The human imperative of stabilizing global climate change at 1.5°C

See allHide authors and affiliations

Science  20 Sep 2019:
Vol. 365, Issue 6459, eaaw6974
DOI: 10.1126/science.aaw6974

The need to stabilize global climate

Climate change will be the greatest threat to humanity and global ecosystems in the coming years, and there is a pressing need to understand and communicate the impacts of warming, across the perspectives of the natural and social sciences. Hoegh-Guldberg et al. review the climate change–impact literature, expanding on the recent report of the Intergovernmental Panel on Climate Change. They provide evidence of the impacts of warming at 1°, 1.5°, and 2°C—and higher—for the physical system, ecosystems, agriculture, and human livelihoods. The benefits of limiting climate change to no more than 1.5°C above preindustrial levels would outweigh the costs.

Science, this issue p. eaaw6974

Structured Abstract


The United Nations Framework Convention on Climate Change (UNFCCC) was established in 1992 to pursue the “stabilization of greenhouse gas concentrations at a level that would prevent dangerous anthropogenic interferences with the climate system.” Since 1992, five major climate change assessment cycles have been completed by the UN Intergovernmental Panel on Climate Change (IPCC). These reports identified rapidly growing climate-related impacts and risks, including more intense storms, collapsing ecosystems, and record heatwaves, among many others. Once thought to be tolerable, increases in global mean surface temperature (GMST) of 2.0°C or higher than the pre-industrial period look increasingly unmanageable and hence dangerous to natural and human systems.

The Paris Climate Agreement is the most recent attempt to establish international cooperation over climate change. This agreement, ratified or acceded to by 185 countries, was designed to bring nations together voluntarily to take ambitious action on mitigating climate change, while also developing adaptation options and strategies as well as guaranteeing the means of implementation (e.g., climate finance). The Agreement is aimed at “holding the increase in the global average temperature to well below 2.0°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C above pre-industrial levels, recognizing that this would significantly reduce the risks and impacts of climate change.” Many unanswered questions regarding a 1.5°C target surround the feasibility, costs, and inherent risks to natural and human systems. Consequently, countries invited the IPCC to prepare a Special Report on “the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty.” The Special Report was completed and approved by the 48th Session of the IPCC in October 2018.


Multiple lines of evidence indicate that the next 0.5°C above today (which will take GMST from 1.0°C to 1.5°C above the pre-industrial period) will involve greater risks per unit temperature than those seen in the last 0.5°C increase. This principle of “accelerating risk” is also likely to drive proportionally and possibly exponentially higher risk levels in the transition from 1.5°C to 2.0°C above the pre-industrial period. We argue that this is a consequence of impacts accelerating as a function of distance from the optimal temperature for an organism or an ecosystem process. Coral reefs, for example, often appear healthy right up until the onset of mass coral bleaching and mortality, which can then destroy a reef within a few months. This also explains the observation of “tipping points” where the condition of a group of organisms or an ecosystem can appear “healthy” right up to the point of collapse, suggesting caution in extrapolating from measures of ecosystem condition to predict the future. Information of this nature needs to be combined with an appreciation of organisms’ distance from their optimal temperature.

Finally, we explore elements of the costs and benefits associated with acting in response to climate change, and come to the preliminary conclusion that restraining average global temperature to 1.5°C above the pre-industrial period would be much less costly than the damage due to inaction on global climate change.


As an IPCC expert group, we were asked to assess the impact of recent climate change (1.0°C, 2017) and the likely impact over the next 0.5° to 1.0°C of additional global warming. At the beginning of this exercise, many of us were concerned that the task would be hindered by a lack of expert literature available for 1.5°C and 2.0°C warmer worlds. Although this was the case at the time of the Paris Agreement, it has not been our experience 4 years later. With an accelerating amount of peer-reviewed scientific literature since the IPCC Special Report Global Warming of 1.5°C, it is very clear that there is an even more compelling case for deepening commitment and actions for stabilizing GMST at 1.5°C above the pre-industrial period.

Climate change and nonlinear responses.

(A) Reef-building corals can bleach, losing (B) dinoflagellate symbionts (~10 mm across) and dying, thus exhibiting (C) a nonlinear response to impacts/risks from climate change. H (high) and VH (very high) are the confidence for transition from one impact/risk level to another: white, no climate change–related impacts; yellow, some detectable climate change impacts/risks; red, severe and widespread impacts/risks; purple, very high impacts/risks with significant irreversibility or persistence.



Increased concentrations of atmospheric greenhouse gases have led to a global mean surface temperature 1.0°C higher than during the pre-industrial period. We expand on the recent IPCC Special Report on global warming of 1.5°C and review the additional risks associated with higher levels of warming, each having major implications for multiple geographies, climates, and ecosystems. Limiting warming to 1.5°C rather than 2.0°C would be required to maintain substantial proportions of ecosystems and would have clear benefits for human health and economies. These conclusions are relevant for people everywhere, particularly in low- and middle-income countries, where the escalation of climate-related risks may prevent the achievement of the United Nations Sustainable Development Goals.

Climate change is one of the greatest challenges for humanity. Global mean surface temperature (GMST) is increasing at the rate of 0.2° ± 0.1°C per decade, reaching 1.0°C above the pre-industrial period (reference period 1850–1900) in 2017 (1). GMST is projected to reach 1.5°C above the pre-industrial period between 2030 and 2052, depending on the model and assumptions regarding projected changes to atmospheric greenhouse gas (GHG) levels and climate sensitivity (1). At the same time, growing awareness of impacts beyond 1.5°C has focused international attention on the feasibility and implications of stabilizing temperatures at this level (2).

