Adapting to the Impacts of Climate Change (2010)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

CHAPTER TWO Vulnerabilities and Impacts A daptation is intended to reduce climate change vulnerabilities and impacts. That means any consideration of adaptation planning must begin with consid- eration of risks associated with climate change vulnerabilities and impacts, to the extent that these can be anticipated. More specifically, adaptation includes (1) the strategies, policies, and measures imple- mented to avoid, prepare for, and effectively respond to the adverse impacts of climate change on natural and human systems (to the extent that they can be anticipated), and (2) the social, cultural, economic, geographic, ecological, and other factors that de- termine the vulnerability of places, systems, and populations. Climate-related changes can create new or interact with existing vulnerabilities to cause impacts, including changes in: • Temperature, both averages and extremes; • Precipitation, both averages and extremes; • The intensity, frequency, duration, and/or location of extreme weather events; • Sea level; and • Atmospheric carbon dioxide (CO2) concentrations. Vulnerability is often defined as the capacity to be harmed. It is a function of the character, magnitude, and rate of climate variation to which a system is exposed, its sensitivity, and its adaptive capacity (Clark et al., 2000; IPCC, 2007a; Turner et al., 2003). Vulnerabilities can be reduced by limiting the magnitude of climate change through actions to limit greenhouse gas (GHG) emissions (ACC: Limiting the Magnitude of Future Climate Change; NRC, 2010c), reducing sensitivity (the underlying social, cultural, economic, geographic, ecological, and other factors that interact with exposures to determine the magnitude and extent of impacts), or improving coping capacity (the ability to avoid, prepare for, and respond to an impact so that it is not seriously disrup- tive). Actions to reduce sensitivity and increase coping capacity are keys to effective adaptation to climate change. A risk perspective (Chapter 4) considers the probability of an exposure and its con- sequences, including uncertainties in projecting the magnitude, rate, and extent of climate change. It also considers factors that shape sensitivities and coping capacities, which are as important as exposures in determining impacts. Later chapters of this report consider options for reducing risks by reducing sensitivities and improving cop-  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E ing capacities. This chapter provides the context by summarizing what is known about current and projected climate change impacts and vulnerabilities in the United States. PROJECTED u.S. CLIMATE CHANgES THAT COuLD REQuIRE ADAPTIvE RESPONSES Climate-related impacts that require adaptation are already being observed in the United States and its coastal waters (USGCRP, 2009), and empirical evidence suggests that many of these and other impacts will grow in severity in the future (USGCRP, 2009). Over the past 50 years: • Average temperature in the United States increased more than 2°F (1°C). • Precipitation in the United States increased an average of 5 percent, and the intensity of precipitation events also increased. • Many types of extreme weather events increased in frequency and intensity; hurricanes, although not more frequent, increased in destructive energy. • Sea level increased along most of the U.S. coast over the past 50 years, with some areas along the Atlantic and Gulf coasts experiencing increases of greater than 8 inches. • Arctic sea ice extent decreased 3 to 4 percent per decade, with end-of-summer ice declining at 11 percent per decade. These changes are causing impacts that should promote adaptation regardless of whether the trends are permanent. In many circumstances, projected increases in the frequency and intensity of many extreme weather events over the next several de- cades will initially drive adaptation more than changes in mean weather variables. Im- ages from Alaska provide a vivid example of observed climate change impacts in the United States (Figure 2.1). These impacts already require adaptation in many locations and economic sectors. Effective adaptation depends on understanding projected climatic changes at geo- graphic and temporal scales appropriate for the needed response. The report Global Climate Change Impacts in the United States (USGCRP, 2009) was based on two pro- jected climate change scenarios: one of relatively moderate changes in the event that GHG emissions peak before the middle of the century and decline thereafter (lower emissions scenario), and another of relatively severe changes in the event that GHG emissions continue to grow at current rates without aggressive actions to limit them (higher emissions scenario). Prospects for adaptation to keep disruption from climate change impacts at socially acceptable levels depend very substantially on what hap- pens with efforts to limit emissions. At moderate rates and levels of climate change, adaptation can be very effective. At severe rates and levels of climate change, limits of 0 

Vulnerabilities and Impacts FIguRE 2.1 The Arctic village of Shismaref: Rising sea levels and fierce storms have eroded the shoreline near this coastal Inupiat village, breaking down sea walls and washing away homes. Residents decided to relocate farther inland for safety, giving up their traditional fishing, sealing, and home-building sites. SOURCE: Photo by Edward W. Lempinen/AAAS. © 2006 AAAS. many adaptation options are likely to be reached, and resulting adaptations are likely to be much more disruptive. A key fact about climate change impacts is that stabilization of atmospheric GHG con- centrations will not immediately stabilize the climate, which will continue to change for some time because of the delayed response of the climate to the buildup of GHGs emitted in the recent past. A companion to this report (ACC: Limiting the Magnitude of Future Climate Change; NRC, 2010c) details the challenges and choices the nation and the world face in sufficiently limiting GHG emissions to keep climatic changes at a relatively moderate level.1 It also concludes that stabilizing emissions at moder- ate levels is becoming increasingly difficult in the face of U.S. and global inaction. The U.S. Global Change Research Program’s (USGCRP’s) characterizations of two possible futures (lower or higher emissions) show that an effective response to climate change must include both adaptation and mitigation (e.g., Wilbanks and Sathaye, 2007). Much of the current knowledge about projected climate changes in the United States comes from an assessment process mandated by the U.S. Congress in the Global 1 For a discussion linking emission rates and atmospheric concentrations of GHGs to changes in global mean temperature, see Chapter 2 of ACC: Limiting the Magnitude of Future Climate Change (NRC, 2010c).  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E Change Research Act of 1990 (P.L. 101-606), including the U.S. National Assessment (USGCRP, 2001) and conclusions from 21 widely peer-reviewed Synthesis and Assess- ment Products (SAPs) produced by the U.S. Climate Change Science Program between 2006 and 2009 on specific topics ranging from knowledge of the physical climate system to the interface between climate change and society. The SAPs were sum- marized and updated in Global Climate Change Impacts in the United States (USGCRP, 2009). According to this summary report, future climate change impacts in the United States will include warmer average temperatures, changes in precipitation patterns, more frequent heat waves and severe storms, rising sea level, and decreases in sea ice and permafrost, which will be particularly rapid in the Arctic. The average temperature in the United States will continue to rise with climate change, but the magnitude of the increase depends primarily on the amount of heat- trapping GHGs emitted globally and how sensitive the climate is in responding to those emissions. Figure 2.2 shows projected temperature change under the higher and lower emissions scenarios in midcentury and at the end of the century. The brack- ets on the thermometers represent the likely range of model projections, although lower or higher outcomes are possible. By the end of the century, the average U.S. temperature is projected to increase approximately 7°F to 11°F (4°C to 6°C) under the higher emissions scenario and approximately 4°F to 6.5°F (2°C to 4°C) under the lower emissions scenario (USGCRP, 2009). Projections of future precipitation generally indicate that northern areas (higher latitudes) will receive more precipitation, and southern areas, particularly in the West, will become drier (USGCRP, 2009). However, the mechanisms by which human-induced climate change affects precipitation are subtler than those of temperature and paint a more complex picture (e.g., Zhang et al., 2007). Figure 2.3 shows projected changes by 2080-2099 under the higher emissions scenario; these are sample results from climate models, not projections of certainty for the future. The amount of rain falling in the heaviest downpours has already increased approxi- mately 20 percent on average in the past century, and this trend is very likely to con- tinue, with the largest increases in the wettest places (USGCRP, 2009). Figure 2.4 shows projected changes from the 1990s average to the 2090s average in the amount of precipitation falling in light, moderate, and heavy events. The lightest precipitation is projected to decrease, while the heaviest will increase, continuing the observed trend. Many types of extreme weather events, such as heat waves, have become more fre- quent and intense during the past 40 to 50 years, while cold extremes have become less frequent (USGCRP, 2009). In the future, currently rare extreme events (for example a 1-in-20-year event) are projected to become more commonplace (Figure 2.5), al-  

