Chapter 2 explored a number of potential abrupt climate changes from the point of view of examining the processes. In this chapter, the committee reframes the discussion to look at the issue of abrupt climate changes from the point of view of how they might affect human society. This chapter synthesizes the previous material into how it relates to food security, water security, ecosystem services, infrastructure, human health, and national security.
Abrupt climate impacts may have detrimental effects on ecological resources that are critical to human well-being. Such resources are called “ecosystem services” (Box 3.1), which basically are attributes of ecosystems that fulfill the needs of people. For example, healthy diverse ecosystems provide the essential services of moderating weather, regulating the water cycle and delivering clean water, protecting and keeping agricultural soils fertile, pollinating plants (including crops), providing food (particularly seafood), disposing of wastes, providing pharmaceuticals, controlling spread of pathogens, sequestering greenhouse gases from the atmosphere, and providing recreational opportunities (Cardinale et al., 2002; Daily and Ellison, 2002; Daily et al., 2000; Ehrlich et al., 2012; Tercek and Adams, 2013). As explained in Box 3.1, ecosystem services can generally be categorized as provisioning services, regulatory services, cultural services, and supporting services. Currently recognized trajectories of climate change have the potential to cause abrupt changes in each of these categories, in three different ways.
First, gradual changes in the climate system can result in crossing ecologically important threshold values in certain climatic parameters that suddenly cause species to disappear from an area (Chapter 2, Extinctions: Marine and Terrestrial). Examples include soils becoming too dry to support forests, corals dying back because water becomes too warm or acidic, or lizards becoming locally extinct because it is too hot for them to forage (Sinervo et al., 2010). A growing number of studies document that ecosystem transformations that result in loss of biodiversity—as can happen from extinction, or from ecological regime shifts that do not necessarily involve global
BOX 3.1 ECOSYSTEM SERVICES
Humans receive a wide variety of benefits from ecosystem resources and processes (Figure A). The term “ecosystem services” has been used to encapsulate these benefits. Although the notion of human dependence on the services that Earth’s ecosystems provide is not new, the definition and categorization of such services were formalized and popularized by the Millennium Ecosystem Assessment in 2005 (MEA, Reid et al., 2005).
The MEA divides ecosystem services into four categories: supporting services, provisioning services, regulating services, and cultural services (Reid et al., 2005):
- Provisioning Services are “products obtained from ecosystems,” including: food, fuel, freshwater, natural medicines, and pharmaceuticals.
- Regulating Services are “benefits obtained from the regulation of ecosystem processes,” including: regulation of climate, water, air quality, erosion, pests, and diseases.
- Cultural Services are the “nonmaterial benefits people obtain from ecosystems through spiritual enrichment, cognitive development, reflection, recreation, and aesthetic experiences,” including: intellectual and spiritual inspiration, ecotourism, and scientific discovery.
- Supporting services are ecosystem services required “for the production of all other ecosystem services.” These include soil formation, photosynthesis, and nutrient cycling.
FIGURE A Agriculture, specifically corn (A), and fresh seafood (B) are two examples of provisioning services. Regulating services include coastal wetlands, like those of Ashe Island, NC (C), which stabilize shorelines and help buffer against storm erosion, and pollinators such as bees (D), birds, and bats, which are depended upon for thirty-five percent of global crops (Klein et al., 2007). The Catskill Mountains (E) act as a water filtration “plant” for New York City—another example of a regulating service. SOURCE: (A) NSF, (B) The George F. Landegger Collection of District of Columbia Photographs in Carol M. Highsmith’s America, Library of Congress, Prints and Photographs Division, (C) NOAA, (D) USDA/Stephen Ausmus, (E) FWS.
BOX 3.2 DETRIMENTAL EFFECTS OF BIODIVERSITY LOSS IN GENERAL
An emerging body of literature is documenting that reducing biodiversity is detrimental to ecological function, and ultimately to ecosystem services that benefit humans, in at least six ways (Cardinale et al., 2012; Collen et al., 2012; WRI, 2005) (Box 2.3). The following list of the costs of losing biodiversity is extracted from the review by Cardinale et al. (2012), as based on their vetting of published studies.
- “Reductions in the number of genes, species and functional groups of organisms reduce the efficiency by which whole communities capture biologically essential resources (nutrients, water, light, prey), and convert those resources into biomass.” This implies that biodiversity-poor systems require more resources to be input by humans in order to maintain them, for example, addition of artificial fertilizers to maintain productivity.
- “There is mounting evidence that biodiversity increases the stability of ecosystem functions through time.” This means that biodiverse systems are more dependable in providing benefits to humanity, for example, carbon sequestration in forests, because they better withstand unanticipated perturbations.
