Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 49
4 How and Where Will 10 Billion People Live on Earth? W e live in the century of the city. In 2008, economic opportunities and, in many places, are the humanity crossed a milestone as it marked engines of economic growth. the first time that more people lived in Most urban population growth will be concentrated urban areas than any other type of settlement. The in the developing world, particularly in Africa and Asia. U nited Nations forecasts that most of the global Historically, urbanization has taken place primarily in population growth in the coming decades will occur in more developed regions. In 1990, the populations of urban areas. World population is expected to increase Europe, North America, Latin America, and Oceania by 1 billion to 5 billion between 2007 and 2050 (UN, were already more than 70 percent urban (Figure 4.2). In 2009),1 yielding a global total of between 7.9 and contrast, even by 2010, the urban population of Africa 12 billion people by 2050 depending on fertility and will barely approach 40 percent and Asia’s urban popula- mortality trends (Figure 4.1).2 Under the medium- tion will be less than half of the total population. growth scenario, world population will be almost In addition to an urbanizing population, globally, 10 billion by 2050, with an estimated 3.1 billion new the number of households is growing faster than urban dwellers.3 Notwithstanding uncertainties around population (Liu et al., 2003). In the United States, the effect of HIV/AIDS and economic downturns on the average household size has been steadily declining, mortality and urban migration, the current scale of from 5.5 in 1850 to 4.5 in 1915 to 2.56 in 2008 (U.S. urbanization is unparalleled in history (Cohen, 2004). Census Bureau, 2004). In particular, the proportion of the population living in single-person households An urbanizing global population has significant has increased significantly, from 7.7 percent in 1940 consequences—potentially positive as well as nega- to 25.8 percent in 2000. The resulting increase in the tive—for resource use, environmental sustainability, number of households relative to population growth and ultimately, the well-being of humanity. The poses significant threats to biodiversity and natural concentration of people and resources in dense urban resources because per capita consumption of resources settlements can reduce the energy consumed for build- is more closely aligned with the number of households ings and transportation and can lower carbon dioxide than with the population per se (Liu et al., 2003). Not emissions (NRC, 2009). Urban areas also provide only does each household maintain its own residence, but total per capita living space has been increasing 1United Nations Population Database, World Population Pros- in many countries. In the United States, the average pects, 2008 Revision. Available at www.esa.un.org/unpp/index. size of single-family homes increased from 1,500 ft2 asp?panel=1 (accessed February 11, 2010). 2The high degree of variance reflects high levels of uncertainty (139 m2) in 1970 to 2,519 ft2 (234 m2) in 2008, an in making population projections. increase of more than 60 percent (U.S. Census Bureau, 3The 0.6 billion difference between the increase in world popula- 2004). Similarly, in China, per capita residential living tion and urban population will be due to population growth occur- ring mainly in urban areas and rural to urban migration. 

OCR for page 49
50 UNDERSTANDING THE CHANGING PLANET FIGURE 4.1 Depending on fertility and mortality rates, total world population could reach nearly 12 billion by 2050, but could be as low as 7.9 billion. SOURCE: United Nations. FIGURE 4.2 Percentage of urban population by region, 1950­2050. SOURCE: United Nations.

