swanson-image  biodiversity-image

Geography 423/523: Advanced Biogeography: Landscape Ecology

Winter 2019

Meeting times: Monday, Wednesday: 10:00-11:20 in 106 Condon

Instructor: Daniel Gavin (dgavin@uoregon.edu)
Office: 110 Condon Hall; Phone: 346-5787
Office Hours: Wednesdays 3:00-4:30.
Required Texts ():

Landscape Ecology in Theory and Practice, 2nd Edition. 2015. By Monica Turner and Robert Gardner. Springer

Note: Use the link above to access a free online copy of the book through the UO Libraries.
Plus additional readings.

Course Overview

The current patterns in the natural world are strongly linked to processes occurring at a range of spatial scales, from centimeters to continents. They are also the result of a series of contingent events that have played out over time periods of days to millennia. In this class we examine the types patterning occurring across landscapes (scales of 10's to 100's of kilometer) and how they have emerged from changing disturbance regimes (fire, disease), climate, and effects of humans. The subject matter will deal mostly with biodiversity, forest vegetation, and disturbance regimes. The class will serve as an introduction to Landscape Ecology, and therefore some background in ecology or biogeography is important. Two special emphases in this class are 1) use of stochastic modeling to understand patterns of landscape structure, and 2) use of paleoecological methods to explore the question of "what is natural" with respect to changing environments of the past 500 years.

Topics covered will include:

  • Introduction to concepts of scale and drivers of landscape pattern.
  • Quantifying landscape pattern (spatial statistics)
  • Stochastic simulation modeling, especially state-and-transition models.
  • Variability (heterogeneity) of landscapes and influence on ecosystems and species interactions (e.g., invasions).
  • Paleoecological methods to reconstruct landscapes of the past.

Prerequisite: GEOG 323 (Biogeography), or permission of the instructor. For permission, contact me by email and list relevant course experience including ecology.

Each week of the course will address a chapter in the textbook, with a few exceptions. Mondays will be a general lecture (but I always encourage discussion) and Wednesdays will involve addressing a few broad questions in small groups and later as a full class.

Differentiation of Geog 523 and 423: Geog 523 students will read one journal article per week and meet at a separate time to discuss this article. This time is to be determined soon.


  1. Problem sets: four problem sets through the term will involve some sort of written or analysis problem related to the topic of the previous week. 40% of grade.
  2. Midterm exam: Several short-answer questions. End of week 4. One page of notes is allowed (to be submitted with the exam). 15% of grade.
  3. Final exam: Addresses the second half of the class. One page of notes is allowed. 10:15 Tuesday, March 19. 25% of grade.
  4. Landscape model project: Group project that will be due Wednesday of Finals Week. 20% of grade. See bottom of this page for details.
schedule in development

Week 1

  • Topic: Introduction to landscape ecology, concepts of scale and pattern.
  • Readings:
    • Chapter 1.
    • Daubenmire, R. 1980. Mountain Topography and Vegetation Patterns. Northwest Science 54:146-152.
  • Discussion topics
    • Class discussion (Wed:): See questions on pages 31 and 32 of the textbook.
    • A common issue in landscape ecology is to meaningfully address patterns that occur at different spatial scales. A pattern of forest vs. grassland may be evident on a map at 1-km grid cell size. But much finer patterns exist at 30-m grid cell sizes. One challenge is to downscale observations at a coarse resolution to a much finer resolution, often by using additional information at the finer resolution (such as elevation, aspect, etc). Which concepts in the Daubenmire paper have implications for the issue of downscaling? What ways might this be overcome?
    • Geog 523 discussion: Schneider, D.C. Applied Scaling Theory.

Week 2

  • Topic: Causes of landscape pattern
  • Readings:
    • Chapter 2
  • Discussion topics
    • See questions at end of Chapter 2.
    • Geog 523 discussion: Dobrowski, S. Z. 2011. A climatic basis for microrefugia: the influence of terrain on climate. Global Change Biology 17:1022–1035.
  • Problem set 1: Southwest Oregon vegetation gradients.

Week 3

  • Monday: no class (MLK day)
  • Topic: Introduction to models
  • Readings:
    • Chapter 3

Week 4

Week 5

Week 6

  • Topic: Landscape Disturbance Dynamics
  • Readings:
    • Chapter 6
  • Discussion topics
  • Problem set 3 (Landscapemetrics). See Canvas under 'Files'. Due Friday Feb 15th.