In broad terms, limiting warming to 1.5°C will require an annual investment in the energy sector between 2016 and 2050 of $1.46 to $3.51 trillion (US$2010) in energy supply and $640 to $910 billion in energy demand measures in order to reach net zero GHG emissions by 2050 [(3), p. 154]. On the other hand, the mean net present value (in 2008) of the damages that would be avoided by 2200 by making these investments is estimated as totaling $496 trillion (US$2010) (35). This, together with other damages that are difficult to quantify (e.g., disruption and migration of human communities; reductions in ecosystem services associated with biodiversity loss), suggests that the potential economic benefits arising from limiting warming to 1.5°C may be at least four or five times the size of the investments needed in the energy system until 2050 (see supplementary materials) (3).

Here, we explore the near-term, mostly unmonetized impacts projected for 1.5°C of global warming, along with the associated risks and adaptation options for natural and human (managed) systems. To better understand the implications of reaching 1.5°C, we compare it to recent conditions (i.e., 1.0°C warming above the pre-industrial period; Fig. 1) and to the conditions that are projected to emerge as we approach 2.0°C of warming. This comparison helps to clarify the benefits (or lack of benefits) of stabilizing GMST at 1.5°C as compared to 2.0°C or higher, as well as providing a framework for societal responses and consequences.

Fig. 1 Impacts are already serious at 1.0°C of global warming.

Increases in global mean surface temperature (GMST) of 1.0°C have already had major impacts on natural and human systems. (A) Increased temperatures and dryness in the Mediterranean region are driving longer and more intense fire seasons with serious impacts on people, infrastructure, and natural ecosystems. Image shows tragic devastation of fire in the Greek village of Mati, 25 July 2018. (B) Evidence of ice sheet disintegration is increasing [here showing a 30-km fracture across the Pine Island Glacier which is associated with the Western Antarctic Ice Sheet (WAIS)]. The fracture (see arrow) appeared in mid-October 2011 and has increased concern that we may be approaching a tipping point with respect to disintegration of the WAIS. (C) Many low-lying countries such as the Maldives experience flooding and will be at an increased threat from sea level rise and strengthening storms over time. (D) Many insects and birds have shifted reproductive events or migration to early times in the season as conditions have warmed. [Image credits: (A) “Lotus R,”; (B) NASA/GSFC/METI/ERSDAC/JAROS and U.S./Japan ASTER Science Team; (C) male, Maldives (O. Hoegh-Guldberg); (D) semipalmated sandpiper (Calidris pusilla), Creative Commons (CC BY-SA 3.0, GNU Free Documentation License)]

Crossing the 1.0°C threshold has already had severe impacts on natural and human systems

The incidence of extremes has increased sharply as GMST has warmed from 0.5°C to 1.0°C (~1980 to 2018) relative to the pre-industrial period, with the intensity and/or frequency of extremes projected to change further with another 0.5°C of warming (5). As GMST has increased, for example, the average temperature of cold days and nights (i.e., the coldest 10%) has also increased overall, as has the average temperature of warm days and nights (i.e., the warmest 10%) globally (5). These changes have also been accompanied by increases in the frequency and/or duration of heatwaves for large parts of Europe, North America, and Australia. Increases in GMST have been accompanied by increases in the frequency, intensity, and/or amount of heavy precipitation in more regions than those with decreases, especially in Northern Hemisphere mid-latitude and high-latitude areas (5, 6). There is also evidence of increasing rainfall associated with recent tropical cyclones (6, 7) and increasingly heavy precipitation during storms in the Central Sahel (8, 9). The total annual number of tropical cyclones has decreased, while the number of very intense cyclones has increased, for many areas (5). There is less confidence regarding trends in the length of drought, although a significant increasing trend has been detected in the Mediterranean region (particularly southern Europe, North Africa, and the Near East) (1012).

As on land, coastal and marine habitats have also experienced an increased frequency, intensity, and duration of underwater heatwaves, with a factor of 3 increase in the number of marine heatwave days globally since 1980 (13). The differential heating of the water column has also led to increased thermal stratification in some coastal and oceanic regions, which decreases ocean-atmosphere gas exchange as well as the turnover of nutrients between the photic and nonphotic layers of the ocean. The annual mean Arctic sea ice extent decreased by 3.5 to 4.1% per annum from 1979 to 2012 (6). The melting of land-based ice includes potentially unstable regions such as the Western Antarctic Ice Sheet (WAIS; Fig. 1B), which contributed 6.9 ± 0.6 mm to global mean sea level (GMSL) from 1979 to 2017. Together with glacial meltwater, thermal expansion of the ocean has accelerated the rate of GMSL increase by up to 0.013 (range, 0.007 to 0.019) mm year–2 since the early 20th century (14). Changes in ocean temperature have also decreased the oxygen concentration of the bulk ocean, interacting with coastal pollution to increase the number and extent of low-oxygen dead zones in many deep-water coastal habitats (15). In addition to increasing GMST, anthropogenic CO2 also enters the ocean, causing a reduction in pH (ocean acidification), which in turn has a negative impact on processes such as early development, calcification, photosynthesis, respiration, sensory systems, and gas exchange in organisms as different as algae and fish (5).

