5
Impacts in the Next Few Decades and Coming Centuries

5.1
FOOD PRODUCTION, PRICES, AND HUNGER

Even in the most highly mechanized agricultural systems, food production is very dependent on weather. Concern about the potential impacts of climate change on food production, and associated effects on food prices and hunger, have existed since the earliest days of climate change research. Although there is still much to learn, several important findings have emerged from more than three decades of research.

It is clear, for example, that higher CO2 levels are beneficial for many crop and forage yields, for two reasons. In species with a C3 photosynthetic pathway, including rice and wheat, higher CO2 directly stimulates photosynthetic rates, although this mechanism does not affect C4 crops like maize. Secondly, higher CO2 allows leaf pores, called stomata, to shrink, which results in reduced water stress for all crops. The net effect on yields for C3 crops has been measured as an average increase of 14% for 580 ppm relative to 370 ppm (Ainsworth et al., 2008). For C4 species such as maize and sorghum, very few experiments have been conducted but the observed effect is much smaller and often statistically insignificant (Leakey, 2009).

Rivaling the direct CO2 effects are the impacts of climate changes caused by CO2, in particular changes in air temperature and available soil moisture. Many mechanisms of temperature response have been identified, with the relative importance of different mechanisms varying by location, season, and crop. Among the most critical responses are that crops develop more quickly under warmer temperatures, leading to shorter growing periods and lower yields, and that higher temperatures drive faster evaporation of water from soils and transpiration of water from crops. Exposure to extremely high temperatures (e.g., > 35ºC) can also cause damage in photosynthetic, reproductive, and other cells, and recent evidence suggests that even short exposures to high temperatures can be crucial for final yield (Schlenker and Roberts, 2009; Wassmann et al., 2009).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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

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

OCR for page 159
5 Impacts in the Next Few Decades and Coming Centuries 5.1 FOOD PRODUCTION, PRICES, AND HUNGER Even in the most highly mechanized agricultural systems, food produc- tion is very dependent on weather. Concern about the potential impacts of climate change on food production, and associated effects on food prices and hunger, have existed since the earliest days of climate change research. Although there is still much to learn, several important findings have emerged from more than three decades of research. It is clear, for example, that higher CO2 levels are beneficial for many crop and forage yields, for two reasons. In species with a C3 photosynthetic pathway, including rice and wheat, higher CO2 directly stimulates photosyn- thetic rates, although this mechanism does not affect C4 crops like maize. Secondly, higher CO2 allows leaf pores, called stomata, to shrink, which results in reduced water stress for all crops. The net effect on yields for C3 crops has been measured as an average increase of 14% for 580 ppm rela- tive to 370 ppm (Ainsworth et al., 2008). For C4 species such as maize and sorghum, very few experiments have been conducted but the observed effect is much smaller and often statistically insignificant (Leakey, 2009). Rivaling the direct CO2 effects are the impacts of climate changes caused by CO2, in particular changes in air temperature and available soil moisture. Many mechanisms of temperature response have been identified, with the relative importance of different mechanisms varying by location, season, and crop. Among the most critical responses are that crops develop more quickly under warmer temperatures, leading to shorter growing periods and lower yields, and that higher temperatures drive faster evaporation of water from soils and transpiration of water from crops. Exposure to extremely high temperatures (e.g., > 35°C) can also cause damage in photosynthetic, reproductive, and other cells, and recent evidence suggests that even short exposures to high temperatures can be crucial for final yield (Schlenker and Roberts, 2009; Wassmann et al., 2009). 159

OCR for page 159
160 CLIMATE STABILIZATION TARGETS A wide variety of approaches have been used in an attempt to quantify yield losses for different climate scenarios. Some models represent indi- vidual processes in detail, while others rely on statistical models that, in theory, should capture all relevant processes that have influenced histori- cal variations in crop production. Figure 5.1 shows model estimates of the combined effect of warming and CO2 on yields for different levels of global temperature rise. It is noteworthy that although yields respond nonlinearly to temperature on a daily time scale, with extremely hot days or cold nights weighing heavily in final yields, the simulated response to seasonal warm- ing is fairly linear at broad scales (Lobell and Field, 2007; Schlenker and Roberts, 2009). Several major crops and regions reveal consistently nega- tive temperature sensitivities, with between 5-10% yield loss per degree warming estimated both by process-based and statistical approaches. Most of the nonlinearity in Figure 5.1 reflects the fact that CO2 benefits for yield saturate at higher CO2 levels. For C3 crops, the negative effects of warming are often balanced by positive CO2 effects up to 2-3°C local warming in temperate regions, after which negative warming effects dominate. Because temperate land areas will warm faster than the global average (see Section 4.2), this corresponds to roughly 1.25-2°C in global average temperature. For C4 crops, even mod- est amounts of warming are detrimental in major growing regions given the small response to CO2 (see Box 5.1 for discussion of maize in the United States). The expected impacts illustrated in Figure 5.1 are useful as a measure of the likely direction and magnitude of average yield changes, but fall short of a complete risk analysis, which would, for instance, estimate the chance of exceeding critical thresholds. The existing literature identifies several promi- nent sources of uncertainty, including those related to the magnitude of local warming per degree global temperature increase, the sensitivity of crop yields to temperature, the CO2 levels corresponding to each temperature level (see Section 3.2), and the magnitude of CO2 fertilization. The impacts of rainfall changes can also be important at local and regional scales, although at broad scales the modeled impacts are most often dictated by temperature and CO2 because simulated rainfall changes are relatively small (Lobell and Burke, 2008). In addition, although the studies summarized in Figure 5.1 consider several of the main processes that determine yield response to weather, several other processes have not been adequately quantified. These include responses of weeds, insects, and pathogens; changes in water resources available for irrigation; effects of changes in surface ozone levels; effects of

