1. Introduction
Climate change increases the stress on urban areas through increasing the number of extreme events and hazards such as heat waves, inland floods, and storm surges which are affecting inhabitant’s lives and property, essential infrastructure and ecosystems [
1]. Among these hazards, heat waves are the most deadliest, as in many countries they cause more fatalities than floods and hurricanes combined [
2,
3].
Globally there is strong evidence that heat waves will increase in duration and frequency over most land areas [
4]. Impacts of heat waves on human and natural systems include direct effects caused by the direct exposure to higher temperatures and indirect effects, such as those on urban sectors such as water, energy, transportation and telecommunication [
5]. Such indirect impacts may include a reduction of drinking water supply, increasing energy demand as well as heat-related mortality (death) and morbidity (illness) [
6]. Furthermore, psychological health problems can occur and cause increasing violence and crime during heat events [
7].
Heat wave impacts are stratified across the population, with certain socio-demographic groups being stronger affected than others. These vulnerable groups include infants, elderly, people with disabilities and health problems [
2,
6,
8,
9,
10] and poor people and low-income groups. Low-income groups are among the most vulnerable to heat events due to their limitation to meet energy costs, their dependency on public facilities and problems in access to proper health care services [
6]. Moreover, vulnerable and low-income populations may be concentrated in areas with increased exposure to risk. For instance, compared to higher-income populations, low-income communities tend to live in lower-standard or older buildings without the capacity to regulate temperature and humidity [
8,
11].
Whereas the relation between heat vulnerability and socio-demographic factors, such as age, race, gender, and poverty, is well examined [
12,
13,
14,
15,
16], the role of economic factors such as different income levels (not only considering the poor and low-income, but also middle- and high-income people) is under-researched. Being a prominent vulnerability factor, income may play an important role in how citizens experience heat waves. It is the aim of this research to close this research gap, and compare the impacts and adaptation perceptions of different income groups in New York City. This information is then used to deduce socially sensible adaptation options, evaluating the effect of different adaptation scenarios on income groups.
The vulnerability concept is useful to understand and evaluate impacts and adaptation of climate change on human and environment system [
6]. According to the IPCC Fifth Assessment Report (AR5), vulnerability is generally defined as “
The propensity or predisposition to be adversely affected. Vulnerability encompasses a variety of concepts and elements including sensitivity or susceptibility to harm and lack of capacity to cope and adapt” [
17] (p. 5). Vulnerability is also generally seen from two perspectives [
17], described as:
- -
Contextual vulnerability (starting-point vulnerability)
- -
Outcome vulnerability (end-point vulnerability)
Contextual vulnerability assumes that certain socio-economic groups in society are more vulnerable than others even before an event happens. According to that understanding, groups such as the elderly, children, women, the poor and people with health issues belong to the most vulnerable, due to their existing circumstances making them more vulnerable than others. Outcome vulnerability assumes people only becoming vulnerable after a hazard occurred [
18]. This can affect different groups in society and is not necessarily confined to the groups mentioned. Focusing on differences across income groups this research is based on the contextual vulnerability (starting-point vulnerability) concept, which is defined as: “
A present inability to cope with external pressures or changes, such as changing climate conditions. The contextual vulnerability is a characteristic of social and ecological systems generated by multiple factors and processes” [
19] (p. 1762).
For example, Rosenthal et al. [
15] evaluated the socio-economic and build environment characteristics of places with high heat-related mortality in New York City. According to their results, there is a significant positive association between heat-related mortality and neighborhood characteristics, comprising less access to air conditioning (A/C), poor housing conditions, and poverty status. Energy costs associated with the use of air conditioning are also one of the major concern for low-income families during heat events. Low-income households who have access to A/C, do not use it due to the concerns about energy costs [
20].
