Suilven Mountain pokes above Assynt’s glacially scoured "knock and lochan" landscape. The term comes from the Gaelic for a small rock hill (cnoc) and a little remnant lake (lochan). Credit: Nadine Hallman

The postglacial landscape of northwestern Scotland was the backdrop for a workshop on maximizing the potential of data on past ice sheet and sea level variability. Scientists gathered in Lochinver, Scotland, in mid-September for the 2014 Paleo Constraints on Sea Level Rise (PALSEA2) meeting, which consisted of 3 days of field discussions followed by 2 days of presentations.

The Assynt Mountains provided an inspirational view for debates on the evolving interpretations of physical evidence for ice sheet development in the region. Here trimlines were once thought to delineate the maximum height of the British and Irish Ice Sheet during the last glaciation but are now interpreted as indicating a thermal boundary in the ice sheet, above which the cold-based ice will act to protect rather than erode the mountain surface it is in contact with.

The workshop focused on reducing uncertainty about how ice sheet and sea level data are compiled, structured, and shared.

The workshop focused on reducing uncertainty about how ice sheet and sea level data are compiled, structured, and shared. For instance, creation of a database of sea level indicators around Britain and Ireland led to the conclusion that the ice models, with a maximum elevation constrained by the trimlines, did not contain enough ice volume at the Last Glacial Maximum to explain patterns of subsequent sea level change. Comprehensive data sets on other now-gone ice sheets showed that only through such databases could continental-scale questions be addressed.

In another more recent example, a compilation of GPS data from Antarctica indicates unstable advance and retreat of portions of the West Antarctic Ice Sheet during the late Holocene, where previously monotonic advance had been assumed. In addition, widely distributed records are required to constrain variations in sea level caused by growth and melt of past ice sheets.

Workshop attendees hand-core sediments in Loch Laxford, a salt marsh within Assynt’s post-glacial landscape. Credit: Tom Bradwell, BGS
Workshop attendees hand-core sediments in Loch Laxford, a salt marsh within Assynt’s postglacial landscape. Credit: Tom Bradwell, BGS

How, then, should these data sets be structured for analysis? Different approaches were presented over the course of the meeting, from categorizing a landscape by exploring it on foot to turning to extensive literature reviews that capture untapped potential in metadata.

Presentations that brought disparate data sets together, clarified assumptions, and applied transparent and consistent models of uncertainty demonstrated exciting results. Specific talks included one on a recent grant award to create a portal that will standardize uranium-thorium age dating calculations, allowing comparison of data across individual studies. Another presentation offered analysis of a data set developed from an extensive literature review that describes new species-specific effects on sea level reconstruction using fossil corals.

All of those involved in the generation of data, from field observations to models, share a responsibility to ensure that the work is as transparent as possible and is communicated via publication vehicles that recognize and support the diverse needs to which database content may be directed. To support this aim, workshop participants committed to production of a best practices document and working protocols for collating sea level and ice sheet indicators. Those involved anticipate that the documents and protocols will be ready for public discussion in 2015.


Antony Long and Natasha Barlow of the Department of Geography, Durham University, organized the meeting, with funding from the Past Global Changes program and the International Union for Quaternary Research.

—Felicity Williams, Ocean and Earth Science, National Oceanography Centre, University of Southampton, Southampton, UK; email:; Nadine Hallmann, Aix-Marseille Université, Centre National de la Recherche Scientifique, Institut de Recherche pour le Développement, Centre Européen de Recherche et d’Enseignement des Géosciences de l’Environnement, Aix-en-Provence, France; and Anders Carlson, College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis

Citation: Williams, F., N. Hallmann, and A. Carlson (2015), Developing databases of ancient sea level and ice sheet extents, Eos, 96, doi:10.1029/2015EO024295. Published on 17 February 2015.

Text © 2015. The authors. CC BY 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.