McSaveney, M.J. 1992 A manual for weathering-rind dating of grey sandstones of the Torlesse Supergroup, New Zealand. Lower Hutt: Institute of Geological & Nuclear Sciences. Institute of Geological & Nuclear Sciences science report 92/04. 52 p.
Abstract: This manual details use of weathering rinds on surface-exposed grey sandstones of New Zealand's Torlesse Supergroup to estimate ages of suitable geomorphic surfaces. Rind thickness, to the innermost limits of pale discolouration, increases with age, of exposure following a relationship modelling erosion and chemical diffusion. The measure of thickness is the mode, determined from chips sampled from many boulders. Sample size is a variable set by the sequence of events which contributed rocks to the surface: fifty chips is the recommended minimum sample; samples of 100 are common; some analyses use a thousand or more. A mode is adequately defined if it is identifiable in subsets of the original sample. Cleaning of samples is recommended: 24-hr leaching in dilute hydrochloric acid, followed by bleaching in household bleach. Digital-reading callipers reading to 0.01 mm are a rapid, objective, rind-measuring tool with a precision of about +/-10%. Measurements are made in bright light under low magnification (2X). For rinds < 0.5 mm, a 2OX binocular microscope is needed to obtain consistent measurements. Measurement is aided by a great depth of field, and wide angle of view, obliquely along the rind. Where thickness variation is not obviously due to age differences, thinner segments of rind on a chip are ignored (rind-age calibration uses the thickest significant modal thickness). Rind-thicknesses are plotted as frequency distributions (normalised frequency per 0.2 mm thickness). A high reading resolution of the callipers (0.01 mm) allows high-resolution graphs to be produced in which modes may be resolved to better than +/-0.05 mm. The precision of the mode is significantly higher than the +/-10% precision of a thickness measurement, because it is inversely proportional to the square root of the number of defining measurements. For a polymodal sample, modal precision is estimated by synthesizing the distribution as the sum of normal sub-populations for which mean, standard deviation and population are known. The precision of the mode then is the standard error of the mean for the subpopulation of the mode of interest. Example applications include dating past movements of the Irishman Creek Fault, comparisons between radiocarbon dates and rind dates at a large rock avalanche in the Craigiebum Range, and at an earthflow which crosses the Ostler Fault site disturbance by tree growth and decay at a tectonically raised beach ridge, and dating of late Quaternary moraines, where mathematical filtering is used to compensate for detectable operator bias in rind measuring.