Changing weather patterns (e.g., temperature, rainfall, dryness, storms) have increased negative impacts on natural and managed systems (Fig. 1, A to D). Changes to coral reefs (5), forests (e.g., changing drought/fire regimes) (16, 17), and low-lying islands and coasts (5), as well as impacts on agriculture production and yield (18, 19), are threatening resources for dependent human communities. There are also many gradual changes that have occurred as GMST has increased, and many of these are no less important than the more abrupt changes. Land-based biomes (i.e., major natural and agricultural ecosystem types) have also shifted to higher latitudes and elevation in boreal, temperate, and tropical regions (5, 15), with similar shifts reported for marine and freshwater organisms. Marine organisms and some ecosystems have also shifted their biogeographical ranges to higher latitudes at rates up to 40 km/year. Rates are highest for pelagic organisms and ecosystems such as plankton, and are lowest for more sedentary benthic organisms and ecosystems such as seaweeds and kelp forests (5, 15). These types of changes (e.g., temperature, storms, circulation) have also affected the structure and function of ocean ecosystems with respect to biodiversity, food webs, disease incidence, and invasive species (5).

Other changes to biological systems include changes to the phenology of marine, freshwater, and terrestrial organisms (e.g., the timing of key events such as reproduction and migration) (5, 15). The phenology of plants and animals in the Northern Hemisphere, for example, has advanced by 2.8 ± 0.35 days per decade as a result of climate change, with similar changes in the flowering and pollination of plants and crops and in the egg-laying and migration times of birds (5, 20). There are indications that climate change has already contributed to observed declines in insects and arthropods in some regions (21, 22). Variations in these types of changes have also been observed in the phenology of tropical forests, which have been more responsive to changes in moisture stress than to direct changes in temperature (5). Although the intention here is not to catalog all of the changes that are occurring in natural systems, it is important to acknowledge that deep and fundamental changes are under way in biological systems with just 1.0°C of global warming so far (5).

Changes in GMST of 1.0°C have also directly and indirectly affected human communities, many of which depend on natural and managed systems for food, clean water, coastal defense, safe places to live, and livelihoods, among many other ecosystem goods and services (5). Coral reefs clearly illustrate the linkage among climate change, ecosystem services, and human well-being. At 1.0°C, large-scale mortality events driven by lengthening marine heatwaves have already reduced coral populations in many places (5), with prominent coral reef ecosystems such as the Great Barrier Reef in Australia losing as much as 50% of their shallow-water corals in the past 4 years alone (5, 23, 24). These changes have potential implications for millions of people who depend on coral reefs for food, livelihood, and well-being (5).

Understanding climate change over the next few decades: Methods and assumptions

There is a range of strategies for quantifying risks for natural and human systems at 1.5°C and 2.0°C above the pre-industrial period. This requires calculating the future exposure of systems to changes in climatic hazards. Some methods rely on the fact that an equivalent amount of warming (e.g., 0.5°C) occurred in the recent past (e.g., ~1950 to 2000, or ~1980 to 2018; Fig. 2A) (3), potentially providing insights into how risks might change in the near future. In this case, the associated risks of the next 0.5°C of global warming (Fig. 2A) are linearly extrapolated from the impacts associated with the previous 0.5°C increase (~1980 to 2018). This method of projecting future risk is likely to be conservative, given that (i) the pace of climate change is increasing (25) and (ii) the impacts per unit of global mean surface temperature are likely to increase as conditions are pushed increasingly beyond the optimal conditions for a particular organism, physiological process, or system (Fig. 2B) (26). Responses by natural and human systems are likely to also differ if temperature pathways involve a gradual increase to 1.5°C above the pre-industrial period, as opposed to pathways that first exceed 1.5°C before later declining to 1.5°C, which is referred to as an “overshoot” (5) (Fig. 2A). High levels of overshoot involve exceeding 1.5°C by 0.1°C or more (Fig. 2A) (3).

Fig. 2 There is still time to restrain global warming to 1.5°C.

(A) Action on climate change can still result in stable or even decreasing global temperatures, although variability around projections is substantial. Strategies that include “overshoot” (upper dashed line, illustrative of a very high level of overshoot) require as yet early stage technologies to ensure that overshoot is kept as short as possible. Also, the larger the overshoot, the higher the risk of irreversible change in affected systems. (B) Responses to changing conditions (shown here as a thermal performance curve) are typically tilted to the right with a steep decline in performance such as growth, toward high temperature extremes. Beyond a thermal optimum, Topt, performance begins to decline beyond the Pejus temperature, Tp. A critical temperature, Tc, characterizes a low level of performance and time-limited passive endurance when, as in ectothermic animals, oxygen supply capacity becomes insufficient to cover oxygen supply, or, as in corals, a symbiosis between corals and their dinoflagellate symbionts suddenly breaks down (coral bleaching) and corals go from appearing healthy to experiencing large scale mortality over days-to-weeks. Accordingly, the high Tc characterizes a temperature of high responsiveness to small increases in temperature extremes, such as by 0.5°C, especially if some life stages have a narrow thermal range indicating high vulnerability. [Redrawn from (26) with permission of Journal of Experimental Biology]