Vulnerabilities and Impacts FIguRE 2.2 Projected temperature change (°F) from 1961-1979 baseline. NOTE: These results are derived from global models whose spatial resolution is insufficient to resolve important details like mountain ranges. SOURCE: USGCRP (2009) (http://www.globalchange.gov). though these projected increases will not be uniformly distributed over temporal and spatial scales. For example, a day so hot that it is currently experienced once every 20 years would likely occur every other year or more frequently by the end of the century under the higher emissions scenario. Although uncertainties remain about whether the number of hurricanes could increase with climate change, the destructive energy of Atlantic hurricanes is likely to increase in this century as sea surface temperature rises (USGCRP, 2009) (Figure 2.6). In addition, cold-season storm tracks are shifting  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E FIguRE 2.3 Projected change in North American precipitation by 2080-2099. NOTE: Cross-hatching indi- cates areas in which climate models do not agree. SOURCE: USGCRP (2009) (http://www.globalchange.gov). northward, and the strongest storms are likely to become stronger and more frequent (USGCRP, 2009). The ocean is warming and glaciers and polar ice sheets are melting, causing sea level to continue to rise, most likely at a faster rate than in recent history. Globally, under the higher emissions scenarios, average sea level is estimated to rise by 3 to 4 feet (USGCRP, 2009). How much land will become submerged will vary regionally, depend- ing on the regional tectonics and geomorphology (land masses can be in the process  

Vulnerabilities and Impacts FIguRE 2.4 Projected changes in light, moderate, and heavy precipitation from the 1990s average to the 2090s average in North America. As shown here, the lightest precipitation is projected to decrease, while the heaviest will increase, continuing the observed trend. The higher emissions scenario yields larger changes. Projections are based on the models used in the IPCC (2007) Synthesis Report. NOTE: “Lower emissions scenario” refers to IPCC SRES B1, “higher emissions scenario” refers to A2, and “even higher emis- sions scenario” refers to A1FI. SOURCE: USGCRP (2009) (http://www.globalchange.gov). of rising or sinking relative to sea level) and ocean currents (which can cause the ocean surface to rise or sink relative to the average global sea level). DETERMININg vuLNERAbILITIES TO PROJECTED CLIMATE CHANgES As defined earlier, vulnerability is a function of the character, magnitude, and rate of climate change to which a system is exposed, as well as the system’s sensitivity and its adaptive capacity. Therefore, vulnerability can be assessed through the examination of these three factors. Assessing exposure to climate change reveals regional differences in the climate-related impacts that the United States will experience. Table 2.1 summa- rizes climate-related exposures and the regions that will most likely be affected. Vulnerability can also be examined through the sensitivity and adaptive capacities of  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E FIguRE 2.5 Projected frequency of extreme heat (2080-2099 average). SOURCE: USGCRP (2009) (http:// www.globalchange.gov). a particular community, system (i.e., economic, ecosystem, etc.), or sector. Vulnerability encompasses the risk and protective factors that ultimately determine whether a sub- population experiences adverse outcomes due to climate change (Balbus and Malina, 2009). For example, Table 2.2 summarizes various subpopulations that are particularly vulnerable to multiple climate-related exposures to health risks. Vulnerabilities of human systems are shaped by a wide variety of nonclimatic condi- tions. Although important, limited access to financial resources is not the only source of vulnerability. Examples of other sources include population shifts and development choices, such as dense urban development in drought-prone areas; and places and  

Vulnerabilities and Impacts FIguRE 2.6 Projected sea surface temperature change. SOURCE: USGCRP (2009) (http://www.global­ change.gov). communities that are especially dependent on climate-sensitive industries such as agriculture, forestry, and tourism. In the built environment, each city’s residents and infrastructures will be affected in unique ways (USGCRP, 2009). The vulnerability of natural systems, on the other hand, depends primarily on an ecosystem’s resilience to change. Changes in ecosystem function, in turn, affect human communities that depend on natural ecosystems to maintain clean water supplies, soil fertility, and other vital services (USGCRP, 2009).  

 TAbLE 2.1 Summary of regional climate-related impacts Climate-Related Impacts Extreme Urban Rainfall Sea Early Degraded Air Heat Heat Tropical with Level United States Census Regions Snowmelt Quality Island Wildfires Waves Drought Storms Flooding Rise New England • • • • • • • ME VT NH MA RI CT Middle Atlantic • • • • • • • • NY PA NJ DE MD East North Central • • • • • • WI MI IL IN OH West North Central • • • • • ND MN SD IA NE KS MO South Atlantic • • • • • • • • WV VA NC SC GA FL DC East South Central • • • • • • KY TN MS AL West South Central • • • • • • • • TX OK AR LA Mountain • • • • • • • MT ID WY NV UT CO AZ NM Pacific • • • • • • • • • AK CA WA OR HI SOURCE: Adapted from CCSP (2008f ). 

