Random variation of fabric eigenvalues: Implications for the use of A-axisfabric data to differentiate till facies

Fabric 'shape', based on the relative values of three eigenvalues calculate d from fabric data, has been used to differentiate sediment facies and infe r deformation histories. The eigenvalues are based on samples drawn from pa rent populations, and as such are subject to statistical variance due to sa mpling effects. In this paper, the degree of statistical variance in fabric data for two types of subglacial till from Breidamerkurjokull. Iceland, is investigated using 'bootstrapping' techniques, in which empirical 'confide nce regions' are built up by repeated resampling of the original data. The experiments show that, for each till type, the observed between-sample vari ability in the fabrics lies within the boundaries associated with random va riations, indicating that the observed range of fabric shapes within each t ill type is likely to be entirely the product of sampling effects. Differen ces in fabric shape between till types are generally greater than that: ass ociated with random variations, indicating that their fabric shape characte ristics, as measured by eigenvalues, are significantly different. Neverthel ess, the results suggest that great care should be exercised when using a-a xis fabric data to differentiate sedimentary facies, or to infer subtle dif ferences in physical processes or conditions. Copyright (C) 2001 John Wiley & Sons, Ltd.