Other approaches for understanding how the world may change at 1.5°C and 2.0°C of global warming draw on information from sources such as laboratory, mesocosm, and field experiments. These approaches simulate projected conditions for different levels of warming and, in the case of marine systems, levels of acidification (e.g., changes in pH, carbonate, or pollution levels) (5, 26, 27). These experimental approaches also provide calibration as well as insight into future conditions and responses (i.e., 1.5°C versus 2.0°C). Some caution is also required given that global increases of 1.5°C or 2.0°C may involve a broad range of regional responses because of uncertainties in (for example) the likelihood of overshoot, land-atmosphere interactions, biophysical effects of land use changes, and interannual climate variability (28). Several lines of evidence for understanding these complex problems include the analysis of the frequency and intensity of extremes as well as projections based on existing climate simulations and empirical scaling relationships for 1.5°C and 2.0°C of global warming (5). Lines of investigation may also include dedicated experiments (for example) prescribing sea surface conditions consistent with these levels of warming, as done in the HAPPI (Half a Degree Additional Warming, Prognosis and Projected Impacts) project (5). Furthermore, fully coupled climate model experiments can be achieved using GHG forcing consistent with 1.5°C or 2.0°C scenarios (5). These multiple yet different lines of evidence underpin the development of qualitatively consistent results regarding how temperature means and extremes could change at 1.5°C as compared to 2.0°C of global warming.

Projected changes in climate at 1.5°C versus 2.0°C of global warming

Understanding the potential advantages of restraining global warming to 1.5°C requires an understanding of the risks associated with the exposure of natural and human systems to climatic hazards, and how these risks change at 1.5°C relative to 2.0°C (Fig. 3) (29). Increases of GMST to 1.5°C will further increase the intensity and frequency of hot days and nights, and will decrease the intensity and frequency of cold days and nights (Fig. 3, C to E). Warming trends are projected to be highest over land, in particular for temperature extremes, with increases of up to 3.0°C in the mid-latitude warm season and up to 4.5°C in cold seasons at high latitudes. These increases are projected to be greater at 2.0°C of global warming, with increases of up to 4°C in the mid-latitude warm season and up to 6°C in the high-latitude cold season (e.g., Fig. 3, A, C, D, and E) (29). Heatwaves on land, which are already increasing pressure on health and agricultural systems, are projected to become more frequent and longer (Fig. 3, C and D).

Fig. 3 Changes vary geographically at GMST of 1.5°C and 2°C.

(A to F) Projected changes are shown in mean temperature (A), mean precipitation (B), number of hot days (NHD; 10% warmest days) (C), temperature of hottest day (TXx) (D), temperature of coldest night (TNn) (E), and change in extreme precipitation (Rx5day) (F). Conditions are projected for 1.5°C (left-hand column) and 2.0°C (middle-hand column) of global warming compared to the pre-industrial period (1861–1880), with the difference between 1.5°C and 2.0°C of global warming being shown in the third column. Cross-hatching highlights areas where at least two-thirds of the models agree on the sign of change as a measure of robustness (18 or more out of 26). Values were assessed from the transient response over a 10-year period at a given warming level, based on Representative Concentration Pathway (RCP) 8.5 Coupled Model Intercomparison Project Phase 5 (CMIP5) model simulations (3, 5), adapted from (29, 73) [see supplementary material 3.SM.2 in (5)]. Figure is a composite of figure 3.3 [here, (A) and (B)], figure 3.7 [here, (C)], and figure 3.4 [here, (D) to (F)] in (5). [Use of figures with permission of the IPCC]

There is considerable evidence that dryness (overall availability of water) will increase in some regions, especially the Mediterranean as well as southern Africa (5, 3032). Risks of drought, dryness, and precipitation deficits are projected to increase at 1.5°C and even further at 2.0°C for some regions relative to the pre-industrial period (Fig. 3, B and F) (5, 33). Recent studies also suggest similar projections for the western Sahel and southern Africa, as well as the Amazon, northeastern Brazil, and central Europe (5, 34). Projected trends in dryness are uncertain in several regions, however, and some regions are projected to become wetter (Fig. 3, B and F) (5). Reaching GMST of 1.5°C and 2.0°C, for example, would lead to a successive increase in the frequency, intensity, and/or amount of heavy rainfall when averaged over global land area (Fig. 3, B and F). Global warming of 2.0°C versus 1.5°C would increase exposure to fluvial flood risk, particularly at higher latitudes and in mountainous regions, as well as in East Asia, China (35), and eastern North America overall (5). The prevalence of subsequent intense wet and dry spells, in which a prolonged drought is immediately followed by heavy precipitation at the same location (potentially leading to flooding) or vice versa, is projected to be greater at 2.0°C global warming versus 1.5°C (36). These large changes between coupled wet and dry conditions represent a major challenge for adaptation because they will affect water quality and availability as well as increase the rate of soil erosion along many coastal areas. Sea level rise can also amplify problems through damage to coastal infrastructure and the salinization of water supplies for drinking and agriculture (5).

Relatively few studies have directly explored the effect of 1.5°C versus 2.0°C of global warming on tropical cyclones (5). These studies consistently reveal a decrease in the global number of tropical cyclones at 1.5°C versus 1.0°C of global warming, with further decreases under 2.0°C versus 1.5°C of global warming. Simultaneously, very intense cyclones are likely to occur more frequently at 2.0°C versus 1.5°C of global warming, with associated increases in heavy rainfall and damage, further emphasizing the advantages of not exceeding 1.5°C (5).

