Global typology of coastal urban vulnerability under rapid urbanization
Till Sterzel*#, Matthias K.B. Lüdeke*, Carsten Walther*, Marcel T. Kok+, Diana Sietz++, Paul L. Lucas+
*Potsdam Institute for Climate Impact Research, Research Domain II – Climate Impacts and Vulnerabilities, Telegraphenberg, 14473 Potsdam, Germany
#climate-babel, Lindenstraße 11, 11467 Potsdam
+ PBL NetherlandsEnvironmental Assessment Agency, 3720AH Bilthoven, TheNetherlands
++ Wageningen University and Research Centre, Sociology of Development and Change, Hollandseweg 1, Bode 18, 6706 KN Wageningen, The Netherlands
Corresponding author: Till Sterzel,
Abstract
Coastal urban areas are urbanizing at unprecedented rates – particularly in non-OECD countries. Combinations of long-standing and emerging challenges generate vulnerability by threatening human well-being and ecosystems alike. On an intermediate scale of complexity this study provides a first spatially explicit global systematization of these heterogeneous challenges into typical urban vulnerability profiles using largely sub-national data. A cluster analysis of urban expansion, urban population growth, marginalization, government effectiveness, exposures and sensitivities to climate-related extreme events, low-lying settlement, and wetland prevalence reveals a global typology of seven clearly distinguishable urban vulnerability profiles. These profiles have typical combinations of vulnerability-generating mechanisms and processes assigned to them. Using ample city case studies for testing the plausibility, we show that combinations of such mechanisms in extreme forms are driving vulnerability in the profiles, and which coastal locations are similar in this regard. Climate extreme vulnerability and government effectiveness further differentiate the huge asymmetries in coastal urban vulnerability. Against the background of underlying global trends we point out profile characteristics that can have implications for prioritizing systemic responses and future case studies, and then propose entry points for generic profile-based vulnerability reduction. Our findings contribute new insights into opportunities for sharing experience and vulnerability reducing measures between socio-ecologically similar urban areas in the rapidly urbanizing coastal fringe.
Keywords: Coastal cities, cluster analysis, vulnerability-generating mechanism, socio-ecological system, quantitative indication
Introduction
Urbanization is a defining phenomenon of our time (Hoornweg et al. 2011). Having reached 3.2 bn in 2011 (UNDESA 2012), urban population has increased by over factor four since 1950, and has lead to a concentration of 40% of the world’s population on a narrow coastal band that takes up 7% of the Earth’s surface (McGranahan et al. 2007). The locus of the most rapid urbanization is the developing world, where more than 95% of the net global population increase is projected to be concentrated in cities in the coming decades (UNDESA 2012; UNDESA 2008). Cities are disproportionately located along rivers and coastlines (Grimm et al. 2008). Urban areas are generally more coastal than rural areas (McGranahan et al. 2007), and urban areas are growing faster on coasts than inland (Seto et al. 2011a; McGranahan et al. 2007). Although the bulk of world population growth is taking place in small and medium-sized cities (Grubler et al. 2012), they are commonly overlooked in favor of larger and more iconic cities in case studies and meta-analyses.
Rapid urbanization is further contextualized by biophysical and socioeconomic characteristics operating on multiple temporal and spatial scales. Urban life is radically changing its coastal environments through unmanaged population increase (Grimm et al. 2008), urban expansion (Seto et al. 2011a), and resource demand (Bloom 2011). At the same time, coastal cities are heavily influenced by global change, and particularly at risk of flooding due to tropical storms and sea-level rise (Handmer et al. 2012). All of these characteristics can assume extreme forms in coastal urban areas.
Altogether, long-standing and emerging challenges are compounding in coastal cities due to the magnitude and acceleration of transitions (Tanner et al. 2009). These challenges risk to outpace efforts to reduce vulnerability in urban areas in rapidly growing low- and middle-income nations (O’Brien et al. 2012; UN-HABITAT 2011a; UNISDR 2011) by increasing exposures and sensitivities in urban populations and ecosystems alike (McGranahan et al. 2007), and overstretching municipal management and planning capacities (Grubler et al. 2012; Prasad et al. 2009).
Despite these converging characteristics in coastal urban areas across the developing world, and the resulting increases in vulnerability of their populations to them, there is a lack of integrating studies on a global scale that systematize these human-environment interactions. For example, Garschagen & Romero-Lankao (2013) point out the greater scientific attention the linkages between different components of vulnerability deserve under urbanization, because it can identify entry points to enhance adaptive capacity at the national level.