- “The impact of biodiversity on any single ecosystem process is nonlinear and saturating, such that change accelerates as biodiversity loss increases.” Therefore, as biodiversity loss increases, ecological systems are increasingly likely to suddenly hit a “tipping point,” for example, economically important fisheries in marine coastal areas being replaced by systems dominated by algae and jellyfish.
- “Diverse communities are more productive because they contain key species that have a large influence on productivity.” This means that removing a single species can have unexpectedly large effects in transforming an entire ecosystem, such as removal of elephants changing a grassland savannah into a forest.
- “Loss of diversity across trophic levels has the potential to influence ecosystem functions even more strongly than diversity loss within trophic levels.” Loss of diversity is seldom restricted to the trophic level that experiences the loss because the connections between interacting species are altered. For example, loss of top predators has been shown to reduce plant biomass significantly.
- “Functional traits of organisms have large impacts on the magnitude of ecosystem functions, which give rise to a wide range of plausible impacts of extinction on ecosystem function.” This means that removal of species with unique ecological attributes is especially disruptive to ecosystems.
Quoted from Cardinale et al., 2012, pg. 59.
extinction but cause extirpation of species locally or regionally—detrimentally affect ecosystem services (Box 3.2). Second, extreme events (increased frequency of floods or drought, for instance) can trigger sudden, regional catastrophes that wipe out natural ecosystems (coastal salt marshes, for instance) or human-controlled ones (crop fields). Third, cascading effects of abrupt, climatically-triggered changes to the physical environment can cause abrupt transformation in widespread ecosystems, such as
loss of sea ice affecting the marine food chain, or loss of coral reefs impacting fisheries (discussed and cited above).
The spectrum of abrupt disruptions to ecosystem services as a result of climate change is broad, but of particular concern are those that would impact essential provisioning services such as food production and water availability. The potential disruptions of food (and water, see below) supplies could be one of the most serious manifestations of abrupt climate change, especially when put in the context of the changing global food system.
The world currently has a population exceeding 7 billion (and is likely to grow to ~9 billion by 2050) and an estimated ~850 million are already considered food insecure (Godfray et al., 2010). The challenges of food security today are driven mainly by poverty, the lack of access to food, and poor institutions (Godfray et al., 2010). Even a slight added impact from climate change, therefore, could lead to significant, abrupt, and problematic food shortages.
Tilman et al. (2011) have estimated that by 2050, world demand for agricultural products will increase by roughly 100 percent, largely driven by rising incomes and increasing meat consumption. This challenge is made more cogent by realizing that even without potential impacts of climate change, providing food security, meeting growing demands for agricultural products, and ensuring the environmental sustainability of agricultural systems worldwide will require a multi-faceted approach (Foley et al., 2011; Godfray et al., 2010; Tilman et al., 2011). Such an approach will need to concentrate on boosting yields (especially in places where yields are low today), improving the resource efficiency of agriculture (especially the water, nutrients, and energy used per calorie of food delivered), avoiding further deforestation and land degradation, shifting diets and biofuels to more sustainable trajectories, and reducing food waste across the entire supply chain (Foley et al., 2011).
To date, investigations of how climate change will affect crop production and food systems have mainly focused on long-term changes in the mean climate (e.g., annual rainfall, patterns of temperature). Mainly, these studies have split into two broad categories: those that do explicitly consider the adaptation of farmers to climate change, and those that do not.
One of the first studies to consider the impacts of climate change on agriculture was conducted by Rosenzweig and Parry (1994). Adaptations include changes in planting
date, variety, and crop, as well as changes in applications of irrigation and fertilizer. They found that climate change scenarios (for 2060) with adaptation could result in increasing yields (from +3 to +10 percent) in developed countries, while developing countries would see decreasing yields of approximately 6 percent. However, with no adaptation, developed countries would see production changes from –4 percent to +11 percent, and developing country yields would see decreases of 9 to 12 percent. Parry et al. (2004) presented an updated version of this work for a variety of climate change scenarios for 2020, 2050, and 2080, and made similar conclusions: yield increases in developed countries will tend to counteract decreases in developing countries. Typical increases in production for the developed world range from 3 to 8 percent, and typical decreases for the developing world range from 2 to 7 percent.
