Tell Me About the Forest (Dead Can Dance)

Robert Gastaldo, Colby College
Author Profile


Most students entering university have some experience with trees growing in a forest ecology. Most student perspectives are that of a northern hemisphere chauvinist. Several other forest structures now exist but although the "present is key to the past," deep-time fossil forests have not always been the same. The goals of the activity are (1) to introduce quantitative ecological measures to fossil benthic (autochthonous)assemblages, (2) Test assemblage relationships using diversity measures, correlation coefficients, and simple multivariate statistical analyses, and (3)Reconstruct an autochthonous fossil community in space, demonstrating that ancient community structure differs from the Recent.

Used this activity? Share your experiences and modifications



Designed for an undergraduate course in paleontology course for majors, although it could be adapted for a graduate-level offering.

Skills and concepts that students must have mastered

Students must have completed basic mathematical skills and the exercise is enhanced if students have completed an introductory statistics course, but this is not necessary. Students should have some understanding of autochthony vs. allochthony in the paleontological sense.

How the activity is situated in the course

The exercise is used as a laboratory following the introduction of fossil groups in an attempt to introduce students into how fossil assemblages can be used to understand past community structure.


Content/concepts goals for this activity

Students use a dataset obtained from the forest floor litter of an autochthonous Carboniferous forest preserved above the Blue Creek Coal, Black Warrior Basin, Alabama. They are given quantitative data of canopy, understory, and groundcover plants from 5 collection sites from which they will determine ecological metrics (Shannon-Wiener, Simpson's, and Margalef indices) for each part of the forest, and then be introduced to exploratory statistical analyses with which they will learn how to calculate similarity coefficients and use these in a simple cluster analysis. Once similar parts of the forest are determined, students will attempt to reconstruct the forest structure in space using known architectural growth strategies for the major plant groups.

Higher order thinking skills goals for this activity

The exercise includes: analysis of a data set to determine relationships between collections, synthesis of quantitative analytical results with the original data to reconstruct a deep time forest community, and the development of an understanding of how fossil assemblages can be used to assess ancient landscapes.

Other skills goals for this activity

Students will be able to express their creative "inner self" artistically in the development of the forest transect that results from their data analysis.

Description and Teaching Materials

The distribution of faunas and floras can provide an understanding of the temporal and spatial relationships of the paleoenvironments in which they lived and thrived. For example, climatically sensitive assemblages, such as reefs and peat swamps, can reflect where warm, shallow, sediment-free waters existed, in the former case, and everwet, warm and humid terrestrial conditions existed, in the latter. Additionally, the geographic distribution of fossil organisms also may provide evidence about the relationships between ocean basins and continental land masses through time.

This laboratory is designed to test several relationships of communities and assemblages using several basic ecological parameters, correlation coefficients, and simple multivariate statistical analyses. The data provided are taken from a Carboniferous coal mine in Alabama.

The complete exercise including assignment and data set (Excel format) are available in the Supporting Materials.
Laboratory Handout for "Tell Me About the Forest" exercise (Acrobat (PDF) 1.4MB Jun10 14)
Excel file of Blue Creek Fossil Forest for "Tell Me About the Forest" exercise (Excel 2007 (.xlsx) 11kB Jun10 14)

Teaching Notes and Tips

The common areas of confusion seem to arise from: (1) how different ecological metrics can tell the researcher different aspects of the data set, and (2) the idea of composite data (the condensation of the two or more original assemblage data into a single data set based on results of correlation coefficients) being used, again, to evaluate relationships with the remaining collections.


Students who meet the goals of this activity have:
  1. correctly calculated the standard ecological methods by hand and, if allowed, using Excel spreadsheet functions and notation
  2. correctly calculated the correlation coefficient(s) selected by the instructor to be used in the Q-mode cluster analysis
  3. correctly assessed the relationships between the 5 collection sites using Q-mode cluster analysis, and
  4. provided a credible reconstruction of the Blue Creek forest structure (growth habits and proportion of canopy vs. understory vs. ground-cover plants) based on the data provided.

References and Resources

The following publications are provided in support of the activity:

Pashin, J.C., and Gastaldo, R.A., 2009, Carboniferous of the Black Warrior Basin, in Greb, S.F., and Chestnut, D.R., Jr., eds., Carboniferous geology and biostratigraphy of the Appalachian and Black Warrior Basins: Kentucky Geological Survey Special Publication 1, Series 12, p. 10-21.

Gastaldo, R.A., Stevanovic-Walls, I., Ware, W.N., and Greb, S.F., 2004, Community Heterogeneity of Early Pennsylvanian Peat Mires: GEOLOGY, v. 32, no. 8, p. 693-696.

Gastaldo, R.A., Stevanovic-Walls, I., and Ware, W.N., 2004, In Situ, Erect Forests Are Evidence for Large-Magnitude, Coseismic Base-Level Changes within Pennsylvanian Cyclothems of the Black Warrior Basin, USA: in Pashin, J.C., and Gastaldo, R.A., eds., Coal-bearing Strata: Sequence Stratigraphy, Paleoclimate, and Tectonics: AAPG Studies in Geology, v. 51, p. 219-238.