Sea level rise

Sea level rise is not uniform around the globe. Some land masses are moving up or down as a consequence of subsidence (land sinking or settling) or post-glacial rebound (land rising due to the loss of weight from ice melt). Therefore, local relative sea level rise may be higher or lower than the global average. Gravitational effects of changing ice masses also add to differences in the distribution of sea water around the globe.[25][26]

When a glacier or an ice sheet melts, the loss of mass reduces its gravitational pull. In some places near current and former glaciers and ice sheets, this has caused local water levels to drop, even as the water levels will increase more than average further away from the ice sheet. Consequently, ice loss in Greenland has a different fingerprint on regional sea level than the equivalent loss in Antarctica.[27] On the other hand, the Atlantic is warming at a faster pace than the Pacific. This has consequences for Europe and the U.S. East Coast, which receives a sea level rise 3–4 times the global average.[28] The downturn of the Atlantic meridional overturning circulation (AMOC) has been also tied to extreme regional sea level rise on the US Northeast Coast.[29]

Many ports, urban conglomerations, and agricultural regions are built on river deltas, where subsidence of land contributes to a substantially increased relative sea level rise. This is caused by both unsustainable extraction of groundwater and oil and gas, as well as by levees and other flood management practices preventing the accumulation of sediments which otherwise compensates for the natural settling of deltaic soils.[30]^: 638[31]^: 88 Total human-caused subsidence in the Rhine-Meuse-Scheldt delta (Netherlands) is estimated at 3–4 m (10–13 ft), over 3 m (10 ft) in urban areas of the Mississippi River Delta (New Orleans), and over 9 m (30 ft) in the Sacramento–San Joaquin River Delta.[31]^: 81–90 On the other hand, post-glacial isostatic rebound causes relative sea level fall around the Hudson Bay in Canada and the northern Baltic.[32]

Historical sea level reconstruction and projections up to 2100 published in 2017 by the U.S. Global Change Research Program.[35] RCPs are different scenarios for future concentrations of greenhouse gases.

The Intergovernmental Panel on Climate Change provides multiple plausible scenarios of 21st century sea level rise in each report, starting from the IPCC First Assessment Report in 1990. The differences between scenarios are primarily due to the uncertainty about future greenhouse gas emissions, which are subject to hard to predict political action, as well as economic developments. The scenarios used in the 2013-2014 Fifth Assessment Report (AR5) were called Representative Concentration Pathways, or RCPs. An estimate for sea level rise is given with each RCP, presented as a range with a lower and upper limit, to reflect the unknowns. The RCP2.6 pathway would see GHG emissions kept low enough to meet the Paris climate agreement goal of limiting warming by 2100 to 2 °C. Estimated SLR by 2100 for RCP2.6 was about 44 cm (the range given was as 28–61 cm). For RCP8.5 the sea level would rise between 52 and 98 cm (20+1⁄2 and 38+1⁄2 in).[26][36]

Sea level rise projections for the years 2030, 2050 and 2100

The report did not estimate the possibility of global SLR being accelerated by the outright collapse of the marine-based parts of the Antarctic ice sheet, due to the lack of reliable information, only stating with medium confidence that if such a collapse occurred, it would not add more than several tens of centimeters to 21st century sea level rise.[26] Since its publication, multiple papers have questioned this decision and presented higher estimates of SLR after attempting to better incorporate ice sheet processes in Antarctica and Greenland and to compare the current events with the paleoclimate data.[37][38][39] For instance, a 2017 study from the University of Melbourne researchers estimated that ice sheet processes would increase AR5 sea level rise estimate for the low emission scenario by about one quarter, but they would add nearly half under the moderate scenario and practically double estimated sea level rise under the high emission scenario.[40][41] The 2017 Fourth United States National Climate Assessment presented estimates comparable to the IPCC for the low emission scenarios, yet found that the SLR of up to 2.4 m (10 ft) by 2100 relative to 2000 is physically possible if the high emission scenario triggers Antarctic ice sheet instability, greatly increasing the 130 cm (5 ft) estimate for the same scenario but without instability.[42]

A 2016 study led by Jim Hansen presented a hypothesis of vulnerable ice sheet collapse leading to near-term exponential sea level rise acceleration, with a doubling time of 10, 20 or 40 years, thus leading to multi-meter sea level rise in 50, 100 or 200 years, respectively.[39] However, it remains a minority view amongst the scientific community.[43] For comparison, two expert elicitation papers were published in 2019 and 2020, both looking at low and high emission scenarios. The former combined the projections of 22 ice sheet experts to estimate the median SLR of 30 cm (12 in) by 2050 and 70 cm (27+1⁄2 in) by 2100 in the low emission scenario and the median of 34 cm (13+1⁄2 in) by 2050 and 110 cm (43+1⁄2 in) by 2100 in a high emission scenario. They also estimated a small chance of sea levels exceeding 1 meter by 2100 even in the low emission scenario and of going beyond 2 metres in the high emission scenario, with the latter causing the displacement of 187 million people.[44] The other paper surveyed 106 experts, who had estimated a median of 45 cm (17+1⁄2 in) by 2100 for RCP2.6, with a 5%-95% range of 21–82 cm (8+1⁄2–32+1⁄2 in). For RCP8.5, the experts estimated a median of 93 cm (36+1⁄2 in) by 2100, with a 5%-95% range of 45–165 cm (17+1⁄2–65 in).[45]

By 2020, the observed ice-sheet losses in Greenland and Antarctica were found to track the upper-end range of the AR5 projections.[46][47] Consequently, the updated SLR projections in the 2019 IPCC Special Report on the Ocean and Cryosphere in a Changing Climate were somewhat larger than in AR5, and they were far more plausible when compared to an extrapolation of observed sea level rise trends.[48]

The main set of sea level rise projections used in IPCC Sixth Assessment Report (AR6) was ultimately only slightly larger than the one in SROCC, with SSP1-2.6 resulting in a 17-83% range of 32–62 cm (12+1⁄2–24+1⁄2 in) by 2100, SSP2-4.5 resulting in a 44–76 cm (17+1⁄2–30 in) range by 2100 and SSP5-8.5 leading to 65–101 cm (25+1⁄2–40 in). The report also provided extended projections on both the lower and the upper end, adding SSP1-1.9 scenario which represents meeting the 1.5 °C (2.7 °F) goal and has the likely range of 28–55 cm (11–21+1⁄2 in), as well as "low-confidence" narrative involving processes like marine ice sheet and marine ice cliff instability under SSP5-8.5. For that scenario, it cautioned that the sea level rise of over 2 m (6+1⁄2 ft) by 2100 "cannot be ruled out".[7] And as of 2022, NOAA suggests 50% probability of 0.5 m (19+1⁄2 in) sea level rise by 2100 under 2 °C (3.6 °F), increasing to >80% to >99% under 3–5 °C (5.4–9.0 °F)."[16]

If countries cut greenhouse gas emissions significantly (lowest trace), the IPCC expects sea level rise by 2100 to be limited to 0.3 to 0.6 meters (1–2 feet).[49] However, in a worst case scenario (top trace), sea levels could rise 5 meters (16 feet) by the year 2300.[49]

Map of the Earth with a long-term 6-metre (20 ft) sea level rise represented in red (uniform distribution, actual sea level rise will vary regionally and local adaptation measures will also have an effect on local sea levels).