Air conditioning is one of the major and most frequent adaptation strategies to the impacts of heat waves in North American cities. However, to improve the effectiveness of air conditioning as an adaptive measure it is important to ensure access to functional air conditioners and sufficient energy for vulnerable groups [
21]. Lemmen and Warren [
21] suggest monetary support of low-income populations and programs for peak load and or voltage reduction. However, important to note, as long as traditional, non-renewable sources of energy are used to run A/Cs it is not a sufficient nor a sustainable solution, but has to be regarded as mal-adaptation instead [
22]—as it may increase the vulnerability of natural and human systems over the long term. Nonetheless, the subsidization of air conditioning for low-income urban residents may entail new financial outlays and be offset by health-related cost savings due to the reductions in heat-related morbidity and mortality [
6].
Adaptation in the context of climate change is defined as “Initiatives and measures to reduce the vulnerability of natural and human systems against actual or expected climate change effects” [
12] (p. 69). Adaptation practices can take place at a different level from individual and household level to the community and institutional level. According to Jian Zuo et al. “The common mechanisms to deal with heat waves and the associated consequences include: structural/institutional, technological and cultural/behavioral” [
10]. However, as Bolitho and Miller [
23] argue, responses to extreme heat reflect a tension between a risk management paradigm (heat as an emergency) and a social vulnerability perspective (heat as chronic stress), whereas adaptation policy and planning that appreciates the interconnections between the two perspectives would likely reduce vulnerability and contribute to more urban sustainability [
23].
Table 1 summarizes their views.
In addition to these measures, infrastructure investments, particularly in vulnerable urban areas, urban greening programs such as green roofs, and building codes requiring reflective exterior surfaces are among the most effective and sustainable adaptation options, and should—according to the First Assessment Report of the Urban Climate Change Research Network (UCCRN)—therefore be strongly considered [
1].
However, adaptation cannot be delivered by a top-down process, particularly in vulnerable urban communities, where residents’ views to impacts and adaptation needs and perceptions go often unheard and unrecognized [
14]. Much in contrast the perception and views of vulnerable populations have to be integrated and made a vital part of adaptation planning, as citizens’ act on their beliefs and perceptions [
24], which is vitally important for the success of adaptation on all levels—individual to community. To this end, it is also important to scientifically assess the residents’ views, making them part of urban research and community science for sustainable adaptation.
Innovative governance forms can help integrating vulnerable populations in research as well as the urban adaptation decision-making processes. The Organization for Economic Cooperation and Development (OECD) defines governance as “
the formal and informal arrangements that determine how public decisions are made and how public actions are carried out, from the perspective of maintaining a country’s constitutional values in the face of changing problems” [
25] cited in [
26] (p. 5). Accordingly, climate change governance includes a broad spectrum of navigating mechanisms, which may include the collaboration of different actors and institutions according to hierarchical forms of principles and regulation [
27]. Thereby, urban climate governance describes the ways in which private, public, and civil society actors articulate climate goals, exercise influence and authority, and organize urban climate planning and the process of implementation [
28]. More precisely, as Chanza and De wit [
29] argue, decentralization, autonomy, accountability, transparency, responsiveness, flexibility, participation and inclusion are basic elements of climate change governance.
Reviewing the state of the art on climate change governance, it is realized that the majority of studies focus primarily on mitigation and mitigation planning, set by the international and national levels of government. The more local dimension of adaptation and adaptation planning is not yet covered comprehensively [
28,
30,
31].
We aim to assess the social sensibility and effectiveness of adaptation options by evaluating the potential consequences of adaptation scenarios on the reduction of detrimental impacts mentioned by the respondents. Respondents are distinguished by four groups: poverty, low-income, middle-income and high-income. To assess the consequences of adaptation scenarios across income groups, we use Fuzzy Cognitive Mapping (FCM) [
32,
33,
34,
35]. The importance of mental models and cognitive maps in identifying and evaluating the key elements of climate change impacts has been highlighted in adaptation research [
32,
33,
34,
36,
37]. Cognitive maps are a representation of external reality by using individual’s perceptions, experiences, and knowledge structured by the respondents’ reasoning. Capturing groups or individual’s cognitive maps regarding climate change impacts clearly illustrates how individuals understand climate problems [
36], which can be used to develop adaptation strategies. FCM is able to deduce socially sensible adaptation options by way of manipulations of the network denoting if-then-connotations, e.g., by way of adding elements—exemplifying new policies, cutting relations or lowering their link weight—exemplifying ceased or lowered influence, or by changing concept weight—exemplifying reduced importance of an element.
