Sea-level changes in the North Atlantic Ocean: implications for the melting history of the Greenland Ice Sheet

Summary Predicted rates of sea-level rise for the 21st century, such as those in the latest IPCC assessment, are global in scope and do not provide sufficient information for coastal residents, stake holders and planners. Several processes need to be quantified if we want to translate global sea-level predictions into practical local values. These include vertical movements of the coast, regional density changes in the ocean and worldwide ocean-surface variations produced by the Earth's changing gravity field. Information on these processes and their contributions to sea-level rise in the recent past is of great value to climate modellers, because it allows them to test and improve sea-level prediction models. Do we understand past contributions to sea-level rise? We know that some of the sea-level rise in the last 50 years has been from the melting of glaciers, small ice caps and ice sheets. Sea levels have also been rising because sea water expands as it gets warmer. Whilst these processes can account for most of the observed sea-level rise since the 1950s, we do not know what caused sea levels to rise during the first half of the 20th century. This is an important problem, because warming during the 1930s and 1940s was faster than in the last two decades. Unfortunately, direct observations of ice melt and ocean warming are not available. This project attempts to find out what happened with the Greenland Ice Sheet before the 1950s by using an indirect method: it takes advantage of a distinct global pattern of sea-level rise that would have occurred if the Greenland Ice Sheet had been melting significantly. Sea-level rise that results from polar ice melt is not evenly distributed across the globe. The explanation for this is that the gravitational field of the Earth changes its shape in response to ice melt and, as a result, sea-level rise occurs faster the further away one goes from the melting source. Melting of the Greenland Ice Sheet produces sea-level rise that increases in magnitude from north to south along the coasts of western Europe. Turning this concept on its head we can estimate how much Greenland has contributed to past sea-level rise if we can quantify this sea-level gradient. To do this, we have several long instrumental records to our disposal, some of which go back to the early 1800s, but these are located in France and Poland, near the middle of the gradient. It would be very useful to have sea-level records at both ends of the gradient, in northern Norway and southern Portugal. This project will establish past sea-level histories for northern Norway and southern Portugal from analyses of sediments preserved in salt marshes. The marshes collect mud every day when the tides come in and, over a long time, build up at approximately the same pace as sea level rises. Precise sea-level changes will be reconstructed from the analyses of small fossils (foraminifera) and precise dating of the sediments. The two sea-level records need to be combined with all available long sea-level records .Several corrections are then required before it is possible to estimate the contribution of Greenland to the observed sea-level rise pattern. Firstly, long-term land movements need to be considered, but these can be eliminated by calculating the difference between the 19th and the 20th century rates of sea-level rise in all records. Secondly, we need to account for the warming of the oceans, and for this we use data from a computer model that will be provided by colleagues. What remains after these corrections is the 'fingerprint' of Greenland ice melt. With the help of another computer model it will be possible to calculate how much of Greenland would have melted to explain the sea-level data. This information will be very valuable to climate modellers, because it helps to explain sea-level rise during the 20th century and can improve sea-level rise predictions for the 21st century.