Role of particle size in till‐fabric characteristics: systematic variation in till fabric from Vestari‐Hagafellsjökull, Iceland

Till‐fabric analysis has often been used to interpret glacier flow directions and subglacial dynamics using vector‐based statistics, but recent data suggest that such analysis may also effectively indicate former glacier dynamics. The results of a fabric investigation of subglacially strained till d...

Full description

Bibliographic Details
Published in:Boreas
Main Authors: Carr, Simon J., Goddard, Marc A.
Format: Article in Journal/Newspaper
Language:English
Published: Wiley 2007
Subjects:
Online Access:http://dx.doi.org/10.1080/03009480701210261
https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1080%2F03009480701210261
https://onlinelibrary.wiley.com/doi/pdf/10.1080/03009480701210261
Description
Summary:Till‐fabric analysis has often been used to interpret glacier flow directions and subglacial dynamics using vector‐based statistics, but recent data suggest that such analysis may also effectively indicate former glacier dynamics. The results of a fabric investigation of subglacially strained till deposited during a surge of Vestari‐Hagafellsjökull, Iceland, are presented. Till fabric was collected at four sites within a limited area where ice‐flow direction during deposition was known from subglacial bedforms at the site. Analysis was carried out on elongate grains (axial ratio <1.5:1) at seven size fractions, with the a‐axis length ranging from 0.25 to 32 mm. The largest grains tend to be parallel to ice flow, whereas smaller grains reflect a mix of parallel and transverse orientations. The implications of these data for the role and validity of till‐fabric analysis are discussed, with reference to vector analyses and the compilation of fabric shape envelopes. It is noted that, in contrast to laboratory experimental data, neither March nor Jeffery mechanisms explain the fabric configurations reported. It is concluded that a standardized approach is necessary for collecting fabric data, and in many situations analysis of data populations at a range of particle sizes is desirable.