Coastal and oceanic regions are also projected to increase in temperature as GMST increases to 1.5°C, and further to 2.0°C, above the pre-industrial period. Absolute rates of warming are only slightly lower in the ocean than on land, although the shallower spatial gradient of ocean temperature will mean that the velocity of climate change may be higher in many regions of the ocean (5, 37). Increases in ocean temperature associated with 1.5°C and 2.0°C of global warming will increase the frequency and duration of marine heatwaves, as well as reducing the extent of ocean mixing due to the greater thermal stratification of the water column (13, 15). Sea ice is projected to continue to decrease in the Arctic, although restraining warming to 1.5°C will mean that an ice-free Arctic summer will only occur every 100 years, whereas warming to 2.0°C above the pre-industrial period will mean that an ice-free Arctic summer is likely to occur every 10 years by 2100 (5, 38). These and other models indicate that there will be no long-term consequences for sea ice coverage in the Arctic (i.e., no hysteresis) if GMST is stabilized at or below 1.5°C (3).

Impacts on ecosystems at 1.5°C versus 2.0°C of global warming

Multiple lines of evidence (5) indicate that reaching and exceeding 1.5°C will further transform both natural and human systems, leading to reduced ecosystem goods and services for humanity. Risks for terrestrial and wetland ecosystems—such as increasing coastal inundation, fire intensity and frequency, extreme weather events, and the spread of invasive species and diseases—are lower at 1.5°C as compared to 2.0°C of global warming (5). In this regard, the global terrestrial land area that is predicted to be affected by ecosystem transformations at 2.0°C (13%, interquartile range 8 to 20%) is approximately halved at 1.5°C to around 6.5% (the area affected at 1.0°C warming is 4%, interquartile range 2 to 7%). Risks for natural and managed ecosystems are higher on drylands as compared to humid lands (5). The numbers of species that are projected to lose at least half of their climatically determined geographic range at 2.0°C of global warming (18% of insects, 16% of plants, 8% of vertebrates) would be substantially reduced at global warming of 1.5°C (i.e., to 6% of insects, 8% of plants, and 4% of vertebrates) (5). In this regard, species loss and associated risks of extinction are much lower at 1.5°C than at 2°C. Tundra and boreal forests at high latitudes are particularly at risk, with woody shrubs having already encroached on tundra, which will increase with further warming (5). Constraining global warming to 1.5°C would reduce risks associated with the thawing of an estimated 1.5 to 2.5 million km2 of permafrost (over centuries) compared to the extent of thawing expected at 2.0°C (5).

Ecosystems in the ocean are also experiencing large-scale changes, with critical thresholds projected to be increasingly exceeded at 1.5°C and higher global warming. Increasing water temperatures are driving the relocation of many species (e.g., fish, plankton); sedentary organisms, such as kelp and corals, are relatively less able to move. In these cases, there are multiple lines of evidence indicating that 70 to 90% of warm-water tropical corals present today are at risk of being eliminated even if warming is restrained to 1.5°C. Exceeding 2.0°C of global warming will drive the loss of 99% of reef-building corals (5). These nonlinear changes in survivorship are a consequence of the increasing impact of changes as they move away from optimal conditions (Fig. 2B) (26). Impacts on oceanic ecosystems are expected to increase at global warming of 1.5°C relative to today, with losses being far greater at 2.0°C of global warming. Large compound or secondary risks exist with respect to declining ocean productivity, loss of coastal protection, damage to ecosystems, shifts of species to higher latitudes, and the loss of fisheries productivity (particularly at low latitudes) (15). There is substantial evidence that these coastal risks will increasingly threaten the lives and livelihoods of millions of people throughout the world (5).

Increasing risks for human (managed) systems at 1.5°C and 2.0°C of global warming

Many risks for society will increase as environmental conditions change. Water, for example, is central to the success or failure of human communities. The projected frequency and scale of floods and droughts in some regions will be smaller under 1.5°C global warming as opposed to 2°C, with risks to water scarcity being greater at 2.0°C than at 1.5°C of global warming for many regions (5). Salinization of freshwater resources on small islands and along low-lying coastlines is a major risk that will become successively more important as sea levels rise, particularly as they will continue to increase even if temperatures stabilize (5). Depending on future socioeconomic conditions, limiting warming to 1.5°C is projected to reduce the proportion of the world’s population exposed to climate-induced water stress by up to 50% relative to 2°C warming (5), although there is considerable variability among regions, as already discussed. Most regions, including the Mediterranean and Caribbean regions, are projected to experience substantial benefits from restraining global warming to 1.5°C (39), although socioeconomic drivers are expected to play a dominant role relative to climate change for these communities over the next 30 to 40 years.

Limiting global warming to 1.5°C is projected to result in smaller reductions in the yield of maize, rice, wheat, and potentially other cereal crops than at 2.0°C, particularly in sub-Saharan Africa, Southeast Asia, and Central and South America (4042). A loss of 7 to 10% of rangeland stock globally is also projected to occur at an increase of 2.0°C above the pre-industrial period, which will have considerable economic consequences for many communities and regions. Reduced food availability at 2.0°C as compared to 1.5°C of global warming is projected for many regions including the Sahel, southern Africa, the Mediterranean, central Europe, and the Amazon. Few examples exist where crop yields are increasing, and hence food security is at increasing risk in many regions (41). Food systems in future economic and trade environments may provide important options for mitigating hunger risk and disadvantage (5, 43, 44), especially if solutions are found to challenges such as the decline in the nutritional quality of major cereal crops from higher CO2 concentrations (5).