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 161 Local Warming (C˚) for: Low Latitudes 0 1.2 2.4 3.6 Mid-High Latitudes 0 1.5 3.0 4.5 Local Warming (C˚) for: Low Latitudes 0 1.2 2.4 3.6 4.8 Mid-High Latitudes 0 1.5 3.0 4.5 6.0 FIGURE 5.1 Average expected impact of warming + CO2 increase on crop yields, without adaptation, for broad regions summarized in IPCC AR4 (left) and for selected crops and regions with detailed studies (right). Shaded area shows likely range (67%). Impacts are averages for current growing areas within each 5-1 replace panels with tiffs.eps region and may be higher or lower for individual locations within regions. Temperature and CO2 changes 2 bitmaps with vector keys for the IPCC summary (left) are relative to late 20th century, while changes estimated for regions (right) were computed relative to pre-industrial. Estimates were derived from various sources (Matthews et al., 1995; Lal et al., 1998; Easterling et al., 2007; Schlenker and Roberts, 2009; Schlenker and Lobell, 2010) (see methods in Appendix for details).

OCR for page 159
162 CLIMATE STABILIZATION TARGETS BOX 5.1 HOW WILL MAIZE YIELDS IN THE UNITED STATES RESPOND TO CLIMATE CHANGES? Nearly 40% of global maize (or corn) production occurs in the United States, much of which is exported to other nations. The future yield of U.S. maize is therefore important for nearly all aspects of domestic and international agriculture. Higher temperatures speed development of maize, increase soil evaporation rates, and above 35°C can compromise pollen viability, all of which reduce final yields. High temperatures and low soil moisture during the flowering stage are especially harmful as they can inhibit successful formation of kernels. In northern states, warmer years generally improve yields as they extend the frost-free growing season and bring temperature closer to optimum levels for photosynthesis. The majority of production, however, occurs in areas where yields are favored by cooler than normal years, so that warming associ- ated with climate change would lower average national yields. The most robust studies, based on analysis of thousands of weather station and harvest statistics for rainfed maize (>80% of U.S. production), suggest a roughly 7% yield loss per °C of local warming, which is in line with previous estimates (USCCSP, 2008b). Given the rate of local warming in the Corn Belt relative to global average, this implies an 11% yield loss per °C of global warming (Figure 5.1). Whether these losses are realized will depend in large part on the effectiveness of adapta- tion strategies, which include shifts in sowing dates, switches to longer maturing varieties, and increased flood frequencies; and responses to extremely high temperatures. Moreover, most crop modeling studies have not considered changes in sus- tained droughts, which are likely to increase in many regions (Wang, 2005; Sheffield and Wood, 2008), or potential changes in year-to-year variability of yields. The net effect of these and other factors remains an elusive goal, but these are likely to push yields in a negative direction. For example, recent observations have shown that kudzu (Pueraria lobata), an invasive weed favored by high CO2 and warm winters, has expanded over the past few decades into the Midwest Corn Belt (Ziska et al., 2010). Adaptation responses by growers are also poorly understood and could, in contrast, reduce yield losses. For example, temperate growers are likely to shift to earlier planting and longer maturing varieties as climate warms, and models suggest this response could entirely offset losses in certain situations. More commonly, however, these adaptations will at best be able to offset 2°C of local warming (Easterling et al., 2007), and they will be less effec- tive in tropical regions where soil moisture, rather than cold temperatures, limits the length of the growing season. Very few studies have considered the evidence for ongoing adaptations to existing climate trends and quantified the benefits of these adaptations.

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 163 development of new seeds that can better withstand water and heat stress and better utilize elevated CO2. A wide range of maize varieties are currently sown throughout the country, cus- tomized to local factors such as latitude, growing season length, and soil, and new varieties are continually developed by private seed companies. These companies have historically focused on biotic stresses, but are now releasing the first varieties explicitly targeted for drought re- sistance. Heat tolerance has not received much investment outside of drought-related traits, likely because of limited economic incentives in current climate. A comparison of maize yields in northern and southern states suggests minimal historical adaptation to heat, as varieties that are more frequently exposed to temperatures above 30°C exhibit similar sensitivities to varieties grown in the North (Schlenker and Roberts, 2009). A major challenge in developing drought and heat tolerance is that traits that confer these often reduce yields in good years, and growers and seed companies have little economic incentive to accept this trade-off given current markets and insurance programs. Another persistent challenge is the decade or more lag between initial investments and seed release. In short, adaptation could offer large benefits, but only if formi- dable technical and institutional barriers are overcome. To put the challenge in context, global cereal demand is expected to rise by roughly 1.2% per year (FAO, 2006), so that adapting to 1°C global warming (or avoiding 11% yield loss) is equivalent to keeping pace with roughly 9 years of demand growth. The corresponding expected impact of 2°C global warming is 25%, or roughly 20 years of demand growth. Future development of new varieties that perform well in hot and dry conditions may also promote adaptation, but again the extent to which this will help remains unclear. Breeders and geneticists must continually weigh trade-offs between producing ample yield under stressful conditions and producing high yields under favorable conditions (Campos et al., 2004). At the higher warming levels considered in this report, it will be increasingly difficult to generate varieties with a physiology that can withstand extreme heat and drought while still being economically productive. Although most studies have focused on crops, effects of climate change on livestock, aquaculture, and fisheries have also been considered in recent years. Livestock in parts of the world are raised mainly on grain and oilseed crops, in which case impacts will largely follow from the prices of these commodities and the costs of cooling or losing animals during heat waves. In other cases livestock depend on grazing pasture and rangeland grasses, which follow a similar pattern to crops in that temperate regions will see modest gains up to ~2°C local warming, although forage quality may de- crease with higher CO2 (Easterling et al., 2007). Although livestock systems are vulnerable in tropical areas, they may become increasingly relied upon as a strategy to cope with greater risks of crop failures (Thornton et al.,

