AbstractGlobal mean sea level is a potentially sensitive indicator of climate change. Global warming will contribute to worldwide sea‐level rise (SLR) from thermal expansion of ocean water, melting of mountain glaciers and polar ice sheets. A number of studies, mostly using tide‐gauge data from the Permanent Service for Mean Sea Level, Bidston Observatory, England, have obtained rates of global SLR within the last 100 years that range between 0·3 and 3 mm yr−1, with most values concentrated between 1 and 2 mm yr−1. However, the reliability of these results has been questioned because of problems with data quality and physical processes that introduce a high level of spatial and temporal variability. Sources of uncertainty in the sea‐level data include variations in winds, ocean currents, river runoff, vertical earth movements, and geographically uneven distribution of long‐term records. Crustal motions introduce a major source of error. To a large extent, these can be filtered by employing palaeo‐sea‐level proxies, and geophysical modelling to remove glacio‐isostatic changes. Ultimately, satellite geodesy will help resolve the inherent ambiguity between the land and ocean level changes recorded by tide gauges.Future sea level is expected to rise by ∼ 1 m, with a ‘best‐guess’ value of 48 cm by the year 2100. Such rates represent an acceleration of four to seven times over present rates. Local land subsidence could substantially increase the apparent SLR. For example, Louisiana is currently experiencing SLR trends nearly 10 times the global mean rate.These recently reduced SLR estimates are based on climate models that predict a zero to negative contribution to SLR from Antarctica. Most global climate models (GCMs) indicate an ice accumulation over Antarctica, because in a warmer world, precipitation will exceed ablation/snow‐melt. However, the impacts of attritional processes, such as thinning of the ice shelves, have been downplayed according to some experts. Furthermore, not all climate models are in agreement. Opposite conclusions may be drawn from the results of other GCMs. In addition, the West Antarctic Ice Sheet is potentially subject to dynamic and volcanic instabilities that are difficult to predict. Because of the great uncertainty in SLR projections, careful monitoring of future sea‐level trends by upgraded tide‐gauge networks and satellite geodesy will become essential.Finally, because of the high spatial variability in crustal subsidence rates, wave climates and tidal regimes, it will be the set of local conditions (especially therelativesea‐level rise), rather than a single global mean sea‐level trend, that will determine each lo