More recently, Deryng et al (2011) used the PEGASUS process-based crop model and found that global maize production for 2050, under a climate change scenario based on rapid economic growth (A1B; see IPCC, 2007), changes by –15 percent, and under a scenario based on more modest economic growth (B1) changes by -8 percent, if farmlevel adaptation (especially changing planting dates) is taken into account. However, without farm-level adaptation of planting dates, the yields decreases are estimated 30 percent and 20 percent respectively. A new study1 uses a cross-sectional method based on the shifting climate zones and find that global maize production in 2050 (under and A1B scenario) could decrease by 7 percent and under a B1 scenario will decrease by 3.5 percent.
In short, it is clear that changes in climate will have profound impacts on global food production and, in turn, food security (Easterling et al., 2012; Lobell and Gourdji, 2012; Lobell et al., 2011). Modeling and statistical analyses consistently show that climate change could introduce substantial changes to global food production (some positive, many strongly negative). However, the exact magnitude of these changes depends on the assumptions made about the adaptation of farmers to climate change (Easterling et al., 2012). This presents a particular challenge in the face of abrupt climate change (compared to slower changes in climate that may occur over many decades). Will abrupt changes in climate cause more severe dislocations in agriculture, because it leaves less time for farmers and agricultural markets to adapt? This remains a critical area for future research.
1 Personal communication, J.S. Gerber, 2013.
Humans currently withdraw roughly 4,000 km3 of water globally, mainly for irrigation (~70 percent), industry (~20 percent), and domestic use (~10 percent). Water consumption (the net use of water from a watershed, accounting for water return flows and recycling within the same watershed) globally is estimated to be 3,000 km3, with agricultural irrigation taking an even larger share (~90 percent) (World Water Council, 2000). Therefore, climate changes, mainly through changes in precipitation and evapotranspiration over watersheds that people depend upon, can cause serious, abrupt (yearly to decadal) changes to critical freshwater resources. Additional concerns about freshwater resources are linked to snow and ice melt, which provides critical drinking and irrigation water to many people worldwide, including highly populous and/ or politically sensitive areas such as Pakistan, India, and along the border of China and Nepal.
As with other ecosystem services, it is necessary to interpret potential climatic impacts on freshwater resources within the broader context of how water resources are already stressed around the world. Freshwater resources are already reaching limits under the increasing demands of a growing population, rising incomes, and increasing per capita consumption (particularly through food). Vorosmarty et al. (2000), for example, demonstrated that changes to the current patterns from water consumption and withdrawals already exceed the expected changes to the water cycle anticipated from climate change. Furthermore, increasing demands on water (estimated from population growth and economic development) will greatly exceed expected changes from climate change. Vorosmarty et al. summarize the situation as, “We conclude that impending global-scale changes in population and economic development over the next 25 years will dictate the future relation between water supply and demand to a much greater degree than will changes in mean climate.” (emphasis added)
Groundwater aquifers, for example, are being depleted in many parts of the world, including the southeast of the United States. Groundwater is critical for farmers to ride out droughts, and if that safety net reaches an abrupt end, the impact of droughts on the food supply will be even larger. Satellites measuring gravity now reveal that groundwater supplies have decreased rapidly around the world over the past decade, including key aquifers in California, the High Plains, and the southeastern United States (Famiglietti et al., 2011). Groundwater is a key part of successful adaptation to periodic drought, which in turn is a key aspect of maintaining stable food supplies. In many cases it is unknown how long this situation could continue without water availability reaching an end, possibly an abrupt end, although history is clear in showing that groundwater supplies can indeed be depleted, parts of the Ogallala Aquifer (un-
der the US Great Plains) being an example. Questions remain about the future of this potential abrupt change, but the potential impact, especially on national and global food supplies, is substantial.
On a larger scale, changes in atmospheric circulation (e.g., changes to monsoon circulations), precipitation variability (e.g., more high extreme rainfalls) and abrupt changes in the condition of snow and ice packs have high potential of reducing crop productivity in some areas, and raising it in others—such shifts will have downstream impacts on local and national economies. For example, largely due to water-delivery issues related to climate change, cereal crop production is expected to fall in areas that now have the highest population density and/or the most undernourished people, notably most of Africa and India (Dow and Downing, 2007). In the United States, key cropgrowing areas, such as California, which provides half of the fruits, nuts, and vegetables for the United States, will experience uneven effects across crops, requiring farmers to adapt rapidly to changing what they plant (Kahrl and Roland-Holst, 2012; Lobell et al., 2006).
