CASE TEACHING NOTES
for
The Wolf, the Moose, and the Fir Tree:
Who Controls Whom on Isle Royale?
A case study of trophic interactions

by
Gary M. Fortier
Department of Small Animal Science
Delaware Valley College

Source: McLaren, B.E. and R.O. Peterson. 1994. "Wolves, moose, and tree rings on Isle Royale." Science 266:1555-1558.

Introduction:
This case study is intended as an analysis case for a sophomore ecology class. It is assumed that the students will already be familiar with the terminology and concepts fundamental to ecology. This case study may be used to introduce the use of predators as control agents. It might follow a case study on either the reintroduction of wolves into the Yellowstone ecosystem or the proposed introduction of grizzlies into the Bitterroot ecosystem (see Grace A. Wang's case, The Bear Facts: Grizzly Recovery in the Bitterroot Ecosystem, and its accompanying teaching notes). Both of these celebrated introductions involve arguments on the nature of predator-prey dynamics--dynamics that remain poorly understood. Analysis of this case may lead some students to reassess their views on the introduction of predators as control agents.

Case Objectives:
This is an analytical case designed to enhance the ability of students to read and comprehend scientific findings as reported in the literature. It also requires students to evaluate alternative hypotheses. The pedagogical goals of this case include the following analytical and conceptual objectives.

Analytical objectives:

Conceptual objectives:

Classroom Management:
The case will be run in undergraduate ecology laboratories of 12-15 students. The class will be broken down into three collaborative learning groups of 4-5 students that will work together on the case. Fifteen minutes will be allocated to a review of the relevant ecological concepts and an introduction to the case.

The students will be given about 75 minutes to work through the case in their groups. The three handouts will be distributed sequentially, not simultaneously. After receiving each handout, the class will be given 15 minutes to read through the material, write out answers to the accompanying questions, and discuss their answers within their learning group. At that point, one of the groups will be chosen to explain their responses to the rest of the class. Ten minutes will be allocated for each presentation and for a general discussion of any points of contention or confusion.

After the last dataset has been discussed, 30 minutes will be allocated to a final synthesis and an explanation of any follow-up assignments. Thus the total time allocated to this case will be two hours, to be completed within a single laboratory period.

Follow-up Exercises:
This case is intended as a follow-up to a dilemma-style case study on the reintroduction of predators into their native ecosystems. Thus I am primarily interested in how the information from the current case may affect the views students developed previously. Students will be asked to bring together the different elements affecting policy decisions by writing a letter to their state governor either supporting or opposing the reintroduction of predators. They will be required to support their decision with references to the current article or other relevant literature.

Students could also develop their ideas further, without employing a second case study, through several alternative assignments.

  1. Write a position paper defending one of four possible conclusions based on the data presented:
    • Reject the primary productivity hypothesis and accept the trophic cascade hypothesis
    • Accept the primary productivity hypothesis and reject the trophic cascade hypothesis
    • Reject both hypotheses
    • Reject neither hypothesis
  2. Find and evaluate a second article on trophic interactions from the primary literature. Be sure to include the following parameters:
    • What specific hypotheses were evaluated?
    • Can you identify the predictions of each hypothesis? Are they mutually exclusive?
    • What assumptions are made by the authors?
    • Can you identify any caveats to the design or the data presented in the article?
    • Do the authors suggest alternative explanations that would also explain their findings? Can you identify alternative hypotheses?
  3. Many of the questions raised in this case could become the basis for problem-based learning exercises. Data relevant to several of the assumptions in this paper have been published elsewhere. Students should be able to find additional information on the relationship between canopy dynamics, environmental conditions, and tree growth rates (Bartholomay et al. 1997, D'Arrigo and Jacoby 1993, Raison and Meyers 1992, Temple et al. 1993, Thomas 1996), factors that affect foliar biomass (Murray et al. 1996, Niinemets 1997), and the prevalence of trophic cascades in other types of ecosystems (Letourneau and Dyer 1998, Matveev 1995, Mikola and Setala 1998, Moran et al. 1996, Preszler and Boecklen 1996). The learning groups can each identify one point they would like to clarify and then research that issue in the library. Each group can deliver an oral report on their learning objective the following week. Student teams would then be in a better position to evaluate the merit of each of the authors' conclusions in a follow-up discussion.