Models consistent with paleo records of sea level rise[26]^: 1189 indicate that substantial long-term SLR will continue for centuries even if the temperature stabilizes.[50] After 500 years, sea level rise from thermal expansion alone may have reached only half of its eventual level, which models suggest may lie within ranges of 0.5–2 m (1+1⁄2–6+1⁄2 ft).[51] Additionally, tipping points of Greenland and Antarctica ice sheets are expected to play a larger role over such timescales,[52] with very long-term SLR likely to be dominated by ice loss from Antarctica, especially if the warming exceeds 2 °C (3.6 °F). Continued carbon dioxide emissions from fossil fuel sources could cause additional tens of metres of sea level rise, over the next millennia. The available fossil fuel on Earth is enough to ultimately melt the entire Antarctic ice sheet, causing about 58 m (190 ft) of sea level rise.[53]

In the next 2,000 years the sea level is predicted to rise by 2–3 m (6+1⁄2–10 ft) if the temperature rise peaks at its current 1.5 °C (2.7 °F), by 2–6 m (6+1⁄2–19+1⁄2 ft) if it peaks at 2 °C (3.6 °F) and by 19–22 m (62+1⁄2–72 ft) if it peaks at 5 °C (9.0 °F).[6]^: SPM-28 If temperature rise stops at 2 °C (3.6 °F) or at 5 °C (9.0 °F), the sea level would still continue to rise for about 10,000 years. In the first case it will reach 8–13 m (26–42+1⁄2 ft) above pre-industrial level, and in the second 28–37 m (92–121+1⁄2 ft).[54]

As both the models and observational records have improved, a range of studies has attempted to project SLR for the centuries immediately after 2100, which remains largely speculative. For instance, when the April 2019 expert elicitation asked its 22 experts about total sea level rise projections for the years 2200 and 2300 under its high, 5 °C warming scenario, it ended up with 90% confidence intervals of −10 cm (4 in) to 740 cm (24+1⁄2 ft) and −9 cm (3+1⁄2 in) to 970 cm (32 ft), respectively (negative values represent the extremely low probability of very large increases in the ice sheet surface mass balance due to climate change-induced increase in precipitation.)[44] The elicitation of 106 experts led by Stefan Rahmstorf had also included 2300 for RCP2.6 and RCP 8.5: the former had the median of 118 cm (46+1⁄2 in), a 17%-83% range of 54–215 cm (21+1⁄2–84+1⁄2 in) and a 5%-95% range of 24–311 cm (9+1⁄2–122+1⁄2 in), while the latter had the median of 329 cm (129+1⁄2 in), a 17%-83% range of 167–561 cm (65+1⁄2–221 in) and a 5%-95% range of 88–783 cm (34+1⁄2–308+1⁄2 in)[45]

By 2021, AR6 was also able to provide estimates for year 2150 SLR alongside the 2100 estimates for the first time. According to it, keeping warming at 1.5 °C under the SSP1-1.9 scenario would result in sea level rise in the 17-83% range of 37–86 cm (14+1⁄2–34 in), SSP1-2.6 a range of 46–99 cm (18–39 in), SSP2-4.5 of 66–133 cm (26–52+1⁄2 in) range by 2100 and SSP5-8.5 leading to 98–188 cm (38+1⁄2–74 in). Moreover, it stated that if the "low-confidence" could result in over 2 m (6+1⁄2 ft) by 2100, it would then accelerate further to potentially approach 5 m (16+1⁄2 ft) by 2150. The report provided lower-confidence estimates for year 2300 sea level rise under SSP1-2.6 and SSP5-8.5 as well: the former had a range between 0.5 m (1+1⁄2 ft) and 3.2 m (10+1⁄2 ft), while the latter ranged from just under 2 m (6+1⁄2 ft) to just under 7 m (23 ft). Finally, the version of SSP5-8.5 involving low-confidence processes has a chance of exceeding 15 m (49 ft) by then.[7]

In 2018, it was estimated that for every 5 years CO₂ emissions are allowed to increase before finally peaking, the median 2300 SLR increases by the median of 20 cm (8 in), with a 5% likelihood of 1 m (3+1⁄2 ft) increase due to the same. The same estimate found that if the temperature stabilized below 2 °C (3.6 °F), 2300 sea level rise would still exceed 1.5 m (5 ft), while the early net zero and slowly falling temperatures could limit it to 70–120 cm (27+1⁄2–47 in).[55]

Jason-1 continued the sea surface measurements started by TOPEX/Poseidon. It was followed by the Ocean Surface Topography Mission on Jason-2, and by Jason-3.

Since the launch of TOPEX/Poseidon in 1992, an overlapping series of altimetric satellites has been continuously recording the sea level and its changes.[58] Those satellites can measure the hills and valleys in the sea caused by currents and detect trends in their height. To measure the distance to the sea surface, the satellites send a microwave pulse towards Earth and record the time it takes to return after reflecting off the ocean's surface. Microwave radiometers measure and correct the additional delay caused by water vapor in the atmosphere. Combining these data with the precisely known location of the spacecraft determines the sea-surface height to within a few centimetres (about one inch).[59] Rates of sea level rise for the period 1993–2017 have been estimated from satellite altimetry to be 3.0 ± 0.4 millimetres (1⁄8 ± 1⁄64 in) per year.[60]