Degradation of coral reefs by ocean warming and acidification will negatively affect fisheries, because reefs are required as habitat for many important food species, especially in poor parts of the world. For example, in the poorest countries of Africa and south Asia, fisheries largely associated with coral reefs provide more than half of the protein and mineral intake for more than 400 million people (Hughes et al., 2012). On a broader scale, many fisheries around the world can be expected to experience changes as ocean temperatures, acidity, and currents change (Allison et al., 2009; Jansen et al., 2012; Powell and Xu, 2012), with attendant socio-economic impacts (Pinsky and Fogarty, 2012). One study suggests climate change, combined with other pressures on fisheries, may result in a 30–60 percent reduction in fish production by 2050 in areas such as the eastern Indo-Pacific, and those areas fed by the northern Humboldt and the North Canary Currents (Blanchard et al., 2012). Because other pressures, notably over-fishing, already stress fisheries, a small climatic stressor can contribute strongly to hastening collapse (Hidalgo et al., 2012).
Other Provisioning Services
Outside the food and water sector, abrupt changes to other provisioning services also are very likely as a result of in-progress climate change (Reid et al., 2005, see Box 3.1).
Forest diebacks (Anderegg et al., 2013) and reduced tree biodiversity (Cardinale et al., 2012) can be expected to have major impacts on timber production. Such is already the case for millions of square miles of beetle-killed forests throughout the American West. Drought-enhanced desertification of dryland ecosystems may cause famines and migrations of environmental refugees (D’Odorico et al., 2013).
In several documented cases the efficacy of provisioning services correlates positively with the biodiversity of an ecosystem (Cardinale et al., 2012); thus, the loss of biodiversity through climate-caused or climate-exacerbated extinctions is of considerable concern. Among the provisioning services that have been shown to increase with biodiversity are: intraspecific genetic diversity increasing the yield of commercial crops; tree species diversity enhancing production of wood in plantations; plant species diversity in grasslands improving the production of fodder; higher diversity of fish leading to greater stability of fisheries yields; higher plant diversity increasing resistance to invasion by less-desirable exotic species, and in decreasing prevalence of fungal and viral infections (Cardinale et al., 2012). Some studies suggest that increased biodiversity also increases the following ecosystem services: carbon storage, pest reduction, reduction in animal diseases, fisheries yields, flood protection, and water quality. However, the efficacy of higher biodiversity in promoting these services still is under study, with conflicting results for different studies clouding the generality of the relationship (Cardinale et al., 2012).
Also of concern is the potential loss of regulatory services, which buffer the effects of environmental change (Reid et al., 2005). For example, tropical forest ecosystems slow the rate of global warming both by absorbing atmospheric carbon dioxide and through latent heat flux (Anderson-Teixeira et al., 2012). Coastal saltmarsh and mangrove wetlands buffer shorelines against storm surge and wave damage (Gedan et al., 2011). Grassland biodiversity stabilizes ecosystem productivity in response to climate variation (see Cardinale et al., 2012 and references therein). Climate change has the clear potential to exacerbate losses of these critical ecosystem services (for instance, decrease in rainforests, desertification) and attendant impacts on human societies.
Direct Economic Impacts
Some species currently at risk of extinction, and some of those which will be further imperiled by ongoing climate change, provide significant economic benefits to people
who live in the surrounding areas, as well as significant aesthetic and emotional benefits to millions of others, primarily through ecotourism, hunting, and fishing. At the international level, for example, ecotourism—largely to view elephants, lions, cheetahs, and other threatened species—supplies around 14 percent of Kenya’s GDP as of 2013 (USAID, 2013) and supplied 13 percent of Tanzania’s in 2001 (Honey, 2008). Yet in a single year, 2009, an extreme drought decimated the elephant population and populations of many other large animals in Amboseli Park, Kenya. Increased frequency of such extreme weather events could erode the ecotourism base on which the local economies depend. Other international examples include ecotourism in the Galapagos Islands—driven in a large part to view unique, threatened species—which contributed 68 percent of the 78 percent growth in GDP of the Galapagos that took place from 1999–2005 (Taylor et al., 2008).
Within the United States, direct economic benefits of ecosystem services also are substantial; for example, commercial fisheries provide approximately one million jobs and $32 billion in income nationally (NOAA, 2013). Ecotourism also generates substantial revenues and jobs in the United States—visitors to national parks added $31 billion to the national economy and supported more than 258,000 jobs in 2010 (Stynes, 2011). For Yellowstone National Park, which attracts a substantial number of visitors for wildlife viewing, visitors in 2010 contributed $334 million to the local economies, and created 4,900 local jobs (Stynes, 2011). Wildlife in Yellowstone is undergoing substantial changes, as evidenced by the clear amphibian decline as a result of drying up of breeding ponds (McMenamin et al., 2008). Visitors to Yosemite National Park in 2009 created 4,597 jobs, and yielded $408 million in sales revenues, $130 million in labor income, and $226 million in value added (Cook, 2011). Recent work there demonstrates that many of the small mammals are shifting their geographic ranges, with as yet unknown consequences to the overall ecosystem, as a result of climate change over the last century (Moritz et al., 2008).