OCR for page 159
164 CLIMATE STABILIZATION TARGETS 2009b). As with livestock, impacts on fisheries are still very uncertain, but a recent study suggests that if global average warming were to be 2°C, catch potential could rise by 30-70% in high latitudes and fall by up to 40% in the tropics, as commercial species shift away from the tropics as the ocean warms (Cheung et al., 2010). Food Prices and Food Security One of the strengths of a global food system is that shortfalls in one area can be offset by surpluses in another. Models of the global food economy suggest that trade will represent an important but not complete buffer against climate change-induced yield effects (Easterling et al., 2007). Specifically, the comparative advantage will shift toward regions currently below opti- mum temperatures for cereal production (e.g., Canada) and away from hot tropical nations, with greater flows of food trade from north to south. On average, studies suggest small price changes for cereals up to 2.5°C global temperature increase above pre-industrial levels, with significant increases for further warming, but there is considerable uncertainty around these estimates (see Box 5.2). Implications of climate change for hunger, or the more technical term— food insecurity—follow in part from price changes, but also depend criti- cally on how sources of income and other aspects of health are affected by climate. A useful rule of thumb provided by early studies suggested that malnourishment would rise by roughly 1% for each 2-2.5% rise in cereal prices (Rosenzweig, 1993). These and subsequent analyses often make untested assumptions about the ability of poor tropical nations to maintain economic growth in the face of declining agricultural productivity. For example, many African countries rely on agriculture for half or more of all economic activity, and losses in productivity could dampen purchasing power. Conversely, where price rises are greater than yield losses, house- holds dependent on agricultural income could see net gains in food security. In general, rural and urban workers with little or no landholdings are the most vulnerable to price shocks. A new generation of models that explicitly account for income sources among poor populations is emerging but yet to provide robust insights. Also important could be climate-induced changes in the incidence of diarrheal and other diseases, which inhibit food security by reducing utilization of nutrients in food.

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 165 BOX 5.2 CLIMATE CHANGE IMPACTS ON GLOBAL CEREAL PRICES Several modeling groups have analyzed future changes in global cereal markets in response to climate change. All operate by making estimates of yield responses in each region, and then inputting these into a model of global trade that computes the optimal mix of crop areas in different regions and the market-clearing price. Five models summarized by the recent IPCC report suggests small price changes for warming up to 2.5°C, and a nonlinear increase in prices thereafter (Easterling et al., 2007). Two important caveats relate to these estimates, however. First, the yield changes used in these models usually assume considerable levels of farm- level adaptations, which substantially reduce impacts. For example, in one prominent study cereal prices rose by 150% for a 5.2°C global mean temperature rise if farm-level adaptations were not included. When changes in planting dates, cultivar choices, irrigation practices, and fertilizer rates were simulated, these price changes were reduced to roughly 40% (Rosenzweig and Parry, 1994). Other studies often do not estimate impacts without adaptation, making it dif- ficult to gauge assumptions. The costs of adaptation are also not considered in these studies, or reflected in price changes. Second, most assessments have not adequately quantified sources of uncertainty. Although different climate scenarios are often tested, processes related to crop yield changes and eco- nomic adjustments are often implicitly assumed to be perfectly known. An additional source of uncertainty is potential competition with bio-energy crops for suitable land, which could limit the ability of croplands to expand in temperate regions as simulated by most trade models.m 5.2 COASTAL EROSION AND FLOODING Our knowledge of the links between atmospheric concentration limits, trajectories toward equilibrium temperature change, and sea level rise is fraught with uncertainty. As reported in Section 4.8, it is therefore only possible to offer a range of sea level rise between 0.5 and 1.0 m through 2100. Moving down the causal chain to consider coastal erosion and flooding adds yet another layer of complication because both are driven primarily by storm surges, land-use decisions, and other processes whose intensities and frequencies change from place to place. These changes alter the characters of associated risks even if changes in the intensities and frequencies of the storms, themselves, cannot be projected. The social and economic ramifica- tions of these physical manifestations of climate change depend critically on patterns of future development and population growth. It is, therefore, extremely difficult to offer credible broad-based estimates of vulnerabilities and potential adaptation costs. At best, in fact, we can offer only suggestive