Regional and local case study literature shows that vulnerability in fast-growing urban coastal zones is heterogeneous, and that the challenges outlined above are unevenly distributed geographically. This makes it challenging to scale up successful vulnerability-reducing measures (Sietz et al. 2011). At the same time, there is a need for regional or global-coverage vulnerability (reduction) analysis, which requires generic insights that acknowledge local specifics and still constitute a global picture. Highlighting in how far outcomes of case studies are also characteristic or relevant for similar urban areas elsewhere could contribute to scaling up successful vulnerability-reducing measures through transfer between urban areas that are systematically similar.
In view of this scaling challenge and the need for an integrated perspective on the forces shaping the vulnerability of cities, a number of studies have used integrated qualitative and quantitative methods for systematizing generic global overviews of socio-ecological problems. On a subnational level, Kok et al. (2015; 2010) and Sietz et al. (2011) characterize and map the typical mechanisms and processes influencing livelihoods in global drylands regarding their socio-ecological vulnerability. The typical mechanisms and processes provide entry points for scaling up vulnerability reducing measures. Focusing on large, densely populated, rapidly growing urban areas, yet on a national scale, Kropp et al (2001) and Lüdeke et al (2004) characterize a typical cause-effect “favela syndrome” and categorize countries according to the dynamics and intensity of this exemplary non-sustainable human-environment interaction. These approaches are based on the hypothesis that it is possible to identify a limited number of typical dynamic cause-effect relationships at an intermediate level of complexity which allow to subsume comparable case studies of socio-ecological problems across the globe. Other studies focus on human-environment problems based on present or projected future climatic risks in a limited number of coastal cities or urban areas. McGranahan et al (2007) assess the geographic- and climate change-related risks for urban settlement in the heavily populated low elevation coastal zone (LECZ). Hanson et al (2011) and Hallegatte et al (2013) quantify current and future economic impacts of climate change-induced sea level rise and flooding for 3o and 136 coastal cities using future climate projections, respectively. Tanner et al. assess climate change resilience in ten rapidly urbanizing cities in predominantly coastal settings in developing countries (2009). De Sherbinin et al (2007) examine vulnerabilities of three coastal megacities to current and future climate hazards using a vulnerability framework considering multiple synergistic stresses and socio-ecological characteristics.
To our knowledge, this paper is the first spatially explicit study to systematize urban vulnerability under rapid urbanization in the global coastal fringe. We present a global systematization of rapidly growing urban coasts according to typical manifestations of mechanisms and processes that increase urban vulnerability under forces of global change. Using sub-national and national data we systematically profile the current situation in these urban areas on an intermediate level of complexity based on quantifying similarities in their socio-ecological problems. For this we apply a formalized method based on clustering (Kok et al. 2015; Janssen et al. 2012) which has been applied before to identify and interpret general mechanisms which similarly create vulnerability in global drylands on regional and local scales (Sietz 2014; Sietz et al. 2012; Sietz et al. 2011; Kok et al. 2010). We address and answer the following research questions:
(1) In how far do urban coasts share characteristic, typical mechanisms making them vulnerable under rapid urbanization?
(2) How are these areas positioned to deal with these mechanisms and reduce vulnerability?
Methods and data
Our method for systematizing how and where vulnerability is typically generated follows a method of vulnerability analysis on an intermediate level of complexity and spatial scale proposed by Kok et al (2015). It consists of the following steps: Drawing from a multitude of vulnerability-generating mechanisms documented in the literature we list and briefly characterize well-documented mechanisms that have been found to typically generate vulnerability in coastal urban areas. Then we identify indicator datasets that render information on the most important dimensions of the vulnerability-creating mechanisms. Finally, we subject these indicators to an established cluster analysis (Janssen et al. 2012; Lüdeke et al. 2014) to address in how far and where typical combinations of the vulnerability-creating mechanisms occur. Thereby each resulting cluster signifies an urban vulnerability profile.
Vulnerability-generating mechanisms and processes
First we clarify the use of specific terminology. We understand a vulnerability-generating mechanism as a cause-effect relationship in which a vulnerability relevant process a drives b (e.g. rapid urban population increase a drives urban expansion b). Thereby the mechanism is that a drives b, while b can be driven by multiple processes. A vulnerability-generating process is a progressing vulnerability-relevant phenomenon, and can generate vulnerability through one or multiple mechanisms. The introduction of this paper shows that the process of rapid urban population increase (a) generates urban vulnerability through a variety of mechanisms. Through an extensive literature review we now specify seven typical mechanisms that generate vulnerability in situations of rapid coastal urbanization, and give examples of cities where they have been documented. They reflect the complex interplay between socioeconomic and biophysical factors, and show linkages between each other. We continue the numeration from above, notwithstanding rapid urban population increase’s role as the main process and driver.
