6.3.2 Tropical Terrestrial Ecosystems Some model studies project that, by 2100, reduced regional rainfall and longer dry seasons could cause the Amazon rainforest to decrease by up to 70%, and change peatlands in southeast Asia from net sinks to net sources of atmospheric CO2. Such dramatic changes in tropical habitats would have large impacts on tropical species. Many of these species, particularly in the Amazon, are at the upper limit of their optimum temperature range and may have nowhere to migrate to. Such species are much more vulnerable to extinction than species in the extratropics273,788-792.
There is also evidence that most fish species in disturbed tropical coral reefs decline in abundance and diversity within three years of disturbance. Species that live off live coral are most vulnerable, while those dependent on algae and detritus tend to thrive. Hence, long term effects of sustained disturbances that kill coral would be large793.
6.3.3 Arctic ecosystems Arctic temperatures are likely to warm by as much as three times that of global temperature increases. Such changes will allow boreal ecosystems to expand significantly in the region, pushing tundra ecosystems northward. Experimental studies indicate that the productivity of woody species within tundra ecosystem would increase, while that for mosses and lichens would decrease. Concurrent effects on northern lake ecosystems will be complicated by the impacts of enhanced UV radiation, and will vary from region to region and species to species794-797.
As summer seasons become longer and ice cover recedes, marine productivity within the Arctic Ocean is also expected to increase, particularly in the early part of the season. Related changes in the food chain of Arctic marine waters are complex, and will have varying impacts on mammal species that live within these waters or are dependent on sea ice cover. Species such as narwhals, polar bears and hooded seals will be most vulnerable. Ringed and bearded seals and bowhead whales are much less affected and may even benefit, although these may experience increased competition over time from invasive species. Polar bear diets vary from region to region and with age and gender. Impacts on bear populations within the Canadian archipelago are likely to be delayed relative to those in other Arctic regions, since ice is expected to persist there longer. There is also evidence that at least some populations of polar bears adapt their diets as food supplies change682,798-801.
Climate change will also likely provide more food for caribou in Arctic regions. However, freezing rain events can cause loss of access to that food, sometimes with disastrous consequences802.
6.3.4 Canadian ecosystems Higher temperatures will also reduce soil freezing days across Canada and accelerate permafrost decay in northern latitudes, although any concurrent decreases in snow cover thickness could at least partly offset these reductions. These indirect effects will further add to the complexity of the response of Canadian ecosystems to climate change. Related changes in land hydrology, for example, will alter nutrient supply and sediment loading in northern aquatic ecosystems. In some high latitude regions, mid-summer lake temperatures may rise by as much as 18°C. The net impact of these changes on species within or dependent on these aquatic systems (including mammals and birds) will be large. Some existing species will be negatively affected, while new exotic species will thrive803-805.
Most Canadian forest ecosystems are projected to become more productive, particularly in northern latitudes and alpine regions, and to shift northward. Some conifer species in southern regions will decline due to loss of suitable habitat. However, forests are also likely to become more vulnerable to disturbances. In eastern Canada, duration and intensity of spruce budworm outbreaks are projected to increase substantially. Changing dynamics of beaver populations, which appear to explain about 80% of boreal wetland fluctuations in western Canada, will also have important impacts. Furthermore, empirical models suggest the number of large wildfires across Canada’s southern boreal regions will also rise dramatically, particularly in areas where drought activity is projected to increase806-811.
New insect species, such as an exotic blow fly species recently observed in Ontario, are also expected to significantly expand across Canada as climate warms. Such invasive species will compete and potentially displace existing species. A key factor in the spread of these species is the number of intense frost days, which depend as much on changes in the variance of temperatures as on changes in the mean680,812.