Vulnerabilities and Impacts TAbLE 2.2 Summary of vulnerability to climate-sensitive health outcomes by subpopulation Groups with Increased Vulnerability Climate-Related Exposures Infants and children Heat stress, ozone air pollution, water- and food-borne illnesses, Lyme disease, dengue Pregnant women Heat stress, extreme weather events, water- and food-borne illnesses Elderly/Chronic medical conditions Heat stress, air pollution, extreme weather events, water- and food-borne illnesses, dengue Impoverished/Low socioeconomic Heat stress, extreme weather events, air pollution, vector-borne status infectious diseases Outdoor workers Heat stress, ozone air pollution, Lyme disease, other vector- borne infectious diseases HOW CHANgINg CLIMATE CONDITIONS AND vuLNERAbILITIES IMPACT DIFFERENT u.S. SECTORS Climate Change Will Interact with Many Social and Environmental Stresses Society, its infrastructure, and its policies were developed in a relatively stable climate. Although climate change will create advantages for some locations and populations, on average, climate change is expected to adversely affect water resources, ecosys- tems, human health, energy, transportation, and other sectors. The expectation of adverse impacts stems in part from the fact that these systems were designed during a period of relatively stable climate conditions and in part from the accelerating rate of change, which presents a novel challenge for adaptation. Recent events suggest that changes in extreme weather events, including heat waves, floods, droughts, wind- storms, and wildfires, will likely be particularly challenging for communities and sec- tors to adapt to. The combination of climate change and trends in population growth also poses serious adaptation challenges. For example, population growth in the past century has been greatest in the South, along the coast, and in larger cities; this trend aligns somewhat with places where the threats of future heat waves and severe storms are greatest (USGCRP, 2009). With most of the U.S. population residing in urban areas, vulnerabilities associated with aging urban infrastructure, traffic congestion, air quality, social inequities, and other variables exacerbate the challenges of adapting to climate change. Key conclusions about how climate change will interact with social  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E and environmental vulnerabilities are summarized by sector in the following sections that were derived from the SAPs and the USGCRP (2009). Climate Change Will Place Additional burdens on Already Stressed Water Resources Climate change has already altered and will continue to alter the water cycle, affecting where, when, and how much water is available for various uses. Rising temperatures, for example, interact with other components of the climate system to alter patterns of precipitation. More intense droughts and flooding events are projected to become common in some regions. Increased droughts will have direct impacts on water re- sources, agriculture, and ecosystems, as well as leading to an increased risk of wildfires. Changes in precipitation will also alter runoff patterns (USGCRP, 2009) (Figure 2.7). FIguRE 2.7 Projected changes in annual average runoff for 2041-2060 relative to a 1901-1970 baseline by water resource region, based on analyses using emissions that fall between the lower and higher emissions scenarios. Lower average runoff is expected in the Southwest and greater runoff is projected for the Northeast. Colors indicate percentage changes in runoff, with hatched areas indicating greater confidence due to strong agreement among model projections. SOURCE: USGCRP (2009) (http://www. globalchange.gov). 0 

Vulnerabilities and Impacts A greater challenge for much of the western water sector will be a decrease in total snowpack and an altered timing of seasonal flow in snowmelt-dominated river basins. Warmer temperatures already have increased the proportion of precipitation that falls as rain rather than snow in the West, resulting in less snowpack accumulation and earlier snowmelt. In the latter half of the 20th century, peak flows in western streams arrived 1 to 2 weeks earlier. By the late 21st century under the higher emissions sce- nario, peak flows are projected to arrive 2 to 5 weeks earlier than in 1951-1980, leading to lower summer stream flows, generally less water availability, and changes in surface and groundwater quantities (USGCRP, 2009). These changes in precipitation, evaporation, and snowpack will further stress water resource allocations in regions such as the West and Southwest. For example, the Colorado River already has insufficient flow to support demand. Even under the lower emissions scenario, large areas of the Southwest are projected to receive 15 to 25 percent less spring precipitation by the end of this century. Under the higher emis- sions scenario, widespread decreases of 30 percent and more are projected. Greater conflicts over water resource allocations between agriculture, urban areas, and natural ecosystems are likely in many areas. Table 2.3 illustrates the varied impacts possible with changing water resources. As discussed in later chapters, current resource management plans are based on his- torical climatic averages (e.g., stream flow, reservoir size, runoff ) that will not continue TAbLE 2.3 Highlights of water-related impacts by sector Sector Examples of Impacts Human health Heavy downpours increase incidence of waterborne diseases and floods, resulting in potential hazards to human life and health. Energy supply and use Hydropower production is reduced due to low flows in some regions. Power generation is reduced in fossil fuel and nuclear plants due to increased water temperatures and reduced cooling water availability. Transportation Floods and droughts disrupt transportation. Heavy downpours affect harbor infrastructure and inland waterways. Declining Great Lakes levels reduce freight capacity. Agriculture and forests Intense precipitation can delay spring planting and damage crops. Earlier spring snowmelt leads to increased extent of forest fires. Ecosystems Cold-water fish are threatened by rising water temperatures. Some warm-water fish will expand ranges. SOURCE: USGCRP (2009) (http://www.globalchange.gov).  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E under future climate conditions (Milly et al., 2008). In addition, societal vulnerability to future water stress and related conflicts is increased by some current laws and prac- tices such as those governing interstate water allocation and reservoir operations, and by the relatively low price that is paid for water in most regions of the United States. Climate Change Can Lead to Large Ecosystem Changes When Impact Thresholds Are Crossed Changing climate conditions will change the distribution and migration patterns of plant and animal species, species productivity and abundance, and species interac- tions and habitat utilization. For example, wildlife corridors for migrations may shift. Species are already moving pole-ward and to higher elevations to remain within their optimal temperature ranges and, in the process, are invading new habitats. Predator- prey relationships are likewise being altered. Some changes will become irreversible as they cross certain threshold levels (i.e., species extinction; see Box 2.1) (USGCRP, 2009). bOx 2.1 The value of biodiversity Humans depend on biodiversity—the array of plants, animals, fungi, and microorganisms that constitute the fabric of the living world. All of our food comes directly or indirectly from plants and animals. Most people in the world depend on living organisms for their medicines; for those who obtain their drugs from pharmacies, roughly half are based on molecules found first in living organisms. Building materials, fossil fuels, chemical feedstocks, and future products and cures yet to be discovered—all of these are or will be derived from living organisms (Diaz et al., 2006). The communities and ecosystems that these organisms constitute support our lives through a variety of ecosystem services, activities that collectively determine the qualities of the atmosphere, regulate local climates, conserve topsoil, regulate runoff, provide pollination services for cultivated and wild plants, and contribute to the beauty and healthfulness of our lives (Daily, 1997). Combined with human population growth and land use change, climate change is a direct threat to the diversity of plant and animal species in many parts of the world, forcing already stressed species to respond to changes in climatic conditions that exceed the rate of change experienced in the past. The value of biodiversity has been recognized by policy actions such as passage of the Endangered Species Act and creation of national parks and biosphere preserves. Climate change could make it difficult to preserve valued landscapes and many of the species that make them special.  