The built environment is at risk from abrupt change. Examples near coasts include infrastructure such as roads, power lines, sewage treatment plants, and subway systems located close enough to the ocean and at a low enough elevation to be subject to the direct and indirect (e.g., storm surges) impacts of sea level rise. Other examples from northern latitudes include roads built on permafrost in Alaska, where that permafrost is now melting causing the roads to buckle and heave. Less obviously, there are also systems whose useful lifetimes are cut short by gradual changes in baseline climate. Such systems are experiencing abrupt impacts if they are built to last a certain period
of time, and priced such that they can be amortized over that lifetime, but their actual lifetime is artificially shortened by climate change. One example would be a large air conditioning system for computer server rooms. If maximum high temperatures rise faster than planned for, the lifetime of such systems would be cut short, and new systems would need to be installed at added cost to the owner of the servers. Another example is storm runoff drains in cities and towns. These systems are sized to handle large storms that precipitate a certain amount of water in a certain period of time. Rare storms, such as a 1000-year event, are typically not considered when choosing the size of pipes and drains, but the largest storms that occur annually up to once per decade or so are considered. As the atmosphere warms and can hold more moisture, the amount of rain per event is increasing (Westra et al., 2013), changing the baseline used to size storm runoff systems, and thus their utility, generally long before the systems are considered to have reached their useful lifetimes.
Another type of infrastructure problem associated with abrupt change is the infrastructure that does not exist, but will need to after an abrupt change. The most glaring example today is the lack of US infrastructure in the Arctic as the Arctic Ocean becomes more and more ice free in the summer. For example, the United States lacks sufficient ice breakers that can patrol waters that, while seasonally open in many places, will still have extensive wintertime ice cover. Servicing and protecting our activities in this resource-rich region is now a challenge, one that only recently, and abruptly, emerged. This challenge has illustrated a time scale issue associated with abrupt change. Currently, it will take years to rebuild our fleet of ice-breakers, but because of the rapid loss of sea ice in 2007 and more recently, the need for these ships is now (NRC, 2007; O’Rourke, 2013).
Globally, about 40 percent of the world’s population lives within 100 km of the world’s coasts. While complete inventories are lacking, the accompanying infrastructure—from the obvious, such as roads and buildings, to the less obvious but no less critical, such as underground services (e.g., natural gas and electric lines)—is easily valued in the trillions of dollars, and this does not include ecosystem services such as fresh water supplies, which are threatened as sea level rises. A nearly equal percentage of the US population lives in Coastal Shoreline Counties.2 In addition, coastal counties are more densely populated than inland ones. The National Coastal Population Report, Population Trends from 1970 to 2020 (NOAA, 2013), reports that coastal county population
FIGURE 3.1 The percentage of US population living in near the coast has been increasing over the past several decades. In 2010, 39 percent of the US population lived in Coastal Shoreline Counties (less than 10 percent of the total land area excluding Alaska). The population density of Coastal Shoreline Counties is over six times greater than the corresponding inland counties. Source: NOAA, 2013.
density is over six times that of inland counties (Figure 3.1). Consequently, the United States has a large amount of physical assets located near coasts and currently vulnerable to sea level rise and storm surges exacerbated by rising seas (See Chapter 3 and especially Box 2.1 for additional discussion of this issue.) For example, the National Flood Insurance Program (NFIP) currently has insured assets of $527 billion in the coastal floodplains of the United States, areas that are vulnerable to sea level rise and storm surges. Examples of significant payouts include the costs of Hurricane Katrina, which totaled $16 billion from NFIP, and significantly more than that for private insurers, and the recent costs of Superstorm Sandy (Figure 3.2), which are still being totaled, but which will likely exceed Katrina by a large amount. In addition, nearly half of the US gross domestic product, or GDP, was generated in the Coastal Shoreline Counties along the oceans and Great Lakes (see NOAA State of the Coast3). Despite the ongoing rise of sea level, and the frequent, high-profile illustrations of the value and vulnerabil-
FIGURE 3.2 During Superstorm Sandy in 2012, storm surges brought water inland and flooded subway terminals in the New York area—part of the billions of dollars in damages from that storm. SOURCE: Port Authority of New York and New Jersey.
ity of coastal assets at risk, there is no systematic, ongoing, and updated cataloging of coastal assets that are in harm’s way as sea level rises. Overall, there is a need to shift to more holistic planning, investment, and operation for global sea ports (Becker et al., 2013).