6.3.5 Oceans The combined effect of natural climate oscillations (such as ENSO) and long term secular warming of ocean waters can result in increased risks of extreme warm SST episodes during which climate variables exceed critical ecological thresholds and cause serious harm to ocean ecosystem species. Increased influx of freshwaters from land runoff onto continental shelves as well as pollution within runoff waters can also change ocean chemistry in some regions. Furthermore, increased production of organic carbon in the upper oceans will result in greater oxygen consumption when the carbon decays. By 2100, this could potentially increase the volume of oxygen-deficient waters in the oceans by 50%. Studies suggest that about 1/3 of coral species examined could be vulnerable to extinction under these combined pressures. Such risks are much greater than for land species, and are greatest in region such as the Caribbean and the western Pacific. In tropical coral reefs, these changes are likely to favour corals that host multiple algae groups that are more adaptable to changing environments that those dependent on single groups701,813-816.
Another important concern is the effect of increasing CO2 uptake into ocean waters on their acidity. Ocean chemistry models suggest that ocean pH could drop about 0.5 under a high CO2 world. Paleo records from the distant past suggest that elevated acidity may have been a factor in mass ocean species extinctions during past periods of similarly high atmospheric CO2. While the effects of higher pH on marine species is as yet poorly understood and hence difficult to predict, most impact studies suggest that calcareous organisms and coralline algae are particularly vulnerable. Data from high CO2 environments near volcanic vents, for example, show that these species are largely replaced by sea grasses and other more tolerant species. Since CaCO3 also acts as the mortar that holds reefs together, decreases in its abundance reduces new reef formation and leaves existing reefs more vulnerable to erosion. High acidity will also dramatically reduce the productivity of mussels and oysters in coastal waters, affecting local economies that rely on these. Finally, increased ocean acidity reduces sound absorption and, by 2050, may increase the distance that noise travels in ocean environments by 40%817-825.
Some researchers, using results of both laboratory studies and field data, suggest that the production of coccolithophores, which represent about 1/3 of marine calcium carbonate production, actually increases under high CO2 concentrations and that the pessimistic assessments of the effects of ocean acidification from other studies may be exaggerated. Another study also suggests that Holocene data implies a much weaker response of pH to rising atmospheric CO2 concentrations than that implied by chemical models. However, others caution that the protocols used for arriving at these contrary results appear to be flawed. Furthermore, these conclusions are inconsistent with results from numerous other studies, and thus may be erroneous826-830.
6.4 Cryosphere and sea levels 6.4.1 Permafrost and hydrates Transient model studies project that, by 2100, the global extent of permafrost may already have been reduced by more than 20%, but that the response will not yet have reached an equilibrium state. Deep permafrost south of 70°N and shallow permafrost north of 70°N will likely survive the projected warming of the next century. The active layer above the permafrost is projected to increase in depth by 0.3 to 0.8 meters. In the southern MacKenzie Basin, this increase would enhance the active depth by up to 40%. The rate of permafrost decay in this region may be further affected by amplified regional warming associated with sea ice retreat in adjacent Arctic waters, and by local factors such as changes in winter snow depth, topography, and response of surface vegetation to warmer climates. The presence of relics of permafrost in some regions of northern Canada that appear to have survived a number of interglacials over the past three-quarters of a million years suggest that such local effects can help reduce the effects of sustained surface warming on permafrost. Where the permafrost does decay, liquid moisture in the soils increase and land surfaces become unstable. Large scale changes in landscape and hydrology result. Permafrost decay can also result in the release of significant amounts of old carbon stored in the permafrost layer, as well as methane, into the atmosphere. This release of old carbon and methane will be partially offset by increased ecological productivity and related carbon sinks76,239,240,528,595,831-837.
Frozen hydrates in the ocean floor may also be vulnerable to decay under warmer climates. A recent study involving an earth system model coupled to a simple ocean floor hydrate model suggests large decay of hydrates could already occur at current atmospheric CO2 levels. Within the next 200 years, warmer ocean temperatures could cause significant releases of methane from ocean floor hydrates in shallow waters. If the release is slow and only 10% of the hydrate methane reaches the atmosphere, it would be oxidized into CO2, producing a potent climate feedback of 10%. If releases are abrupt, the sudden pulse of potent methane gas directly released into the atmosphere would have a much more significant short term climatic effect838.
