Yes, Venice is sinking, and the sea is rising against it at the same time. Over the thirty years to 2020 the water stood 4.9 mm higher against the city each year, of which about 1.5 mm came from the ground going down and the rest from the sea coming up (Ferrarin et al., 2024). Measured against the pavements, relative mean sea level at Venice has gained more than 30 cm since 1870.
The worst of the sinking was man-made and it is over. Industrial wells on the lagoon edge drew water and gas out of the aquifers beneath it, and the ground settled by about 9 cm between 1930 and 1970 (Madricardo et al., 2019, p. 2). Regulation and alternative water supplies have avoided that kind of subsidence since the 1970s (Lionello et al., 2021, p. 2639). What remains in the historic centre is 1 to 2 mm per year, against 3 to 4 mm per year across the lagoon and 4 to 6 mm per year in poorly consolidated salt marshes (Lionello et al., 2026, p. 11).
The rising half of the problem is the half still growing. Sea level rise at Venice is expected to reach 0.42 m by 2100 under a low emissions scenario and 0.81 m under a very high one, with up to 1.80 m stated as something that cannot be excluded (Lionello et al., 2026, p. 2). This page separates the two components, explains the layer the city is pinned to, and gives the measured rates with their sources.
How much of the rise is sinking land and how much is rising sea?
Venice has two separate problems that arrive as one number. The figure that matters to a pavement is relative sea level, which combines the ocean rising with the ground settling. Over 1991 to 2020 relative mean sea level at Venice rose at 4.9 mm per year, and comparing the city tide gauge with the one out at the lagoon inlet puts the subsidence share of that at about 1.5 mm per year (Ferrarin et al., 2024). Roughly two thirds of the movement is the sea itself.
The long view is slower. Averaged over the past 150 years the Venetian relative rise has run at 2.5 mm per year, the product of mean sea level rise together with natural and human subsidence (Lionello et al., 2021, p. 2634). Relative mean sea level has gained more than 30 cm since 1870 (Ferrarin et al., 2024). The recent rate is close to double the long-term average, which is the Venice problem in one comparison: a city whose tide gauge datum was fixed in 1897 is being read by water that no longer behaves as it did then.
Air and sea have warmed alongside, by about 1.8 and 1.1 degrees Celsius across the same three decades in the northern Adriatic (Ferrarin et al., 2024).
Why Venice sank fastest between 1930 and 1970
The twentieth century added a cause of sinking that had nothing to do with geology. Industrial plants and wells on the mainland edge of the lagoon drew water, and in places natural gas, from the aquifers under it, and the clays compacted as the pressure dropped. Ground water and natural gas extraction lowered the ground by about 9 cm between 1930 and 1970 (Madricardo et al., 2019, p. 2). The effect peaked in the two post-war decades, when relative sea level rise at Venice accelerated to about 5 mm per year across 1950 to 1970 (Lionello et al., 2021, p. 2634).
That phase is closed. Most subsidence in the last hundred years was caused by human action, largely groundwater withdrawal, and regulation plus alternative supplies of water for industrial, agricultural and civil use have avoided it since the 1970s (Lionello et al., 2021, p. 2639). The same review adds two cautions: the regulations have to keep being enforced, because localised subsidence can still be measured, and natural background subsidence from geological trends that run over centuries continues at up to around 1 mm per year and cannot be switched off.
Nine centimetres sounds modest. It is about a sixth of the roughly 55 cm that separates the lowest part of St Mark's Square from present mean sea level (Lionello et al., 2021, p. 2634), and that margin is the one the whole city is measured by.
How fast is Venice sinking now?
Subsidence rates now depend on where you stand. In the historic centre, across the 50 years since groundwater use was regulated, the ground has gone down 1 to 2 mm per year. Across the lagoon the rate rises to 3 to 4 mm per year, and in poorly consolidated salt marshes it reaches 4 to 6 mm per year (Lionello et al., 2026, p. 11). The authors describe extrapolating about 1 to 2 mm per year for the historic centre and 3 to 4 mm per year for the lagoon into the coming centuries as plausible (Lionello et al., 2026, p. 11).