Vulnerabilities and Impacts Of the global ecosystem services assessed by the Millennium Ecosystem Assessment (MEA), 60 percent were already on the decline due to human-driven stresses on natu- ral systems (MEA, 2005). Because of such stresses—including exploitation, contamina- tion, and habitat fragmentation—many ecosystems on land and sea are thought to be less resilient to the additional stress of a changing climate (USGCRP, 2009). Thus, climatic changes might result in further declines in provisioning services (e.g., food or timber production), regulating services (e.g., shoreline protection from storms pro- vided by wetlands), supporting services (e.g., water filtration and contaminant re- moval), and cultural services (e.g., recreation or sacred places). Loss or changes in these ecosystem services would negatively affect human well-being. Many terrestrial and marine ecosystems are changing in character and may look fundamentally different in the future, with unknown consequences. Some ecosystems, such as shallow-water coral reefs, are at risk of disappearing completely in coming decades; other ecosystems are simply changing in unpredictable ways. Significant effects of climate change have been observed in ecosystem processes, such as those that control plant growth and decomposition. These and other changes could cause large-scale shifts in the ranges of species, the timing of the seasons, and animal migra- tion (NRC, 2008a). Because species display great variation in their sensitivity to climate change, mobility, and lifespan, not all members of an ecological community will shift uniformly in response to climate change (USGCRP, 2009). The combination of climate change with other environmental stressors such as human resource exploitation (e.g., fishing or timber harvest) and barriers to migration (e.g., roads, residential develop- ments, and other built environments) will increase the risk of species extinctions. Changes in ecosystems are very likely to continue throughout the century. For exam- ple, Figure 2.8 shows current and projected shifts in forest types, with major changes projected for many regions. In the Northeast, under a midrange warming scenario, the currently dominant maple-beech-birch forest type is projected to be completely displaced by other forest types. Wildfires, outbreaks of insect pests and disease pathogens, and spread of invasive weed species have already increased, with climate implicated in some of these changes. These trends are likely to continue. In the western United States, the fre- quency of large wildfires and the length of the fire season increased substantially in recent decades, due primarily to earlier spring snowmelt and higher spring and sum- mer temperatures (although annual area burned may have been just as high in the 1920s). Insect pests, coupled with pathogens, annually cause an estimated $1.5 billion in damage in the United States. Changes in climate contributed significantly to several  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E FIguRE 2.8 Current and projected shifts in forest types. SOURCE: USGCRP (2009) (http://www.global­ change.gov). major insect pest outbreaks in the United States and Canada over the past several decades, particularly the mountain pine beetle. Deserts and drylands are likely to become hotter and drier, feeding a self-reinforcing cycle of invasive plants, fire, and erosion. The arid Southwest is projected to become even drier in this century, and emerging evidence suggests that this process is already under way. Deserts are also projected to expand to the north, east, and upward in elevation in response to projected warming and associated changes in climate. Arctic marine ecosystems are being severely affected by the loss of summer sea ice, and further changes are expected. The ice currently provides a vital platform for ice- dependent seals (such as the ringed seal), polar bears, and walruses to hunt and rest. It is uncertain how these animals could adapt to significantly less sea ice. As in other sectors, current policies to manage the impacts of human activities on the natural environment—including fisheries, wildlife, and forest management; pollu- tion control; and habitat protection—were based on the assumption that the Earth’s climate was relatively stable. The current rate of climate change suggests that man-  

Vulnerabilities and Impacts agement systems are likely to require substantial revision to maintain current levels of effectiveness in a warmer world (West et al., 2009). Coastal Areas Are at Increasing Risk from Sea Level Rise and Storm Surges The combination of sea level rise and storm surges poses a threat to coastal cities and ecosystems, especially areas that already experience multiple other stressors such as urban growth, human-induced changes in sediment loading and land subsidence, and high nutrient runoff. Uncertainties as to the exact extent of sea level rise in a particular location mean that local, state, and national agencies and organizations involved in coastal zone planning and management need to prepare for a range of possibilities. Coastal counties are among the most densely populated areas in the United States— more than a third of all Americans live near the coast, and activities along or on the ocean contribute more than $1 trillion to the nation’s economy. This intense devel- opment of coastal areas has increased their vulnerability to sea level rise and storm surges by decreasing the extent of natural buffers and causing accelerating rates of subsidence. For example, coastal Louisiana has already lost 1,900 square miles of wetlands in recent decades due to sea level rise and human alterations, weakening its capacity to absorb storm surges from hurricanes. Shoreline retreat has been observed along most U.S. exposed shores. Projected sea level rise could inundate portions of major cities such as Miami or New York during storm surges or even extreme high tides. Sea level rise can also lead to saltwater intrusion of freshwater aquifers in coastal areas that could reduce freshwater supplies (USGCRP, 2009). Projected changes in the timing of spring runoff and associated high nitrogen load- ing from agriculture will combine with increased sea surface temperatures to further reduce available oxygen in coastal waters. An additional threat to marine ecosystem health is ocean acidification (decrease in ocean pH), which is caused directly by ris- ing atmospheric CO2. Already heat-stressed corals will be further damaged by ocean acidification, with impacts that might reverberate across the reef food web (see NRC, 2010d). In addition, island states, territories, and protectorates vulnerable to sea level rise face additional threats due to the potential loss of coral reefs that serve as natural buffers from storm surge. Crop and Livestock Production Will be Increasingly Challenged Agriculture is considered one of the sectors most adaptable to climate change. In the United States, agricultural products contribute more than $200 billion in food com-  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E modities, with livestock accounting for more than half. Plants and animals display a broad range of vulnerability to increased temperature, resulting in diverse impacts. Global market pressures will also determine the ability of the agricultural sector to adapt to climate change. At average global temperature increases of less than 5.4°F (3°C), some agricultural systems will benefit and others will be adversely affected. At higher levels of warming, crop and livestock production is projected to decline in all regions due to increased heat, pests and pathogens, water stress, and weather extremes (Rosenzweig et al., 2008). Many crops grow better at higher atmospheric CO2 levels, but protein and nitrogen content often decline, resulting in less nutritious crops. Higher CO2 levels also improve water-use efficiency in some plants, which could benefit agriculture in water-stressed areas. However, heat-related stresses will increase water demand and require adjustments in agriculture practices. In addition, livestock productivity rates are projected to decrease because many warm-blooded species, including milk cows, are susceptible to heat stress, and the quality of pasture and rangeland forage will decrease with higher CO2 levels. Threats to Human Health Will Increase Climate change directly affects human physical and mental health through changes in the frequency, intensity, and/or duration of extreme weather events. While the fre- quency of extreme cold events will likely decrease, heat waves are increasing. Depend- ing on the extent and effectiveness of adaptation measures, heat-related illnesses and deaths could increase over coming decades as heat waves increase in frequency, intensity, and duration (Kovats and Hajat, 2008) (see Box 2.2). Heat waves are already one of the leading causes of weather-related mortality, as evidenced in Europe during the summer of 2003 when extreme heat was linked to more than 70,000 excess deaths (Robine et al., 2008) and the 1995 Chicago heat wave that caused 696 excess deaths ( Whitman et al., 1997; Semenza et al., 1999). Although data are limited to estimate the health impacts that may result from changes in extreme weather events other than heat waves, such events create poten- tially serious health consequences. For example, flooding not only causes direct inju- ries but, in some regions, also increases the risk of sewage overflows that contaminate drinking water. Urban population growth is expected to exacerbate the health risks associated with extreme events. Warmer air temperatures are associated with higher ozone levels, a known lung ir- ritant. Because half of the U.S. population is already living in counties where air pollu-  