OCR for page 159
166 CLIMATE STABILIZATION TARGETS ranges of aggregate risk and more quantitative estimates only for specific locations. Figure 5.2 offers a portrait of the geographic spread of deltas and mega- deltas where mega-cities are at the greatest risk from rising seas—these are the “hot-spots” of “key vulnerabilities” in the coastal zone. Ericson et al. (2006) estimated that nearly 300 million people currently inhabit a sample of 40 such deltas with an average population density of 500 people per km2. Translating this observation into projections of future vulnerabilities, Table 5.1 shows the sensitivity of estimates of populations subject to coastal flooding in 2080 to assumptions about socioeconomic development as de- scribed in the SRES scenarios—sensitivity generated by differences across the scenarios in population growth and by differences in assumptions about economic development and therefore the capacity to adapt. Figure 5.3 em- phasizes the importance of adaptation when it suggests, for example that 1 m of sea level rise could put between 10 and 300 million more people at risk of coastal flooding each year. It is important to note, in interpreting this figure, that the likelihood of inundation from coastal storms may not be proportional with sea level rise. Moreover, the consequences of these storm events calibrated in millions of people in jeopardy from coastal flooding de- pend on local population densities and geographic features. The result of the FIGURE 5.2 Relative vulnerability of coastal deltas as shown by the indicative population displaced by 5.2.eps current sea-level trends to 2050 (Extreme=>1 million; High=1 million to 50,000; Medium=50,000 to 5,000; following Ericson et al., 2006). Source: Nicholls et al. (2007: Figure 6.6). bitmap

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 167 TABLE 5.1 Regional Distribution of Population Subject to Coastal Flooding in 2080 along Alternative SRES Scenarios. Table 5.1.eps NOTE: Population estimates by region assume proportional population growth within coastal regions. Source: bitmap Nicholls (2004) as displayed in Nicholls et al. (2007: Table 6.5). FIGURE 5.3 Estimates of people flooded in coastal areas attributable to sea level rise along alternative SRES scenarios. Estimates of the number of additional people in jeopardy from coastal flooding along alternative 5-3.eps SRES development scenarios for three gross categories of adaptation intensities are displayed. Constant bitmap protection envisions maintaining current practices, evolving protection envisions increasing protection as local economies grow to preserve the current pattern with respect to national GDP, and enhanced protec- tion envisions accelerating the pace of adaptation so that increasing resources are devoted to protection. Source: Nicholls et al. (2007: Figure 6.8) derived from Nicholls and Tol (2006).

OCR for page 159
168 CLIMATE STABILIZATION TARGETS confluence of these complications is a noticeable threshold of accelerating risk around 0.3 m of sea level rise that is captured even by global aggregates regardless of adaptation effort. As a result, even 50 cm of SLR could put between 5 and 200 million more people annually at risk of flooding. Tol (2007) used a specific integrated assessment model of his own creation to portray aggregate measures of erosion that parallel estimates of populations facing complete displacement and/or significant economic loss derived from economically efficient abandonment and/or growing protec- tion costs across developed and developing countries. Figure 5.4 calibrates his results graphically in relation to sea level rise; they were derived from a socioeconomic portrait that was crafted to be consistent with the IS92a emissions scenario for which seas rise by roughly 60 cm through 2100. This work suggests that 50 cm of sea level rise could permanently displace up to 4 million people and cause more than 250,000 km2 of wetland and dry-land to be lost to erosion worldwide (with 90% of these losses projected to occur in developing countries). The human faces behind the global displacement results portrayed here can, of course, be seen in examples of erosion from coastal storms and rising seas. In the Arctic, Newtok, Alaska is already pre- paring for complete displacement, for example, and several neighboring towns face the same fate in the near future. Meanwhile, many small island states like Tuvalo, the Maldives, and the Cook Islands foresee similar futures this century if sea level rise continues. Geographic detail for physical processes like erosion and inundation FIGURE 5.4 Losses attributable to sea level rise. Estimates of wetland and dry-land losses for developed 5-4.eps (Panel A) and developing countries (Panel B) correlated with sea level rise along a socioeconomic scenario bitmap that tracks IS92a. Source: Derived directly from Tol (2007) as depicted in Nicholls et al. (2007: Figure 6.10).

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 169 TABLE 5.2 Selected Losses from Sea Level Rise and Associated Erosions across Asia SLR Rise (from 2000 levels) Location Magnitude Source 81.4 ⋅ 103 k m2 0.3 m China Du and Zhang (2000) 21.3 ⋅ 103 k m2 Huanghe-Huaihe Delta 54.5 ⋅ 103 k m2 Changjiang Delta 5.5 ⋅ 103 k m2 Zhujiang Delta 2.3 ⋅ 103 k m2 1.0 m Japan Mimura and Yokoki (2004) 1.0 m Korea 1.2% area Madsen and Jakobsen (2004) 478 km2 (11%) 1.1 m India and Bangladesh Loucks et al. (2010) 1,396 km2 (33%) 1.2 m India and Bangladesh 4,015 km2 (96%) 0.3 m India and Bangladesh from sea level rise has been emerging over the past decade. Table 5.2, for example, offers estimates for several locations in Asia. Some are located in important deltas in China where modest sea level rise of 0.3 meters would cause significant loss of land area from inundation and erosion; others are located in eastern and southeastern Asia where 1 m of sea level rise would cause significant loss of land and protective mangroves in addition to putting many people at risk of displacement. The final entry reports recent estimates of associated loss in the habitat of the only tiger population in the world (panthera tigris) that is adapted to living in mangroves; Loucks et al. (2010) report that a nonlinear decline to extinction (at 30 cm) would begin around 15 cm of sea level rise. Turning to specific locations within the United States, where it is pos- sible to focus attention on downstream impacts and the potential adapta- tion, Figure 5.5 first depicts coastal vulnerability to erosion across the mid- Atlantic region at the end of the century for three sea level rise scenarios. Enormous variability from site to site along the coastline is clearly displayed; and so it is obvious that potential risks and the potential for adaptation can be expected to be equally diverse. 5.3 STREAMFLOW Runoff is defined as the difference between precipitation and the sum of evapotranspiration and storage change on or below the land surface. On long term balance, it must be balanced by precipitation minus evapotranspi- ration, which also equals atmospheric moisture convergence. Streamflow is

