PAST - Case study 14 - Biostratigraphy with foraminiferans

Case study 14 - Biostratigraphy with foraminiferans

Data files: alveolinid.dat and rascalve.dat

Drobne (1977) compiled the occurrences of 15 species of Alveolina (a large foraminiferan) in 11 sections in the lower Eocene of Slovenia. Following Guex (1991), we will use this data set as an example of biostratigraphical correlation with the methods of Unitary Associations and Ranking-Scaling.

Alveolina (A.) moussoulensis, 3.9 mm long. Photograph by Ralf Gietl (used with permission).

Unitary Associations

Open the file alveolinid.dat, and look at the data. The samples, in rows, are grouped by section using colors. Within each section, samples are entered in stratigraphical order from bottom to top. Try first to correlate the sections 'by eye', using for example index species of your choice. If you are several students carrying out this exercise, compare your results.

Choose 'Unitary associations' in the 'Strat' menu, and look at the statistics. View the unitary associations and the correlation. How does this correlation compare to your 'subjective' attempt? Also look at the reproducibility matrix to see which unitary associations can be identified in the different sections. Are any of the unitary associations weakly reproducible (observed in only a few sections)?

View the graph of superpositional relationships between maximal cliques, and identify cliques that occur in cycles. What does the presence of cycles imply?

Horizon 4 of section 1 contains a number of reworked specimens. Do you get a better result if you run the analysis again without this horizon?

Finally, with horizon 4 of section 1 removed, view the reproducibility graph, which shows the superpositions of unitary associations that are actually observed in one or more sections. The red lines suggest how unitary associations can be merged in order to make reproducible biozones. How many biozones are suggested, and what unitary associations do they contain?

More information about unitary association analysis can be found in the manual.

Ranking-Scaling

We will now analyze the same data set using the Ranking-Scaling method. Open the file rascalve.dat and look at the data. The sections occupy single rows, and the positions of first appearances of the different species have been coded with numbers (0 means absence).

Choose 'Ranking-scaling' in the 'Strat' menu. You may leave the settings as they are, and click 'Go'. View the ranking solution. How does the order of first appearances compare with that given by the Unitary Associations method?

Also view the Scaling solution. Has the order of events changed from the ranking solution? Interpret the meaning of the standard deviations. Finally, look at the scattergrams, where the positions of events in the different wells are plotted against the sequence after scaling. Identify contradictions between the optimal sequence and the sequence in each well.

How will you compare the results from the Unitary Associations method versus Ranking-Scaling?

More information about Ranking-Scaling can be found in the manual.