Vulnerabilities and Impacts bOx 2.2 urban Heat Waves Throughout much of the Midwest, projections for 2090 suggest increases in nighttime tem- peratures (relative to 1975) of more than 3.6°F (2°C) during the worst heat waves (Ebi and Meehl, 2007). Illnesses caused by exposure to high temperatures include heat cramps, heat-induced fainting, heat exhaustion, heatstroke, and death (Kilbourne, 1997). Heatstroke has a high fatality rate, and even nonfatal heatstroke can lead to long-term illness (Dematte et al., 1998). Although the risk of heat illness exists for the entire population, a number of factors increase the risk: older and younger ages; use of certain drugs; dehydration; low level of fitness; excessive exertion; over- weight; lower socioeconomic status; and living alone. A heat wave of the same magnitude as the 2003 European heat wave in a large American city is projected to increase excess heat-related deaths by more than five times the average (Kalkstein et al., 2008). New York City’s total projected excess deaths would exceed the current national sum- mer average for heat-related mortality, with the death rate approaching annual mortality rates for common causes of death such as accidents. The extent to which death rates would actually increase during an event will depend on adaptation, including the population’s acclimatization to higher temperatures, modifications to the urban environment that reduce urban heat island effects, implementation of heat wave early warning systems, greater access to air conditioning, better education about response options, and other measures. tion exceeds national health standards, further deterioration in air quality is a concern. There is growing evidence that ground-level ozone concentrations would be more likely to increase than decrease in the United States as a result of climate change, if one assumes that emissions of ozone precursors remain constant (Bell et al., 2007). An increase in ozone could cause or exacerbate heart and lung diseases. Warmer temper- atures and higher CO2 levels also are likely to increase pollen production and lengthen the pollen season for some plants, potentially affecting allergies and respiratory health (Beggs, 2004; Kinney, 2008; USGCRP, 2009). The number of cases of climate-sensitive food- and water-borne diseases (i.e., salmo- nella) may increase among susceptible populations (USGCRP, 2009). The very young and old, the poor, those with health problems and disabilities, and certain occupa- tional groups are at greater risk. Vector-borne diseases (i.e., Lyme disease and others) may shift their geographic ranges, although climate will seldom be the only factor (USGCRP, 2009).  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E Energy and Transportation Will be Affected Climate change impacts on the energy industry are likely to be most apparent at subnational scales, such as regional effects of extreme weather events, reduced water availability leading to constraints on energy production, and sea level rise affecting energy production and delivery systems. Warming will be accompanied by decreases in demand for heating energy and increases in demand for cooling energy. This is pro- jected to drive up overall electricity use and create higher peak demands in most re- gions, but it also may reduce the use of heating oil and natural gas in winter. Although the energy industry will be affected in multiple ways by changing weather patterns, it is generally considered to have the financial and the managerial resources to adapt (see also the discussion of impacts of climate change policies below). Sea level rise and storm surges will increase the risk of major coastal impacts on vul- nerable energy and industrial infrastructure and transportation (CCSP, 2007, 2008b), including temporary or permanent flooding of airports, roads, rail lines, and tunnels. More frequent extreme precipitation events would increase the risk of disruptions and delays in air, rail, and road transportation, as well as damage from mudslides in some areas. Increases in the intensity of strong hurricanes would lead to more evacuations, infrastructure damage and failure, and transportation interruptions (NRC, 2008c). Arc- tic warming will lengthen the marine transport season, while permafrost thaw on land will damage infrastructure and reduce the ice road season. As experienced in Melbourne, Australia during a 2009 heat wave, an increase in ex- treme heat can limit some transportation operations (including airports) and cause pavement and track damage when heat compromises construction materials (NRC, 2008c). On the other hand, decreases in extreme cold can provide benefits such as reduced snow and ice removal costs and reduced snow-related road closures, as well as a potential decrease in snow- and ice-related traffic fatalities. Locations, Systems, and Populations Will be Affected by Climate Change Responses As Well As by Climate Change Itself Impacts of climate change that may require adaptation by human and natural systems will not be limited to the direct effects of changes in temperature, precipitation, storm behavior, and sea level. Climate change is also likely to create indirect effects through the impacts of climate change policies. Examples include the effects that limits on GHG emissions may have on energy prices, technology choices, and both regional and institutional comparative advantages. In some cases, such indirect effects present a  

Vulnerabilities and Impacts greater potential concern than climate change per se, and in many cases the literature on such indirect effects is more substantial than that on direct effects (CCSP, 2007). Examples of impacts of climate change policies that might require an adaptive re- sponse include the following: • If climate change policies emphasize reductions in GHG emissions, as ex- pected, then regional economies dependent on fossil fuel production and use (especially coal) are likely to need to transition to different economic bases or rely on new technologies. Studies that examined the effects of policies that limit the impact of climate change on U.S. regions (e.g., Oladosu and Rose, 2007) have generally shown that aggregate economic impacts on coal-pro- ducing regions would be negative but could be quite modest, depending on how policies are implemented and how the regions respond. • If climate change policies tend to favor land-intensive renewable energy alter- natives, especially energy from biomass, then land areas devoted to natural re- source preservation, forestry, agriculture, and ranching may have new oppor- tunities for income generation. Some stresses could develop from competition between food and energy crops for land area, between resource preservation and biofuel production, and between energy crops and other uses of scarce resources such as water. • Proposed climate change policies may raise energy prices as relatively inex- pensive fossil energy sources are replaced by lower-emitting but more ex- pensive alternatives. Analyses of the amount of the increase vary according to assumptions about such issues as ancillary benefits of a switch to alternative fuels and effects of policies designed to stimulate technological change. But some net costs to consumers are likely, which would affect relatively energy- intensive aspects of economies and societies, including costs of both transpor- tation and electricity supplies. • Climate change policies that alter the nation’s portfolio of energy supply and use technologies will inevitably create economic winners and losers, although very little research has been conducted on this topic. Most likely to be affected are industries related to fossil fuels and the structure of the electric utility in- dustry (Richels and Blanford, 2008). There are already signs that the reduction of GHG emissions is a factor in competition and economic health within the automobile industry (Levy and Rothenberg, 2002; Vance and Mehlin, 2009) and this could be a harbinger of impacts in other sectors as well. Such effects are, in fact, only one aspect of complex interactions between adaptation and mitigation (Wilbanks and Sathaye, 2007). Finding ways to integrate mitigation and  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E adaptation at the national level has been difficult; but at a local level, most decision makers and stakeholders contemplating climate change actions find it unwarranted to consider one strategy apart from the other (see also Chapter 5). A key issue is how mitigation and adaptation actions relate to each other. Some options offer comple- mentarities and synergies, while some work at cross-purposes with each other. For example, increasing the efficiency and affordability of space cooling helps to extend the benefits of cooling to a wider range of the residents of warming settlements; at the same time, it also reduces requirements for electricity generation to enable those services. On the other hand, choices between growing biomass for energy produc- tion and growing biomass as a GHG sink, both of which can be mitigation strategies, relate in different ways to adaptation strategies. For instance, bioenergy production can add to challenges in adapting to water scarcity in some regions. In addition, grow- ing biomass intended for long-term carbon storage can be complicated by climate change impacts on regional ecological systems, along with associated adaptive land use strategies. Other possible impacts of climate change policies—not all of them negative—include effects on choices of energy production and use technologies, on environmental emis- sions, and on international energy technology and service markets (for additional de- tails see ACC: Limiting the Magnitude of Future Climate Change; NRC, 2010c). Yet another issue is possible side effects of “geoengineering” options, should they be implemented. In general, geoengineering options intended to reduce the amount of solar radiation reaching the Earth—such as by creating a sulfate cloud in the atmosphere—would be virtually certain to affect vegetation growth and rainfall regimes, although the mag- nitude and geographic distribution of the potential effects are not well understood. Options intended to reduce current levels of CO2 in the atmosphere would require extensive carbon storage in places such as underground geologic formations, a possi- bility that presents a different range of impact and adaptation concerns. In either case, both known and unintended impacts could require adaptations in response. COMPARATIvE METRICS OF IMPACTS AND vuLNERAbILITIES In order to determine tradeoffs between various climate change policies and actions, scientists have attempted to find objective measures for dangerous climate interfer- ence (as prescribed in the United Nations Framework Convention for Climate Change 1992) that might push the system beyond its adaptive capacity. To date, such an ob- jective characterization of “dangerous” climate interference has not been developed. Nevertheless, a framework to consider global key vulnerabilities was developed for the Intergovernmental Panel on Climate Change (IPCC) Third Assessment Report (Smith 0 