OCR for page 159
206 CLIMATE STABILIZATION TARGETS 3,130 km2. Without aid from humans these species are probably not going to be able to persist. Extinction is irreversible. Choices among stabilization targets can be expected to determine the scope of future extinction (e.g., types of spe- cies, geographic regions, etc.) that could be caused by climate change, or alternatively the scale of protective adaptation measures such as species management that could be considered to avoid extinctions. 5.8 BIOLOGICAL OCEAN Impacts of CO2, pH, and Climate Change in the Ocean’s Biology Marine ecosystems will be affected by climate change via physical changes in ocean properties and circulation (Sections 4.1, 4.4, and 4.7), ocean acidification via altered seawater chemistry from rising atmospheric CO2 (Section 4.9), and sea-level rise via coastal habitat loss. Some of the key potential impacts will involve changes in the magnitude and geographi- cal patterns of ecological and biogeochemical rates and shifts in the ranges of biological species and community structure (Boyd and Doney, 2002). Impacts are expected to include both direct physiological impacts on organ- isms through, for example, altered temperature, CO2, and nutrient supply, and indirect effects through altered food-web interactions such as chang- ing seasonal timing (phenology) of phytoplankton blooms or disruptions in predatory-prey interactions. Primary production by upper-ocean phytoplankton forms the base of the marine food-web and drives ocean biogeochemistry through the export flux of organic matter and calcareous and siliceous biominerals from plank- tonic shells. Plankton growth rates for individual species are temperature dependent and tend to increase under warming up to some threshold. When viewed in aggregate, plankton community production rates approximately follow an exponential curve in nutrient replete conditions, which would sug- gest increasing global primary productivity over this century as sea surface temperatures increase (Sarmiento et al., 2004). In most regions of the ocean, however, primary production rates are limited by nutrients such as nitrogen, phosphorus, and iron. Diatoms, a key shell-forming group of phytoplankton, are also limited by silicon. The rates of many other biological processes, such as bacterial respiration and zooplankton growth and respiration, also speed-up as temperature rises, the integrated effect at the ecosystem level is difficult to predict from first principles. Warming also occurs in conjunction

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 207 with other factors (rising CO2, altered ocean circulation), and the potential synergistic or antagonistic effects of multiple stressors must be considered. Satellite observations indicate a strong negative relationship, at inter- annual time scales, between marine primary productivity and surface warm- ing in the tropics and subtropics, most likely due to reduced nutrient supply from increased vertical stratification (Behrenfeld et al., 2006). Satellite data also indicate that the very lowest productivity regions in subtropical gyres expanded in area over the past decade (Polovina et al., 2008), although these trends may be due to interannual variability (Henson et al., 2010). Numerical models project declining low-latitude marine primary produc- tion in response to 21st century climate warming (Sarmiento et al., 2004; Steinacher et al., 2010) (Figure 5.19). Warmer, more nutrient-poor condi- tions in the subtropics could enhance biological nitrogen fixation (Boyd and Doney, 2002), an effect that may be amplified by higher surface water CO2 levels (Hutchins et al., 2009). The situation is less clear in temperate and polar waters, although there is a tendency in most models for increased production due to warming, reduced vertical mixing, and reduced sea-ice cover. For example, the rapid warming and sea-ice retreat along the West Antarctic Peninsula has lead to a poleward shift in the region of strong sea- sonal primary production that has impacts for higher trophic levels including seabirds (Montes-Hugo et al., 2009). In most open-ocean regions, however, the climate signal in primary production and other ecosystem properties may be difficult to distinguish from natural variability for many decades (Boyd et al., 2008; Henson et al., 2010). Changes in atmospheric nutrient deposition (nitrogen and iron) linked to fossil-fuel combustion and agriculture also can alter marine productivity but mostly on regional scales near industrial and agricultural sources (Duce et al., 2008; Krishnamurthy et al., 2009). Subsurface oxygen levels likely will decline due to warmer waters (low- er oxygen solubility) and altered ocean circulation, leading to an enlarge- ment of open-ocean oxygen minimum zones and stronger coastal oxygen depletion in some regions (Keeling et al., 2010; Rabalais et al., 2010). Low subsurface O2, termed hypoxia, occurs naturally in open-ocean and coastal environments from a combination of weak ventilation and/or strong organic matter degradation. Dissolved O2 gas is essential for aerobic respiration, and low O2 levels negatively affect the physiology of higher animals leading to so-called “dead-zones” where many macro-fauna are absent. Coastal hypox- ia can lead to marine habitat degradation and, in extreme cases, extensive fish and invertebrate mortality (Levin et al., 2009; Rabalais et al., 2010). Ex- panded open-ocean oxygen minimum zones would increase denitrification and may contribute to increased oceanic production of the greenhouse gas

OCR for page 159
208 CLIMATE STABILIZATION TARGETS FIGURE 5.19 Model projected change in vertically integrated annual mean primary production (PP) relative 5-19.eps to pre-industrial conditions (decadal mean 1860-1869) for the end of the 21st century under SRES A2. The changes represent the difference between 2090-2099 and 1860-1869 (decadal means). Multi-model means bitmap have been computed for four coupled ocean-atmosphere models using regional skill scores as weights. Where no observation-based data is available to calculate skill scores (e.g., in the Arctic) the arithmetic mean of the model results is shown. The magnitude of the primary production changes are shown in per- cent normalized to global mean areal primary production rate and are presented for a nominal increase in global mean surface air temperature of 1°C. Source: Steinacher et al. (2010). nitrous oxide (N2O). The organic matter respiration that generates hypoxia also elevates CO2, and multiple stressors of warming, deoxygenation, and ocean acidification magnify physiological and microbial responses (Pörtner and Farrell, 2008; Brewer and Peltzer, 2009).

