SHORT TITLE (<50 Characters)
Testing a new interpretation of glacial sediments

100-WORD ABSTRACT
New theory suggests that silt pellets in ancient glacial deposits provide widespread evidence of subglacial supercooling. This could be very important, but remains controversial because we lack comparative data on silt pellets from modern glaciers where supercooling occurs, and alternative processes could also create glacial pellets. This project will provide data on the characteristics of silt pellets from ancient sediments, from modern supercooling and non-supercooling glacial environments and from new experimental simulations of alternative pellet-forming processes. This will make it possible to test the validity of using silt pellets as indicators of supercooling.

BACKGROUND TO THIS PROJECT
A new theory in glaciology suggests that water flowing through subglacial depressions may become supercooled, may freeze rapidly upon depressurisation, and may then entrain huge amounts of subglacial sediment into the glacier (e.g. Alley et al., 1998). This theory has enormous significance for our ability to use landforms and sediments to reconstruct past glacial environments. It has already been used as the basis for major re-interpretations of sediments deposited by the former Laurentide ice sheet in North America (e.g. Ridge, 2002). Central to these re-interpretations are silt pellets that occur within glacial deposits. These pellets have been interpreted as the result of subglacial supercooling, freezing, and the entrainment of lake sediments from subglacial depressions. The presence of silt pellets in glacial sediments is therefore interpreted to demonstrate a very specific subglacial thermal regime, which in turn indicates a very specific glacial environment. However, the glaciological basis of the interpretation of silt pellets to indicate supercooling is unproven, silt pellets exist in glaciers in many different modern glacial environments, and the rapid adoption of the supercooling hypothesis has been described as premature (Spedding and Evans, 2002). Previously established interpretations of silt pellets as products of water flow at ice crystal boundaries remain tenable, and the supercooling/pellet-formation connection has not been adequately tested. 
     There is a very obvious missing link in our ability to interpret silt pellets. In North America there are ancient pellets left behind by long-departed glaciers and researchers wonder whether they may reflect supercooling processes. In Greenland, at an existing ice sheet margin where subglacial supercooling has not been suggested, researchers have described silt pellets that are currently being produced in the ice and deposited in glacial sediments (e.g. Knight et al., 1994; 2000). In southern Iceland supercooling has been demonstrated at numerous sites (e.g. Knudsen et al., 2002), silt pellets exist within the ice (personal observation by PI), but no detailed work has been carried out on the silt pellets or their mode of formation. The missing link is a detailed analysis of silt pellets from these different environments to test (1) whether they are in fact directly comparable and (2) whether they can be unequivocally related to a specific glacial or hydrological process. 

AIM OF THIS PROJECT
The aim of this project is to provide a glaciologically-based test of the proposed relationship between sedimentary silt pellets and freezing of subglacial sediments, by comparing natural pellets from different glacial settings with artificial pellets that we will create in laboratory simulations of subglacial freezing mechanisms.

SPECIFIC OBJECTIVE
The specific objective is to compare the morphology, microstructure and sedimentology of pellets from: (1) moraines of the former Laurentide ice sheet; (2) modern ice-sheet margin in west Greenland; (3) modern “supercooling” environment in south Iceland; (4) new laboratory simulations of (a) supercooling process and (b) grain boundary process.

MATERIALS AND FACILITIES ALREADY AVAILABLE
· pellets have already been collected from the Greenland ice sheet and from moraines in west Greenland, and are available for analysis.
· pellets from sediments deposited by the Laurentide ice sheet have been collected by collaborators at Minnesota Geological Survey and are available for analysis.
· pilot experiments simulating pellet formation in Keele’s low-temperature laboratory have already been completed. The research assistant responsible for those experiments is available to work on this project.
· the host department will supply all technical support, field-equipment, laboratory facilities, and secretarial assistance totalling approx. £5000 funding in kind. Staff time in fieldwork (3 x staff) is self-funded, totalling approx. £4000 funding in kind.