6.4.2 Land ice and sea level rise There continues to be uncertainty about the net rate of both Greenland and Antarctic ice sheet response to warmer climates, partly because of the poor understanding of the processes that affect ice flow at the base and margins of the ice sheets. One recent ice sheet model study suggests that the Greenland ice sheet will decline in volume over the next few centuries, but may gradually recover over millennia provided future global warming remains modest. However, if accumulated CO2 emissions from human activities exceed 2500 GtC, warming will likely exceed the critical threshold for total ice sheet collapse over subsequent millennia. Most of the melt water released from the Greenland ice sheet remains in the Atlantic basin for 50 years, thus delaying slightly the impact on sea level rises around the small island states in the tropical Pacific839-841.
Glacier ice volume in the European Alps will likely be a small fraction of their pre-industrial state by 2100, and could completely disappear if regional warming exceeds 5°C. A similar rapid decline projected for Himalayan glaciers will not only add to sea levels but will seriously impact water flows in regional rivers and the peoples that depend on this water. Reduced ice mass of glaciers located over active magma zones, such as those in Iceland, may also increase the risk of enhanced volcanic activity842-844.
The magnitude of future sea level rise in response to thermal ocean expansion and land ice volume change remains uncertain and controversial. Some model simulations suggest that net land ice melt contribution by 2100 will be only 5 cm, much lower than that projected by the IPCC in its latest assessment report. Others suggest glaciers alone could add 10 to 25 cm by 2100. Semi-empirical model studies, using a global temperature-sea level rise relationship, suggest the total rise from all sources by 2100 could be between 0.5 to 1.4 m, much greater than the recent IPCC estimates. Investigations using physical constraints of ice flow dynamics imply that sea level rise during the next century could not exceed 2 m, but that a 0.8 rise would be plausible. New paleoclimate evidence of sea level rise during the peak of the last interglacial, when global temperatures were similar to that projected for the late 21st century, also indicate that mean rates of sea level rise at that time were about 1.6 m/century. While these higher projections have been met with some skepticism, recent trends in temperature and sea level rise appear to support the argument that the IPCC projections are more likely to be underestimates than overestimates606,845-854.
Figure 15. Observed and projected changes in sea level between 1850 and 2100. Projections use a semi-empirical approach based on best statistical fit of past sea level change with global mean temperature, which is applied to the global mean temperature projections of the IPCC TAR. The gray uncertainty range spans the range of temperature rise of 1.4° to 5.8° C. The dashed gray lines show the estimated added uncertainty due to the statistical error of the statistical fit used to generate the projections. Colored dashed lines are the individual temperature scenarios; ranging from the high SRES A1F1 forcing scenario (light blue line) to the low B1 forcing scenario (yellow line)(Rahmstorf, 2007, ref. #847). Greatest impacts of sea level rise will be in low lying deltas and along sandy shorelines where erosion will exacerbate the rate of land loss. Eroding shores will also cause a redistribution of coastal sediments, impacting deltas and coastal reefs. While efforts to constrain the magnitude of global warming and the adoption of adaptive measures can reduce the impacts of such rises in sea level, significant damages cannot be avoided. Small island states and deltas are most vulnerable. Risks of damage rise dramatically with the magnitude of sea level rise and with increased wave heights associated with more intense tropical storms. They are also higher where local capacity to adapt to sea level rise is constrained by poverty. Changes in storm tracks could also result in exacerbated shoreline erosion from sea level rise in regions where storms become more active. However, erosion would likely decrease where storm activity abates855-858.
While individual model simulations vary considerably in their projections for future sea ice conditions, many now simulate an ice free Arctic in late summer by 2100. Some models project ice free conditions in late summer as soon as 2040. Summer sea ice loss in the marginal seas of the Arctic basin, as well as in the Bering, Okhotsk and Barents Seas could exceed 40% by 2050. Ensemble model results, which replicate current Arctic sea ice behavior well, suggest average annual ice cover could decrease by between 22 and 33% by 2090, with almost all of the multi-year ice having disappeared by the end of the century. Thus the seasonal cycle for ice cover would increase dramatically. Recent observations support the argument that the Arctic ice cover is much more vulnerable than previously thought. A key factor in potentially abrupt ice cover reduction will be increased ocean heat transport into the Arctic Ocean310,317,332,859-861.