Case Analysis:
Handout #1: Introduction

Experimental manipulations can be quite difficult in systems containing large vertebrates. Consequently, long-term observations and correlational analysis are often used to assess interactions between different trophic levels. The limitations of this approach are well known; strong correlations may not be indicative of causation. The Isle Royale data constitute a widely recognized study of predator-prey interactions. The censusing of moose and wolf populations has been thoroughly documented, allowing the authors to infer causation based on the relative timing of events.

The main goal of this study was to determine whether changes in primary productivity or herbivore-carnivore interactions are responsible for the suppression of plant growth in the Isle Royale National Park. Before reviewing any data, students should be able to predict the nature of any relationships between plants, herbivores, and predators.

Questions 1-3
The primary productivity hypothesis suggests that there should be positive correlations between each trophic level; inverse correlations would be predicted under the trophic cascade model. Furthermore, the direction of control is different under each hypothesis. Plant growth may be controlled from the top by herbivory or from the bottom by the availability of resources. Thus the removal of a top predator, such as the gray wolf, would lead to increased moose density and decreased plant growth under the trophic cascade model. If plant growth is limited by primary productivity, wolf removal will have no effect on the growth of the firs.

Question 4
If we use tree-ring analysis as an index of herbivore pressure, we are assuming that any reduction in ring width has been due to a loss of foliar biomass through predation. This will only be true if growth is not being limited by precipitation or climatic conditions.

Regarding the historical impact of moose on the fir, the balsam fir exhibited a strong decline following the colonization of Isle Royale by moose early in this century. However, this only supports the inference that moose herbivory is a significant pressure on firs if no other major changes occurred on the island in the last eighty years. Other factors, in addition to predation pressure, might have accounted for the demise of balsam fir. We must assume that there were no major fires or storms, no significant new diseases or pests were introduced, no new competitors arrived, etc. We need more information before we can confirm or reject these assumptions.

Handout #2: Figure 1


Fig. 1. Population parameters of the Isle Royale ecosystem from 1958-1994. Shaded areas signify periods of forage suppression that may be connected to interactions between herbivores and carnivores.
  1. Population size of wolves each winter (based on aerial counts).
  2. Population size of moose each winter (based on aerial counts and skeletal remains).
  3. Ring-widths from the west end of Isle Royale, N=8.
  4. Ring-widths from the east end of Isle Royale, N=8.
  5. Actual evapotranspiration rates (AET), annual calculations based on data from April-October at a weather station 20 km from Isle Royale. AET is an approximation of primary productivity, it represents water availability as a function of temperature and rainfall.

Image Credit: Regraphed from information published in Science 226 (December 2, 1994): 1557

Question 1
The banded areas on these graphs may be confusing to some students. Different areas of each graph are shaded to show a possible relationship between minima and maxima at different trophic levels. However, the banded areas of the two ring-width plots (graphs C & D) are lined up; these bands represent minima on graph C (west) but are neither minima nor maxima on graph D (east).

Questions 2-3
The authors used these figures to suggest that periods of fir suppression are coupled to maxima in the moose population, though there is a 1-2 year lag in the west end of the island and a 5-year lag on the east end. This may be due to the climatic differences across the island. If more favorable conditions exist on the west end, then firs on that end of the island should have a higher growth rate and should recover more quickly once the moose population has declined. This interpretation would support the trophic cascade hypothesis-- moose populations limit fir growth.

Question 4
The moose and wolf populations appear to be inversely related. The highlighted wolf minima are believed to have caused the moose maxima. This is supported by data on the aging of the moose population (older moose suffer higher predation rates—data presented elsewhere) and the fact that the moose minima follow the wolf maxima in time. In other words, causation is inferred by the timing of events.

Question 5
There is no apparent relationship between AET and ring widths, undermining one of the central predictions of the primary productivity hypothesis. This could be interpreted as indirect support for the trophic cascade model if we assume that these hypotheses are mutually exclusive.

