Multivariate techniques for biostratigraphical correlation

摘要: Multivariate techniques summarize large sets of data and present their major trends in a graphically simplified manner. This paper discusses ways which these can be used to correlate biostratigraphical data. Stratigraphically constrained cluster analyses, highlight distinct zones, may performed on number sections. These plotted together provide concise summary that allows sequences compared. Ordinations, such as principal components detrended correspondence several sections simultaneously extract new set axes represent the combined various taxa. The first few against depth ‘biostratigraphical logs’; rest ignored insignificant variation or noise. Sections then correlated based logs, either manually using procedures sequence slotting. have been tested artificial from Edwards (1984), consisting taxa known spatial temporal ranges. Correlations between were found match closely true ages. Inconsistencies data, time-transgressive taxa, isolated one axis, allowing them removed studied further.