Satellites are useful for measuring regional variations in sea level, such as the substantial rise between 1993 and 2012 in the western tropical Pacific. This sharp rise has been linked to increasing trade winds, which occur when the Pacific Decadal Oscillation (PDO) and the El Niño–Southern Oscillation (ENSO) change from one state to the other.[61] The PDO is a basin-wide climate pattern consisting of two phases, each commonly lasting 10 to 30 years, while the ENSO has a shorter period of 2 to 7 years.[62]

Between 1993 and 2018, the mean sea level has risen across most of the world ocean (blue colors).[63]

The global network of tide gauges is another important source of sea-level observations. Compared to the satellite record, this record has major spatial gaps but covers a much longer period of time.[64] Coverage of tide gauges started primarily in the Northern Hemisphere, with data for the Southern Hemisphere remaining scarce up to the 1970s.[64] The longest running sea-level measurements, NAP or Amsterdam Ordnance Datum established in 1675, are recorded in Amsterdam, Netherlands.[65] In Australia, record collection is also quite extensive, including measurements by an amateur meteorologist beginning in 1837 and measurements taken from a sea-level benchmark struck on a small cliff on the Isle of the Dead near the Port Arthur convict settlement in 1841.[66]

This network was used, in combination with satellite altimeter data, to establish that global mean sea-level rose 19.5 cm (7.7 in) between 1870 and 2004 at an average rate of about 1.44 mm/yr (1.7 mm/yr during the 20th century).[67] By 2018, data collected by Australia's Commonwealth Scientific and Industrial Research Organisation (CSIRO) had shown that the global mean sea level was rising by 3.2 mm (1⁄8 in) per year, at double the average 20th century rate,[68][69] while the 2023 World Meteorological Organization report found further acceleration to 4.62 mm/yr over the 2013–2022 period.[3] Thus, these observations help to check and verify predictions from climate change simulations.

Regional differences are also visible in the tide gauge data. Some are caused by the local sea level differences, while others are due to vertical land movements. In Europe for instance, only some land areas are rising while the others are sinking. Since 1970, most tidal stations have measured higher seas, but sea levels along the northern Baltic Sea have dropped due to post-glacial rebound.[70]

Changes in sea levels since the end of the last glacial episode

An understanding of past sea level is an important guide to where current changes in sea level will end up once these processes conclude. In the recent geological past, thermal expansion from increased temperatures and changes in land ice are the dominant reasons of sea level rise. The last time that the Earth was 2 °C (3.6 °F) warmer than pre-industrial temperatures was 120,000 years ago, when warming due to Milankovitch cycles (changes in the amount of sunlight due to slow changes in the Earth's orbit) caused the Eemian interglacial; sea levels during that warmer interglacial were at least 5 m (16 ft) higher than now.[71] The Eemian warming was sustained over a period of thousands of years, and the magnitude of the rise in sea level implies a large contribution from the Antarctic and Greenland ice sheets.[26]^: 1139 According to Royal Netherlands Institute for Sea Research, levels of atmospheric carbon dioxide similar to today's ultimately increased temperature by over 2–3 °C (3.6–5.4 °F) around three million years ago. This temperature increase eventually melted one third of Antarctica's ice sheet, causing sea levels to rise 20 meters above the present values.[72]

Since the Last Glacial Maximum, about 20,000 years ago, sea level has risen by more than 125 metres (410 ft), with rates varying from less than 1 mm/year during the pre-industrial era to 40+ mm/year when major ice sheets over Canada and Eurasia melted. meltwater pulses are periods of fast sea level rise caused by the rapid disintegration of these ice sheets. The rate of sea level rise started to slow down about 8,200 years before present; sea level was almost constant for the last 2,500 years. The recent trend of rising sea level started at the end of the 19th century or at the beginning of the 20th.[73]

There has been an increase in the ocean heat content during recent decades as the oceans absorb most of the excess heat created by human-induced global warming.[77]

The oceans store more than 90% of the extra heat added to Earth's climate system by climate change and act as a buffer against its effects. The amount of heat needed to increase average temperature of the entire world ocean by 0.01 °C (0.018 °F) would increase atmospheric temperature by approximately 10 °C (18 °F):[78] a small change in the mean temperature of the ocean represents a very large change in the total heat content of the climate system.

When the ocean gains heat, the water expands and sea level rises. The amount of expansion varies with both water temperature and pressure. For each degree, warmer water and water under great pressure (due to depth) expand more than cooler water and water under less pressure.[26]^: 1161 Consequently cold Arctic Ocean water will expand less than warm tropical water. Because different climate models present slightly different patterns of ocean heating, their predictions do not agree fully on the contribution of ocean heating to SLR.[79] Heat gets transported into deeper parts of the ocean by winds and currents, and some of it reaches depths of more than 2,000 m (6,600 ft).[80]

Processes around an Antarctic ice shelf

The large volume of ice on the Antarctic continent stores around 70% of the world's fresh water.[81] There is constant ice discharge along the periphery, yet also constant accumulation of snow atop the ice sheet: together, these processes form Antarctic ice sheet mass balance. Warming increases melting at the base of the ice sheet, but it is likely to increase snowfall, helping offset the periphery melt even if greater weight on the surface also accelerates ice flow into the ocean.[82] While snowfall increased over the last two centuries, no increase was found in the interior of Antarctica over the last four decades.[83] Further, sea ice, particularly in the form of ice shelves, blocks warmer waters around the continent from coming into direct contact with the ice sheet, so any loss of ice shelves substantially increases melt raises and instability.[83]

The Ross Ice Shelf, Antarctica's largest, is about the size of France and up to several hundred metres thick.

Different satellite methods for measuring ice mass and change are in good agreement, and combining methods leads to more certainty about how the East Antarctic Ice Sheet, the West Antarctic Ice Sheet, and the Antarctic Peninsula evolve.[84] A 2018 systematic review study estimated that the average annual ice loss across the entire continent was 43 gigatons (Gt) during the period from 1992 to 2002, acceletating to an annual average of 220 Gt from 2012 to 2017.[85] The sea level rise due to Antarctica has been estimated to be 0.25 mm per year from 1993 to 2005, and 0.42 mm per year from 2005 to 2015, although there are significant year-to-year variations.[4]

In 2021, limiting global warming to 1.5 °C (2.7 °F) was projected to reduce all land ice contribution to sea level rise by 2100 from 25 cm to 13 cm (from 10 to 6 in.) compared to current mitigation pledges, with mountain glaciers responsible for half the sea level rise contribution,[86] and the fate of Antarctica the source of the largest uncertainty.[86] By 2019, several studies have attempted to estimate 2300 sea level rise caused by ice loss in Antarctica alone: they suggest 16 cm (6+1⁄2 in) median and 37 cm (14+1⁄2 in) maximum values under the low-emission scenario but a median of 1.46 m (5 ft) metres (with a minimum of 60 cm (2 ft) and a maximum of 2.89 m (9+1⁄2 ft)) under the highest-emission scenario.[7]

