Sea-level rise exponentially increases coastal flood frequency

www.nature.com/scientificreports OPEN Sea-level rise exponentially increases coastal flood frequency Mohsen Taherkhani 1, Sean Vitousek 1,2*, Patrick L. Barnard Tiffany R. Anderson3 & Charles H. Fletcher3 2 , Neil Frazer 3 , Sea-level rise will radically redefine the coastline of the 21st century. For many coastal regions, projections of global sea-level rise by the year 2100 (e.g., 0.5–2 meters) are comparable in magnitude to today’s extreme but short-lived increases in water level due to storms. Thus, the 21st century will see significant changes to coastal flooding regimes (where present-day, extreme-but-rare events become common), which poses a major risk to the safety and sustainability of coastal communities worldwide. So far, estimates of future coastal flooding frequency focus on endpoint scenarios, such as the increase in flooding by 2050 or 2100. Here, we investigate the continuous shift in coastal flooding regimes by quantifying continuous rates of increase in the occurrence of extreme water-level events due to sealevel rise. We find that the odds of exceeding critical water-level thresholds increases exponentially with sea-level rise, meaning that fixed amounts of sea-level rise of only ~1–10 cm in areas with a narrow range of present-day extreme water levels can double the odds of flooding. Combining these growth rates with established sea-level rise projections, we find that the odds of extreme flooding double approximately every 5 years into the future. Further, we find that the present-day 50-year extreme water level (i.e., 2% annual chance of exceedance, based on historical records) will be exceeded annually before 2050 for most (i.e., 70%) of the coastal regions in the United States. Looking even farther into the future, the present-day 50-year extreme water level will be exceeded almost every day during peak tide (i.e., daily mean higher high water) before the end of the 21st century for 90% of the U.S. coast. Our findings underscore the need for immediate planning and adaptation to mitigate the societal impacts of future flooding. Sea-level rise is slow, yet consequential1 and accelerating2. Upper-end sea-level rise scenarios could displace hundreds of millions of people by the end of the 21st century3. However, even small amounts of sea-level rise can disproportionately increase coastal flood frequency4,5. A multitude of oceanic processes affect both mean and extreme water levels, such as the tide, tropical and extratropical storms, climatic cycles (e.g., El Nino/Southern Oscillation), oceanic eddies, and circulation patterns6–11. Hence, the frequency and severity of coastal flooding varies on a multitude of time scales. Yet, the persistent trend and acceleration of sea-level rise have a profound interaction with transient extreme events12. In theory, sea-level rise progressively increases the frequency and severity of flooding5. In practice, the monotonic increase in flooding, driven by elevating long-term mean sea level, is often overshadowed by interannual variability in extreme events13, which will likely continue through the middle of the 21st century14. Many have quantified future increases in potential coastal flood frequency by deriving ‘multiplying factors’15, ‘amplification factors’16, or ‘factors of increase’5 in exceedance probability or, equivalently, reductions in return period of extreme water-level events due to sea-level rise by 2050 or 210013,17,18. Large-scale studies of future ‘flooding’ typically investigate potential increases in the water-level hazard in the absence of site-specific exposure (as is the case in the current paper as discussed below in Application). The reported factors of increase in flood hazard potential are often exceedingly large, ranging from 10 to 1000 for even modest sea-level rise scenarios of 0.5 m or less. Yet, focusing on SLR scenarios and their impacts by 2050 or 2100 is perhaps inappropriate, given that significant changes in coastal flooding have been observed in recent years19,20 and are expected to change dramatically in the coming decades5,21–23, and planning horizons rarely exceed thirty years. While the incremental (e.g., ‘stair-cased’) factors of increase are staggering, they do not effectively illuminate the continuous, 1 University of Illinois at Chicago, Department of Civil & Materials Engineering, Chicago, IL, 60607, USA. 2U.S. Geological Survey, Pacific Coastal and Marine Science Center, Santa Cruz, CA, 95060, USA. 3University of Hawai’i at Mānoa, School of Ocean and Earth Science and Technology, Department of Earth Sciences, Honolulu, HI, 96822, USA. *email: Scientific Reports | (2020) 10:6466 | https://doi.org/10.1038/s41598-020-62188-4 1 www.nature.com/scientificreports/ www.nature.com/scientificreports time-dependent transition between present and future flood hazard regimes that we will inevitably experience in the coming decades. Taking a new approach, Sweet et al. (2017)24 and Stephens et al. (2018)25 identified the decade in the future that present-day, extreme water-level events become common-place (e.g., ~5 times per year). They found that many coastal cities transition to dramatically higher flood hazard regimes before 2050 for even moderate sea-level rise scenarios based on probabilistic projections. This new approach, establishing new flood hazard regimes based on a calendar date, effectively communicates the urgency for sea-level rise planning and adaptation. This concept has been further expanded by introducing the concept of a ‘trigger’22 or an ‘adaptation pathway’26,27, i.e., the combination of an intermediate sea-level threshold and an associated time frame when decisions must be made, in order to provide sufficient lead-time to efficiently adapt in a cost-effective manner before more critical flood hazard thresholds are exceeded. Here, we estimate the timing of dramatic shifts in coastal flooding frequency by considering the rate of increase in flood frequency. We argue that the rate of flooding increase is a critical yet poorly understood component to address future sea-level rise impacts: sea-level rise is a continuous process, and thus increment-based assessments may misrepresent the underlying issue. As considered in previous works15,28 and based on the theoretical arguments presented in Methods, we focus on characterizing exponential rates of increase in extreme events driven by persistent shifts in mean water level due to sea-level rise. Note that an exponential increase implies a doubling in the frequency of extreme events over a given amount of sea-level rise or a given period of time, although the particular form of exponential growth used here has not been explicitly considered until now. Given well-established sea-level projections29 and water-level records at a number of long-running tide gauges around the U.S., we estimate sea-level and time scales associated with doubling the odds of exceeding extreme water-level thresholds, defined here (...truncated)