OCR for page 49
5 SETTLEMENT space has tripled over a 27-year period, from 8.1 m2 in and interactions that link people, places, and processes 1978 to 26 m2 in 2005 (National Bureau of Statistics together. These linkages, which span environmental, of China, 2009). cultural, social, economic, and political realms, mean It follows that we need to understand how 10 bil- that, for example, urbanization in one location affects lion people will be allocated among households and demand for resources or waste disposal in another. distributed geographically across the world over the The geographical sciences have a long tradition next 40 years. What processes create diverse patterns of examining where, within urban areas, various kinds of human settlement? How is accelerating urbaniza- of people live, of investigating the processes that help tion changing social, environmental, and economic to create such patterns, and of assessing the implica- conditions? The environmental challenges, resource re- tions of residential patterning for variation in access quirements, infrastructure needs, energy demands, and to various opportunities such as jobs, medical care, or governance issues associated with the unfolding growth recreation. Since the 1960s, for example, studies have in urban population raise fundamental, policy-relevant documented the ways in which urban settlements are questions—and pose unprecedented opportunities for distinguished by segregation along the lines of stage in moving toward sustainability—that cannot be ignored the life course (e.g., singles, couples without children, as society navigates the urbanizing world of the 21st families with children, and so on), socioeconomic sta- century. Thus, one of the biggest challenges facing tus, and race and ethnicity. The spatial patterning of humanity is how and where 10 billion people will live so these dimensions is different in different regions of the as to reduce their environmental footprint. Given that world (Abu-Lughod, 1969), but in every place, the pat- most of these 10 billion people will live in cities, what terns of where people live within cities are the outcome are the consequences of an urbanizing Earth, and how of a mix of public policies and household preferences. can we reduce the negative impacts, while enhancing Modern geographical approaches and methods are the positive impacts? generating insights into urban land-use patterns, from intracity to regional and global scales. The routine collec- tion of imagery for most of Earth’s land areas by satellites role oF geograPhical scieNces provides an invaluable historical record covering more The study of human settlements is inherently an than three decades. This revolutionary development investigation of human–environment interactions, makes it possible to monitor human modification and which requires spatially explicit data and analysis, an urbanization of Earth’s surface across a range of spatial understanding of the interaction among places and resolutions, from <1 m to the global scale (Sawaya et al., 2003; Zhang et al., 2004). Satellites such as Terra, Aqua, across scales, and knowledge of the trade-offs among Landsat, and Tropical Rainfall Measuring Mission all different land uses. Studies of the city, urban growth, urban-land-use theory, and the development of human provide data on the urban environment (Box 4.1). settlements all have long traditions in the geographi- The growing inventory of geographically indexed cal sciences (Marsh, 1864). Much of the work on data makes it possible to combine satellite images urban areas, their form and function in urban plan- with census and other information to develop analyti- ning, urban economics, urban geography, and urban cally useful maps that show the distribution of human sociology, has drawn on the spatial land-use models population around the world. For example, the Global Rural-Urban Mapping Project4 has generated a glob- of von Thünen (1826/1966), Burgess (1925), Muth ally consistent and spatially explicit dataset of urban (1961), Alonso (1964), and others. population distribution. Furthermore, advances in the Although cities have always depended on com- development of analytical methods for geovisualization, plex linkages with their immediate surroundings, as geosimulation, and spatially explicit process models well as with more distant places, the speed, reach, and have simulated urban growth (Torrens, 2006); shown impact of these interconnections have become truly the linkages between places and scales over time (Kwan, global, and will continue to be so. Understanding the processes and consequences of accelerating urbaniza- 4 See sedac.ciesin.columbia.edu/gpw (accessed January 20, tion therefore requires tracing the web of connections 2010).

OCR for page 49