The world's largest potential source of sea level rise is the East Antarctic Ice Sheet (EAIS). It holds enough ice to raise global sea levels by 53.3 m (174 ft 10 in)[87] Historically, it was less studied than the West Antarctica as it had been considered relatively stable,[83] an impression that was backed up by satellite observations and modelling of its surface mass balance.[85] However, a 2019 study employed different methodology and concluded that East Antarctica is already losing ice mass overall.[83] All methods agree that the Totten Glacier has lost ice in recent decades in response to ocean warming[88][89] and possibly a reduction in local sea ice cover.[90] Totten Glacier is the primary outlet of the Aurora Subglacial Basin, a major ice reservoir in East Antarctica that could rapidly retreat due to hydrological processes.[38] The global sea level potential of 3.5 m (11 ft 6 in) flowing through Totten Glacier alone is of similar magnitude to the entire probable contribution of the West Antarctic Ice Sheet.[91]

The other major ice reservoir on East Antarctica that might rapidly retreat is the Wilkes Basin which is subject to marine ice sheet instability.[38] Ice loss from these outlet glaciers is possibly compensated by accumulation gains in other parts of Antarctica.[85] In 2022, it was estimated that the Wilkes Basin, Aurora Basin and other nearby subglacial basins are likely to have a collective tipping point around 3 °C (5.4 °F) of global warming, although it may be as high as 6 °C (11 °F), or as low as 2 °C (3.6 °F). Once this tipping point is crossed, the collapse of these subglacial basins could take place as little as 500 or as much as 10,000 years: the median timeline is 2000 years. On the other hand, the entirety of the EAIS would not be committed to collapse until global warming reaches 7.5 °C (13.5 °F) (range between 5 °C (9.0 °F) and 10 °C (18 °F)), and would take at least 10,000 years to disappear.[92][93] It is also suggested that the loss of two-thirds of its volume may require at least 6 °C (11 °F) of warming.[94]

Even though East Antarctica contains the largest potential source of sea level rise, West Antarctica ice sheet (WAIS) is substantially more vulnerable. In contrast to East Antarctica and the Antarctic Peninsula, temperatures on West Antarctica have increased significantly with a trend between 0.08 °C (0.14 °F) per decade and 0.96 °C (1.73 °F) per decade between 1976 and 2012.[95] Consequently, satellite observations recorded a substantial increase in WAIS melting from 1992 to 2017, resulting in 7.6 ± 3.9 mm (19⁄64 ± 5⁄32 in) of Antarctica sea level rise, with a disproportionate role played by outflow glaciers in the Amundsen Sea Embayment.[96]

In 2021, AR6 estimated that while the median increase in sea level rise from the West Antarctic ice sheet melt by 2100 is ~11 cm (5 in) under all emission scenarios (since the increased warming would intensify the water cycle and increase snowfall accumulation over the ice sheet at about the same rate as it would increase ice loss), it can conceivably contribute as much as 41 cm (16 in) by 2100 under the low-emission scenario and 57 cm (22 in) under the highest-emission one.[7] This is because WAIS is vulnerable to several types of instability whose role remains difficult to model. These include hydrofracturing (meltwater collecting atop the ice sheet pools into fractures and forces them open),[37] increased contact of warm ocean water with ice shelves due to climate-change induced ocean circulation changes,[97][98] marine ice sheet instability (warm water entering between the seafloor and the base of the ice sheet once it is no longer heavy enough to displace the flow, causing accelerated melting and collapse)[99] and even marine ice cliff instability (ice cliffs with heights greater than 100 m (330 ft) collapsing under their own weight once they are no longer buttressed by ice shelves). These processes do not have equal influence and are not all equally likely to happen: for instance, marine ice cliff instability has never been observed and was ruled out by some of the more detailed modelling.[100]

Thwaites Glacier, with its vulnerable bedrock topography visible.

The Thwaites and Pine Island glaciers are considered the most prone to ice sheet instability processes. Both glaciers' bedrock topography gets deeper farther inland, exposing them to more warm water intrusion into the grounding zone.[101][102] Their contribution to global sea levels has already accelerated since the beginning of the 21st century, with the Thwaites Glacier now amounting to 4% of the global sea level rise.[103][104][105] At the end of 2021, it was estimated that the Thwaites Ice Shelf can collapse in three to five years, which would then make the destabilization of the entire Thwaites glacier inevitable.[106] The Thwaites glacier itself will cause a rise of sea level by 65 cm (25+1⁄2 in) if it will completely collapse,[107][102] although this process is estimated to unfold over several centuries.[103]

Since most of the bedrock underlying the West Antarctic Ice Sheet lies well below sea level, it is currently buttressed by Thwaites and Pine Island Glaciers, meaning that their loss would likely destabilize the entire ice sheet.[38][108] This possibility was first proposed back in the 1970s,[37] when a 1978 study predicted that anthropogenic CO₂ emissions doubling by 2050 would cause 5 m (15 ft) of SLR from the rapid WAIS loss alone.[109][37] Since then, improved modelling concluded that the ice within WAIS would raise the sea level by 3.3 m (10 ft 10 in).[110][111] In 2022, the collapse of the entire West Antarctica was estimated to unfold over a period of about 2000 years, with the absolute minimum of 500 years (and a potential maximum of 13,000 years). At the same time, this collapse was considered likely to be triggered at around 1.5 °C (2.7 °F) of global warming and would become unavoidable at 3 °C (5.4 °F). At worst, it may have even been triggered already:[92][93] subsequent (2023) research had made that possibility more likely, suggesting that the temperatures in the Amundsen Sea are likely to increase at triple the historical rate even with low or "medium" atmospheric warming and even faster with high warming. Without unexpected strong negative feedbacks emerging, the collapse of the ice sheet would become inevitable.[112][113]

While it would take a very long time from start to end for the ice sheet to disappear, it has been suggested that the only way to stop it once triggered is by lowering the global temperature to 1 °C (1.8 °F) below the preindustrial level; i.e. 2 °C (3.6 °F) below the temperature of 2020.[94] Other researchers suggested that a climate engineering intervention aiming to stabilize the ice sheet's glaciers may delay its loss by centuries and give more time to adapt, although it's an uncertain proposal, and would necessarily end up as one of the most expensive projects ever attempted by humanity.[114][115]