ACTIVITY REQUIRING FUNDS
1. Sample collection from south Iceland (requires 2-week field expedition). Samples of englacial silt clots will be collected from known supraglacial and basal ice exposures at the glaciers, Skaftafellsjokull and Skeidararjokull (where supercooling is known to occur), and Breidamerkurjokull. Samples will also be sought at Kviarjokull, which has recently been identified as a “supercooling” site (Knudsen et al., 2002). 
2. Laboratory simulations of pellet formation by supercooling and grain boundary processes (requires research assistant 50% FTE for 4 months, experimental apparatus, laboratory consumables). Procedures already established by this research team (Knight and Knight, 1994; 1999) for experimental reproduction of subglacial freezing processes and flow of water through the vein network of polycrystalline ice will be applied to create artificial englacial silt pellets reflecting these different subglacial processes. 
3. Laboratory analysis of samples (requires research assistant 50% FTE for 8 months, sample preparation costs and laboratory consumables). Analysis will compare natural pellets from the USA, Iceland and Greenland, and artificial pellets created in our laboratory simulations. Pellets will be resin-fixed and thin-sectioned for morphological and microstructural analysis. Particle size and grain texture analyses will be completed on disaggregated pellets. Facilities are available at Keele University.

DISSEMINATION OF FINDINGS
Results will be disseminated in international journals (targets: Quaternary Science Reviews, Journal of Glaciology), a dedicated project web page, appropriate national and international conferences and Keele’s programme for widening participation and public understanding of science. Research assistant time includes allowance for contribution to writing.

REFERENCES
Alley, R.B. and 5 others (1998) Glaciohydrologic supercooling: A freeze-on method to create stratified, debris-rich basal ice. II. Theory. Journal of Glaciology 44 (148), 563-569.
Knight, P.G., and 2 others (1994) Ice flow around large obstacles as indicated by basal ice exposed at the margin of the Greenland ice sheet.  Journal of Glaciology 40 (135), 359-367.
Knight, P.G. and Knight D.A. (1994) Glacier sliding, regelation water flow, and development of basal ice. Journal of Glaciology 40 (136), 600-601.
Knight, P.G. and Knight, D.A. (1999) Experimental observations of subglacial debris entrainment into the vein network of polycrystalline ice. Glacial Geology and Geomorphology  (online) http://boris.qub.ac.uk/ggg
Knight, P.G., and 4 others (2000)  Preservation of basal-ice sediment texture in ice sheet moraines. Quaternary Science Reviews 19 (13), 1255-1258.
Knudsen, Ó., and 7 others (2002) Five ‘supercool’ Icelandic glaciers. in Jónsson, S. (ed.). 25th Nordic Geological Winter Meeting January 6th- 9th, 2002 Reykjavik, Iceland. Abstract Volume
Ridge, J. C. (2002) Dynamics of the eastern Ontario lobe: calving, basal freezing, and till deposition in deep lacustrine troughs. Geological Society of America, Abstracts with Programs Denver Annual Meeting, Paper No. 208-2.
Spedding, N. and Evans, D.J.A. (2002) Sediments and landforms at Kviarjokull, southeast Iceland: a reappraisal of the glaciated valley landsystem. Sedimentary Geology 149 (1-3), 21-42

Knight, P.G. and Knight, D.A. (2004) (Invited conference paper) "Field observations and laboratory simulations of basal ice formed by freezing of supercooled subglacial water". Invited contribution to AMICS (Antarctic ice-sheet dynamics and climatic change: Modelling and Ice Composition Studies) workshop Dynamic Interaction between the Antarctic Ice Sheet and the Subglacial Environment, Vrije Universiteit Brussel, Brussels, April 2004. Sponsored by the Belgian Federal Science Policy Office (BELSPO).

Knight P.G. and Knight D.A. (2004) (Conference paper) "Examining the hypothesis of basal ice formation by freezing of supercooled subglacial water." Contribution to International Glaciological Society (British Branch) meeting, Sheffield, UK. September 2004

Cook, S.J.; Knight, D.A.; Knight, P.G.; Waller, R.I. (2005) (Conference paper) "Field and laboratory investigations into basal ice formation by freeze-on of supercooled water." Contribution to General Assembly of the European Geosciences Union, Vienna. Abstract CD-ROM: Geophysical Research Abstracts, Volume 7, (2005) abstract EGU05-A-02240.

Knight, P.G. and Knight D.A. (2005) (Peer-reviewed journal article). Laboratory observations of debris-bearing ice facies frozen from supercooled water. JOURNAL OF GLACIOLOGY   51, 337-9

Knight P.G. and Knight D.A. (2006) (Research case-study in edited volume) "Case Study: Laboratory observations of ice formation and debris entrainment by freezing turbid supercooled water." in Knight, P.G. (ed.) "Glacier Science and Environmental Change" (Blackwell, Oxford).