Ecosystem Simulation Model

Introduction:

This model simulates the different interactions of a simple ecosystem consisting of turtles and plants. There is also an option where oxygen and carbon dioxide can become factors.

This model lets the user explore and learn how the various checks and balances that are part of natural ecosystems interact with each other. Whether the ecosystem fails or not is the ultimate goal, but to find out at what levels does the ecosystem balance itself fairly well and at what levels does it fail quickly. For example, it should be clear that having 100 turtles and 1 plant in an ecosytem will certainly lead to extinction.

Rationale:

Brookstone sells for $80 a small glass ecosystem that consists of two male shrimp, one plant, and water. It is a self-contained system that is stated to last over 30 years. We felt it was interesting how every thing could be in perfect balance and wanted to see if we could create our own stable ecosystem.

How it works:

The INITIAL-TURTLES slider indicates the number of turtles at the beginning.
The DENSITY_PLANTS slider indicates the percentage of plants on the screen
The TIME-TO-EAT slider indicates how often a turtle will become hungry
The HOURS_SUNLIGHT slider indicates how many hours of sunlight there are each day

If the user wishes to acticate oxygen/carbon dioxide monitoring, he/she can turn on the OXYGEN switch.

Each turtle is randomly initialized to be either hungry or not hungry. Turtles become hungry at a specified time (time-to-eat slider) and if they land on a patch with food, they will consume 1 or 2 units of food. If they do not find food within 48 ticks (hours), they will die. When the oxygen switch is turned on, each turtle will consume a small amount of oxygen and release a certain amount of carbon dioxide every tick. If oxygen levels drop below 18%, turtles begin to die and below 16% they die with greater frequency.

Each patch is initialized as being eaten or not eaten. A not eaten patch means there is a plant there. Plants initially start with 5 units of food and either grow a new unit of food or create a new plant at a dead patch every 24 hours of sunlight exposure. Plant color varies with the amount of food (more food => darker color). When oxygen is turned on, plants consume a certain amount of carbon dioxide and release a certain amount of oxygen. If carbon dioxide levels drop below 0.8, plants start dying slowly and below 0.6 they die faster.

To Setup the Model:
1) specify the intial number of turtles and the intial percentage of plants on the screen.
2) set the amount of time turtles get hungry (time-to-eat)
3) set the hours of sunlight received each day (hours_sunlight)
4) decide whether or not oxygen/carbon dioxide monitoring should be activated.

created with NetLogo

view/download model file: ecosystem.nlogo

Discussion and analysis:

The main goal is to find some range for the initial number of turtles compared to the initial percentage of plants where we can create a fairly good balance in the ecosystem.

Set time-to-eat at 12 hours and hours_sunlight to 12 hours. By doing this, we are assuming turtles eat twice a day and that half of each day has sunligh, which are reasonable assumptions. We are also assuming that oxygen monitoring is on because it more accurately represents the real world.

If oxygen levels fall below 18%, turtles will begin dying. When they die, less oxygen is being consumed, so oxygen levels begin to rise or decrease at a slower rate. The same effect occurs with carbon dioxide at the 0.8% level.

If oxygen first begins to decrease, this means there are too many turtles compared to the percentage of plants. If carbon dioxide first begins to decrease, this means there are too many plants compared to turtles.

When either oxygen or carbon dioxide decrease/increase at a high rate, the proportion of turtles to plants is not in balance. If the decrease is too great, the ecosystem will die off quickly because the counter measures that at in place will not be able to restore balance quick enough.

We find that when the initial density of plants is set to approximately double that of the initial number of turtles, the ecosystem stays faily consistant. The graphs slowly diverge, until oxygen or carbon dioxide fall below a certain level, prompting either turtles or plants to die. It is a good sign when carbon dioxide begins to fall first, because plants can die and regenerate. Thus, it will lead to a balance. The slower the rate of initial divergence, the better the system balances in the long run. If the oxygen levels decrease at first, turtles will eventually start dying to counteract the decreating oxygen. This will lead to a short run balance, but all the turtle will eventually die off because this model does not factor in turtle reproduction.

It can be seen that this model is faily simple in its outcomes. Based on the rate of increase/decrease and whether it is oxygen or carbon dioxide that is decreasing, the user is able to predict the outcome fairly well.

Summary of results:

Oxygen decreases quickly : ecosystem dies quickly (all turtles dead)

Oxygen decreases slowly : ecosystem eventually dies after many fluctuations (all turtles dead)

Carbon dioxide decreases quickly: most plants die and then turtles die because of no food.

Carbon dioxide decreaes slowly: ecosytem balances itself well, but will eventually die in long run.

The only senario whcih accurately produces a well balanced ecosystem is where carbon dioxide is decreasing slowly. Because it decreases slowly, once it reaches the 0.8% level, plants begin dying off so the CO2 graph begins to rise again and oxygen decreases. When oxygen drops below the 18% level, turtles begin dying, which makes oxygen increase again. This cycle goes on until the number of turtles reach 0. This system can go on for a long time, but due to limitations in the design, the ecosystem will ultimately die. A modification that might balance the ecosystem indefinetely would be to give the turtles a chance of reproduction.

Extending the Model:

What if a natural disaster occured such as meteor striking the earth? All turtles and plants in the immediate area will be instantly killed. How would this change the balance in the ecoystem? At the equilibrium level of turtles and plant, would this event wipe out the ecosystem or would the natural levels gradually come to be?

What if weather was factored in? There could be rain or periods of extreme heat? In these instances, there more be more oxygen in the air or the rain could make the plants grow faster.

What if turtles could reproduce? This might create a better balance when oxygen first begins to decrease, killing off turles.

Credits and References:

This model was developed by Eric Cheng and Mazen Al-khafi.

http://pubweb.nwu.edu/~ewc142/