Coastal zones are among the most densely populated areas in the world1, but they are threatened by rising sea levels caused by climate change. Writing in Earth’s Future, Brown et al.2 estimate the impact of sea-level rise in terms of the land area and the number of people exposed, for several scenarios in which global warming is limited to different temperature increases. Their study shows that the amount of exposure to sea-level rise depends on our ability to cap global temperature changes, and that the main benefits of this cap will be seen only after 2100.
Brown and colleagues began their investigation by simulating how global temperature will change in response to different scenarios of greenhouse-gas emissions, using a simple computational model of the Earth system3. They then calculated the global mean sea-level rise that would occur as a result of the projected temperature changes, assuming that the main causes of sea-level rise are the expansion of the volume of ocean water associated with warming, and the melting of land-based ice (that is, ice over land in glaciers, Antarctica and Greenland). To account for the fact that sea-level rise does not occur uniformly across Earth, they then scaled their time series of global mean sea levels with previously reported projections4 of regional patterns of sea-level change for 2100.
In the scenario in which global temperature increases are capped at 1.5 °C before 2100, they find a median sea-level rise of 0.4 metres by 2100 and of 1 m by 2300. By contrast, in the scenario in which temperatures continue to increase as they are doing now, sea-level rises are 0.8 m in 2100 and a staggering 4.5 m in 2300. These results show that there will be some sea-level rise regardless of efforts to mitigate climate change, because sea levels will not immediately stop rising when the temperature targets are met. However, the effect of capping global temperatures early will be increasingly felt after 2100 and lead to significantly less sea-level rise by 2300.
In a second step, Brown et al. considered the area of land that has an average chance of being flooded once every 100 years (that is, a 1% chance in any given year); such flooding events can be severe (Fig. 1). To do this, they used a database of the world’s coastline characteristics5, which includes land-elevation data measured by radar. The authors combined the data with their scenarios of sea-level rises, and found that an area of 540,000 square kilometres is already at risk of 1-in-100-year coastal flooding events. For the scenario in which the global temperature rise is mitigated to 1.5 °C, this area increases to 620,000 km2 by 2100, and to 702,000 km2 by 2300 (values correspond to the 50th percentile of the range of predicted values for sea-level rises). In the absence of mitigation, they find that the area at risk by 2100 (708,000 km2) is not much different from that in the mitigation scenario, but increases to 1,630,000 km2 by 2300, which is three times the area at risk today.
In the third component of Brown and colleagues’ study, the authors consider the number of people at risk from coastal flooding. If global warming is kept to 1.5 °C, they find that 1.5–2.1% of the global population will be exposed to a 1-in-100-year coastal flooding by 2100, compared with 4.3–5.4% of the global population in the non-mitigation scenario. In other words, more than half of the potential population exposure can be avoided by 2100 if global warming is capped. It is important to keep in mind, however, that population exposure does not depend only on the amount of sea-level rise — the number of people exposed could decrease if people move away from the coast, for example.
Given that sea levels will rise irrespective of future global temperature changes, Brown et al. stress that at least some action will need to be taken to adapt. However, their calculations of the land area exposed to sea-level rise do not consider the effects of coastal-protection measures, such as the construction of dykes or dunes. If such measures were considered in the analysis, then the area and number of people exposed would change. Moreover, the construction of coastal defences will be largely driven by economic considerations, which will be different for different countries.
The authors define land at risk from sea-level rise as the 1-in-100-year coastal floodplain. But will people be driven away from such land by the infrequent floods, or will they accept the occasional inundation, moving away only temporari ly as needed? It might be better to consider the area of land that will become permanently flooded to make a more-direct estimate of the population that will be exposed to sea-level rise.
A limitation of the study is that Greenland, Antarctica and all glaciers elsewhere are lumped together as the land ice that contributes to sea-level rise. However, ice will probably melt at different rates in each of these regions. This will cause the relative contributions of the different sources to change over time. One way to improve the regional estimates of sea-level rise would therefore be to scale the contributions of the ice sheets according to their individual effects.
Nevertheless, studies such as those of Brown and colleagues are essential, because they show the complexity of the climate system’s response to change and how this affects society. By directly connecting the effects of climate change to the consequences for humans, the authors clearly show that climate mitigation needs to happen now for a better future.
Nature 558, 196-197 (2018)