5 UNDERSTANDING THE CHANGING PLANET BOX 4.1 Remote Sensing Applications Related to Human Settlements The longest continuous observations of Earth are available through the Landsat Program, which has remotely monitored urban areas and urban growth since 1972. From the rate and magnitude of urban expansion, to population density and the global distribution of nighttime lights, satellite data have provided baseline information about the characteristics and geographical distribution of human settlements around the world. The long temporal record of satellite data is producing a clearer picture of the evolution of urban form (Herold et al., 2003; Seto and Fragkias, 2005), the impact of cities on prime agricultural land (Seto et al., 2000; Imhoff et al., 2004a), and the footprint of urban areas on local and global climate (Voogt and Oke, 2003; Jin et al., 2005; Shepherd, 2005). Data from the Moderate Resolution Imaging Spectroradiometer (MODIS) provide daily global coverage at 500-m to 1-km spatial resolution on urban characteristics such as land cover, surface albedo, aerosols, and land surface “skin” temperature (Engel-Cox et al., 2004; Zhou et al., 2004). Since 1992, the Defense Meteorological Satellite Program, Operational Linescan System (DMSP/OLS) has recorded low levels of visible and near-infrared radiance at night, making it possible to detect lights from cities, towns, and industrial sites (see Figure) (Small et al., 2005; Amaral et al., 2006). Below are examples of urban applications using satellite remote sensing data. Nighttime lights in North America. SOURCE: National Atlas online database. Available at www.nationalatlas.gov/atlasftp.html#nitelti (accessed January 20, 2010). 2000); and provided new visual representations of data the characteristics of settlements. Contemporary urban for hypothesis generation (Carr et al., 2005). settlements in coastal China, for example, must be un- As discussed in the introduction to this report, a derstood in the context of flows of people from villages hallmark of the geographical perspective is the rec- in the countryside, flows of capital from the United ognition that human–environment interactions and States and elsewhere, and flows of raw materials and responses to changed environments are “place-based,” finished products that sustain manufacturing in these meaning that they depend in part on particular cir- coastal cities (see Chapter 5). At the intraurban scale, cumstances and characteristics that coalesce in places. point location data from cell phones can be aggregated Especially important to a place-based analysis is the at various scales to map the pattern and intensity of effort to understand how linkages between places shape urban activities throughout the day, with enormous

OCR for page 49
5 SETTLEMENT Satellite Spatial Resolution Repeat Cycle Sample Urban Applications Landsat 15-60 m 18 days Urban morphology (Herold et al., 2003; Seto and Fragkias, 2005); urban growth and agricultural land loss (Seto et al., 2000) MODIS 250 m to 1 km 16 days Urban air quality (Engel-Cox et al., 2004); urban area mapping (Doll et al., 2001; Schneider et al., 2003); footprint of urban climates on vegetation phenology (Zhang et al., 2004) ASTER 15-30 m 16 days Urban land cover (Netzband and Stefanov, 2004) MISR 250-275 m 16 days Aerosol optical thickness (Jiang et al., 2007); urban land cover (Doll et al., 2001) AVHRR 1.1 km 9 days Urban heat island effect (Gallo et al., 1993); urban surface temperatures (Dousset and Gourmelon, 2003) SPOT 2.5-20 m 26 days Urban land cover (Dousset and Gourmelon, 2003); urban detection (Baraldi and Parmiggiani, 1990) IKONOS 0.8-4 m 3 to 5 days off-nadir, Urban features detection (Weydahl et al., 2005); urban road detection 144 days true-nadir (Haverkamp, 2002) SAR 10-100 m 24 days Urban features detection (Weydahl et al., 2005); human settlement detection, population estimation, and urban land-use pattern (Henderson and Zong- Guo, 1997); urban road detection (Tupin et al., 2002) DMSP/OLS 0.56-2.7 km Twice daily Population estimates (Amaral et al., 2006); urban extent (Small et al., 2005); urban energy consumption (Elvidge et al., 1997; see also Figure. NOTES: ASTER = Advanced Spaceborne Thermal Emission and Reflection Radiometer; AVHRR = Advanced Very High Resolution Radiometer; DMSP/OLS = Defense Meteorological Satellite Program, Operational Linescan System; MISR = Multi-angle Imaging SpectroRadiometer; MODIS = Moderate Resolution Imaging Spectroradiometer; SAR = synthetic aperture radar; SPOT = Satellite Pour l’Observation de la Terre; potential for urban and transportation planning (Ratti known about which urban forms have fewer negative et al., 2006). environmental consequences. From the design of local neighborhoods and the layout of a city, to the regional configuration of city clusters and the global geographi- research suBQuesTioNs cal distribution of urban areas, how and where urban areas develop affects resource use, biodiversity, carbon What forms of urbanization are most emissions, and ultimately environmental sustainability. environmentally sustainable? Although it is well recognized that denser, more com- Although forecasts suggest that most popula- pact settlements reduce urban growth and the physical tion growth will occur in urban areas, much less is footprint of cities, the environmental and social trade-

OCR for page 49