Food production from marine fisheries and aquaculture is of growing importance to global food security but is facing increasing risks from ocean warming and consequently ocean acidification (5). These risks increase at 1.5°C of global warming and are projected to have an impact on key organisms such as finfish, corals, crustaceans, and bivalves (e.g., oysters), especially at low latitudes (5). Small-scale fisheries that depend on coastal ecosystems such as coral reefs, seagrass, kelp forests, and mangroves are expected to face growing risks at 1.5°C of warming as a result of the loss of habitat (5). Risks of impacts, and subsequent risks to food security, are projected to become greater as global warming reaches 1.5°C (5, 43, 44). Tropical cyclones have major impacts on natural and human systems and are projected to increase in intensity in many regions, with the damage exacerbated by rapid sea level rise (14, 45). The tropical cyclones in the North Atlantic basin in 2017 had major, widespread effects on the small islands of Caribbean as well as the United States, resulting in many deaths, displacement of communities, elevated rates of morbidity and mental health issues, as well as the long-term loss of electricity generation and distribution. These impacts have resulted in major economic damage, which has exceeded the annual GDP of some small island developing states (46, 47).

Millions of people are already exposed to coastal flooding due to sea level rise and storms, particularly in cities. Projections of sea level rise remain uncertain (5) and may include large nonlinear responses, in part attributable to the contribution of land-based ice (4850). Because of the time lag between increased emissions and higher sea levels, differences in mitigation at 1.5°C and 2.0°C are small relative to the uncertainty in the projections at 2050 or even 2100. Small differences can, however, have big impacts: An increase of 0.1 m of sea level rise, for example, will expose an additional 10 million people to flooding by 2100 (5), particularly those living in low-lying deltas and small islands (5, 51). Even with mitigation, adaptation remains essential, particularly as multi-meter sea level rise remains possible over several centuries for higher levels of temperature rise (5). Estimates of the net present value in 2008 of global aggregate damage costs (which would be incurred by 2200 if global warming is limited to 2.0°C) reach $69 trillion (5) but are reduced to $54 trillion if global warming is limited to 1.5°C. Damages from sea level rise alone would contribute several trillion dollars annually for a 2°C constrained scenario (52).

Warming of 1°C has increased the frequency and scale of impacts on human health through changes in the intensity and frequency of heatwaves, droughts, floods, and storms, as well as impacts on food quantity and nutritional quality (through increasing CO2 concentrations) resulting in undernutrition or malnutrition in some regions (5, 43, 44). Multiple lines of evidence indicate that any further increases in GMST could have negative consequences for human health, mainly through the intensification of these risks (5, 53). Lower risks are projected at 1.5°C than at 2.0°C of global warming for heat-related morbidity and mortality, as well as for ozone-related mortality if ozone precursor emissions remain high. Limiting global warming to 1.5°C would result in 420 million fewer people being frequently exposed to “extreme heatwaves” [defined by duration and intensity (54)] and about 65 million fewer people being exposed to “exceptional” heatwaves as compared to conditions at 2.0°C GMST warming (55). Human health will also be affected by changes in the distribution and abundance of vector-borne diseases such as dengue fever and malaria, which are projected to increase with warming of 1.5°C and further at 2.0°C in most regions (5). Risks vary according to human vulnerability, development pathways, and adaptation effectiveness (43, 44, 56). In some cases, human activities can lead to local amplification of heat risks from urban heat island effects in large cities (57, 58). More specific impacts of, and solutions to, climate change on cities are discussed elsewhere (43, 56).

Global warming of 1.5°C will also affect human well-being through impacts on agriculture, industry, and employment opportunities. For example, increased risks are projected for tourism in many countries, whereby changes in climate have the potential to affect the attractiveness and/or safety of destinations, particularly those dependent on seasonal tourism including sun, beach, and snow sport destinations (5, 15). Businesses that have multiple locations or markets may reduce overall risk and vulnerability, although these options are likely to be reduced as stress and impacts increase in frequency and areal extent. Risks and adaptation options may lie in developing alternative business activities that are less dependent on environmental conditions. These risks become greater as warming increases to 2.0°C and pose serious challenges for a large number of countries dependent on tourism and related activities for national income (5).

Multiple lines of evidence reveal that poverty and disadvantage are also correlated with warming to 1.0°C above the pre-industrial period, with the projection of increasing risks as GMST increases from 1.0°C (today) to 1.5°C and higher (43, 44). In this regard, out-migration from agriculturally dependent communities is positively correlated with global temperature, although our understanding of the links between human migration and further warming of 1.5°C and 2.0°C is at an early stage (5). Similarly, risks to global aggregate economic growth due to climate change impacts are projected to be lower at 1.5°C than at 2.0°C by the end of the century (5). The largest reductions in economic growth at 2.0°C compared to 1.5°C are projected for low- and middle-income countries and regions (the African continent, Southeast Asia, India, Brazil, and Mexico). Countries in the tropics and Southern Hemisphere subtropics are projected to experience the largest negative impacts on economic growth if global warming increases from 1.5°C to 2.0°C above the pre-industrial period (5, 43, 44). The most perceptible impacts of climate change are likely to occur in tropical regions as GMST increases to 1.5°C and eventually to 2.0°C above the pre-industrial period (59).