Vulnerabilities and Impacts Reasons For Concern for the United States 5 Large Risks to Net Negative High Large Increase in Global Mean Temperature above circa 1990 (˚C) Increase Many Negative for Most Increase in All Regions Metrics 4 3 Future 2 Negative 1 for Some Positive or Regions; Negative Positive Market Risks to for Impacts Low Increase Increase Some Others 0 Past -0.6 Risks to Aggregate Distribution Risks of Large National Risk of Unique Impacts of Impacts Scale Security Extreme and Discontinuities Concerns Weather Threatened Events Systems FIguRE 2.9 Risks from climate change for the United States. Climate change consequences for the United States are plotted against increases in global mean temperature (°C) after 1990. Each column represents country-specific outcomes associated with increasing global mean temperature for each of the six reasons for concern. The color scheme represents progressively increasing levels of risk: white in- Figure 2-9 dicates neutral or small negative or positive impacts or risks, yellow indicates negative impacts for some systems or low risks, and red means negative impacts or n that are more widespread and/or greater in vector versio risks magnitude. Orange indicates a range of transition from risks calibrated in the modest risks of yellow and replaced from original source those calibrated in more severe and/or widespread risks of red. SOURCE: Yohe (2010); for details related to the assumptions in this figure see Appendix D. et al., 2001; IPCC, 2001b; and updated in Smith et al., 2009a). Following Yohe (2010), the panel responded to Woolsey (2009), Peters (2009), and Burke et al. (2009) by adding a sixth “reason for concern” related to the national security interests of the United States. To be precise, the aggregate metrics, as applied to the United States in Figure 2.9, include the following: 1. Risk of extreme weather events. The likelihood of extreme events with sub- stantial consequences for societies and natural systems such as increases in frequency or intensity of heat waves, floods, droughts, wildfires, or tropical cyclones, etc. 2. Risk to unique and threatened systems. The likelihood of imposing increased damage or irreparable loss to unique and threatened systems such as coral  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E reefs, tropical glaciers, endangered species, unique ecosystems, biodiversity hotspots, indigenous communities, etc. 3. Aggregate impacts. The likelihood of recognizing damages in aggregate. There are impacts distributed across the economy that can be aggregated into a single metric. Again, this reason for concern traditionally reported aggregate economic damages reports as, for example, the social cost of carbon. 4. Distribution of impacts. The likelihood of disparities of impacts (positive or negative) across regions or sectors. Some regions, sectors, or communities could face harm from climate change while others could even benefit. While this reason for concern historically focused primarily on economic metrics, recent work has aggregated subnational alternative metrics. 5. Risks of large­scale discontinuities. The likelihood of certain “threshold” phe- nomena that may have very large impacts. Examples include partial or com- plete deglaciation of the West Antarctic or Greenland ice sheets (that could lead to rapid increases in sea level), substantial reduction in the strength of the North Atlantic Meridional Overturning Circulation (that could result in a relatively rapid change in the climate system due to redistribution of heat in the oceans), and a “runaway greenhouse effect” (featuring more rapid warm- ing) driven by methane emissions from melting permafrost. 6. National security concerns. The likelihood that growing attention to climate change risks and vulnerabilities that will occur beyond national borders will nonetheless require response by defense and other mission agencies within the U.S. government (for additional details, see Chapter 6). It should be emphasized that this figure calibrates risks to increases in global mean temperature. It follows that the depicted transitions from low to high levels of concern do not necessarily reflect how risks might change at different rates of warming, nor do they necessarily indicate when impacts might be realized and how vulnerabilities might be influenced by alternative development pathways and the exercise of adap- tive capacity. When applied to the globe and here to the United States, the underlying “reasons for concern” framework nonetheless continues to be a viable mechanism with which to describe key climate risks and thus help to identify priorities with regard to ongoing and future research initiatives and attractive foci for policy discussion and implementation (IPCC, 2001a, 2007a). In summary, two qualitative conclusions emerge from the expert judgments that are embodied in Figure 2.9. Both reflect the evolving changes in climate variability that will be driven by long-term climate change.  

Vulnerabilities and Impacts 1. If policy makers were provided with only aggregate economic metrics when they asked to be informed about the significance and timing of impacts, then they would miss many if not all of the other risks that are captured in five other equally appropriate reasons for concerns. Indeed, Figure 2.9 suggests that decision makers could, as a result, come to the erroneous conclusion that it may take quite some time for the country as a whole to experience the rami- fications of “dangerous anthropogenic interference” with the climate system that has attracted the attention of many other countries. Of course, distinct localities and regions within the United States will have to cope with a diverse set of climate-driven vulnerabilities, and decision makers who work in these arenas will have an incentive to consider more focused economic aggregates. Even in these cases, though, interpreting aggregate economic indicators can be difficult, especially if those indicators ignore economic damages that will occur beyond specified borders—damages that are certainly part of a full and complete characterization of the potential economic risk to the nation. 2. Conversely, dangerous anthropogenic interference in the climate system will likely be discovered at all levels as climate change alters the intensities, fre- quencies, and regional distributions of extreme weather events. It is in these areas where investing in adaptive capacity and exercising adaptation options at the local level play their most critical roles; and it is through these manifes- tations that diversity in the climate risks facing various geographic regions scattered across the country and various climate-sensitive sectors scattered throughout the economy supports bottom-up approaches to evaluating ad- aptation needs. The first conclusion is almost a corollary of the observation that aggregate economic estimates of damages too often ignore low-probability risks. Indeed, this deficiency is just one of a growing list of concerns about relying too heavily on monetary estimates—estimates that, for the most part, miss many nonmarket damages and nearly all consequences from social contingent consequences (see, e.g., Yohe, 2009a; Yohe and Tirpak, 2008). The second follows from the expectation that reasons for concern, by offering alternative but nonetheless aggregate metrics, communicate the diversity of those risks more effectively. Care needs to be taken in interpreting these conclusions. Authors of the various ver- sions of the reasons for concern have emphasized that they cannot be the sole basis of policy; too many assumptions and limits of knowledge are either buried or missing. It is important to understand, for example, that these six aggregate “reasons for concern” reflect adaptation only to the extent the capacity to respond is included in the under-  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E lying literature. It has long been understood that the capacity to adapt depends on development pathways that cannot be reflected in simple calibrations of changes in global mean temperature; it is now understood that the ability to exercise the capacity to adapt is very site specific. It follows that it was impossible, in this and other “embers” exercises, to include depictions of where existing or potential “coping ranges” might be exceeded by changes in global mean temperature. Reasons for concern are best viewed as suggestions of where one might discover vul- nerabilities and impacts that some or even most might consider “dangerous.” Superim- posed against ranges of temperature trajectories (as in Figure SPM-2 or SPM-3 in IPCC, 2001a), they might even suggest when such danger might begin to occur. It follows that reasons for concern, when properly applied, can help scientists and decision mak- ers identify areas where more detailed analyses of vulnerabilities, and the associated opportunities for effective adaptation, might be most productive in directing research and informing policy design and implementation. MAJOR SCIENTIFIC CHALLENgES IN ASSESSINg CLIMATE CHANgE IMPACTS AND vuLNERAbILITIES AND THEIR IMPLICATIONS FOR ADAPTATION At this early stage in analyzing adaptation needs and potentials, many scientific chal- lenges remain in assessing vulnerabilities and impacts associated with climate change (ACC: Advancing the Science of Climate Change; NRC, 2010b). Six of the most significant of these challenges include: 1. The level of scientific confidence in understanding and projecting climate change increases with spatial scale while the relevance and value of the projections for society declines. A branch of climate science called detection and attribution (D&A) seeks to under- stand the causes of observed changes in climate by comparing observed changes with those simulated by climate models under rising atmospheric GHG concentrations and against background climate variability in model simulations with no rising GHGs. For statistical reasons, D&A is most successful at large spatial scales. It was first used to identify human influence on globally averaged temperature over the 20th century. Difficulties remain in attributing temperature changes on smaller than continental scales and over time scales of less than 50 years (IPCC, 2007b). Although the level of scientific confidence in climate change projections decreases at smaller spatial and temporal scales, the societal value increases. For example, while there is limited value in using the global averaged temperature for planning adaptation, information such  

