Increased Exposure of Coastal Cities to Sea-Level Rise Due to Internal Climate Variability
Distinguishing the human activity signature in past and ongoing sea-level changes is a challenging topic of intense studies. A major obstacle for the reliable detection of anthropogenic signals originates from significant contribution of natural fluctuations to sea-level changes. These fluctuations mirror the complex climate-system dynamics, commonly called internal climate variability (ICV). The ICV is generated by chaotic interactions within and between the atmosphere, hydrosphere, cryosphere, biosphere, and the solid-Earth processes occurring over an extremely wide range of time and space scales. These interactions can manifest as cycles, instabilities or irregular oscillations of the climate system. One example of a well-known manifestation of the internal variability in the climate system is the El Nino/Southern Oscillation. The practical importance of the ICV is rooted in its chaotic nature, leading to an irreducible uncertainty in the sea-level projections.
Combining the Community Earth System Model version 1 Large Ensemble experiments with power-law statistics, we show that, by 2100, if the ICV uncertainty reaches its upper limit, new sea-level-rise hotspots would appear in Southeast Asian megacities (Chennai, Kolkata, Yangon, Bangkok, Ho Chi Minh City and Manila), in western tropical Pacific Islands and the Western Indian Ocean. The better the ICV uncertainty is taken into account and correctly estimated, the more effective adaptation strategies can be elaborated with confidence and actions to follow.
The sea-level projections are accompanied by inherent uncertainties, of which a significant part arises from complex and unpredictable interactions within and between climate-system components, rendering them irreducible. Neglecting the uncertainty induced by the ICV inevitably results in an underestimation of future SLR and the associated flooding risks. Along with the spread in the sea-level changes derived from CESM1-LE, we used a realistic statistical model of large deviations in sea-level fluctuations based on previous analyses of the observed sea-level records. The power-law model provides a simple and efficient framework for estimating ICV uncertainty in sea-level projections. The power-law statistics seem to indicate a wider range of probable sea-level changes generated by the ICV than the spread among CESM1-LE members. This effect is pronounced by 2100 and under RCP8.5 along the coasts of East Africa, west Australia, insular Southeast Asia, and western United States. If the ICV uncertainty upper limit is reached, some SLR hotspots would be created in Southeast Asian megacities, making a part of at least ~20% in the expected SLR at Chennai, Kolkata, Yangon, Bangkok, Ho Chi Minh City and Manila, reaching ~30% in the western tropical Pacific Islands and in Western Indian Ocean.