5 UNDERSTANDING THE CHANGING PLANET offs between certain shapes and sizes of cities and among take decades to be realized. Furthermore, the life span distributions of cities across different ecosystems and of the built environment and the cumulative effects of geographical locations remain poorly understood. Cities land-use change results in an increase in the cumulative represent the most human-dominated landscapes on savings in energy and emissions (NRC, 2009). Earth, but what are the consequences of different paths In the next 40 years, patterns and rates of urbaniza- of urban development? What urban forms minimize tion will vary by region and within regions. The ways resource use such as water, energy, and building materi- in which these settlements develop will affect lifestyles, als? What are the consequences for ecosystem services, landforms, and livelihoods. For example, there is grow- habitat fragmentation, or biodiversity of clustered urban ing evidence that low-density urban development is development vs. noncontiguous metropolitan regions? positively correlated with automobile dependence, which What are the environmental trade-offs of 3 billion new in turn affects walkability, physical activity, and obesity urban dwellers being housed in megacities such as Seoul (Ewing et al., 2006; Frank et al., 2006). An epidemio- and Shanghai vs. smaller urban centers such as Chonju logical study of more than 1,500 research papers on the and Harbin? relationship between the built environment and obesity Urban areas and urban expansion are prime causes found that 84 percent of the studies reported a positive of habitat fragmentation, habitat loss, and species association between low-density urbanization and obe- extinction (McKinney, 2002). Worldwide, the trans- sity (Papas et al., 2007). By definition, urban expansion formation of landscapes for urban development has transforms land cover, but where urban growth occurs driven plant extinctions (Hahs et al., 2009). In the also affects local ecosystem processes. The growth of Phoenix, Arizona, for example, has dramatically altered United States, urbanization affects more species than the Indian Bend Wash watershed through restructuring does any other human activity (Czech et al., 2000). The stream channels and creating artificial water lakes in the wildland-urban interface, where developed urban areas city (Roach et al., 2008; see also Chapter 3). meet undeveloped wildland, is a focal zone for habitat Worldwide, the fastest growing cities will be in fragmentation, the introduction of exotic species, and Africa and Asia, and many of these fast-growing cities loss of biodiversity (Radeloff et al., 2005). of tomorrow are just small towns today. There is, there- Not only do cities transform landscapes, but also fore, significant opportunity to direct the form of urban urban form and urban densities affect resource use. development and for science to help shape policies that Extensive low-density suburbanization in the United can lead toward more sustainable and less environmen- States has contributed to the growth in total vehicle tally and socially disruptive cities. An important part miles traveled (see Chapter 7, Figure 7.1) and reduced nonmotorized travel by transit, bicycle, or foot. Exten- of the science needed to support such policies is the sive urban development also leads to the conversion incorporation of uncertainty into models and scenarios of productive agricultural land and environmentally of urban development. important ecosystems. Given increasing urbanization, A key question is which urban forms are more which pathways of urban development will reduce sustainable under different environmental and socio- the demand for energy and allow for more ecosystem cultural conditions. For example, cellular automata services? It has long been suspected that urban land research has generated insight into the evolution of development patterns such as the density of employ- urban land-use patterns and dynamics (White and ment and population, diversity of mixed uses, and the Engelen, 1993; Batty, 1997; Clarke et al., 1997). design of neighborhoods and streets affect the demand Coupling cellular automata models and agent-based for travel and energy use (Cervero and Kockelman, models can simulate decision making and represent 1997). More compact development through higher complex spatial interactions of stakeholders (Parker et al., 2004). Spatially explicit multiscale models of urban residential and employment densities can reduce energy expansion and land-use change can provide environ- consumption and carbon emissions. However, the en- mental indicators such as carbon sequestration, habitat ergy and emissions savings are likely to be small in the fragmentation, and biodiversity (Alberti and Waddell, short term, because the benefits of land-use changes 2000; Theobald, 2005; Verburg et al., 2008). A growing and the reversal of current development patterns will

OCR for page 49