Greenland 2007 melt, measured as the difference between the number of days on which melting occurred in 2007 compared to the average annual melting days from 1988 to 2006[116]

Most ice on Greenland is part of the Greenland ice sheet which is 3 km (10,000 ft) at its thickest. Other Greenland ice forms isolated glaciers and ice caps. The sources contributing to sea level rise from Greenland are from ice sheet melting (70%) and from glacier calving (30%). Average annual ice loss in Greenland more than doubled in the early 21st century compared to the 20th century,[117] and there was a corresponding increase in SLR contribution from 0.07 mm per year between 1992 and 1997 to 0.68 mm per year between 2012 and 2017. Total ice loss from the Greenland Ice Sheet between 1992 and 2018 amounted to 3,902 gigatons (Gt) of ice, which is equivalent to the SLR of 10.8 mm.[118] The contribution for the 2012–2016 period was equivalent to 37% of sea level rise from land ice sources (excluding thermal expansion).[119] This rate of ice sheet melting is also associated with the higher end of predictions from the past IPCC assessment reports.[120][47] In 2021, AR6 estimated that under the SSP1-2.6 emission scenario which largely fulfils the Paris Agreement goals, Greenland ice sheet melt adds around 6 cm (2+1⁄2 in) to global sea level rise by the end of the century, with a plausible maximum of 15 cm (6 in) (and even a very small chance of the ice sheet reducing the sea levels by around 2 cm (1 in) due to gaining mass through surface mass balance feedback). The scenario associated with the highest global warming, SSP5-8.5, would see Greenland add a minimum of 5 cm (2 in) to sea level rise, a likely median of 13 cm (5 in) cm and a plausible maximum of 23 cm (9 in).[7]

Certain parts of the Greenland ice sheet are already known to be committed to unstoppable sea level rise.[121][122][123] Greenland's peripheral glaciers and ice caps crossed an irreversible tipping point around 1997, and will continue to melt.[124][125] A subsequent study had found that the climate of the past 20 years (2000–2019) would already result of the loss of ~3.3% volume in this manner in the future, committing the ice sheet to an eventual 27 cm (10+1⁄2 in) of SLR, independent of any future temperature change.[126] There is also a global warming threshold beyond which a near-complete melting of the Greenland ice sheet occurs.[127] Earlier research has put this threshold value as low as 1 °C (1.8 °F), and definitely no higher than 4 °C (7.2 °F) above pre-industrial temperatures.[128][26]^: 1170 A 2021 analysis of sub-glacial sediment at the bottom of a 1.4 km Greenland ice core finds that the Greenland ice sheet melted away at least once during the last million years, even though the temperatures have never been higher than 2.5 °C (4.5 °F) greater than today over that period.[129][130] In 2022, it was estimated that the tipping point of the Greenland Ice Sheet may have been as low as 0.8 °C (1.4 °F) and is certainly no higher than 3 °C (5.4 °F) : there is a high chance that it will be crossed around 1.5 °C (2.7 °F). Once crossed, it would take between 1000 and 15,000 years for the ice sheet to disintegrate entirely, with the most likely estimate of 10,000 years.[92][93]

Based on national pledges to reduce greenhouse gas emissions, global mean temperature is projected to increase by 2.7 °C (4.9 °F), which would cause loss of about half of Earth's glaciers by 2100—causing a sea level rise of 115±40 millimeters.[131]

There are roughly 200,000 glaciers on Earth, which are spread out across all continents.[132] Less than 1% of glacier ice is in mountain glaciers, compared to 99% in Greenland and Antarctica. However, this small size also makes mountain glaciers more vulnerable to melting than the larger ice sheets. This means they have had a disproportionate contribution to historical sea level rise and are set to contribute a smaller, but still significant fraction of sea level rise in the 21st century.[133] Observational and modelling studies of mass loss from glaciers and ice caps indicate a contribution to sea level rise of 0.2-0.4 mm per year, averaged over the 20th century.[134] The contribution for the 2012–2016 period was nearly as large as that of Greenland: 0.63 mm of sea level rise per year, equivalent to 34% of sea level rise from land ice sources.[119] Glaciers contributed around 40% to sea level rise during the 20th century, with estimates for the 21st century of around 30%.[4] The IPCC Fifth Assessment Report estimated that glaciers contributing 7–24 cm (3–9+1⁄2 in) to global sea levels.[26]^: 1165

In 2023, a Science paper estimated that at 1.5 °C (2.7 °F), one quarter of mountain glacier mass would be lost by 2100 and nearly half would be lost at 4 °C (7.2 °F), contributing ~9 cm (3+1⁄2 in) and ~15 cm (6 in) to sea level rise, respectively. Because glacier mass is disproportionately concentrated in the most resilient glaciers, this would in practice remove between 49% and 83% of glacier formations. It had further estimated that the current likely trajectory of 2.7 °C (4.9 °F) would result in the SLR contribution of ~11 cm (4+1⁄2 in) by 2100.[135] Mountain glaciers are even more vulnerable over the longer term. In 2022, another Science paper estimated that almost no mountain glaciers can be expected to survive once the warming crosses 2 °C (3.6 °F), and their complete loss largely inevitable around 3 °C (5.4 °F): there is even a possibility of complete loss after 2100 at just 1.5 °C (2.7 °F). This could happen as early as 50 years after the tipping point is crossed, although 200 years is the most likely value, and the maximum is around 1000 years.[92][93]

Sea ice loss contributes very slightly to global sea level rise. If the melt water from ice floating in the sea was exactly the same as sea water then, according to Archimedes' principle, no rise would occur. However melted sea ice contains less dissolved salt than sea water and is therefore less dense, with a slightly greater volume per unit of mass. If all floating ice shelves and icebergs were to melt sea level would only rise by about 4 cm (1+1⁄2 in).[136]

Trends in land water storage from GRACE observations in gigatons per year, April 2002 to November 2014 (glaciers and ice sheets are excluded).