Vulnerabilities and Impacts as the projected ranges in possible changes in 100-year flood risk on a particular river at a given location can be very useful even if highly uncertain. 2. A finer­scale understanding of climate change risks and vulnerabilities is needed. Impacts and adaptation are often local issues because the actual climate change impacts experienced will result from interactions of a specific climatic exposure with a specific population, sector, or system sensitive to that exposure, as well as the ability of that population, sector, or system to avoid, prepare for, and effectively respond to the risk. Thus, the same climatic exposure can have different consequences in differ- ent locations, and even in the same location at different time periods. Improvements are needed in the ability to project climate change at local and regional scales and to increase understanding of risks and the ability to design efficient and effective responses. Significant scientific challenges also remain in our limited understand- ing of the social, environmental, economic, institutional, and other factors that could interact with climatic changes to create impacts in any given location. Although Hur- ricane Katrina cannot be attributed to climate change, it demonstrated how hazard predictions with high certainties might fail to elicit proactive, necessary adaptations, partially due to a lack of understanding of local vulnerabilities and partially due to the difficulties of incorporating the latest natural and social sciences knowledge into practice and pre-disaster planning (NRC, 2006). Social groups particularly vulnerable to extreme weather events include the elderly, pregnant women, children, people with chronic medical conditions, people with mobility and cognitive constraints, and the urban and rural poor—all groups that are disproportionately represented within low- income communities (Balbus and Malina, 2009). Understanding such vulnerabilities in advance, and implementing appropriate strategies to increase resilience, affects the magnitude and extent of climate impacts (Kates et al., 2006; NRC, 2006). Projected increases in the frequency and intensity of extreme weather events high- light the need to increase understanding of those most at risk, both today and in future societies with possibly different risk profiles. However, research on vulnerability and impacts (and associated sustainability indices) has frequently been at scales too large to incorporate social justice issues. One counterexample is heat waves, where re- search at finer scales has shown that poorer areas of cities often have higher tempera- tures because of fewer green spaces, and that the residents of these areas may have less access to air conditioning or may not open windows during heat waves for fear of crime, thus increasing vulnerability. 3. Multiple stresses will interact with the impacts of climate change, leading to differ­ ent vulnerabilities to the same climate condition in different locations and a need for different adaptive responses.  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E Actual impacts will depend not only on climate but also on changes in other stresses over the same period. For example, impacts of climate change on vulnerable coast- lines in 2070 will be shaped not only by changes in sea level, storm tracks, and storm intensities but also by land subsidence, changes in population size and distribution, economic activities and wealth, technology, and institutional structures. Understand- ing how interactions with these other factors accentuate or ameliorate climate change impacts is important for adaptation planning. For example, detailed projections of socioeconomic scenarios are often not available beyond several decades into the future (and if they are available, they are highly speculative), which limits our ability for integrated modeling to understand how interactions could play out over time. 4. Adapting to changes in averages versus changes in extremes results in a funda­ mental scientific and policy challenge. Projections of the impacts of climate change tend to focus on changes in average weather variables, particularly changes in average temperature. As important as these changes are likely to be—for example, how increasing average temperature affects the suitability of particular cereal crops for a given region—the actions required for adapting to averages can be different from the actions required for adapting to extreme events. Strategies to manage the risks of climate change need to address projected increases in the frequency and intensity of extreme weather events, as well as unexpected threshold events. Depending on the cost of adaptation options versus the cost of impacts they are designed to avert, it may be helpful to prepare for low- probability/high-consequence events. Science and engineering needs include reeval- uation of boundaries of flood plains, better flood maps, and redesign or retrofitting of hospitals and other critical infrastructure so that services would not be disrupted during an extreme weather event. Effective adaptation will thus require consideration of climate change risks along multiple dimensions: increasing resilience to warmer temperatures and average changes in the water cycle while at the same time increas- ing resilience to extreme weather events—and doing both while considering current and future changes in other driving forces. 5. Interactions and integration across regions and sectors cause considerable complexity and will lead to unanticipated consequences of both impacts and adaptations. Climate change impacts in one sector or region usually spread secondary and ter- tiary impacts elsewhere. For instance, reducing impacts of summer warming on the quality of life of urban populations is likely to call for more air conditioning in homes and places of work. This will impact the energy sector by adding peak demands for electricity production, which can in turn impact the water sector by requiring more  

