Open AccessThe use of mini‐samples in palaeomagnetism
Author(s) -
Böhnel Harald,
Michalk Daniel,
Nowaczyk Norbert,
Naranjo Gildardo Gonzalez
Publication year - 2009
Publication title -
geophysical journal international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.302
H-Index - 168
eISSN - 1365-246X
pISSN - 0956-540X
DOI - 10.1111/j.1365-246x.2009.04260.x
Subject(s) - standard deviation , wilcoxon signed rank test , geology , paleomagnetism , statistics , mineralogy , sampling (signal processing) , lava , confidence interval , sample size determination , drill , mathematics , volcano , materials science , seismology , paleontology , physics , optics , mann–whitney u test , detector , metallurgy
SUMMARY Rock cores of ∼25 mm diameter are widely used in palaeomagnetism. Occasionally smaller diameters have been used as well which represents distinct advantages in terms of throughput, weight of equipment and core collections. How their orientation precision compares to 25 mm cores, however, has not been evaluated in detail before. Here we compare the site mean directions and their statistical parameters for 12 lava flows sampled with 25 mm cores (standard samples, typically 8 cores per site) and with 12 mm drill cores (mini‐samples, typically 14 cores per site). The site‐mean directions for both sample sizes appear to be indistinguishable in most cases. For the mini‐samples, site dispersion parameters k on average are slightly lower than for the standard samples reflecting their larger orienting and measurement errors. Applying the Wilcoxon signed‐rank test the probability that k or α 95 have the same distribution for both sizes is acceptable only at the 17.4 or 66.3 per cent level, respectively. The larger mini‐core numbers per site appears to outweigh the lower k values yielding also slightly smaller confidence limits α 95 . Further, both k and α 95 are less variable for mini‐samples than for standard size samples. This is interpreted also to result from the larger number of mini‐samples per site, which better averages out the detrimental effect of undetected abnormal remanence directions. Sampling of volcanic rocks with mini‐samples therefore does not present a disadvantage in terms of the overall obtainable uncertainty of site mean directions. Apart from this, mini‐samples do present clear advantages during the field work, as about twice the number of drill cores can be recovered compared to 25 mm cores, and the sampled rock unit is then more widely covered, which reduces the contribution of natural random errors produced, for example, by fractures, cooling joints, and palaeofield inhomogeneities. Mini‐samples may be processed faster in the laboratory, which is of particular advantage when carrying out palaeointensity experiments.