Detailed model studies into the future fate of ice within the confines of the Canadian Archipelago suggest this region may be one of the last refuges for multi-year ice. Land fast ice thickness in the region is projected to decline by 50 cm by 2100, and annual ice cover duration to be reduced by two months862.
Such changes in Arctic ice cover appear to be unprecedented within the past million years. Hence there are no good analogues to help understand the net effects on the physical Arctic system, or the ecological life and the cultures of northern people that depend on it. Potential consequences include altered Arctic hydrology and surface energy budgets, and changes in atmospheric circulation that could affect precipitation patterns across North America863-864.
6.5 Extremes 6.5.1 Temperature and Precipitation Although models differ in their projections of the rate of future temperature change, all consistently indicate that extreme temperature events will change much more dramatically than mean temperatures. On average, they project that, by mid-century, a one in twenty year warm event will likely occur every 10 years in high latitudes and every 5 years in mid latitudes, and much more often in tropical regions. Once CO2 concentrations double, hot days that now occur once every 100 days could increase to 20-30 days per 100 days. Important factors include decreased soil moisture (which reduces evaporative heat loss), fewer wet days and reduced night-time cooling. By 2100, the 1 in 100 year temperature extremes will exceed 40°C in southern Europe and the southern United States, and reach 50°C in parts of the tropics. Cold extremes are projected to warm even more rapidly. For example, the frequency of extreme cold outbreaks in most areas of mid to low latitudes in the Northern Hemisphere is projected to decrease by 50 to 100% over the next century. Models also suggest that such cold outbreaks will weaken over warm oceans and ice margins, although only slightly over the Nordic Sea region. Although weaker outbreaks over the Atlantic Ocean tend to weaken ocean circulation, this may be offset by a reduction in precipitation associated with these outbreaks292,865-870.
Climate models agree that average precipitation intensity will increase globally, although there is disagreement on changes in intensity at the regional scale. In general, model results suggest that, by 2100, the average return period of extreme precipitation events will be one-third that of today. The frequency of the most extreme (greater than the 99th percentile) events will increase significantly more than more moderate (90th percentile) events300,865,870.
Figure 16. Projected changes by latidude zone in the time interval between extreme precipitation events that now occur once every 20 years (Kharin et al., 2007, ref. # 300). A Canadian study also notes that freezing rain events will become more frequent for some regions of central Canada during cold months of winter, particularly northern Ontario. However, for southern Ontario, they are projected to decrease during early and late winter periods871.
More details on precipitation extremes are provided in section 6.1.
6.5.2 Tropical and extra tropical cyclones Projections for future response of tropical cyclone activity to warmer climates vary significantly from study to study. Most experts agree that, everything else being equal, it would be surprising if their activity did not increase as SSTs increase. Furthermore, there would be a higher risk of increased storm intensity in regions that warm more than the mean, and vice versa. However, higher temperatures may also increase the role of wind shear in tropical cyclone development. Since wind shear over the tropical Atlantic is likely to increase if the regional pattern of future warming becomes El Niño-like, total number of tropical cyclones over the Atlantic may actually decrease. Despite this, the intensity of winds and associated rainfall for those Atlantic cyclones that do develop into strong hurricanes may increase. One estimate suggests a 39% increase in the number of intense storms for each degree of rise in SSTs. Rising sea levels would further aggravate the impacts of such storms on coastal regions. Several studies also suggest that there will be more frequent and more intense tropical storms in the northwest Pacific Ocean, but a decline in such storms in the Southern Hemisphere and over the northern Indian Ocean634,635,668,872-877.
Model projections for changes in sea level pressures are generally consistent with a decrease in the total number of mid-latitude winter storms over the next century, but an increase in the number of intense storms. Canadian model simulations, using a high resolution atmospheric component, projects similar results for average mid-latitude autumn storm activity over the North Atlantic by 2050. These simulations also suggest that these storms are likely to move across the North Atlantic faster, and that the storm track will be somewhat wider and further north than today878-882.