Vulnerabilities and Impacts cooling water for thermal power plant operation. Likewise, agricultural adaptations in a region may call for increased use of irrigation, while water resources adaptation in that region may call for decreased water availability for irrigation. Impacts can cross regional boundaries as well. For instance, more intense storms in vulnerable areas can mean flows of evacuees to other regions, along with at least temporary shortages (or increases in the price) of products and services disrupted by the storms, as was the case with energy products after Hurricane Katrina (Bamberger and Kumins, 2005). Further research is needed to improve our understanding of how to effectively develop cross-regional and cross-sectoral adaptation plans. 6. The types of impacts, vulnerabilities, and adaptation options are different for natural and human systems. Both national and international assessments have described the broad patterns of recent and projected responses of natural systems and biodiversity to climate change (IPCC, 2007a; MEA, 2005; USGCRP, 2009). It is highly likely that most natural systems are sensitive to climate change. Much of our current understanding of ecosystem dynam- ics, however, is based on observations and models that assume less dramatic direc- tional changes in environment and ecosystem dynamics. These models and theories provide an important starting point for understanding rapid change, but they will undoubtedly require reassessment as new patterns of environmental and ecological controls emerge. These changes are likely to result in the loss of some ecosystems and the formation of novel ecosystems due to the loss of some species and arrival of others. Loss of biodi- versity is quite likely, including both loss of rare species and loss or reduced impor- tance of keystone species. During these times of rapid biological adjustment, meta- population dynamics (i.e., interactions among partially isolated subpopulations) and migration of species across increasingly fragmented and human-modified landscapes are likely to exert greater influence on ecosystem structure and functioning than in the past. All of these changes could reduce the resilience of natural ecosystems and make them more vulnerable to threshold changes. These and other broad changes in the ground rules by which ecosystems operate create significant scientific challenges in understanding and predicting the patterns and consequences of changes in natural ecosystems. The prospect of widespread ecological change also raises two pragmatic questions that represent additional research challenges: How can the rates of undesirable ecological change (e.g., loss of biodiversity) be minimized? And how can the flow of essential ecosystem services on which society depends be sustained in the face of  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E rapid ecological change? Both questions will require improved understanding of the dynamics of social-ecological systems and collaborations. ADAPTATION AND uNCERTAINTy Adapting to climate change impacts will require doing our best to understand the factors that drive both the impacts and our ability to respond. This reality has led to urgent calls for more information about the range of possible impacts and the level of certainty in our projections of the future. It is clear that society cannot avoid the risks of climate change entirely. One challenge for decision makers will be the limits to our ability to identify and reduce uncertainties related to climate change. Major uncertainties in determining future climate include the natural internal vari- ability of the climate system, the trajectories of future emissions of GHGs and aerosols, and the response of the global climate system to any given set of future emissions (see also Chapter 4 and Meehl et al., 2007; NRC, 2010c). The magnitude and sources of these uncertainties can be explored using global climate models. These models have become more sophisticated and accurate over time in replicating the historical record. However, it is unlikely that climate models will be able to predict the future on fine spatial scales with a high degree of accuracy on long time scales. At best, climate models can provide insights about the range of possible futures. Lack of certainty about future conditions is commonly, but often inappropriately, used as a rationale for inaction. In fact, improving our understanding of the kinds of uncer- tainties that we face will be helpful in risk-management decisions, even if the uncer- tainties cannot be readily quantified. For example, some uncertainties result from pro- cesses that are still missing from the climate models but are potentially resolvable in the future (e.g., changes in climate that result from changing land use and land cover). There are other uncertainties that are inherent in the complexity of the climate system itself, and it is unlikely that those kinds of uncertainty will be reduced significantly. For example, the uncertainty of the long-term trajectory of GHGs is very likely not to be reducible (CCSP, 2009c). Another source of uncertainty comes from the fact that current global climate mod- els operate at relatively coarse spatial scales (hundreds of kilometers or miles), and thus do not accurately represent conditions in specific places; rather, they represent average conditions across broad regions such as the entire Southwest.2 This prob- 2 However, it should be noted that the resolution of global models is increasing, and some of the simula- tions for the next IPCC report may be run at 50 km for the near-term future (next 20 years) .  

Vulnerabilities and Impacts lem of spatial scale is overcome through the application of various “downscaling” techniques—ways to generate information at higher spatial resolution from coarse- scale global model output. There are three primary methods of downscaling: (1) simple downscaling, where the coarse-resolution information is simply interpolated to higher resolution, or the coarse-scale changes in climate are used in the context of higher-resolution observed data; (2) statistical downscaling, which relies on statisti- cal relationships between historically observed large-scale climate variables and local climate (e.g., daily temperature in a specific city) that are then applied to the climate change context; and (3) dynamical downscaling techniques, such as regional mod- eling, where a higher-resolution climate model is applied to just part of the Earth’s surface (e.g., the western United States) and is “nested” in the global models. Regional climate models can better represent smaller-scale processes such as those related to complex terrain (e.g., mountains) and provide data at scales closer to those at which decisions are made (within a watershed, for example). Regional climate mod- els are useful in trying to understand the physical processes that control regional cli- mate and the likely impacts of climate change within regions and sectors for risk-man- agement planning. However, “downscaled” climate data can introduce other sources of uncertainty and are not yet the panacea that many resource managers hope they will be (Wang et al., 2004).3 Making adaptation decisions in the context of uncertainty will remain a challenge, but one that can be overcome with careful attention to improv- ing the understanding and characterization of—and the ability to communicate—the nature and sources of uncertainty. CONCLuSIONS The United States is already experiencing impacts of climate change that require adaptation. Some of these impacts are already testing, or soon will seriously test, the nation’s coping mechanisms. In summary, the panel finds that climate change impacts are certain to increase throughout this century, requiring significant effort to adapt in order to avoid socially, economically, and environmentally disruptive changes in systems with high value to society. Adaptation options need to address current and projected changes in mean weather variables as well as increases in the frequency and intensity of many extreme events. 3 Different regional climate models produce different responses to the boundary conditions from the global models, presenting another source of uncertainty. Also, high-resolution modeling must be considered in the context of the other uncertainties mentioned above. Nesting one regional model inside one global model, regardless of how high the resolution, will not provide important information about the larger-scale uncertainties, and can even be misleading and create a “false certainty.”  

A D A P T I N G T O T H E I M PA C T S O F C L I M AT E C H A N G E Impacts later this century will be notably greater if GHG emissions are not stabilized at a moderate level. If the magnitude of climate change is relatively severe, as depicted in the USGCRP higher projection, then regions, sectors, and systems will be hard pressed to cope with impacts and their costs. In addition, impacts of climate change are highly diverse and disaggregated, in many cases playing out at localized geographic, sectoral, and societal scales. As a result, effective approaches to adaptation will likely vary from case to case. In most cases, impacts are imbedded in interactions between climatic changes per se and other driving forces, such as changes in demographics, economics, land use, and technology, which also vary from case to case. Therefore, impacts and vulnerability are place-based and fundamentally driven by the scale at which the impact occurs. Many scientific challenges remain in assessing impacts and vulnerabilities and providing the specific and localized information needed to guide adaptation decisions. Conclusion: Many current and future climate change impacts require immediate actions to improve the ability of the nation to adapt. Because some impacts may not require immediate attention, possible adaptation options need to be prioritized based on where and when urgent action is needed. This highlights the need to identify vulnerabilities, impacts, and adaptation options across the nation, at all levels of decision making. Conclusion: Gaps in the knowledge required to link anticipated impacts with appropriate adaptation strategies and actions need to be addressed as a high national research priority. Conclusion: It is inadequate to provide policy makers with only aggregate economic metrics to convey the significance and timing of climate change impacts. Aggregated data miss most nonmarket damages and nearly all social contingent consequences that society might deem unacceptable, including those from outside our borders. Conclusion: Uncertainty about the nature of future climate change impacts in specific locations is not a rationale for inaction but a call for better understanding and communication of the sources and nature of uncertainty in the context of decision making. 0