Human activity impacts how much water is stored on land. Dams retain large quantities of water, which is stored on land rather than flowing into the sea (even though the total quantity stored will vary somewhat from time to time). On the other hand, humans extract water from lakes, wetlands and underground reservoirs for food production, which often causes subsidence. Furthermore, the hydrological cycle is influenced by climate change and deforestation, which can lead to further positive and negative contributions to sea level rise. In the 20th century, these processes roughly balanced, but dam building has slowed down and is expected to stay low for the 21st century.[137][26]^: 1155

Water redistribution caused by irrigation from 1993 to 2010 caused a drift of Earth's rotational pole by 78.48 centimetres (30.90 in), causing an amount of groundwater depletion equivalent to a global sea level rise of 6.24 millimetres (0.246 in).[138]

Bramble Cay melomys, the first known mammal species to go extinct due to sea level rise.

When seawater reaches inland, coastal plants, birds, and freshwater/estuarine fish are threatened with habitat loss due to flooding and soil/water salinization.[153] So-called ghost forests emerge when coastal forest areas become inundated with saltwater to the point no trees can survive.[154][155] Starting around 2050, some nesting sites in Florida, Cuba, Ecuador and the island of Sint Eustatius for leatherback, loggerhead, hawksbill, green and olive ridley turtles are expected to be flooded, and the proportion would only increase over time.[156] And in 2016, Bramble Cay islet in the Great Barrier Reef was inundated, flooding the habitat of a rodent named Bramble Cay melomys.[157] In 2019, it was officially declared extinct.[158]

An example of mangrove pneumatophores.

While some ecosystems can move land inward with the high-water mark, many are prevented from migrating due to natural or artificial barriers. This coastal narrowing, sometimes called 'coastal squeeze' when considering human-made barriers, could result in the loss of habitats such as mudflats and tidal marshes.[23][159] Mangrove ecosystems on the mudflats of tropical coasts nurture high biodiversity, yet they are particularly vulnerable due to mangrove plants' reliance on breathing roots or pneumatophores, which might grow to be half a metre tall.[160][161] While mangroves can adjust to rising sea levels by migrating inland and building vertically using accumulated sediment and organic matter, they will be submerged if the rate is too rapid, resulting in the loss of an ecosystem.[162][163][161] Both mangroves and tidal marshes protect against storm surges, waves and tsunamis, so their loss makes the effects of sea level rise worse.[164][165] Human activities, such as dam building, may restrict sediment supplies to wetlands, and thereby prevent natural adaptation processes. The loss of some tidal marshes is unavoidable as a consequence.[166]

Likewise, corals, important for bird and fish life, need to grow vertically to remain close to the sea surface in order to get enough energy from sunlight. The corals have so far been able to keep up the vertical growth with the rising seas, but might not be able to do so in the future.[167]

Aerial view of the Tanzanian capital Dar Es Salaam

In Africa, risk from sea level rise is amplified by the future population growth. It is believed that 54.2 million people lived in the highly exposed low elevation coastal zones (LECZ) around 2000, but this number will effectively double to around 110 million people by 2030, and by 2060 it will be around 185 to 230 million people, depending on the extent of population growth. While the average regional sea level rise by 2060 will be around 21 cm (with climate change scenarios making little difference at that point), local geography and population trends interact to increase the exposure to hazards like 100-year floods in a complex way.[21]

Abidjan, the economic powerhouse of Ivory Coast

Maputo, the capital of Mozambique

A man looking out over the beach from a building destroyed by high tides in Chorkor, a suburb of Accra. Sunny day flooding caused by sea level rise, increases coastal erosion that destroys housing, infrastructure and natural ecosystems. A number of communities in Coastal Ghana are already experiencing the changing tides.

In the near term, some of the largest displacement is projected to occur in the East Africa region, where at least 750,000 people are likely to be displaced from the coasts between 2020 and 2050. It was also estimated that by 2050, 12 major African cities (Abidjan, Alexandria, Algiers, Cape Town, Casablanca, Dakar, Dar es Salaam, Durban, Lagos, Lomé, Luanda and Maputo) would collectively sustain cumulative damages of USD 65 billion for the "moderate" climate change scenario RCP4.5 and USD 86.5 billion for the high-emission scenario RCP8.5: the version of the high-emission scenario with additional impacts from high ice sheet instability would involve up to 137.5 billion USD in damages. Additional accounting for the "low-probability, high-damage events" may increase aggregate risks to USD 187 billion for the "moderate" RCP4.5, USD 206 billion for RCP8.5 and USD 397 billion under the high-end instability scenario.[21] In all of these estimates, the Egyptian city of Alexandria alone amounts for around half of this figure:[21] hundreds of thousands of people in its low-lying areas may already have to be relocated in the coming decade.[151] Across sub-Saharan Africa as a whole, damages from sea level rise could reach 2–4% of GDP by 2050, although this is strongly affected by the extent of future economic growth and adaptation.[21]

The remains of Leptis Magna amphitheater, with the sea visible in the background

In the longer term, Egypt, Mozambique and Tanzania are also projected to have the largest number of people affected by annual flooding amongst all African countries if global warming reaches 4 °C by the end of the century (a level associated with the RCP8.5 scenario). Under RCP8.5, 10 important cultural sites (Casbah of Algiers, Carthage Archaeological site, Kerkouane, Leptis Magna Archaeological site, Medina of Sousse, Medina of Tunis, Sabratha Archaeological site, Robben Island, Island of Saint-Louis and Tipasa) would be at risk of flooding and erosion by the end of the century, along with a total of 15 Ramsar sites and other natural heritage sites (Bao Bolong Wetland Reserve, Delta du Saloum National Park, Diawling National Park, Golfe de Boughrara, Kalissaye, Lagune de Ghar el Melh et Delta de la Mejerda, Marromeu Game Reserve, Parc Naturel des Mangroves du Fleuve Cacheu, Seal Ledges Provincial Nature Reserve, Sebkhet Halk Elmanzel et Oued Essed, Sebkhet Soliman, Réserve Naturelle d'Intérêt Communautaire de la Somone, Songor Biosphere Reserve, Tanbi Wetland Complex and Watamu Marine National Park).[21]

Matsukawaura Lagoon, located in Fukushima Prefecture of Honshu Island

As of 2022, it is estimated that 63 million people in the East and South Asia are already at risk from a 100-year flood, in large part due to inadequate coastal protection in many countries. This will be greatly exacerbated in the future: Asia has the largest population at risk from sea level and Bangladesh, China, India, Indonesia, Japan, Pakistan, the Philippines, Thailand and Vietnam alone account for 70% number of people exposed to sea level rise during the 21st century.[17][168] This is entirely due to the region's densely populated coasts, as the rate of sea level rise in Asia is generally similar to the global average. Exceptions include the Indo-Pacific region, where it had been around 10% faster since the 1990s, and the coast of China, where globally "extreme" sea level rise had been detected since the 1980s, and it is believed that the difference between and of global warming would have a disproportionate impact on flood frequency. It is also estimated that future sea level rise along the Japanese Honshu Island would be up to 25 cm faster than the global average under RCP8.5, the intense climate change scenario. RCP8.5 is additionally associated with the loss of at least a third of the Japanese beaches and 57–72% of Thai beaches.[17]

