Menu Close

Improving sea-level projections

We need to know more about how ice sheets interact with the warming oceans and warming atmosphere. Greenpeace/ADavies

Satellite and in situ observations show sea level is continuing to rise.

In the last Intergovernmental Panel on Climate Change report, earth system model projections indicated global sea-level rise by 2100 ranges from 20 to 60cm above 1990 values, plus an additional component from a potential dynamic response of the ice sheets.

This rise was estimated to be a further 10 to 20cm if the recently observed contribution to sea-level rise from increased ice-sheet discharge was to grow linearly with temperature. This would give a total range in 2100 of about 20-80 cm.

Current understanding of ice-sheet processes is incomplete, however, and larger rises in sea level cannot be categorically excluded. Sea level will continue to rise beyond 2100. The changes we cause in the atmosphere today will continue to affect sea levels for hundreds of years to come. Mass loss from the Greenland and Antarctic ice sheets remains the largest uncertainty in projections of sea-level rise for the 21st Century and beyond.

In a Perspective in today’s edition of Science, we note that for civilisation, the stakes are high. In Australia there will be impacts, too, and scientists and decision-makers are grappling with the degree of action that needs to be taken and the time available for them to respond. Put simply, how we respond is about risk management.

We are confident in model projections that show sea-level will continue to rise. There has been substantial progress made in the extent and accuracy of observations against which we can test the models. However, there are still uncertainties in the rate of ocean heat uptake, Antarctic snowfall, ocean circulation beneath the ice shelves and the resulting basal melting, and glacier and ice-sheet response. Contributions from accelerated ice sheet outflow or the rapid collapse of glaciers that are in direct contact with the oceans could potentially lead to larger rises than in current projections.

The science is progressing; for example, a new glacier inventory has just been completed. It will allow better assessment of glacier contributions. Also, global climate models approximately reproduce the part of sea-level caused by thermal expansion of upper ocean waters over recent decades.

The international upper ocean monitoring program Argo is generating volumes of data, and technology is now permitting observations in the sea-ice zones. However, the ocean’s regional dynamic response to climate in the models remains poorly understood.

Science needs to focus on improving projections of sea-level rise at the regional level: this is what influences communities and this is what decision-makers are most concerned about. This will require the global scientific community to pursue a wider multi-disciplinary approach to address the current uncertainties and to achieve a greater accuracy in sea-level rise projections.

We have sufficient information to make adaptation decisions that have long time horizons. But models must be critically compared with observations to get a more accurate representation of regional projections. One of the largest challenges remains understanding how the ice sheets interact with the warming oceans and warming atmosphere.

Simple models (so-called semi-empirical models) scale historical sea-level data, mostly from about 1850, with global temperatures. Estimates of future sea-level rise using these models give substantially larger values. Given the difficulties and time required to develop fully-coupled Earth system models that include components representing the ocean, atmosphere and ice sheets that can reliably project sea-level rise, it is important to question and understand the accuracy and fidelity of these simple semi-empirical models.

Science is about seeking greater certainty. Resolving the discrepancies between Earth system models and semi-empirical models may lead to new projections that exploit advantages of both and may provide more reliable scenarios in a timely fashion.

Want to write?

Write an article and join a growing community of more than 180,900 academics and researchers from 4,919 institutions.

Register now