Week 7

  • Topic: Organisms and Landscape Pattern
  • Readings:
    • Chapter 7
  • Discussion topics
    • ST-SIM demonstration.
    • Geog 523 discussion: Frelich, L.E. 2002. Forest dynamics and disturbance regimes. Cambridge Studies in Ecology. Chapter 8: Forest stability over time and space.

Week 8

  • Monday: Snow Day
  • Wednesday: Snow Day

Week 9

  • Problem set 4 (Tuesday): ST-SIM disturbance simulation.
  • Topic: Reconstructing Past Landscapes
  • Readings:
    • Schoonmaker, P.K. 1998. Paleoecological perspectives on ecological scale.
  • Discussion topics
    • How do we define a 'historical baseline'. How do we quantify what it was? What is the value of it?
    • How do we recontruct past landscapes?
    • What lessons come from the paleo-record that are relevant to interpreting landscapes today?

Week 10

  • Topic: Landscape Dynamics in a Rapidly Changing World
  • Readings:
    • Chapter 9
  • Discussion topics
    • Geog 523 discussion: Morelli, T. L., C. Daly, S. Z. Dobrowski, D. M. Dulen, J. L. Ebersole, S. T. Jackson, J. D. Lundquist, C. I. Millar, S. P. Maher, W. B. Monahan, K. R. Nydick, K. T. Redmond, S. C. Sawyer, S. Stock, and S. R. Beissinger. 2016. Managing Climate Change Refugia for Climate Adaptation. PLOS ONE 11:e0159909.
Final Exam: 10:15 Tuesday, March 19. Same room location.

Final project

Basic guidelines of the project:

  • Group projects encouraged....two to three people per group. OK to mix 423/523 students.
  • Use the ST-SIM simulation model for the Applegate (download from Canvas, along with the PDF file showing the model design and input maps).
  • Design a model experiment in which two parameters are varied. The parameters should be varied over a wide range of values. Possible parameters to change include:
    • One or more transition probabilities for fire, and/or
    • The fire size distribution for replacement-severity fire, and/or
    • The rate of harvest and/or the target levels for harvest.
  • Get inspired to explore certain parameters in the Applegate model. You could even read the issues of fire management in the Applegate Valley by reading parts of the Applegate Fire Plan (in Canvas).
  • Be sure to run the model long enough so that initial conditions are no longer affecting output. This occurs, for the model as distributed, at about 300 years. Then run the model longer in order to capture variability. Thus the period of 300 to 600 years may be the correct time period to view results.
  • Examine the response of some important aspect of the environment to the chosen values. For example, the amount of timber harvest, the amount of chaparral, the variability and amount of mature forest, etc.
  • Ideally, collect data from many runs of the model (across a range of parameter values) of the response of the model. Make a scatter plot of the response of the model to the parameter value chosen.

To meet these guidelines, the following will be needed:

  • A minimum of four maps showing how spatial outputs vary among scenarios
  • About four line plots showing variability of state-classes over time.
  • A minimum of three pages of text, double-spaced, that explains your model experiment and your interpretation of output.
  • It may be possible to export the output maps as a .tif image to be analyzed using landscapemetrics in R. However, I have not tried this and I suggest attempting this only if your group has a set of results for the model and can afford to spend the effort on this.
    • Be sure to explain the rationale for the choice of parameters you are varying. What real-world situation would cause those transition rates to change?
    • Please ensure roughly equal contribution of all members of the group. Do this by including a table like below (with actual names). Everyone must contribute to writing. I would like submission comments added to Canvas from each group member stating "I agree with the reported level of my contribution to the project." The project itself needs to be submitted by only one person in a group. The table should show that no person contributed less than 60% of the average contribution. If not then points cannot be distributed evenly among students.
Contribution Student1 Student2 Student3
Brainstormed and contributed to experiment idea 10%60%30%
Ran the model simulations, saved the output 80%0%20%
Helped process the output files and provided interpretation of results 30%60%10%
Wrote the first draft and edited the text 20%10%70%
Total 140%130%130%
Grading rubric

30%: Idea of the experiment: explanation for the reasoning for the experiment(s) conducted. Why choose the two parameters, what might they represent in the real world?

20%: Presentation of the results of the experiments. Neatness of graphs, use of figure captions explaining content of each figure.

30%: Interpretation of the results...are the results expected from what you may have predicted? What did you learn about the model?

20%: Clarity of the writing. Addressing the implications of the results for land cover change in this region. Use of citations when required.

Department of Geography, University of Oregon
Modified Feb 27, 2019