables and ability to adapt policies to reflect changes in those variables over time. While both cities had comprehensive plans, neither city had plans that addressed system-specific slow variables, such as increasingly extreme mid-latitude weather events due to climate change, or severe soil degradation. However, meaningfully assessing the acknowledgment of slow variables is subjective and difficult to report as a metric, making any meaningful resilience comparison between the cities challenging.

When system variables are connected by very short feedbacks, by contrast, changes to one part of the system triggers observable change in other parts of the system, providing immediate information regarding the effects of decision making. As a result, midsize cities with tight feedbacks have the capacity to gather, disseminate, and act on relevant data regarding city functioning. However, part of the tight feedback loop includes acting on new knowledge as it arises. For example, cities typically have water quality testing devices in place to monitor the effects of seasonal land use change, alterations to water treatment processes, new water sources, etc. When decision makers fail to monitor or incorporate new information about water quality appropriately, the value of short feedbacks is squandered. Such was the case in Flint, Michigan, where decision makers switched to a new municipal water supply without corrective measures in place to immediately detect and reverse elevated lead levels due to reactions between the more acidic water source and aging city pipes. The failure in this tight-feedback loop led to irreversible lead poisoning effects on Flint residents who consumed city water [45]. Thus, our analysis is based on both (1) the presence of tight feedbacks; and (2) the willingness to act on these feedbacks to ensure access to safe drinking water in the two cities.

Our investigation of tight feedbacks in Lincoln, NE and Springfield, IL examined the strength of water quality feedback mechanisms. We found critical infrastructures at the nexus of safe water supply, food production, and energy generation that are difficult to capture with a single point in time. Generally, however, both cities overall have apparently strong water quality mechanisms in place, including high frequency monitoring. Both have this monitoring data available to the public, and encourage participation in water quality issues with open meetings and water quality concern hotlines. It is not immediately evident how a major disturbance, such as suddenly elevated nitrate levels in drinking water, would be addressed or handled.

Social capital in this context is the net sum of benefits generated from relationships among components in a system’s social network. High social capital in the system contributes to knowledge generation and diffusion and facilitates actors in the system to conduct complex problem solving [46]. This may bolster the ability of the social system to respond to disturbances, therefore contributing to the system’s overall social-ecological resilience. In our investigation of social capital, we use a combination of social-economic and governance factors to present a general overview of social capital in the two midsize cities. Differences in gross domestic product, funding for the arts and education, and unemployment rates may indicate slightly higher social capital in Lincoln, NE. However, this metric (similar to “acknowledges slow variables”) is subjective and difficult to reliably assess for comparison.

Innovation is a measure of how well the social system learns, embraces change, and develops local rules to address problems as they arise. This requires a diverse, educated workforce with high social mobility to allow everyone equal access to decision-making roles. For this evenly represented workforce to innovate, it must have access to high-tech jobs. The technology created can be used to test solutions to existing problems or generate new ideas in the midsize city system. Overall, we found that the two cities are nearly identical in the comparative measures of innovation we chose (see Appendix A), with the exception of high-tech employment. Lincoln, NE has a larger high-tech economic sector than Springfield, IL, indicating greater capacity to innovate and thus withstand unforeseen disturbances.

Governance overlap describes the degree to which critical services provided by government in the social-ecological systems are provided by multiple entities/institutions. If a single institution is responsible for a critical service, that service may be vulnerable to loss or damage in the event of a disturbance, reducing the system’s overall resilience. We examined governance overlap of critical services by reviewing both cities’ disaster response plans. Both had high overlap in essential services, ensuring continuity in services in the face of a variety of natural and human emergencies. Due to the especially structured nature of Springfield’s disaster response plan as compared to Lincoln, a clear framework of responsibilities can be assumed by a number of different persons or agencies under multiple contingencies. This could confer slightly higher city resilience against disasters for Springfield, IL relative to Lincoln, NE.

Ecosystem services are the essential and nonessential and benefits obtained by humans from ecosystems. They include tangibles, like agricultural products, and intangibles, like the spiritual value gained from recreational spaces. In modernizing and expanding cities, ecosystem services, like water and food provisioning or air purification, are increasingly generated remotely. Midsize cities, however, remain reliant on their immediate landscape for many ecosystem services. A major disturbance or disaster event that severs the pipeline delivering these ecosystem services into a city could cut off access to food supplies, water or access to remote recreation spaces. Therefore, we assessed land cover information from within and immediately surrounding the midsize cities to determine the flow of ecosystem services to each city. These assessments include information about land cover diversity, land uses that generate valuable agricultural and mining products, wetland cover, access to hunting and fishing, and green space. Due largely to differences in biodiversity and green space, there is some evidence that Lincoln, NE may have higher ecosystem service production from the landscape within and surrounding the city, contributing to overall system resilience.

3.5. Resilience Assessment Synthesis

These nine aspects of resilient systems are interdependent and represent a holistic approach to describing systems and their tradeoffs (Figure 6). They do not capture every aspect of Lincoln, NE’s or Springfield, IL’s whole system resilience to each type of disturbance, but instead provide an overview of these complex social-ecological midsize city systems’ resilience. From this, a comparison of where the systems diverge can emerge to inform practitioners and provide targeted areas for transformation to more sustainable paths.

While attempting to quantify resilience is operationally dependent and system specific, and thus relative rather than absolute, our approach provides one way to visualize and pinpoint areas where resilience may be deficient. This type of approach can serve as a starting point for other practitioners interested in examining the resilience of their system, by tailoring their investigation to their specific question, finding a comparable midsize city on a path that should be emulated/avoided, and then assessing a set of a priori resilience indicators to gain new insights into the inner workings of their midsize cities.