That spread explains something visitors see without being able to name it. The stone city is the most stable part of the lagoon, because it sits on compacted ground and was pinned into it on purpose. The marshy fringe and the outer flats are settling two to four times faster, which is one reason the lagoon's vegetated edges are under pressure well before the city's pavements are. The loss is already documented: sea-floor surveys of successive dates record a 50 per cent fall in salt marsh surface, alongside the deepening that followed the dredging of the tanker canal between 1960 and 1970 (Madricardo et al., 2019, p. 3).
What Venice is built on: the caranto
Under the paving stones there is mud, and under the mud there is a hard layer the city is pinned to. The foundations of Venice rest on wooden bearing piles driven into the resistant caranto, a word mainland Venetian farmers used for the hard cemented crusts that surfaced in their fields at ploughing. It entered the scientific literature through the early geotechnical studies of the ground under Venice (Mozzi et al., 2003).
The caranto is a buried landscape. It is the weathered surface of a Late Pleistocene alluvial plain, in physical continuity with soils that still lie behind the lagoon, drowned under lagoon deposits by Holocene sea level rise coupled with local subsidence (Mozzi et al., 2003). Above it, under the city itself, the lagoon deposits are 2 to 6 m thick with a maximum age of 4,670 radiocarbon years, plus or minus 70. Below it, the river sediments laid down since the last interglacial are about 55 m deep. The lagoon as a landscape began forming around 6000 BP (Mozzi et al., 2003).
So what Venice stands on is a drowned farmland crust with a few metres of lagoon silt on top, nailed through with timber. That is firmer than it sounds, and it is also why draining the aquifers below it mattered so much.
How much higher will the water be by 2100?
Two peer-reviewed estimates, framed differently, agree on the shape of the century. Sea level rise at Venice is expected to reach 0.42 m by 2100 under a low emissions scenario (RCP2.6) and 0.81 m under a very high one (RCP8.5), measured against the beginning of the century, with up to 1.80 m stated as something that cannot be excluded (Lionello et al., 2026, p. 2). The earlier review gives a relative sea level range of about 30 to 110 cm by the end of the century, widening to 10 to 120 cm once subregional uncertainty is included, and above 180 cm under an unlikely but plausible high-end scenario (Lionello et al., 2021, p. 2638).
The consequences are easier to grasp than the metres. Without defence structures, those levels would leave 15, 70 and 98 per cent of the city centre flooded daily at the astronomical tide maxima (Lionello et al., 2026, p. 2). The mobile barriers at the three inlets are what currently stands between those percentages and the present.
The adaptation study's own conclusion is about timing rather than height. Large interventions, whether permanent barriers, ring dikes or major structural reinforcement, need lead times of 30 to 50 years, so the window for starting a transition to a different strategy is already approaching (Lionello et al., 2026, p. 10).
Why a few centimetres matter in a city this low
Venice is built at the height of its own tide. As of 2020 more than half the city lay 0.80 to 1.20 m above mean sea level, in a lagoon with a tidal range of about 1 m (Lionello et al., 2026, p. 2). The lowest part of St Mark's Square sits approximately 55 cm above present mean sea level (Lionello et al., 2021, p. 2634). A rise of 10 cm, which the 1991 to 2020 rate delivers in about two decades, takes a real share of that margin.
Tide heights quoted in Venice are measured against a local datum, Punta Salute 1897, which lies about 26 cm below mean sea level (Madricardo et al., 2019, p. 10). That is why a tide announced at 110 cm does not mean 110 cm of water standing in the street. It is also why the datum itself records the problem: it was fixed at the mean sea level of its day, and the sea has moved since.
For a visitor the practical effect is limited to footwear and timing. For the city it is a question of which defence strategy gets chosen, and how soon.