Arctic Transportation and Infrastructure
Some of the most apparent infrastructure impacts are in the Arctic, owing to both the rapidity of summer sea ice loss in the Arctic Ocean and the non-linear rise of air temperatures there relative to the global mean (“Arctic climate amplification”). For human transportation systems, these trends have both positive and negative impacts, with rising maritime access in seasonally frozen rivers and seas but declining overland access to seasonally frozen ground (Smith, 2010; Stephenson et al., 2011).
Permafrost, or permanently frozen ground, is ubiquitous around the Arctic and sub-Arctic latitudes and the continental interiors of eastern Siberia and Canada, the Tibetan Plateau and alpine areas. As such, it is a substrate upon which numerous pipelines, buildings, roads and other infrastructure have (or could be) built, so long as these structures are properly designed to not thaw the underlying permafrost. For areas underlain by ice-rich permafrost, severe damage to permanent infrastructure can result from settlement of the ground surface as the permafrost thaws (Nelson et al., 2001, 2002; Streletskiy et al., 2012). These terrestrial problems are driven by lessened ground freeze owing to milder winters and/or deeper snowfall (which insulates the ground) that are hallmarks of the Arctic climate amplification.
Numerous engineering problems are associated with thawing of ground permafrost, including loss of soil bearing strength, increased soil permeability, and increased potential for thermokarsting, differential thaw settlement, and heave (Shiklomanov and Streletskiy, 2013). Over the past 40 years, significant losses (>20 percent) in ground load-bearing capacity have been computed for large Arctic population and industrial centers, with the largest decrease to date observed in the Russian city of Nadym where bearing capacity has fallen by more than 40 percent (Streletskiy et al., 2012). Numerous structures have become unsafe in Siberian cities, where the percentage of dangerous buildings ranges from at least 10 percent to as high as 80 percent of building stock in Norilsk, Dikson, Amderma, Pevek, Dudina, Tiksi, Magadan, Chita, and Vorkuta (ACIA, 2005). Problems are also apparent on the Tibetan Plateau, where mean annual ground temperatures have risen as much as 0.5°C in the past 30 years with damages to built infrastructure caused by thaw settlement and slumping in the affected regions (Yang et al., 2010).
The second way in which milder winters and/or deeper snowfall reduce human access to cold landscapes is through reduced viability of winter roads (also called ice roads, snow roads, seasonal roads, or temporary roads). Like permafrost, winter roads are negatively impacted by milder winters and/or deeper snowfall (Hinzman et al., 2005; Prowse et al., 2011). However, the geographic range of their use is much larger, extending to seasonally frozen land and water surfaces well south of the permafrost limit. They are most important in Alaska, Canada, Russia, and Sweden, but also used to a lesser extent (mainly river and lake crossings) in Finland, Estonia, Norway, and the northern US states. These are seasonal features, used only in winter when the ground and/or water surfaces freeze sufficiently hard to support a given vehicular weight. They are critically important for trucking, construction, resource exploration, community resupply and other human activities in remote areas. Because the construction cost to build a winter road is <1 percent that of a permanent road (e.g., ~$1300/km
versus $0.5–1M/km, Smith, 2010) winter roads enable commercial activity in remote northern areas that would otherwise be uneconomic.
Since the 1970s, winter road season lengths on the Alaskan North Slope have declined from more than 200 days/year to just over 100 days/year (Hinzman et al., 2005). Based on climate model projections, the world’s eight Arctic countries are all projected to lose significant land areas (losses of 11 percent 82 percent) currently possessing climates suitable for winter road construction (Figure 3.3), with Canada (400,000km2) and Russia (618,000km2) experiencing the greatest losses in absolute land area terms (Stephenson et al., 2011).
Figure 3.3 also presents a first attempt to quantify navigation potential for ships. In the Arctic Ocean, climate model projections of thinning sea ice thickness, lower sea ice concentration, lower multi-year ice (MYI) fraction, and shorter ice-covered season all enable increased accessibility to ships. Using the CCSM4 climate model, Stephenson et al. (2011; 2013) quantified these trends for three different ship classes (Polar Class 3, Polar Class 6, and common open-water ships) from present-day to the late 21st century. In general, the Russian Federation is projected to experience the greatest increase (both in percent change and total marine-accessible area) in accessibility to its offshore Exclusive Economic Zone (EEZ), followed by Greenland and Norway. Offshore accessibility increases for Canada and the United States are projected to be less than for the Russian EEZ, owing to greater ice persistence in the Canadian Archipelago and already high accessibility off the North Slope of Alaska today. The timing and magnitude of these projected marine accessibility increases are likely conservative, both because most GCM projections of sea ice loss generally lag behind observations and the CCSM4 model in particular has weaker Arctic climate amplification than previous versions (e.g., ~16 percent less than CCSM3, despite higher global warming; Vavrus et al., 2012). When compared to other GCMs, the CCSM4 model also tends to project greater sea ice cover throughout the 21st century relative to other models (Massonnet et al., 2012).