Handout #3: Figure 2


Fig. 2. Ring-widths of balsam firs from Isle Royale. Each line represents data from an individual tree harvested in 1992. Note that moose are able to browse as high as 3m.
  1. Location RH (N=10), firs from this area were 26-48 years old and exceeded 3m in height during the late 1970s.
  2. Location SS (N=9), firs from this area were 48-60 years old and were less than 2m in height.

Image Credit: Regraphed from information published in Science 226 (December 2, 1994): 1557

Questions 1-3
This figure captures several differences between the RH and SS sample sites; the mere number of parameters involved may create confusion for some students. These sites differed with respect to the precipitation, the degree of canopy enclosure, the type of forest present, the age of the trees, and the height of the trees. These data are included to demonstrate that producers may respond to changes in primary productivity, but only when large-scale disturbances override the effects of herbivore-carnivore interactions. The RH site is in the eastern section of Isle Royale; the previous handout explains that this end of the island has less precipitation, lower temperatures, and probably experiences lower growth rates overall. However, it is also in a disturbed, open-canopy area; the dramatic increase in light and space available should probably override climatic conditions and accelerate new growth. Furthermore, these trees were tall enough to be above the browse line in the late 1970s, thus they were largely released from suppression by moose. Trees at the SS site are well within the reach of moose -- they are in a closed-canopy area with reduced light availability. Thus only the RH site trees should be able to escape suppression by moose and respond to the increase in primary productivity that has occurred recently.

Questions 4-5
The data from the two sites obviously diverge after 1975; they may support either hypothesis depending on the assumptions we are willing to accept. Trees from the RH site appear to escape suppression from moose herbivory after the moose begin to decline, trees at the SS site do not. Furthermore, the RH firs continue to experience high growth rates even after the moose population rebounds in the late 1980s. Both of these observations suggest that herbivores do not exert significant pressure on the growth of balsam firs. However, the authors claim that these findings do not undermine the trophic cascade hypothesis. Instead, they suggest that external disturbances may eliminate the effects of higher trophic interactions and allow producers to respond to underlying changes in productivity.

The importance of each factor may be difficult to tease apart due to the large number of differences between the two sites. It is difficult to attribute the upsurge in graph A to local habitat effects because of the confounding effects of tree height. If the trees at the RH site escape herbivory due to their height, then trees > 3m should respond to changes in productivity regardless of their location on the island. Without these data it is difficult to assess the relative importance of local habitat effects and their influence on higher order trophic interactions. To recapitulate, the data support the trophic cascade model but only if we assume:

  1. The deviations in tree ring width shown in Fig. 2 are exceptional and are due to local habitat effects. Thus any correspondence between AET and ring width will limited and will only be observed in disturbed areas.
  2. The methodologies employed are appropriate. AET is fair measure of primary productivity throughout this ecosystem; tree ring width is a reasonable estimate of plant growth over time.
  3. If population changes follow one another, then the latter change was likely to have been caused by the former.

Question 6
Many of the confounding factors listed above could be eliminated in a well designed experiment. One obvious solution is to remove or exclude the wolves from one portion of the island to observe any corresponding changes in fir growth rates or the densities of moose. Large-scale experiments can be quite difficult to execute; however, they reduce our dependence on correlation and allow us to test specific causal mechanisms.

Relevant Literature

Bartholomay, G.A., R.T. Eckert, and K.T. Smith. 1997. Reductions in tree-ring widths of white pine following ozone exposure at Acadia National Park, Maine, U.S.A. Canadian Journal of Forest Research 27:361-368.

D'Arrigo, R.D. and G.C. Jacoby. 1993. Tree growth-climate relationships at the northern boreal forest tree line of North America: evaluation of potential response to increasing carbon dioxide. Global Biogeochemical Cycles 7:525-535.

Letourneau, D.K. and L.A. Dyer. 1998. Density patterns of Piper ant-plants and associated arthropods: top-predator trophic cascades in a terrestrial system? Biotropica 30:162-169.

Matveev, V. 1995. The dynamics and relative strength of bottom-up vs. top-down impacts in a community of subtropical lake plankton. Oikos 73:104-108.

McLaren, B.E. and R.O. Peterson. 1994. Wolves, moose, and tree rings on Isle Royale. Science 266: 1555-1558.