55 SETTLEMENT research community is coupling historical approaches particular active cooling systems, was the strongest with modern geographical methods to reconstruct the protective factor against heat-related death (Naughton spatial history and urban ecology of cities (e.g., the et al., 2002). Existing transportation infrastructure may Mannahatta Project5 and the Spatial History Project6). not have been designed to withstand extreme events For example, a spatial historical analysis that compares associated with climate change. The vulnerability of the growth of Bangalore, India, with Silicon Valley transportation and energy infrastructure will depend in the United States shows that roads drive expansive on the rate of warming, the types of extreme events urban development and farmland loss in Silicon Valley, that occur, and new design standards for buildings and but not in Bangalore (Reilly et al., 2009). Satellite infrastructure. How can urban areas adapt to the suite image analysis can provide information on what con- of global environmental changes? figurations, forms, and size of human settlements have The ability of urban areas to cope with changing the least ecological impact. For example, nighttime global environmental circumstances depends on geo- city lights derived from satellite data have been used to graphic location in relation to a range of physical and assess the effect of urban development on soil resources human circumstances, the capacity of local political (Imhoff et al., 1997) and net primary productivity and economic institutions to deal with disruption, and (Milesi et al., 2003). Satellite data have also been used the organization and physical character of the built to map the effect of urban expansion on long-term eco- environment. Different types of global environmental logical changes (Ellis et al, 2006), agricultural land loss changes (e.g., sea-level rise, extreme heat, or prolonged (Seto et al., 2000), and forest fragmentation (Wang and drought) can also affect urban adaptive capacity in Moskovits, 2001). Geographical data and geographi- variable ways. Around the world, evidence is growing cal analysis can also help quantify the spatial structure that there is an increase in extreme weather and climate of urban development (Seto and Fragkias, 2005; Keys events (Alexander et al., 2006). In North America, et al., 2007) and evaluate the effects of urban form on droughts are becoming more severe, heavy precipitation water use ( Jantz et al., 2004; Guhathakurta and Gober, events are more frequent and intense, and severe storms 2007), energy demand (Ewing and Rong, 2008), and are increasing in power and frequency (CCSP, 2008). air quality (Stone et al., 2007). These intense weather and climate events could pose major threats to urban systems and disrupt urban activi- ties. Furthermore, depending on the nature of extreme how can urban areas adapt to and become more events, they could exacerbate existing local environ- resilient to global environmental change? mental conditions—reducing water quality, threatening Human-induced alterations and transformations of sanitation and public health, worsening local air quality, Earth drive environmental changes that are global in and intensifying the urban heat island effect. scale (Vitousek, 1992). At the same time, global envi- However, climate change is not the only type of ronmental change will have a wide spectrum of effects global environmental change to which urban areas will on urban areas. Climate-related effects on urban areas need to adapt. Changes in the hydrological cycle and include increases in temperature, heat stress, sea-level water availability will constrain future urban growth rise, storm surges, threats to building stock, energy and require water quantity management across multiple and transportation infrastructure, and urban flooding, jurisdictions of metropolitan areas (Holway, 2009). Cities near rivers and in delta systems will be particu- drainage, and landslides (NRC, 2008c). The increase in larly threatened by flooding and landslides. Thirteen the frequency and magnitude of extreme events such percent of the world’s urban population lives in low- as wind, snow, ice storms, hurricanes, and heat waves elevation coastal zones, defined as coastal regions less will threaten building stock, energy and transportation than 10 m above sea level (McGranahan et al., 2007). infrastructure, and ultimately the well-being of urban These coastal communities will be particularly vulner- populations. Studies of the 1995 and 1999 Chicago able to sea-level rise and storm surges. heat waves conclude that housing infrastructure, in More comprehensive knowledge and understanding 5See www.mannahattaproject.org (accessed January 20, 2010). of global environmental change and its consequences 6See spatialhistory.stanford.edu (accessed January 20, 2010).

OCR for page 49
56 UNDERSTANDING THE CHANGING PLANET for urban areas will be needed if local decision makers, 2050. Accurate and timely mapping and forecasting communities, and individuals are to prepare, cope, of urban settlements will be important for a range of and adapt to these changes (Sánchez-Rodríguez et al., applications, including anticipating infrastructure needs; 2005). Adaptation research has focused on two issues planning for emergency response systems (Kwan and (Burton et al., 2002): (1) How can adaptation reduce Lee, 2005); identifying regions and communities at risk the impacts of climate change, and (2) what types of from storm surges, sea-level rise, and natural hazards; adaptation policies are needed? More than 20 cities and mapping urban growth and loss of natural and around the world—particularly in Europe and North agricultural land (Hasse and Lathrop, 2003). Cellular America—are in the process of developing local climate automata models that were initially developed by Von change adaptation and action plans. For example, the Neumann (1966) are now used widely to predict urban Chicago Climate Action Plan outlines strategies for growth in a range of cities around the