6.6 Societal impacts 6.6.1 Health Globally, the impact of climate change on human health will be greatest for poor populations that have minimal adaptive capacities. Scientists caution that these socio-economic-health linkages need to be better understood in order to develop appropriate adaptive strategies883-884.
Within North America, warmer climates are expected to decrease background surface ozone levels, partly due to interaction with higher levels of water vapour. Hence there will be little change in related health risks in rural regions upwind of major polluting sources. However, in heavily industrialized regions, increased production of ozone pre-cursors combined with warmer temperatures projected for 2050 are expected to increase the annual frequency of days with high ozone concentrations above acceptable thresholds. For example, such events could increase in the eastern United States by up to 12 days. A similar study for four major cities in central and eastern Canada indicates that local high ozone days could increase by 40-100%. These increases would in turn cause a rise in related mortality rates and risks of cancer. Scavenging of dust and aerosols by precipitation may increase, partially alleviating these effects. Reduced dry deposition of ozone due to effects of elevated CO2 concentrations on plant stomatal conductance could aggravate them. There could also be increased emissions of ozone and its pre-cursors from wild fires765,885-891.
Rising temperatures are expected to enhance the rate of transmittal of viruses such as the West Nile virus and Lyme disease into cooler regions, as well as the risks of epidemics. For example, by 2080, temperatures throughout most of the southern parts of central and eastern Canada will be suitable for the expansion of the range for the tick that transmits Lyme disease. The risk of food-borne stomach ailments also appears to increase with rising temperatures892-894.
In general, mortality rates linked to extreme cold events depend on how unusual such events are, while those for extreme heat events depends on the severity. Therefore, increased mortality due to more frequent heat waves is not likely to be significantly offset by reduced mortality from fewer cold extremes895.
6.6.2 Social-Economic Historical studies suggest that social unrest tends to follow significant changes in regional climate and related impacts on food and natural resources. For example, the cooling during the Little Ice Age impeded agricultural production and contributed to major social problems, inflation, war and population decline. Water scarcity due to drought can also be important in this regard. Due to these historical precedences, some argue that societal impacts of future climate change may be greater than anticipated. Many tropical countries, where human development indexes are low, are particularly vulnerable. However, others counter that there is little reliable data to substantiate such linkages, and that political and economic factors are likely far more important in triggering social unrest than resource scarcity. They agree, however, that these linkages need to be better understood in order to assess and mitigate related risks329,896-900.
One study suggests that, if future sea level rise can be restricted to 25 cm, impacts on the national economies of coastal countries around the world will likely be modest. One of the more vulnerable economic sectors will be tourism. Most significantly affected will be small island states, which could lose 0.5% of their GDP. Some coastal countries could see increased tourism and hence actually benefit from a modest increase in sea levels901.
Traditional knowledge, strong social networks and flexibility in hunting patterns are all factors that could help Inuit communities adapt to dramatic changes in climate projected for Nunavut. However, the added effects of other stresses on the natural ecosystems, particularly the quality and abundance of freshwater, may significantly increase the socio-economic risks associated with climate change alone. Furthermore, a rapidly changing culture is undermining the traditional adaptive capacity in Canada’s north. These complex relationships need to be better understood in order to properly assess how climate change will affect food supply and security in the north. In the Canadian Prairies, one of the key challenges facing communities will be the water supply crisis that is likely to emerge from the combined effects of climate change and human mismanagement of resources902-905.
Warmer winters will reduce the economic viability of the outdoor winter sports industry across much of southern Canada. Major ski centers in regions such as southern Quebec could become economically unviable by the 2050s. Offsetting this will be longer summer recreational seasons and related tourism. In the Greater Toronto Area, for example, the number of rounds of golf played each year could increase by more than 70% if appropriate adaptive measures are implemented. Tourism in national parks is also likely to benefit from longer seasons, although changes in landscapes (e.g. disappearing glaciers) and wildfire risks could eventually offset some of these benefits906-908.
Longer ice free seasons could also significantly extend shipping seasons in rivers such as the MacKenzie River, although volume could be affected in mid-summer by reduced flow rates909.