McRoberts, R.E., L.D. Mech, and R.O. Peterson. 1995. The cumulative effect of consecutive winter's snow depth on moose and deer populations: A defense. Journal of Animal Ecology 64:131-135.

Mikola, J. and H. Setala. 1998. No evidence of trophic cascades in an experimental microbial-based soil food web. Ecology 79:153-164.

Moran, M.D., T.P. Rooney, and L.E. Hurd. 1996. Top-down cascade from a bitrophic predator in an old-field community. Ecology 77:2219-2227.

Murray, P.J., D.J. Hatch, and J.B. Cliquet. 1996. Impact of insect root herbivory on the growth and nitrogen and carbon contents of white clover (Trifolium repens) seedlings. Canadian Journal of Botany 74:1591-1595.

Niinemets, U. 1997. Energy requirements for foliage construction depends on tree size in young Picea abies trees. Trees 11:420-431.

Peterson, R.O., N.J. Thomas, J.M. Thurber, J.A. Vucetich, and T.A. Waite. 1998. Population limitation and the wolves of Isle Royale. Journal of Mammalogy 79:828-841.

Preszler, R.W. and W.J. Boecklen. 1996. The influence of elevation on tri-trophic interactions: opposing gradients of top-down and bottom-up effects on a leaf-mining moth. Ecoscience 3:75-80.

Raison, R.J. and B.J. Myers. 1992. The biology of forest growth experiment: linking water and nitrogen availability to the growth of Pinus radiata. Forest Ecology and Management 52:279-308.

Temple, P.J., G.H. Riechers, P.R. Miller, and R.W. Lennox. 1993. Growth responses of ponderosa pine to long-term exposure to ozone, wet and dry acidic deposition, and drought. Canadian Journal of Forest Research 23:59-66.

Thomas, S.C. 1996. Asymptotic height as a predictor of growth and allometric characteristics in Malaysian rain forest trees. American Journal of Botany 83:556-566.

Thurber, J.M. and R.O. Peterson. 1993. Effects of population density and pack size on the foraging ecology of gray wolves. Journal of Mammalogy 74:879-889.

Vucetich, J.A., R.O. Peterson, and T.A. Waite. 1997. Effects of social structure and prey dynamics on extinction risk in gray wolves. Conservation Biology 11:957-965.

Further Reading

Fortin, D., H.L. Beyer, M.S. Boyce, D.W. Smith, T. Duchesne, and J.S. Mao. 2005. Wolves influence elk movements: Behavior shapes a trophic cascade in Yellowstone National Park. Ecology 86(5):1320-1330.

Robbins, J. October 18, 2005. Hunting Habits of Yellowstone Wolves Change Ecological Balance in Park. The New York Times Section F, Column 1, Science Desk, p. 3.

Internet Sites

Predator-prey relationships
http://www.globalchange.umich.edu/globalchange1/current/lectures/predation/predation.html
(qualitative explanation of trophic interactions, including the trophic cascade model)

North American Wolf Association
http://www.nawa.org/
(emphasizes education and restoration)

Timberwolf Information Network - wolf information
http://www.timberwolfinformation.org/info/info.htm
(includes ecological reports from Isle Royale wolf population)

Our Living Resources (U.S. Deparment of the Interior)
http://biology.usgs.gov/s+t/index.htm
(information on wolves, elk, and deer; also includes an overview of the Yellowstone Ecosystem and species recovery plans, including wolves)

Restoring America's Wolves
http://www.nwf.org/wildlife/graywolf/
(From the National Wildlife Federation)

Moose Decline on Isle Royale
http://www.sas.it.mtu.edu/urel/PressReleases/97releases/moose.html
(press release of recent population changes)

Predator control of ruffe
http://www.glsc.usgs.gov/main.php?content=research_invasive_ruffe&title=Invasive%20Fish0&menu=research_invasive_fish
(snapshot of a potential top-down system of control in an aquatic system; why it won't work)

Acknowledgements: This case was developed as part of a National Science Foundation-sponsored Case Studies in Science Workshop (NSF Award #9752799) held at the State University of New York at Buffalo on June 1-5, 1998.

Last Updated:  10/20/05 nas
This file is also available in Adobe Portable Document Format (PDF).  PDF Version

comments page Top page Isles Royale