Table 1 summarizes the emergence of potential climate change “hotspots” (i.e., areas where risks are large and growing rapidly) for a range of geographies and sectors (5). In all cases, these vulnerable regions show increasing risks as warming approaches 1.5°C and higher. Not all regions, however, face the same challenges. In the Arctic, for example, habitat loss is paramount, while changing temperature and precipitation regimes represent primary risks in the Mediterranean, southern Africa, West Africa, and the Sahel. These rapidly changing locations represent interactions across climate systems, ecosystems, and socioeconomic human systems, and are presented here to illustrate the extent to which risks can be avoided or reduced by achieving the 1.5°C global warming goal (as opposed to 2.0°C).

Table 1 Climate change “hotspots” are expanding.

Emergence and intensity of climate change “hotspots” under different degrees of global warming [summary, updated, table 3.6 in (5)]. Calibrated uncertainty language is as defined by the Intergovernmental Panel on Climate Change (3). SIDS, small island developing states.

View this table:

Trajectories toward hotspots can also involve nonlinearities or tipping points. Tipping points refer to critical thresholds in a system that result in rapid systemic change when exceeded (5). The risks associated with 1.5°C or higher levels of global warming reveal relatively low risks for tipping points at 2.0°C, but a substantial and growing set of risks as global temperature increases to 3°C or more above the pre-industrial (Table 2) (5). For example, increasing GMST to 3°C above the pre-industrial period substantially increases the risk of tipping points such as permafrost collapse, Arctic sea ice habitat loss, major reductions in crop production in Africa as well as globally, and persistent heat stress that is driving sharp increases in human morbidity and mortality (Table 2) (5).

Table 2 Increased warming increases risks of exceeding tipping points.

Summary of enhanced risks in the exceedance of regional tipping points under different global temperature goals [summary, table 3.7 in (5), not intended to be exhaustive].

View this table:

Solutions: Scalability, feasibility, and ethics

GMST will increase by 0.5°C between 2030 and 2052 and will multiply and intensify risks for natural and human systems across different geographies, vulnerabilities, development pathways, as well as affect adaptation and mitigation options (1, 43, 44, 56). To keep GMST to no more than 1.5°C above the pre-industrial period, the international community will need to bring GHG emissions to net zero by 2050 while adapting to the risks associated with an additional 0.5°C being added to GMST (3, 5) The impacts associated with limiting warming to 1.5°C, however, will be far less than those at 2.0°C or higher (Tables 1 and 2). Aiming to limit warming to 1.5°C is now a human imperative if escalating risks of dangerous if not catastrophic tipping points and climate change hotspots are to be avoided (2, 5).

An important conclusion of the IPCC Special Report is that limiting GMST to 1.5°C or less is still possible (3, 60). This will require limiting GHG emissions to a budget of 420 Gt CO2 for a 66% or higher probability of not exceeding 1.5°C (44). As global emissions are currently around 42 Gt CO2 per year, pathways should bring CO2 emissions to net zero over the next few decades (i.e., phase out fossil fuel use) alongside a substantial reduction (~35% relative to 2010) in emissions of methane and black carbon over the same time scale (44). The current set of national voluntary emission reduction pledges [nationally determined contributions (NDCs)], however, will not achieve the goals of the Paris Agreement (2, 61), particularly when considering the land-use sector (62). Instead, GMST is projected to increase by 3° to 4°C above the pre-industrial period (1, 44), posing serious levels of risk for natural and human systems (3, 5, 20).

The majority of pathways for achieving 1.5°C also require carbon dioxide removal from the atmosphere. Delays in bringing CO2 emissions to net zero over the next 20 to 30 years will also increase the likelihood of pathways that exceed 1.5°C (i.e., overshoot scenarios) and hence a greater reliance on net negative emissions after mid-century if GMST is to return to 1.5°C (Fig. 2A). Technologies designed to remove CO2 from the atmosphere are at an early stage of development, with many questions as to their feasibility and scalability (5). For example, bioenergy with carbon capture and storage (BECCS), afforestation and reforestation, blue carbon (i.e., carbon sequestration by marine ecosystems and processes), soil carbon sequestration, direct capture, biochar (i.e., charcoal for burial in soils), and enhanced weathering variously incur problems associated with feasibility, scalability, and/or acceptability. These strategies are potentially in competition with each other. For example, BECCS would require approximately 18% of global land to sequester 12 Gt CO2/year (5). This requirement is likely, however, to drive an accelerating loss of primary forest and natural grassland, which in turn would increase GHG emissions (5). Early emission reductions plus measures to conserve land carbon stocks may reduce these effects. Policy options might limit the expansion of agriculture at the expense of natural ecosystems, and/or safeguard agricultural productivity from reductions due to BECCS and/or biofuel production (5).

Some carbon dioxide removal options do not rely as extensively on BECCS, but rather focus on afforestation and/or the restoration of natural ecosystems. It is feasible, for example, to limit warming to 1.5°C using strategies such as changing diets and promoting afforestation to remove CO2 (3, 5, 43, 44). Negative consequences of afforestation, such as monoculture plantations on local biodiversity, might be countered by preferentially restoring natural ecosystems and by reestablishing the ability of native grasslands, peatlands, forests, mangroves, kelp forests, and salt marshes to sequester carbon. This creates a “win-win” scenario in which both climate and biodiversity benefit, contributing to UN Sustainable Development Goal (SDG) 15, “Life on Land,” and hence simultaneously making a major contribution to the goals of both the Convention on Biological Diversity (CBD) and the United Nations Framework Convention on Climate Change (UNFCCC). Compatible with this idea is the recent UN establishment of the 2020s as the “Decade of Restoration,” with the intention to build a global resolve to conserve biodiversity, increase its resilience to climate change, and use the more functional ecosystems that result to sequester up to a total of 26 Gt C (63).