world, includ- mitigating climate and actions to prepare for climate ing San Francisco (Clarke et al., 1997); Beijing, China change adaptation. Similarly, the London Climate (Chen et al., 2002); Chillán, Chile (Henríquez et al., 2005); and Lagos, Nigeria (Barredo and Dermicheli, Change Adaptation Strategy identifies key climate risks 2003). Other methods are being developed that do not and prioritizes adaptation strategies. However, recent have large data requirements and will be especially ap- research shows that local adaptation strategies may plicable in developing country contexts (Fragkias and conflict with mitigation efforts. One community on the Seto, 2007). Understanding where and how intra- and north coast of Australia’s New South Wales has a policy interurban settlements are changing, or are likely to that requires new residential developments to maintain change, provides fundamental information and insights vegetative cover and preserve wildlife habitat for the local to urban planners and policy makers. Innovations in the koala population. To achieve these goals, however, the geographical and computer sciences—especially online developments must be low density and car dependent, mapping and real-time geographic information systems which in turn increases driving and energy consumption. (GIS)—are bringing together different types of spatial Furthermore, siting homes in natural vegetation also data from multiple sources (from governments to indi- increases exposure to other climate change risks such as viduals) to increase their timeliness and utility for societal fire (Hamin and Gurran, 2009). benefit. Rather than using static risk or hazard maps that Advances in spatially explicit modeling and geo- suggest that vulnerability is unchanging, dynamic GIS graphical simulation can deepen understandings of with real-time information and real-time computing can how cities can be more resilient to global environment provide up-to-date information on the distribution of change. A first step in evaluating possible adaptation vulnerability across landscapes and can differentiate risk and mitigation strategies is knowledge about where among different communities within cities (Bankoff et cities are growing and their vulnerabilities to global al., 2004). For example, as in other parts of the world, change (Chapter 3). In many African countries, urban most of the largest Asian cities are located on a river population statistics are out of date or lacking, making bank, in a delta, or along the coast (UN Habitat, 2008), the task of mapping the location and growth of urban making them particularly vulnerable to climate change, settlements particularly challenging (Cohen, 2004). sea-level rise, flooding, and other extreme events. In the In these regions, much of the housing is informal (Satterthwaite, 2007). Because censuses are absent, case of the 2004 Asian tsunami, GIS-based vulnerability infrequent, or unreliable in many regions and because models coupled with demographic methods of tsunami- informal settlements often lack official recognition, displaced populations were used to quantify mortality many of these settlements become “invisible towns” estimates, methods that could also be used in the early (Stickler, 1990), highlighting the need for studies using stages of disaster relief (Doocy et al., 2007). modern geographical tools to map and understand The uncertainties associated with climate change, settlement locations and growth patterns. as well as with the formation of human settlements, There is a similar need to understand the pat- pose challenges for analysts and policy makers alike. terns of urbanization in Asia, where 16 of the world’s Advances in geovisualization and geosimulation can 27 largest urban agglomerations will be located by help inform adaptation strategies under different

OCR for page 49
5 SETTLEMENT TaBle 4.1 Various Pathways Through Which Urbanization Affects the Climate System anthropogenic greenhouse gas urban land cover urban aerosols emissions Urban heat island Surface energy budget Insolation, direct aerosol effect Radiative warming and feedbacks (UHI) and mean surface temperature record Wind flow and turbulence Surface energy budget, urban Direct and indirect aerosol effects and Radiative warming and feedbacks morphological parameters, mechanical related dynamic/thermodynamic response turbulence, bifurcated flow Clouds and precipitation Surface energy budget, UHI Aerosol indirect effects on cloud- Radiative warming and feedbacks destabilization, UHI mesocirculations, precipitation microphysics, insolation UHI-induced convergence zones effects Land surface hydrology Surface runoff, reduced infiltration, Aerosol indirect effects on cloud- Radiative warming and feedbacks less evapotranspiration microphysical and precipitation processes Carbon cycle Replacement of high net primary Black carbon aerosols Radiative warming and feedbacks, productivity land with impervious fluxes of carbon dioxide surface Nitrogen cycle Combustion, fertilization, sewage Acid rain, nitrates Radiative warming and feedback, release, and runoff NOx emissions SOURCE: Seto and Shepherd (2009). urbanization and global change scenarios. For example, The urban heat island effect is further affected by the the WaterSim Project7 simulates water