PrefaceGlobal mean sea-level rise has increased from a rate of a few centimetres per century over the past few millennia to a few decimetres per century in the past few decades. This ten fold increase in rate is due to climate change and is dominated by melting of land ice and warming of ocean water. The current warming trend is expected to continue and so global mean sea level will continue to rise. In this article, we review recent advances in our understanding of past sea-level changes on decadal to millennial timescales to consider how well future changes can be constrained. The majority of studies suggest that global mean sea-level rise will most likely be less than 1 m over the 21 st century. Importantly, there will be significant departures from the global mean by several decimetres in many areas. As a consequence, future research should be targeted at better constraining the spatial variability in future changes so that high-risk areas can be identified.
IntroductionWith about 200 million people living within coastal floodplains and 2 million km 2 of land and $1 trillion worth of assets lying less than 1 m above current sea level, sea-level rise is one of the major socio-economic hazards associated with global warming 1 . The expected rate is, however, extremely uncertain. While the latest IPCC report 2 suggests a range of 0.18-0.59 m of sea-level rise between 1980-1999 and 2090-2099, it emphasises that the contribution from ice dynamic changes is highly uncertain and provides three "illustrative" scenarios suggesting a possible addition of up to 0.17 m from this source. Since then, several studies 3,4 have suggested that a rise larger than 1 m cannot be ruled out. Sea-level rates of this magnitude (m per century) are not uncommon in reconstructions of past sea-level change using geological evidence and it has recently been suggested that similar rates occurred during the previous interglacial warm period 120,000 years ago 5,6 when the volume of land ice was similar to that at present.A "headline" figure of 1 m during the 21 st century represents only the global average sealevel rise. Many different physical processes contribute to sea-level change (see Box 1) and none of these produce a spatially uniform signal. Indeed, one of the few statements that can be made with certainty is that future sea-level change will not be the same everywhere. Thus, the development of regional and local estimates of future sea-level rise -required for effective risk assessment -is one of the primary challenges for the coming years 2 . Prediction relies on models, and the veracity of model output is based on verification against historical and geological data.. However, interpretation of these data requires great care in light of the large spatial and temporal variability in sea-level change. In this article we summarise recent progress in understanding the variability in a suite of sea-level related observations at timescales ranging from decades to millennia. We conclude with an estimate of our current ability to predict...