One estimate finds that Asia will suffer direct economic damages of 167.6 billion USD at 0.47 meters of sea level rise, 272.3 billion USD at 1.12 meters and 338.1 billion USD at 1.75 meters (along with the indirect impact of 8.5, 24 or 15 billion USD from population displacement at those levels), with China, India, the Republic of Korea, Japan, Indonesia and Russia experiencing the largest economic losses. Out of the 20 coastal cities expected to see the highest flood losses by 2050, 13 are in Asia. For nine of those (Bangkok, Guangzhou, Ho Chi Minh City, Jakarta, Kolkata, Nagoya, Tianjin, Xiamen and Zhanjiang) sea level rise would be compounded by subsidence. By 2050, Guangzhou would see 0.2 meters of sea level rise and the estimated annual economic losses of 254 million USD - the highest in the world. One estimate calculates that in the absence of adaptation, cumulative economic losses caused by sea level rise in Guangzhou under RCP8.5 would reach ~331 billion USD by 2050, ~660 billion USD by 2070 and 1.4 trillion USD by 2100, while the impact of high-end ice sheet instability would increase these figures to ~420 billion USD, ~840 billion USD and ~1.8 trillion USD, respectively. In Shanghai, coastal inundation amounts to ~0.03% of local GDP; but would increase to 0.8% (confidence interval of 0.4–1.4%) by 2100 even under the "moderate" RCP4.5 scenario in the absence of adaptation. Likewise, failing to adapt to sea level rise in Mumbai would result in the damages of 112–162 billion USD by 2050, which would nearly triple by 2070. As the result, efforts like the Mumbai Coastal Road are being implemented, although they are likely to affect coastal ecosystems and fishing livelihoods.[17] Nations with extensive rice production along the coasts like Bangladesh, Vietnam and China are already seeing adverse impacts from saltwater intrusion.[169]

It is estimated that sea level rise in Bangladesh may force the relocation of up to one-third of power plants as early as 2030, while a similar proportion would have to deal with the increased salinity of their cooling water by then. Research from 2010s indicates that by 2050, between 0.9 and 2.1 million people would be displaced by sea level rise alone: this would likely necessitate the creation of ~594,000 additional jobs and ~197,000 housing units in the areas receiving the displaced persons, as well as to secure the supply of additional ~783 billion calories worth of food.[17] in 2021, another paper estimated that 816,000 would be directly displaced by sea level rise by 2050, but this would be increased to 1,3 million when the indirect effects are taken into account.[170] Both studies assume that the majority of the displaced people would travel to the other areas of Bangladesh, and attempt to estimate population changes in different localities.

2010 estimates of population exposure to sea level rise in Bangladesh

In an attempt to address these challenges, the Bangladesh Delta Plan 2100 has been launched in 2018.[171][172] As of 2020, it was seen falling short of most of its initial targets.[173] The progress is being monitored.[174]

In 2019, the president of Indonesia, Joko Widodo, declared that the city of Jakarta is sinking to a degree that requires him to move the capital to another city.[175] A study conducted between 1982 and 2010 found that some areas of Jakarta have been sinking by as much as 28 cm (11 inches) per year[176] due to ground water drilling and the weight of its buildings, and the problem is now exacerbated by sea level rise. However, there are concerns that building in a new location will increase tropical deforestation.[177][178] Other so called sinking cities, such as Bangkok or Tokyo, are vulnerable to these compounding subsidence with sea level rise.[179]

King's Beach at Caloundra

In Australia, erosion and flooding of Queensland's Sunshine Coast beaches is projected to intensify by 60% by 2030, with severe impacts on tourism in the absence of adaptation. Adaptation costs to sea level rise under the high-emission RCP8.5 scenario are projected to be three times greater than the adaptation costs to low-emission RCP2.6 scenario. For 0.2- to 0.3-m sea level rise (set to occur by 2050), what is currently a 100-year flood would occur every year in New Zealand cities of Wellington and Christchurch. Under 0.5 m sea level rise, the current 100-year flood in Australia would be likely to occur several times a year, while in New Zealand, buildings with a collective worth of NZ$12.75 billion would become exposed to new 100-year floods. A metre or so of sea level rise would threaten assets in New Zealand with a worth of NZD$25.5 billion (with a disproportionate impact on Maori-owned holdings and cultural heritage objects), and Australian assets with a worth of AUD$164–226 billion (including many unsealed roads and railway lines). The latter represents a 111% rise in Australia's inundation costs between 2020 and 2100.[180]

An aerial view of São Paulo's Port of Santos

By 2100, a minimum of 3-4 million people in South America would be directly affected by coastal flooding and erosion. 6% of the population of Venezuela, 56% of the population of Guyana (including in the capital, Georgetown, much of which is already below the sea level) and 68% of the population of Suriname are already living in low-lying areas exposed to sea level rise. In Brazil, the coastal ecoregion of Caatinga is responsible for 99% of its shrimp production, yet its unique conditions are threatened by a combination of sea level rise, ocean warming and ocean acidification. The port complex of Santa Catarina had been interrupted by extreme wave or wind behavior 76 times in one 6-year period in 2010s, with a 25,000-50,000 USD loss for each idle day. In Port of Santos, storm surges were three times more frequent between 2000 and 2016 than between 1928 and 1999.[181]

Beach nourishment in progress in Barcelona.