Urban expansion (b) - Urban land expansion is driving large-scale land cover change in developing countries across the globe (Seto et al. 2012; Seto et al. 2011a). Less-developed countries are generally experiencing much higher levels of both urban expansion and its main driver – rapid population growth (Angel 2011). Importantly so, urban expansion is faster in low elevation coastal zones than in other places (Seto et al. 2011a). Thereby exposure to climate extremes is typically increased (UN-HABITAT 2011a). This mechanism has been exemplarily documented for coastal cities including Accra, Bangkok (Shlomo 2011), Tel Aviv, Algiers, and Manila (Angel et al. 2010).
Wetland loss (c) - Wetlands and floodplains provide important functions and services for the urban and surrounding populations, e.g. flood regulation functions for attenuating negative consequences of climate extremes (Costanza et al. 2008). Unchecked urban expansion is leading to unprecedented degradation and destruction of such coastal ecosystems through increased demand for land (Baird 2009) and encroachment (Seto et al. 2012; McGranahan et al. 2007; Bravo de Guenni et al. 2005). The degradation of these functions through encroachment has increased urban inhabitants’ exposure and sensitivity to floods (Bravo de Guenni et al. 2005; Hardoy et al. 2001), and has serious implications for the livelihoods of societies dependent on their functions or services (Nicholls 2004). This was illustrated in New Orleans through Hurricane Katrina in 2005 (Törnqvist & Meffert 2008).
d) Management overstretch - Rapid urbanization is increasing the vulnerabilities in urban populations by overstretching municipal management and planning capacities (Grubler et al. 2012; Prasad et al. 2009; Tanner et al. 2009). The rapid endogenous and exogenous urban growth can overwhelm basic urban services, especially if municipal adaptive capacity is initially low. This comes as new challenges arise before long-standing ones have been dealt with (Tanner et al. 2009). For example, future increase of climate change is poised to further stretch management capacities in coastal areas (Alam & Rabbani 2007, Dodman et al. 2011, De Sherbinin et al. 2007). This mechanism of overstretched management and planning has exemplarily been illustrated in case studies from Dhaka (Alam & Rabbani 2007), Dar es Salaam (Dodman et al. 2011), and Mumbai (De Sherbinin et al. 2007). We subsume management and planning under “management” in the following.
e) Marginalization - Rapid population increase is often absorbed into the urban fabric through an increase of densely populated informal settlements (Prasad et al. 2009). The growth of informal settlements has been largely driven by poverty and marginalisation of poor and less equipped populations in and around megacities in many developing countries (Bravo de Guenni et al. 2005; Douglas et al. 2008) and is frequently underestimated (Kit & Lüdeke 2013). Such settlements have less capacities to deal with shocks, e.g. climate extremes (Handmer et al. 2012; Huq et al. 2007; Hardoy et al. 2001) such as tropical cyclones and floods (Handmer & Dovers 2007), and less capacities to deal with subsequent negative impacts (Bull-Kamanga 2003). This particularly holds true where poor management, and low building and infrastructure quality coincide with densely populated areas (Prasad et al. 2009). This mechanism has been observed in numerous coastal cities in India (Revi 2008), or Iliolo City in the Phillipines (Rayos Co 2010).
f) Vulnerable settlement development – Under inadequate urban planning and unchecked growth informal settlements outlined above commonly encroach more risk-prone areas where exposure to floods and cyclones is high (Cardona et al. 2012; Kit et al. 2011; Satterthwaite 2007). These areas are avoided by wealthier groups due to their higher exposure (Prasad et al. 2009). This leads to an increase in vulnerable populations and population density with low building qualities in floodplains. This mechanism has been observed in cities such as Lagos (Adelekan 2010), Mumbai (Chatterjee 2010), and Esmeraldas, Ecuador (Luque et al. 2013), and for many urban poor in large African coastal cities (Douglas et al. 2008).
g) Generation of exposure and sensitivity to climate extremes - Rapid and unplanned urbanization is a key driver of vulnerability to climate extremes (Cardona et al. 2012). Coastal cities are already disproportionately exposed and sensitive to climate extremes, e.g. cyclones (Hanson et al. 2011; Nicholls et al. 2007) and floods (Mondal & Tatem 2012; Hanson et al. 2011), which threaten human well-being (Handmer et al. 2012). Sections of most of the largest cities on the African coast are currently at risk of flooding (IPCC 2012; Adelekan 2010; Awuor et al. 2008). Exemplary cities subject to climate extremes include Dhaka (storm surges, UN-HABITAT 2011a) Sorsogon City, Philippines (Taifuns, Button et al. 2013), Mumbai, Rio de Janeiro, Shanghai (Floods, De Sherbinin et al. 2007).