Assignment 2: Learning about patterns from simulations
This assignment is worth 10 points. This will be due by Thursday Oct 22. You should have time to start in during week3 lab. Since you will not be meeting at the lab during week 4 (you will be working on your specific site with your group during a time of your choice) you can hand in this assignment to the ESM office (SB2 218) or to Professor Rueter during the lectures). See the late work policy
Background - review from lecture concepts
- Simulations are scenarios or potential paths that a system might take.
- They are based on models which can be very simple and yet generate complex patterns.
- Usually you would start by picking a set of parameters (such as temperature and light availability) and a probability distribution of factors (such as chance of a rainy month) and predict crop growth during a summer.
- With many interacting parameters it is impossible to know the which particular scenario is going to happen.
- Instead of trying to predict one exact outcome, you will be looking for the patterns of behavior and how the system responds to changes in the parameters (such as how much grass would grow if the average temperature were 1 degree higher or lower).
- When you see a pattern, you can't absolutely link it to one set of causes because there might be other sets of parameters that give the same outcome.
- You might be able to say that if you observe a particular pattern that you would think it was more or less likely to be caused by a particular model or set of parameters.
Landscape ecology and biodiversity simulations
There are five simulations that are similar to the ones presented in lecture. You will run each simulation under several different conditions and then make some comments (see below).
All work that you submit must be in your own words.
There are two points of this exercise:
1) You will explore some simple animated models of factors that affect biodiversity (fire, predation/refuge, immigration and extinction). These are related to your lab projects, but don't try to apply any of these simulations too closely to your particular project.
2) You should consider how you can learn about interactive processes from using tools such as these. It should be very easy to see the patterns but it may take some time to understand how changing the parameters change the outcomes.
For each of the models listed below, you can open it in your browser window and read the background information. We have provided a few notes on what we want you to focus on first (simple starting points). You can keep exploring any of these models as much as you want. These models are constructed in NetLogo which is a program that allows individual agents (which they call "turtles") to be given simple rules for behavior when interacting with other "turtles".
Read the information sheet for each model and then open the simulation in another window.
The models, starting settings, and questions are listed below.
Answer each of the questions and hand in for the assignment. You can do this in your handwriting if it is legible.
NetLogo community models:
1. fire sweeping through a landscape
http://ccl.northwestern.edu/netlogo/models/Fire
We used this in lecture. Adjust the proportion of trees at the start to explore what the lowest % of trees leads to almost complete burning of the forest. Remember, each run is going to be slightly different, so you may want to run the simulation several times at
set the forest percentage
hit setup
hit go
Questions:
1a. What is the range of critical forest densities that leads to complete burn. What was the lowest % density that lead to a > 90% burn?
1b. Describe the pattern of the advancing forest burn. Why do you think it looks like that?
2. landscape changes and fire suppression
http://ccl.northwestern.edu/netlogo/models/community/Fire%20Ecology
This is a more complicated version of fire and recovery. We will focus on the impact of suppressing forest fire.
Compare fire size and pattern with no fire suppression with higher levels of suppression.
2a. Describe the pattern of mature trees, regrowing areas and fire. How does it change from run to run and what can you say in general about the pattern.
2b. What are the effects of increased fire suppression?
3. Refuge and predator prey interactions:
http://ccl.northwestern.edu/netlogo/models/community/Wolf%20Sheep%20Predation%20Refuge
This simulation is very abstract. The refuge is a region at both sides of the screen in which the wolves can't go.
Increase the size of the refuge and notice the pattern of interaction between wolves and sheep with time.
3a. What refuge conditions lead to collapase of the populations, oscillation, boom and bust?
3b. Find a setting that leads to oscillation. Describe why this happens, i.e. what is the sequence of factors that seems to repeat.
4. Island biogeography
explanation:
http://faculty.etsu.edu/JONESTC/Virtualecology/PDFs/IslandBiogeoInfo.pdf
model:
http://faculty.etsu.edu/jonestc/Virtualecology/Models/Island.html
Be sure to read the background explanation on this model. This model simulates the effect of size and distance from the mainland on the equilibrium species diversity on islands. This is related to the cooridors and size of green space "islands" in the middle of an urban environment.
You should focus on comparing two different island sizes or two different distances to see how those factors work. Don't change other factors until you've worked on that.
4a. Describe, in your own words, the processes that lead larger and closer islands to have higher biodiversity. Give an example of a comparison you ran with the model.
4b. How might you modify this model to examine the immigration and extinction of amphibians in a small Portland Park?
General statements:
5a. Describe how these simulations pertain to the study of factors that limit biodiversity in Portland Parks. Which simulation do you think is most relevant in this sense?
5b. How might you incorporate one of these simulations (or something similar) to study your particular study site?
|