Experiential learning on the Tangled Bank: plant traits and ecological succession
An opportunity for experiential learning
In 1967 Lehigh University Professor Francis Trembley convinced the university to stop mowing a small area of the campus. Professor Trembley named the area the “Tangled Bank,” and it became a place where students could observe nature right outside the classroom. At one point in time there was even a “Tangled Bank” sign on the slope. One year after the mowing stopped, Professor Trembley had the foresight to encourage an undergraduate student to collect and identify all plant species growing on the Tangled Bank, and these collections were archived in the Lehigh Herbarium. Read the full story of the Tangled Bank here (including some great recollections of Lehigh alumni in the comments).
In the fall of 2013, the Lehigh University ecology class (EES-152) completed a botanical survey of the “Tangled Bank” to document what species occupy the site today. The primary objective of this project was for the students to use their data in conjunction with the 1968 plant collections to assess how plant characteristics, such as functional and life history traits, change from early to mid succession. In addition to learning about plant traits and succession, the project allowed students to learn how to apply some commonly used statistical tests to assess differences between groups. They also compared their tree data to a similar dataset collected earlier in the semester from the Lehigh University Experimental forest, a forest that has undergone secondary succession for approximately twice as long as the Tangled Bank. The goal of this comparison was to assess how tree density and biomass change with succession.
Functional traits and succession
What are functional traits? Functional traits are characteristics of species that strongly influence performance, and are therefore fundamental to survival and reproduction. They can be physiological, morphological, or represent reproductive strategies (the latter are oftentimes referred to as life history traits). Plant functional traits include things like photosynthetic pathway (C3, C4, CAM), growth rate, shade tolerance, number of seeds, size of seeds, growth form, life span, seed-dispersal method, and seed viability. Functional diversity (i.e., the total diversity of these traits represented by an ecological community) is increasingly used as an important measure of biodiversity.
What functional traits might be expected to change with ecological succession? Shade tolerance, growth form (herb versus tree), and lifespan might be a few of the more obvious traits that would be expected to change, as plants of later succession include many long-lived trees that compete for light. However, changes in other traits with succession may be less obvious. For example, how might you expect seed viability (i.e., the length of time a seed can survive in the soil before germination) to change with succession?
The Tangled Bank was divided into nine plots, with 3 or 4 students responsible for a complete botanical inventory of each plot. The students identified and estimated the abundance of all plant species, and collected voucher specimens for archival in the Lehigh Herbarium. For trees, diameter at breast height (dbh) was measured and used to calculate total basal area of the forest and the average basal area per tree. The total density of trees (per 1000 m2) was also calculated. Each group submitted an excel spreadsheet with their inventory results and dbh measurements as the first deliverable for the project.
Data compilation and research on plant traits
The data from each group were combined into a class dataset. Students developed lists of the dominant species that grew on the bank in 1968 and 2013 (Table 1, above). Each student was then assigned two plant species (one from 1968 and one from 2013), and required to gather information about the attributes of these species, particularly with respect to life history and functional traits. The objective was to compile as much quantitative and semi-quantiative information on the traits of these species as possible.
The students could use any source of information for this research, as long as they documented their sources; but, we anticipated that they would rely heavily on some of the publicly available databases that have been developed to describe plant characteristics. Unfortunately the shutdown of the US government prevented access to several of these resources, and limited the number of traits that we could examine. However, the students made the best of the situation and found as much information as possible using a variety of sources. They submitted their research in the form of an excel spreadsheet (a template was provided to help standardize the data collection), and this was the second deliverable for the project. Information on sixteen characteristics were found for most species, so we focused our analyses on these (Table 2).
Data analysis and results
Students then performed chi-squared and t-tests using excel, for categorical and continuous data respectively, to assess differences in functional traits of the plant communities in 1968 and 2013 (Table 2). The results of these statistical tests were submitted by each student as the third deliverable of the project.