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 209 Open-ocean deoxygenation has been observed in the thermocline of the North Pacific and tropical oceans over decadal periods, perhaps due to natural climate variability (Mecking et al., 2008). Models project long- term reductions of 1-7% in the global oxygen inventory and expansions of open-ocean oxygen minimum zone over the 21st century (Frölicher et al., 2009; Keeling et al., 2010). The duration, intensity, and extent of coastal hypoxia has also been increasing substantially over the last half-century, but primarily due to elevated fertilizer run-off and atmospheric nitrogen deposition that contribute to coastal eutrophication, enhanced organic mat- ter production, and export and subsurface decomposition that consumes O2. Climate change could accelerate coastal hypoxia via surface warming and regional increases in precipitation and river runoff that increase wa- ter-column vertical stratification; on the other hand, more intense tropical storms could disrupt stratification and increase O2 ventilation (Rabalais et al., 2010). Expanding coastal hypoxia is also induced in some regions by reorganization in ocean-atmosphere physics. Off the Oregon-Washington coast, increased wind-driven upwelling is linked to the first appearance of hypoxia, and even anoxia, on the inner-shelf after five decades of hypoxia- free observations (Chan et al., 2008). Further south in the California Current System, the depth of hypoxic surface has shoaled along the coast by up to 90 m (Bograd et al., 2008). The same physical phenomenon, along with the penetration of fossil-fuel CO2 into off-shore source waters, are introducing waters corrosive to aragonite (W < 1) onto the continental shelf (Feely et al., 2008). There is conflicting evidence on how coastal upwelling may respond to climate change, and impacts may vary regionally (Bakun et al., 2010). Laboratory and mesocosm experiments indicate that many marine or- ganisms are sensitive to elevated CO2 and ocean acidification, with both positive and negative physiological responses (Fabry et al., 2008; Doney et al., 2009a,b; NRC, 2010). The projected rates of change in global ocean pH and W over the next century are a factor of 30-100 times faster than temporal changes in the recent geological past, and the perturbations will last many centuries to millennia. Although there are spatial and temporal variations in surface seawater pH and saturation state, projected future surface water pH values for the open-ocean are below the range experienced by contempo- rary populations, and the ability of marine organisms to acclimate or adapt to the magnitude and rate of change is unknown. The largest identified negative impacts are on shell and skeleton growth by calcifying species including corals, coralline algae, and mollusks. Cor- als utilize the aragonite mineral form of calcium carbonate, and the rate of coral calcification declines with falling aragonite saturation state even when waters remain supersaturated, and corals appear to need saturation

OCR for page 159
210 CLIMATE STABILIZATION TARGETS states (W > 3) for healthy growth (Langdon and Atkinson, 2005; Kleypas and Yates, 2009). Decreased calcification is observed for corals with symbiotic zooxanthella (photosynthetic algae living within coral animals), and CO2 fertilization of zooxanthella does not alleviate acidification effects. Studies of net community calcification rates for coral reef ecosystems indicate that overall net calcification also decreases with rising CO2 (Silverman et al., 2007), and model studies suggest a threshold of about 500-550 ppm CO2 where coral reefs would begin to erode rather than grow, negatively im- pacting the diverse reef-dependent taxa (Silverman et al., 2009). Observed physiological responses for mollusks, such as pteropods, oysters, clams, and mussels, include reduced calcification, increased juvenile mortality and reduced larval settlement, and smaller, thinner, and malformed shells (Orr et al., 2005; Green et al., 2009; Miller et al., 2009). Crustaceans also utilize calcium carbonate in their shells, but the response to elevated CO2 is less well-understood with studies reporting both increased and decreased calcification rates (Fabry et al., 2008). Decreased calcification rates with rising CO2 are observed as well for key planktononic calcifiers including foramaniferia and most strains or coccolithophores. Some organisms may benefit in a high-CO2 world, in particular pho- tosynthetic organisms that are currently limited by the amount of dissolved CO2. In laboratory experiments with elevated CO2, higher photosynthesis rates are found for certain phytoplankton species, seagrasses, and macroal- gae, and enhanced nitrogen-fixation rates are found for some cyanobacteria (Hutchins et al., 2009). Indirect impacts of ocean acidification on non- calcifying organisms and marine ecosystems as a whole are possible but more difficult to characterize from present understanding. A limited number of field studies that have been carried out in mostly benthic systems with naturally elevated CO2 are broadly consistent with the laboratory studies in terms of predicted changes in community structure (e.g., decrease in calci- fiers; increase in non-calcifying algae) (Hall-Spenser et al., 2008; Wootton et al., 2008). Polar ecosystems also may be particularly susceptible when surface waters become undersaturated for aragonite, the mineral form used by many mollusks including pteropods, which are an important prey species for some fish. Socioeconomic impacts from degraded fisheries and other marine resources are possible but poorly known at this point (Cooley and Doney, 2009). Based on historical survey data, the geographic range of many marine species has shifted poleward and into deeper waters due to ocean warm- ing (Perry et al., 2005; Nye et al., 2009). Model projections indicate that poleward expansion and equatorial contraction of geographical ranges

