Back to Search Start Over

Stratified Sampling to Define Levels of Petrographic Variation in Coal Beds: Examples from Indonesia and New Zealand.

Authors :
MOORE, TIM A.
Source :
Indonesian Journal on Geoscience. Apr2016, Vol. 3 Issue 1, p29-51. 23p.
Publication Year :
2016

Abstract

Stratified sampling of coal beds for petrographic analysis using block samples is a viable alternative to standard methods of channel sampling and particulate pellet mounts. Although petrographic analysis of particulate pellets is employed widely, it is both time consuming and does not allow variation within sampling units to be assessed - an important measure in any study whether it be for paleoenvironmental reconstruction or in obtaining estimates of industrial attributes. Also, samples taken as intact blocks provide additional information, such as texture and botanical affinity that cannot be gained using particulate pellets. Analysis of variance tests (ANOVA) on stratified samples from an Eocene coal bed in Kalimantan, Indonesia showed that the largest amount of microscopic variation occurs between coal types indicating they are petrographically dissimilar. The ANOVA tests also indicated that only a maximum of 125 point counts are needed to make precise estimates of block composition. Confidence limits of estimates from stratified sampling were calculated to be comparable to what would have been obtained using standard crushed particulate pellets. This means less work is needed for the same accuracy of estimates using a stratified sampling method over that of crushed particulate pellets for coal beds. Macroscopic point counting of vitrain bands can accurately account for those particles >1 mm within a coal interval. This point counting method is conducted using something as simple as string on a coal face with marked intervals greater than the largest particle expected to be encountered (although new technologies are being developed to capture this type of information digitally). Comparative analyses of particulate pellets and blocks on the same interval show less than 6% variation between the two sample types when blocks are recalculated to include macroscopic counts of vitrain. Therefore even in coarse-grained coals, stratified sampling can be used effectively and representatively. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
23559314
Volume :
3
Issue :
1
Database :
Academic Search Index
Journal :
Indonesian Journal on Geoscience
Publication Type :
Academic Journal
Accession number :
119398956
Full Text :
https://doi.org/10.17014/ijog.3.1.29-51