4. Discussion

As there is no absolute measure of resilience, we proposed a general framework for assessing relative resilience of midsize cities. In addition to the assessment example provided above, measures of subsystem resilience can be combined into an integrative measure of resilience using tools such as Multi-Criteria Decision Analysis (see [12]), which allows city stakeholders to integrate scientific data and value tradeoffs inherent in resilience and sustainability. Where longitudinal data is available, one can compare relative resilience within a system over time.

Relevant questions that can be addressed with such approaches include: How do we achieve sustainable societal goals that accommodate ecological complexity and uncertainty (i.e., adaptive governance) [36]? What are the sustainable prospects for these cities, and how do we assess their relative resilience over time? What are the conditions and processes that gave rise in some cases to adaptation and resilience? Where, and when, should resilience be fostered, and under what conditions might more basic transformative change be necessary? How do we measure the preconditions, structures, processes, and outcomes in order to generalize variables and indicators? How do we address impending challenges related to water quality and quantity? What is the role of these cities’ agricultural or public land watersheds? What is the role of infrastructure development in resilient cities?

Governance capacity and response to the science of vulnerable systems is a critical limiting factor in urban systems, shaping the response of midsize cities to challenges arising in both the ecological and infrastructural subsystems. Social-economic decisions affecting infrastructure viability and decisions regarding tradeoffs between development and the maintenance of ecosystem services are made through the process, authority, and norms of governance. Thus, we hypothesize that the adaptive capacity of governance structures and processes is a key factor in predicting the sustainability of midsize city systems.

The integration of concepts of social-ecological resilience and governance uses five components to analyze the legal and institutional aspects of governance: (1) structure, including redundancy, polycentricity, and networks among urban decision makers and stakeholders and the managers of the surrounding landscape and watershed; (2) scale, including the ability to respond across jurisdictional boundaries at the scale of a disturbance; (3) adaptive capacity, including both the capacity to experiment, learn, and adjust, and the capacity of a diverse array of local actors to participate in the decision-making process; (4) legitimacy in the process and substance of decision making; and (5) power, considering both the distribution and balance of power as indicators of the sustainability and desirability of the governance regime [47].

In the context of midsize cities, the key governance factors affecting resilience are likely to include the capacity to participate and adapt, the enhancement of that capacity through networks and access to resources at scales above and below regional governance, and the legitimacy of urban planning and decision making as identified through the participatory processes. In addition, understanding the feedbacks between the ecosystem and urban decision making, is key in assessing the adaptive capacity of local and regional governance.

4.1. Scale

The components and processes that compose a midsize city represent a vast array of scales, both in size and time horizon (e.g., building life spans, gentrification and poverty cycles in neighborhoods, etc.). While the city itself may experience fluctuations in its desirability, so too do its component parts. Panarchy theory is a lens through which we can conceptualize these fluctuations occurring at different scales in a complex system [48,49]. For example, a single neighborhood in an urban system could experience fluctuations between high and low crime and unemployment rates without affecting the overall rates of the whole city. Alternatively, some combination of internal system dynamics and/or an external stressor could result in synchronous high crime and unemployment rates among a threshold plurality of neighborhoods. Then, the undesirable state of these small components scale up, putting larger scales, or the entire system, at risk of reorganizing into an undesirable state, even if other system components at other scales are desirable and resilient.

This contrasts with the more conventional view that posits top-down or external disturbances as the greatest threats to maintaining a desirable system state. In complex systems, both may occur, and the result of these cross-scale dynamics cannot be ignored in resilience assessments. Thus, the integrative approach to measuring system city resilience that we detail above is somewhat incomplete without a consideration of the state of each of the subsystems, their individual scales, and the potential for these components to scale up and affect collapse.

4.2. Identifying Undesirable States and Fostering Transformation

Midsize cities exist along a continuum of sustainability, crossing key thresholds into persistent, unsustainable states at some point along the spectrum. Management efforts should focus on enhancing the desirability of the current state, and ensuring that the desirable state is resilient and on the sustainable side of thresholds. However, some cities are in undesirable states. Here, we encounter a basic problem: for cities in undesirable states, how do we lessen the resilience of the current undesirable state, foster a transformation to a desirable state, and enhance the resilience of the desirable state when achieved? We wish to understand and identify indicators of a negative shift in resilience for midsize cities. Are there leading indicators that presage the loss of resilience in a city? If so, what are those leading indicators? Indicators currently identified in the literature include increasing variance in key parameters or processes, critical slowing, Fisher Information [50], spatial variance [51], and other variance indices. For cities that have shifted into an undesirable state, longitudinal data can be used to assess the potential to identify leading indicators of those regime changes post hoc. This is a critical research area in need of further development so that resilience assessments can be used as predictive—and not just descriptive—tools for sustainable urban planning and development.

5. Conclusions

In midsize cities, the major ecological, infrastructure, and social-economic subsystems have distinctive characteristics. They are also interdependent, influencing one another at multiple scales in ways that are often not apparent until closer examination. An integrative approach is therefore necessary for capturing properties that emerge from the complex interactions of these subsystems and their component parts. Here, we suggested such an approach, using Walker and Salt’s nine “properties of a resilient world” [13] to operationalize comparisons of midsize cities with similar cultural, demographic and geographic settings. From this, a visualization of comparative system resilience indicators emerges that allows us to ask how one midsize city can be on sustainable trajectory while a seemingly analogous city is on an unsustainable path. It also provides us with valuable insight into how we can use midsize cities’ flexibility and stability relative to their larger and smaller counterparts, respectively, to transform more of our urban areas into long-term, sustainable systems. From this, we can begin to avoid critical thresholds in order to preserve these valuable social-ecological systems so many of us call home.