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 211 for particular species will continue and that at any particular location the frequency of replacement of “cool-water” species by “warm-water” species will likely increase. Individual marine species will be impacted differen- tially; for example pelagic fish ranges may be impacted more than demer- sal ranges, which will lead to changes at the community and ecosystem model. Few studies have looked comprehensively across many marine taxa and geographic regions, but a recent pilot model projection suggests the potential for significant changes in community structure in the Arctic and Southern Ocean biodiversity due to invasion of warm water species and high local extinction rates in the tropics and subpolar domains. Fish stock size may either grow or decline due to altered primary production, prey abun- dance, and temperature-dependent growth rates, the trend for each species depending on its particular biology and habitat (Brown et al., 2010; Hare et al., in press). Complex predation and competition interactions may re- verse the expected responses for some species (Brown et al., 2010). Climate change may also disrupt larval dispersal and development patterns as well as existing predator-prey interactions through altered currents and seasonal phenologies for spawning and plankton blooms (Parmesan, 2006). Specific marine habitats may be particularly sensitive to changing cli- mate. Rising sea-level would impact, and in many cases degrade, coastal wetlands and estuaries, coral reefs, mangroves, and salt-marshes through inundation and enhanced coastal erosion rates; these coastal environments serve as important nursery habitats for larval and juvenile life-stages. Re- gional impacts depend on local vertical land movements and would be exacerbated where the inland migration of ecosystems is limited by coastal development and infrastructure. The thermal tolerance of many coral spe- cies is limited, and over the past several decades, warmer sea surface tem- peratures have led to widespread tropical coral bleaching events (loss of algal zooxanthella) and increased coral mortality. Warming and more local human impacts have been associated with declines in the health of coral reef ecosystems worldwide. Bleaching can occur for sea surface tempera- ture changes as small as +1-2°C above climatological maximal summer sea surface temperatures, and more frequent and intense bleaching events are anticipated with further climate warming (e.g., Veron et al., 2009). Sea-ice dependent species are also at risk, and rapid warming in the Arctic and parts of Antarctica has resulted in substantial shifts in whole food-webs (Ducklow et al., 2007; Montes-Hugo et al., 2009).

OCR for page 159
212 CLIMATE STABILIZATION TARGETS 5.9 ILLUSTRATIVE ADDITIONAL FACTORS There are many more climate impacts that could be very important but are not as well understood as those described above. Some illustrative examples are briefly provided here. National Security Many processes could plausibly connect climate change to national security concerns. For instance, military experts have pointed to the poten- tial for climate-induced food and water shortages to contribute to political instability, which can then be exploited by extremists (CNA Corporation, 2007). The potential for mass migrations associated with resource shortages or flooding are also potential “threat multipliers.” Climate changes will also likely affect military operations, such as via inundation of low-lying military bases, and introduce new geopolitical dilemnas, such as the opening of sea routes in the Arctic. Yet perhaps because of the complex nature of national security threats and the paucity of relevant data, there are relatively few quantitative exam- ples that document the climate sensitivity of phenomena related to national security. Some empirical evidence suggests an important role for climate in domestic and international conflict. Long-term fluctuations of global wars and death rates since 1400 are correlated with shifts in temperature (Zhang et al., 2007a). In Africa, civil wars since 1980 have been roughly 50% more likely in years 1°C warmer than average (Burke et al., 2009). Precipitation decreases are also associated with conflict in Africa, although projected rainfall changes are not large relative to historical variability (Miguel et al., 2004; Hendrix and Glaser, 2007). Obviously more work is needed to advance understanding of national security threats from climate change. Specifically, although the implica- tions of climate change for resource scarcity are uncertain, the complex relationship between resource scarcity and conflict is even more tenuously understood (Barnett, 2003; Nordås and Gleditsch, 2007). At the same time, military experts routinely caution that waiting for quantitative precision can be very risky, and intuition alone is often used to make major strategic decisions for national security (CNA Corporation, 2007). Dynamic Vegetation Changes in climate and CO2 beyond 2100 will likely be sufficient to cause large-scale shifts in natural ecosystems. Although relatively few

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 213 modeling studies extend beyond 2100, many show substantial changes occurring by 2100 that indicate the potential for even greater changes in the following centuries. Indeed, some major shifts such as the expansion of shrublands in Arctic regions are already evident in recent decades (Sturm et al., 2001; Tape et al., 2006), and this shift is consistent with results from warming experiments in the region (Walker et al., 2006). One important consequence of this expansion is that the resulting decrease in surface al- bedo can amplify local summer warming in the future by a factor of two or more (Chapin et al., 2005). Major biome shifts also appear likely in some temperate and tropical regions by 2100 (Scholze et al., 2006). The Eastern part of the Amazon rain- forest, for example, may shift to a seasonally dry forest or even a savanna due to likely rainfall decreases in the dry season by 2100 (Cox et al., 2004; Malhi et al., 2009). Beyond 2100, these shifts become more likely. High CO2 levels will likely promote expansion of vegetation into currently bar- ren tundra and desert ecosystems, because of higher water-use efficiencies, which again would amplify local warming because of albedo effects (Bala et al., 2006). Most models used to simulate future vegetation changes rest on strong empirical relationships between current climate and the distribution of major biomes. Less is known about how transitions between equilibrium states occur, and for instance whether deep roots of established trees limit their sensitivity to climate shifts. Another source of uncertainty in projections of vegetation change is potential interactions with local land use, which for instance could accelerate regional climate change in tropical forests (Malhi et al., 2008). Despite these uncertainties, higher emissions scenarios will almost certainly result in climate shifts that are large enough to cause major vegetation shifts by 2100 and beyond. Some Climate Changes Beyond 2100 More is known about the very long term (millennia) and the present century, but there is a gap in understanding and more limited knowledge of climate system behavior over the next few centuries. Here we present two examples of areas where information on the next few centuries is available. Circulation About half of the AR4 climate models were used to project the future climate beyond 2100 to 2200. These simulations were performed using the