Extensive adaptation to 1.5°C of global warming or higher will be very important, especially if we have underestimated climate sensitivity. Developing socially just and sustainable adaptation responses will be increasingly necessary to help natural and human systems to prepare and respond to rapid and complex changes in risk (43). The global adaptation stocktake instigated by the Paris Agreement will help accountability through documentation and mechanisms that inform enhancement at national levels (64, 65). It must also be acknowledged that there are limits to adaptation for natural and human systems (66) and that loss and damage will follow (5, 6769). For example, actions to restore ecosystems may not always be possible given available resources, and it may not be feasible to protect all coastal regions from erosion and loss of land. These challenges mean that identifying, assessing, prioritizing, and implementing adaptation options are very important for reducing the overall vulnerability to increasing climate-related risks as GMST increases. It has become increasingly clear that long-term solutions to climate change must also reduce disadvantage and poverty. Consequently, the recent IPCC Special Report pursued its findings in the context of “strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty” (3). Although previous reports recognized the importance of not aggravating disadvantage, few have specifically focused on solutions that involve multiple elements of climate change, sustainable development, and poverty alleviation. For example, greater insights and knowledge are required to understand how multiple SDGs interact with each other, although many of these interactions are beneficially synergistic (70). Note that SDGs are far more easily reached at 1.5°C than at 2.0°C or more of global warming (43).

The important issue of “loss and damage” also highlights the inequity between nations that have largely caused climate change (and have received the greatest benefits) and those that have not. This inequity is particularly important for least developed countries (LDCs) and small island developing states that have contributed relatively little to global GHG emissions but now face disproportionate risks and harm from climate change, even at 1.5°C (6769, 71). UNESCO has also emphasized the importance of ethics within a nonbinding Declaration of Ethical Principles in Relation to Climate Change in 2017 (72). Specifically, this declaration states that “decision-making based on science is critically important for meeting the mitigation and adaptation challenges of a rapidly changing climate. Decisions should be based on, and guided by, the best available knowledge from natural and social sciences including interdisciplinary and transitionary science and by considering (as appropriate) local, traditional and indigenous knowledge.” These types of initiatives are especially important in the development of policies and actions that avoid inequalities that arise through exclusion and misinformation (61). A transformation toward climate-resilient and low-carbon societies needs to be done in a way that addresses the issue of justice and equity, through ensuring that trade-offs and synergies are identified and actioned (43).


Warming of 1.0°C since the pre-industrial period has fundamentally transformed our planet and its natural systems. Multiple lines of evidence reveal that a 1.5°C world will entail larger risks to both human and natural systems. The risks of a 2°C world are much greater. This places us at a critical time in human history where proportionate action taken today will almost certainly minimize the dangerous impacts of a changing climate for hundreds of millions of people.

Our preliminary estimates suggest that the benefits of avoided damage by the year 2200 may greatly exceed energy sector investment costs to 2050. Current NDCs for 2030 are insufficient to drive this even if followed by “very challenging increases in the scale and ambition of mitigation after 2030” [(44), p. 95], because models based on the current understanding of economic and technical dynamics cannot identify how to reduce GHG emissions to net zero by 2050 from the current NDC starting point in 2030. Rather, these ambitions are consistent with a global warming level of 3° to 4°C, which means that immediate and transformative action is required between now and 2030 in order to greatly scale up current nationally stated plans for GHG reductions. Strategies for responding to climate change must be scalable to the challenges of climate change being faced today and into the future, while at the same time being feasible and fair. Given the scope and threats associated with climate change, there is an increasing need for large scale strategies such as the UN Climate Resilient Development Pathways (CRDP) or “Green New Deal” [UN Environment Programme (UNEP)] if society is to avoid potentially catastrophic circumstances over the next few decades and century.

Supplementary Materials

References and Notes

Acknowledgments: We are grateful for the support provided by the Intergovernmental Panel on Climate Change (IPCC), particularly that of the Technical Support Units for Working Groups I and II, as well as the large number of contributing authors and science officers involved in the IPCC Special Report (3). Author contributions: All authors volunteered and contributed their time in producing this publication and the underlying Special Report from the IPCC (3). Funding: Supported by the Australia Research Council (ARC) Laurette Fellowship and Centre for Excellence in Coral Reef Studies (O.H.-G.) and by the European Research Council (ERC) “DROUGHT-HEAT” project funded by the European Community’s Seventh Framework Programme (grant agreement FP7-IDEAS-ERC-617518) (S.I.S.). Competing interests: The findings, interpretations, and conclusions expressed in the work do not necessarily reflect the views of the World Bank, its board of executive directors, or the governments they represent. K.E. established a sole-proprietor consulting company (ClimAdapt LLC) more than a decade ago, and occasionally consults with international entities and development partners on managing the health risks of climate change. No other competing interests have been declared by the authorship.

Stay Connected to Science

Navigate This Article