The students found that plants of early succession generally had smaller seeds, longer seed viability in seedbanks, lower age of first flowering, shorter lifespans, smaller maximum height, and smaller average leaf area (Table 2, Figures 1 & 2). The mid-successional plant community had a greater percentage of species with seed dispersal via mammals and birds (Table 2, Figure 1). Early successional species also tended to be less tolerant of shade and were likely more readily eaten by vertebrate herbivores than those of mid succession (Table 2, Figure 1). Plants of early succession also tended to produce more seeds per plant and have faster growth rates, although differences in these two traits did not meet our threshold for statistical significance (p<0.05). Also, no significant differences were found between N-fixation capacity, frost tolerance, fire tolerance, and drought tolerance of plants in the two communities.
Connecting functional traits to Grime’s life history classification
Various life-history classification methods have been developed by ecologists to facilitate thinking about how species are adapted to environmental conditions. J.P. Grime proposed one such classification scheme for plants, which focused on adaptations to the amount of disturbance (i.e., processes that destroy biomass) and stress (i.e., external constraints that limit the rate of production) in the environment. Under conditions of frequent disturbance, ruderal (i.e., “weedy”) species tend to be favored. Under conditions of high stress, plants tolerant of environmental extremes (e.g., cacti, carnivorous plants) tend be favored. Under conditions of low stress and infrequent disturbance, competitive species tend to be favored.
As an example of how species during succession may fit into Grime’s classification scheme, the functional trait data from the Tangled Bank was used to develop indices related to competition, stress, and disturbance. Traits that would likely be selected for in these different environments were grouped and each species was given a composite score for each of the axes shown in Figure 3 based on the total standardized score of the grouped traits. The figure highlights the shift from ruderal species to better competitor species that has taken place between 1968 and 2013 on the Tangled Bank (Figure 3).
Changes in tree density and biomass with succession
Tree data from the Tangled Bank was compared with data that the students collected earlier in the semester from the Lehigh Experimental Forest. Tree density (trees/1000m2) was higher on the Tangled Bank than the Lehigh Experimental Forest, and the average size (basal area) of trees was larger in the Lehigh Experimental Forest (Figure 4A, 4B). Total basal area of trees, which takes into account both density and size, was greater on the Tangled Bank (Figure 4C).
For the students, deliverable #4. Due 8 November. Complete the following tasks/questions:
1) In less than one page (single-spaced), summarize the differences in functional traits of plant communities in early and mid succession on the Tangled Bank. Your summary should include descriptions of why the observed differences likely occur (i.e., processes).
2) What other functional traits, not included in our analyses, might be expected to differ between 1968 and 2013?
3) For many of the traits we were only able to obtain categorical values, and the number of categories we defined varied among traits. For example, dispersal mode had six defined categories whereas nitrogen fixation capacity only had two. Do you think that the number of categories and how we defined them affected our statistical results and interpretation? Using the original excel spreadsheet, create new, broader categories for a trait of your choice by merging data into a smaller number of categories. Does reducing the number of categories impact the chi-squared test result? Does it change the interpretation?
4) What are the assumptions of a t-test? Did our data violate any of these assumptions?
5) Describe the position of the 1968 and 2013 species in the Grime’s life history diagram (Figure 3). What traits do you think were used to develop the species scores along each axis? In other words, which particular traits would likely be most related to each axis? Some traits could be favored under more than one environmental condition (i.e., high stress, high competition, high disturbance). Justify your answers.
6) Where would the species of the Lehigh Experimental Forest likely be place on Figure 3?
7) Assuming that the differences in successional age are solely responsible for the differences in the density, size of trees, and total tree basal area of the Tangled Bank and the Lehigh Experimental Forest, what specific processes likely resulted in these differences?
8) Your answer to #7 assumed that the only difference between the two sites was age, a common approach often referred to as space-for-time substitution. However, what other factors might contribute to the differences between the two sites?
9) Using the data in Table 3 (below), calculate the rate of change of tree density and average tree basal area in a) early succession (first 45 years) and in b) mid-succession (45 to 98 years). Explain what processes likely contribute to these different rates of change.
Posted on October 27, 2013, in Ecology (EES-152), Fieldwork, Original Posts, Teaching and tagged Ecological succession, Ecology, experiential learning, Forest history, Francis Trembley, Lehigh Experimental Forest, Nature, Plants, Science, student activity, Tangled Bank. Bookmark the permalink. 2 Comments.