Concepts of Ecology

Discussion topics for September 17, 1998
[You should have read the assigned readings in Gotelli and Crouse et al. (1987). McPeek and Peckarsky's paper might also help you think about the more general issues.]

  1. A variety of approaches can be used to explore the sensitivity of population growth (e..g, l ) to changes in the transition matrix (A, comprised of elements aij). Two that you've already seen including the standard measure of sensitivity, d l /d a, and elasticity, d (ln l )/ d (ln a). How do these two measures differ? What are the advantages and disadvantages of each. If you are a manager charged with devising a single type of intervention (focused on a single life stage) that will best improve the future dynamics of sea turtles (using the data in Crouse et al.'s paper), which measure would you choose (i.e., to find the most "sensitive" stage). 

  2. A stable age distribution (SAD) bears considerable similarity to the survivorship schedule. Why? If there is a stable age distribution and l =0, the SAD exactly mirrors the survivorship schedule (you know this, right?). In these age-structured cases, the SAD is monotonically decreasing. How does this contrast with a stable stage distribution (SSD) in a stage-structured population, and what influences the SSD's shape? 

  3. What is reproductive value, RV (no not, "the left eigenvector" -- rather, what is it in biological terms?). 

  4. Assume an asexual species with several closed sub-populations that are homogeneous genetically. Within each sub-population, a single beneficial mutatation arises that differs from the wildtype in a way that alters one element of the transition matrix, A (each sub-population has a different mutation arise). Assume the wildtype A is the same as the sea turtle A of Crouse et al. Can you a priori assess which mutant will increase (spread through the population) most rapidly? Might this be assessed by knowing the reproductive values of different stages, or sensitivites, or elasticities, or something else? More generally, how might knowledge of RV, d l /d a, and d (ln l )/ d (ln a) help you assess life history evolution? For example, in a frog species, might you expect beneficial mutations to be "concentrated" in a particular stage (tadpole vs. adult), or deleterious mutations to be concentrated in particular stages? Why?
osenberg@zoo.ufl.edu