Many sandy coastlines in Europe are vulnerable to erosion caused by sea level rise. In Spain, Costa del Maresme is anticipated to retreat by 16 meters by 2050 relative to 2010, and potentially by 52 meters by 2100 under RCP8.5[182] Other vulnerable coastlines include Tyrrhenian Sea coast of Italy's Calabria region,[183] Barra-Vagueira coast in Portugal[184] and Nørlev Strand in Denmark.[185]

In France, it was estimated that 8,000-10,000 people would be forced to migrate away from the coasts by 2080.[186] The Italian city of Venice is located on islands. It is highly vulnerable to flooding and has already spent $6 billion on a barrier system.[187][188] A quarter of the German state of Schleswig-Holstein, inhabited by over 350,000 people, is at low elevation and has been vulnerable to flooding since the preindustrial times. Many levees already exist, but to its complex geography, a flexible mix of hard and soft measures was chosen, which is intended to support a safety margin of >1 meter rise per century.[189] In the United Kingdom, sea level at the end of the century would increase by 53 to 115 centimetres at the mouth of river Thames and 30 to 90 centimetres at Edinburgh.[190] To address this reality, it has divided its coast into 22 areas, each covered by a Shoreline Management Plan. Those are further sub-divided into 2000 management units in total, spanning across three "epochs" (0–20 years, 20-50 and 50–100 years).[189]

The Netherlands is a country that sits partially below sea level and is subsiding. It has responded by extending its Delta Works program.[191] Drafted in 2008, the Delta Commission report said that the country must plan for a rise in the North Sea up to 1.3 m (4 ft 3 in) by 2100 and plan for a 2–4 m (7–13 ft) rise by 2200.[192] It advised annual spending between €1.0 and €1.5 billion for measures such as broadening coastal dunes and strengthening sea and river dikes. Worst-case evacuation plans were also drawn up.[193]

Tidal flooding in Miami during a king tide (October 17, 2016). The risk of tidal flooding increases with sea level rise.

As of 2017, around 95 million Americans lived on the coast: for Canada and Mexico, this figure amounts to 6.5 million and 19 million people. Increased chronic nuisance flooding and king tide flooding is already an issue in the highly vulnerable state of Florida,[194] as well as alongside the US East Coast.[195] On average, the number of days with tidal flooding in the USA increased 2 times in the years 2000-2020, reaching 3–7 days per year. In some areas the increase was much stronger: 4 times in the Southeast Atlantic and 11 times in the Western Gulf. By the year 2030 the average number is expected to be 7–15 days, reaching 25–75 days by 2050.[196] U.S. coastal cities have responded to that through beach nourishment or beach replenishment, where mined sand is trucked in and added, in addition to other adaptation measures such as zoning, restrictions on state funding, and building code standards.[197][198] Along an estimated 15% of the US coastline, the majority of local groundwater levels are already below the sea level. This places those groundwater reservoirs at risk of sea water intrusion, which renders fresh water unusable once its concentration exceeds 2-3%.[199] The damages are also widespread in Canada and will affect both major cities like Halifax and the more remote locations like Lennox Island, whose Mi'kmaq community is already considering relocation due to widespread coastal erosion. In Mexico, the damages from SLR to tourism hotspots like Cancun, Isla Mujeres, Playa del Carmen, Puerto Morelos and Cozumel could amount to 1.4–2.3 billion USD.[200] The increase in storm surge due to sea level rise is also a problem. For example, due to this effect Hurricane Sandy caused additional 8 billion dollars in damage, impacted 36,000 more houses and 71,000 more people.[201][202]

In the future, northern Gulf of Mexico, Atlantic Canada and the Pacific coast of Mexico would experience the greatest sea level rise. By 2030, flooding along the US Gulf Coast could cause economic losses of up to 176 billion USD: around 50 billion USD may be avoided through nature-based solutions like wetland restoration and oyster reef restoration.[200] By 2050, the frequency of coastal flooding in the US is expected to rise tenfold to four "moderate" flooding events per year, even without storms or heavy rainfall.[203][204] In the New York City, current 100-year flood would occur once in 19–68 years by 2050 and 4–60 years by 2080.[205] By 2050, 20 million people in the greater New York City area would be threatened, as 40% of the existing water treatment facilities would be compromised and 60% of power plants will need to be relocated. By 2100, sea level rise of 0.9 m (3 ft) and 1.8 m (6 ft) would threaten 4.2 and 13.1 million people in the US, respectively. In California alone, 2 m (6+1⁄2 ft) of SLR could affect 600,000 people and threaten over 150 billion USD in property with inundation, potentially representing over 6% of the state's GDP. In North Carolina, a meter of SLR inundates 42% of the Albemarle-Pamlico Peninsula, costing up to 14 billion USD (at 2016 value of the currency). In nine southeast US states, the same level of sea level rise would claim up to 13,000 historical and archaeological sites, including over 1000 sites eligible for inclusion in the National Register for Historic Places.[200]

Malé, the capital island of Maldives.

Small island states are nations whose populations are concentrated on atolls and other low islands. Atolls on average reach 0.9–1.8 m (3–6 ft) above sea level.[206] This means that no other place is more vulnerable to coastal erosion, flooding and salt intrusion into soils and freshwater caused by sea level rise. The latter may render an island uninhabitable well before it is completely flooded.[207] Already, children in small island states are encountering hampered access to food and water and are suffering an increased rate of mental and social disorders due to these stressors.[208] At current rates, sea level would be high enough to make the Maldives uninhabitable by 2100,[209][210] while five of the Solomon Islands have already disappeared due to the combined effects of sea level rise and stronger trade winds that were pushing water into the Western Pacific.[211]

Surface area change of islands in the Central Pacific and Solomon Islands[212]

Adaptation to sea level rise is costly for small island nations as a large portion of their population lives in areas that are at risk.[213] Nations like Maldives, Kiribati and Tuvalu are already forced to consider controlled international migration of their population in response to rising seas,[214] since the alternative of uncontrolled migration threatens to exacerbate the humanitarian crisis of climate refugees.[215] In 2014, Kiribati had purchased 20 square kilometers of land (about 2.5% of Kiribati's current area) on the Fijian island of Vanua Levu to relocate its population there once their own islands are lost to the sea.[216]

While Fiji is also impacted by sea level rise,[217] it is in a comparatively safer position, and its residents continue to rely on local adaptation like moving further inland and increasing sediment supply to combat erosion instead of relocating entirely.[214] Fiji has also issued a green bond of $50 million to invest in green initiatives and use the proceeds to fund adaptation efforts, and it is restoring coral reefs and mangroves to protect itself flooding and erosion as a more cost-efficient alternative to building sea walls, with the nations of Palau and Tonga adopting similar efforts.[214][218] At the same time, even when an island is not threatened with complete disappearance due to flooding, tourism and local economies may end up devastated. For instance, a sea level rise of 1.0 m (3 ft 3 in) would cause partial or complete inundation of 29% of coastal resorts in the Caribbean, while a further 49–60% of coastal resorts would be at risk from resulting coastal erosion.[219]