OCR for page 159
214 CLIMATE STABILIZATION TARGETS A1B emission scenario until 2100, and then holding the forcing fixed to 2200. In some of the models, the MOC was seen to maintain an equilibrium strength that was similar to that projected for 2100; in other models the MOC strengthened somewhat from their nadir early in the 22nd century. Too few of the higher end AR4 climate models have been integrated far enough into the future to assess whether persistently high greenhouse gas concentrations will cause a permanent change to the strength of the MOC. Sea Ice Beyond 2100 Climate model simulations suggest that in the decades following 2100 the Arctic may be perennially ice-free (Winton, 2006a,b; Eisenman and Wettlaufer, 2009). However, on the millennium scale the system may oscil- late between being totally ice-covered or having ice only along the land margins (Ridley et al., 2008). Only two IPCC models predict a year-round ice-free Arctic in the decades after 2100. However, these are the models that have the most sophisticated sea ice components. The scenario within which these models lose their Arctic ice is the 1% per year CO2 increased to quadrupling, a concentration of 1,120 ppm, after which although atmo- spheric CO2 is kept constant temperatures continue to rise. These models, one initiated from pre-industrial conditions and the other from present-day conditions are run for nearly 300 years; quadrupling occurs at 140 years. Both models exhibit a gradual linear decline in September sea-ice loss be- coming ice free when the average polar temperature is –9°C. In March, the transition to an ice-free state is also linear until polar temperatures reach –5°C, at which point one model experiences an abrupt transition, associ- ated with that model’s ice-albedo feedback mechanism, while in the other it remains linear and the ocean heat flux plays a larger role. The tempera- ture at which the Arctic becomes ice free in these models is 13°C above present-day values (Winton, 2006a,b). The ice-albedo, convective cloud, and ocean heat transport feedbacks all play necessary roles in the loss of the winter sea ice (Abbot et al., 2009b). However, the ice-albedo feedback plays a key role. While an ice-free Arctic may present new economic opportunities it will also likely have profound impacts on climatological and ecological systems locally and globally. A few of these are mentioned here. From the physical standpoint, the loss of Arctic sea ice means that the mediating influence of sea ice on energy flux exchanges between the atmosphere and ocean will no longer prevail and the Arctic atmosphere will warm. Model studies sug- gest that these two impacts will affect the effectiveness of the overturning

OCR for page 159
IMPACTS IN THE NEXT FEW DECADES AND COMING CENTURIES 215 in the thermohaline circulation (e.g., Broecker, 1997; Lemke et al., 2007; Levermann et al., 2007), an impact with global consequences for climate variability. The reduced latitudinal temperature gradients that result from the Arctic warming will modify the atmospheric circulation dynamics in the Northern Hemisphere. Mid-latitude storm tracks may shift (e.g., Deser and Teng., 2008), the westerlies may weaken, and storm intensities may decrease poleward of 45 N (e.g., Royer et al., 1990; Honda et al., 1999). Large-scale pressure systems such as the Azores High (Raymo et al., 1990) as well as the Asian monsoon and the Hadley Cell circulation systems may be affected (Liu et al., 2007). Along with these impacts on the atmospheric and oceanic circulation, loss of Arctic sea ice has the potential to enhance the rates of surface melt of Greenland’s glaciers. Present-day enhanced melting of Greenland’s ice sheet is associated with increased advection of ocean heat onto the ice sheet from a warmer ocean, resulting in enhanced melt (e.g., Rennermalm et al., 2009). The warmer ocean surface temperatures that will occur in the absence of sea ice can be expected to enhance the rates of warming. The increased melt will contribute to sea level rise. Sea ice in the Arctic is of major ecological importance; it is a habitat for a variety of species. An ice-free Arctic will promote large scale changes in Arctic marine ecosystems. Already in the Arctic, loss of sea ice has been associated with polar bear population decrease (e.g., DeWeaver, 2007); seasonal or perennial loss of sea ice will only exacerbate this situation. Sea ice protects the shorelines from erosion and helps maintain continuous per- mafrost. Lawrence et al. (2008b) show that loss of Arctic sea ice speeds the degradation of permafrost. Warming of the permafrost has already led to the destabilization of infrastructure in the Arctic, and removal of the protective cover of ice has already led to increased shoreline erosion (IPCC, 2007a,b); this can only worsen as sea ice cover is lost. Additionally, warming of the permafrost may lead to the emission of methane to the atmosphere, which has the potential to enhance greenhouse gas-related warming (Macdonald, 1990).

OCR for page 159