supply and interaction among building geometry, land use, and demand in a desert city. The simulation and forecast urban materials (Oke, 1973; Arnfield, 2003). tools being developed in conjunction with this project Urban areas have been shown to produce a warming can help cities evaluate policies to adapt to a water- trend over regional climate (Kalnay and Cai, 2003), and constrained environment. there is mounting evidence that urban areas also affect precipitation (Lowry, 1998; Shepherd and Jin, 2004). It has long been known that urban areas alter their regional What are the impacts of accelerating and large-scale and microclimates, resulting in an increase in rainfall urbanization on local and regional climate patterns? downwind (Landsberg, 1970). Empirical evidence There is a growing scientific understanding of the shows that in some regions, urban expansion and urban relationship between urbanization and climate (Voogt air pollution result in a decline in rainfall (Amanatidis and Oke, 2003; Shepherd, 2005). Urbanization alters et al., 1993; Rosenfeld, 2000; Kaufmann et al., 2007). In climate through multiple pathways (Table 4.1). other places, urbanization has induced precipitation and Locally, the conversion of vegetated surfaces to possibly created thunderstorms, leading to significant urban areas modifies surface energy balance dynamics anomalies in precipitation patterns (Figure 4.3) (Dixon (Lo et al., 1997; Banta et al., 1998). Altering the and Mote, 2003). In places where it has been shown exchange of heat, water, trace gases, aerosols, and that urban areas increase rainfall, the difference can be momentum between the land surface and overlying as much as 5 percent to 25 percent (Mote et al., 2007). atmosphere leads to the urban heat island effect, which In addition to the quantity of rainfall, cities can also is characterized by elevated daytime and nighttime affect the timing and formation of thunderstorms and temperatures in and near urban areas compared to non- the severity of precipitation, as has been found for Tokyo urban or rural areas (Arnfield, 2003; Crutzen, 2004). (Yonetani, 1982), Beijing (Guo et al., 2006), Atlanta (Bornstein and Lin, 2000), Mexico City ( Jauregui and Romales, 1996), among others. These changes imply 7See watersim.asu.edu/ (accessed January 20, 2010).

OCR for page 49
5 UNDERSTANDING THE CHANGING PLANET FIGURE 4.3 Annual precipitation anoma­ lies, 1900–1988. Regions in blue show an increase in precipitation over the mean during the 1900–1988 period. Regions in red have become relatively drier during the same period. Areas without data are shown in white. SOURCE: Dai et al. (1997). that urban land uses and urban expansion through Continued advances in understanding the link land-cover change can affect local, regional, and global between urban land use and regional climate will re- climate at diurnal, seasonal, and long-term scales quire mining a variety of data—from remote sensing of (Stohlgren et al., 1998; Zhou et al., 2004). clouds, aerosols, and land to field-based meteorological A clearer picture is emerging that urban areas data on temperature and precipitation. It will require can affect climate at different scales. However, we modeling urban climates at fine spatial scales to under- lack a comprehensive understanding of the dynamics stand the effect of building materials, street geometry, by which urban expansion will affect climate; nor do and building geometry on local temperatures (Oke, we understand the interactions between local-scale 1973, 1981), explicit treatment of urban land use in dynamics and regional and global patterns. We also climate models (Bonan et al., 2002; Jin et al., 2005), have a fragmented picture of urban land-use patterns at and techniques for downscaling general circulation a global scale. Most analyses of urban land use are based models (Wilby and Wigley, 1997). on individual case studies of city or metro regions, and there are few comparative, regional, or global studies summarY (Seto and Shepherd, 2009). We have a good under- Urbanization in the 21st century will have far-reaching standing of some of the ways that urban areas affect effects ranging from the local to the global. Under- climate at several scales, but we lack comprehensive standing where people will live and how cities will understanding, and more importantly, an understand- develop in the future has implications for all aspects of ing of how urban growth—or different forms of urban human and environmental well-being discussed in this growth—will affect climate. Given the magnitude of report, from the provision of food for a growing urban the global urban transition of the 21st century, there population to safeguarding our planet’s biodiversity is an urgent need to understand the impact of urban and ecological services. New geographical data and land-use change on precipitation and temperature, and emerging analytical methods, combined with existing to forecast scenarios of urban expansion and their pos- research tools and techniques will help develop a more sible impacts on precipitation changes. How will the coherent and complete understanding of the patterns, growth of settlements affect rainfall and temperatures implications, and uncertainties of urbanization. at local, regional, and global scales?