It should be highlighted that, by using perception data (not focusing on scientific facts) and by using online questionnaires to collect those, “hard to reach” populations such as people in poverty and low income groups are included as active stockholders in the research on adaptation planning. By that, we also hope to give these groups a voice in the climate change adaptation and planning process.
The main objective of this research is to develop an evaluation tool to simulate heat wave adaptation according to different income group’s perceptions especially hard to reach population such as in poverty group and low-income citizens. The research is structured along the following research questions in order to reach the main objective.
- -
What are the main differences across income groups in regards to their concerns about future impacts of heat waves?
- -
What are the main differences across income groups in regards to their opinion about citizens’ responsibility in heat wave adaptation and urban sector(s) most in need of adaptation actions during future heat waves in NYC?
- -
What are the main differences between different income groups’ cognitive maps in regards to impacts of heat waves in NYC?
- -
How do prominent adaptation options affect different income groups in NYC, i.e. lower the impacts of heat waves for each group?
4. Discussion
The aim of this research was to assess the perception of NYC residents’ regarding impacts of heat waves and aspects of adaptation. By using citizen’s perception we aimed to concentrate on the local level and to develop relevant information for socially sensible adaptation options to heat waves as bottom-up process—and in contrast to top-down approaches in the governance hierarchy. Doing so this study also aimed to use and prioritize non-scientific local knowledge as the main driver in developing adaptation options for the local level. Using perception data and citizen’s cognitive maps with regard to the impacts of heat wave, residents’ understanding about heat waves become apparent, which is vitally important for individual and autonomous adaptation.
Moreover, using online interviews to collect residents’ perception data has proven to be a useful method and channel to reach in particular people of low incomes and people living in poverty—usually regarded as hard to reach. Using paid questionnaire surveys their views and perceptions can be elicited and theoretically be integrated and respected in the urban planning process. By that, people of lower income (could) become systematically involved as active stockholders in the urban governance and decision making processes. The developed tool is therefore useful, as considers the views and perception of vulnerable groups alongside other citizens, e.g., those belonging to higher income classes. Only if views of all income groups are respected adaptation measures can hypothetically be fully effective.
Our results show that residents living in poverty and of low income are more worried by heat wave impacts than higher income groups. They also perceive impacts to be larger in the future and a larger adaptation responsibility with themselves. The scenario analyses showed that investments in the transit sector show the highest positive impacts for all income groups, but for the lower income groups most. These results should be very useful for the decision makers in New York City, allowing aligning adaptation options with regard to future heat waves. According to the results of FCM analysis, focusing on the transit sector would have a potentially positive effect on concepts related to the health sector and water and electricity sectors as well and will lead to more effective and comprehensive answers to citizens needs when they face the negative impacts of heat waves.
There are also some limitations of the study. Compared to the population size of NYC (according to the American Community Survey projected to be at 8,405,837 in 2013) the sample size of 762 respondents is relatively small. However, for a social study it is quite comprehensive and particularly rich, with more than 60 variables to different subjects which can provide a useful overview of differences between various income groups in New York City.
The other limitation is in regards to gathering the FCM data through an online questionnaire. Eliciting networks via questionnaires is a complex task, increasing the risk of misunderstandings and mistakes, especially about the relation between concepts. According to Özesmi and Özesmi (2004) and Olazabal and Reckien (2015) face to face interview method should be favoured. However, using online questionnaires has also advantages, as it allows to reach more participant in a shorter timeframe and selected participants of particular characteristics or large diversity. Online FCM samples can therefore provide a more comprehensive sample, e.g., increasing the spatial scope of sampling from all New York City.