A second impact of declining sea ice thickness and concentration is decreased shipping distance and travel time through summer trans-polar routes linking the Atlantic and Pacific oceans (Figure 3.4). The shipping distance between Shanghai and Rotterdam, for example, is approximately ~19,600 and ~25,600 km, respectively via the Suez or Panama canals, but only ~15,800 km over the northern coast of Russia (the Northern Sea route) or ~17,600 km through the Canadian archipelago (the Northwest Passage). Although the prospect of such trans-Arctic routes materializing has attracted considerable media attention (and indeed, 46 vessels transited the Northern Sea Route during the 2012 season), it is important to point out that these routes would
FIGURE 3.3 Changes in marine and land-based transportation accessibility by midcentury, calculated by subtracting midcentury (2045-2059) from baseline (2000-2014) conditions. Green indicates where new maritime access to moderately ice-strengthened ships (Canadian Type A icebreaker) will become enabled. Red indicates where conditions presently suitable for building temporary winter roads (assuming 2000 kg weight vehicles) will be lost. All eight Arctic states are projected to suffer steep declines (11 to 82 percent) in winter road potential, caused by by milder winters and deeper snow accumulation (from Stephenson et al., 2011).
FIGURE 3.4 Fastest Trans-Arctic navigation routes during the peak shipping month of September at present (Septembers 2006-2015) and by midcentury (Septembers 2040-2059) as driven by ensemble-averaged GCM projections of sea ice concentration and thickness (from the ACCESS1.0, ACCESS1.3, GFDL-CM3, HadGEM2-CC, IPSL-CM5A-MR, MPI-ESM-MR, CCSM4 models), for hypothetical ships seeking to cross the Arctic Ocean between the North Atlantic (Rotterdam, The Netherlands and St. Johns, Newfoundland) and the Pacific (Bering Strait). Red lines indicate fastest available routes for Polar Class 6 icebreakers; blue lines indicate fastest available routes for common open-water ships. Where overlap occurs, line weights indicate the number of successful transits using the same navigation route. These particular simulations assume a “mediumlow” increase in greenhouse warming (+4.5 Watts/m2 increase in radiative forcing, called the RCP 4.5 scenario), further simulations assuming a “high” increase in greenhouse warming (+8.5 Watts/m2, the RCP 8.5 scenario) show further increases in navigability. Dashed lines indicate national 200-nautical mile Exclusive Economic Zone (EEZ) boundaries; white backdrops indicate period-averaged sea ice concentrations (figure adapted from Smith and Stephenson, 2013).
operate only in summer, and numerous other non-climatic factors remain to discourage trans-Arctic shipping including lack of services, infrastructure, and navigation control, poor charts, high insurance and escort costs, unknown competitive response of the Suez and Panama Canals, and other economic factors (AMSA, 2009; Liu and Kronbak, 2010; Brigham, 2010, 2011).
There are a number of potential adverse effects to human health that may be brought on by changes in the climate. Related issues of food and water security have been discussed in previous sections. This section briefly describes several other human health-related impacts—heat waves, vector-borne and zoonotic diseases, and waterborne diseases—but there are others, including potential impacts from reduced air quality, impacts on human health and development, impacts on mental health and stress-related disorders, and impacts on neurological diseases and disorders (see for example Portier et al., 2010; NRC, 2001; WHO, 2000; WHO/WMO, 2012). The committee stresses that this brief discussion is intended to make the point that human health issues are in many ways tied to abrupt change, and its brevity should not be construed as an indication of the importance of the topic. A full treatment of this subject would be much more extensive, but is beyond the scope of this study as well as the expertise of this committee.
Heat waves cause heat exhaustion, heat cramps, and heat stroke; heat waves are one of the most common causes of weather-related deaths in United States (USGCRP, 2009). Summertime heat waves will likely become longer, more frequent, more severe, and more relentless with decreased potential to cool down at night. Increases in heat-related deaths due to climate change are likely to outweigh decreases in deaths from cold snaps (Åström et al., 2013; USGCRP, 2009). In general, heat waves and the associated health issues disproportionately affect more vulnerable populations such as the elderly, children, those with existing cardiovascular and respiratory diseases, and those who are economically disadvantaged or socially isolated (Portier et al., 2010). Increasing temperature and humidity levels can cross thresholds where it is unsafe for individuals to perform heavy labor (below a direct physiological limit). Recent work has shown that environmental heat stress has already reduced the labor capacity in the tropics and mid-latitudes during peak months of heat stress by 10 percent, and another 10 percent decrease is projected by 2050 (Dunne et al., 2013) with much larger decreases further into the future.
Heavy rainfall and flooding can enhance the spread of water-borne parasites and bacteria, potentially spreading diseases such as cholera, polio, Guinea worm, and schistosomiasis. “Outbreaks of waterborne diseases often occur after a severe precipitation event (rainfall, snowfall). Because climate change increases the severity and frequency of some major precipitation events, communities—especially in the developing world—could be faced with elevated disease burden from waterborne diseases” (Portier et al., 2010). Individual extreme events (see section on Climate Extremes in Chapter 2) could result in abrupt changes in the spread of these diseases, but overall, the impact of climate change on these diseases is not well established.
Vector-borne diseases are those in which an organism carries a pathogen from one host to another. The carrier is often an insect, tick, or mite, and well-known examples include malaria, yellow fever, dengue, murine typhus, West Nile virus, and Lyme disease. Zoonotic diseases are those that are transmitted from animals to humans by either contact with the animals or through vectors that carry zoonotic pathogens from animals to humans; examples include Avian Flu, and H1N1 (swine flu). Changes in climate may shift the geographic ranges of carriers of some diseases. For example, the geographic range of ticks that carry Lyme disease is limited by temperature. As air temperatures rise, the range of these ticks is likely to continue to expand northward (Confalonieri et al., 2007). Overall, the spread of vector-borne and zoonotic diseases that are climate-sensitive will depend heavily on both climate and non-climate factors.
The question for this report is whether any of these effects on human health are likely to change abruptly in the coming decades. One can imagine a gradual migration of insect species over decades, or abrupt outbreaks of waterborne diseases triggered by extreme weather events like floods. Health impacts have the potential to increase the costs and the abruptness of the human health impacts of climate change.
The topic of climate and national security has been discussed elsewhere (see for example Busby, 2007; Fingar, 2008; McElroy and Baker, 2012; Sullivan et al., 2007; Youngblut, 2010), including a recent review entitled Climate and Social Stresses: Implications for Security Analysis (NRC, 2012b). Consequently, remarks here on the subject will be brief, but as with health issues (above), brevity should not be interpreted as an indication of importance. The topic is of vital concern, and interested readers are directed to the aforementioned NRC report and references therein, as well as the excellent discussion of this topic by Schwartz and Randall (2003).
Overall, the links between climate and national security are indirect, involving a complicated web of social and political factors. Climate effects discussed earlier in this report, including food and water security, have the potential to drive national security concerns. Although international cooperation is more typical than conflict in confronting water security issues, conflicts over water issues may become more numerous as droughts become more frequent. In addition, famine and food scarcity have the potential to cause international humanitarian issues and even conflicts, as do health security issues from epidemics and pandemics (also see previous section). These impacts from climate change may present national security challenges through humanitarian crises, disruptive migration events, political instability, and interstate or internal conflict. The impacts on national security are likely to be presented abruptly, in the sense that the eruption of any crisis represents an abrupt change.
An example of an abrupt change that affects the national infrastructure of a number of countries is the opening of shipping lanes in the Arctic as a result of the retreating sea ice. There are geopolitical ramifications related to possible shipping routes and territorial claims, including potential oil, mineral, and fishing rights. The Arctic Council, which was formerly a relatively unknown international body, has become the center of vigorous negotiations over some of these issues. This is a change that is occurring over the course of a couple of decades, well within a generation.
It is important to recognize that abrupt climate change as it affects national security presents opportunities as well as challenges. For example, the United States is still heavily dependent on foreign oil, despite the recent increase in fossil fuel supplies made available by hydrologic fracturing of source rocks. Also, greenhouse gases enter a shared and well-mixed atmosphere, and thus solutions will afford an opportunity to enhance international cooperation and build transparency and trust among nations.
While it may not be possible to predict the exact timing of abrupt climate events and impacts, it is prudent to expect that they will occur at some point. The NRC report on Climate and Social Stresses (NRC, 2012b) recommends a scenario approach for preparing for abrupt climate impacts that may have ramifications for national security. The report recommends the use of stress testing, where “a stress test is an exercise to assess the likely effects on particular countries, populations, or systems of potentially disruptive climate events.” The material presented in this report could inform these types of stress tests by presenting the types of abrupt climate impacts that are possible (see Chapter 2 and Table 4.1).