Model Plant Competition After Fire With Simulation

ISEF Category: Environmental Engineering

Subcategory: Land Reclamation  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

After a wildfire, the land does not reset itself. Plant species race for space, water, and sunlight, and the winners can shape the whole recovery. A simulation lets you test that race without waiting for a real fire season. You can change rainfall, species traits, and recovery rules, then watch the outcome change.

What Is It?

An agent-based simulation is a computer model where each plant acts like a small decision maker. Each agent follows rules. A native seedling may grow best in one kind of soil moisture, while an invasive plant may spread faster after disturbance. When you run many agents together, bigger patterns appear, like who dominates a burned area and who fades out.

Think of it like a strategy game for plants. Each plant cannot see the whole landscape, but it responds to nearby space, water, and competition. NetLogo and Mesa are two tools that let you build this kind of model. You can create a grid that stands in for a post-fire site, then test how rainfall patterns change survival, spread, and recovery.

This topic mixes ecology, coding, and environmental engineering. You are not guessing what should happen. You are building a system and checking how the system behaves under different climate scenarios.

Why This Is a Good Topic

This is a strong science fair topic because you can test clear variables, like rainfall level, drought timing, or invasive growth rate, and measure clear outputs, like percent cover or time to native recovery. It connects to real land reclamation problems after fire, where managers need to know which plants help recovery and which ones take over. You can learn modeling, data analysis, and how to turn an ecological question into code and graphs.

Research Questions

• How does lower average rainfall change the balance between invasive and native plant cover after fire?
• What is the effect of drought timing on native seedling survival in a post-fire landscape?
• Does higher invasive dispersal rate increase long-term dominance under the same rainfall scenario?
• To what extent do native regrowth speed and invasive spread speed interact to change recovery time?
• Which rainfall pattern, steady, pulsed, or erratic, produces the greatest loss of native plant cover?
• How does initial planting density change the chance that native plants recover under climate stress?

Basic Materials

• Laptop or desktop computer with enough memory to run simulations.
• NetLogo or Mesa installed.
• Spreadsheet software such as Google Sheets or LibreOffice Calc.
• Basic graphing tool in the spreadsheet software.
• Notebook for model rules, parameter choices, and test runs.
• Access to published papers or government reports on post-fire vegetation recovery.

Advanced Materials

• Laptop or desktop computer with Python installed.
• Mesa package in Python.
• Jupyter Notebook for building and testing the model.
• NetLogo for cross-checking results against a second platform.
• GIS or raster data source for turning a real burned landscape into a grid.
• R or Python libraries for statistical testing and visualization.
• Version control software such as Git for tracking model changes.

Software & Tools

• NetLogo: Lets you build and run agent-based ecology models with a visual interface.
• Mesa: Supports agent-based modeling in Python and works well for custom rules and data output.
• Jupyter Notebook: Helps you organize code, notes, plots, and model runs in one place.
• Google Sheets: Lets you clean results and make simple comparison charts.
• Python: Handles simulation logic, repeated runs, and statistical analysis.

Experiment Steps

1. Define the post-fire question you want to answer and the exact output you will measure, such as native cover, invasive cover, or recovery time.
2. Choose the plant traits and landscape rules your agents will follow, including growth, spread, survival, and competition.
3. Build a simple baseline model first, then add rainfall scenarios one at a time so you can see what each change does.
4. Plan control runs and repeated trials so random chance does not dominate your results.
5. Test how sensitive your outcomes are to key assumptions, such as dispersal distance or drought tolerance.
6. Compare scenario outcomes with published ecology or land management findings to judge whether your model behavior makes sense.

Common Pitfalls

• Making the model too complex too early, which hides whether rainfall or species traits actually drive the result.
• Using vague plant rules, which makes the simulation look realistic but impossible to interpret.
• Running only one trial per scenario, which lets random variation distort the pattern.
• Changing several parameters at once, which prevents you from telling which rainfall pattern caused the outcome.
• Ignoring real ecological constraints, which can make the model predict impossible spread or recovery speeds.

What Makes This Competitive

A stronger project goes beyond a basic yes or no answer. You can compare multiple rainfall patterns, test sensitivity to model assumptions, and check whether your results stay stable across many runs. You can also compare model output to published post-fire recovery studies or remote sensing trends. That kind of careful analysis makes your project feel like real research, not just a class demo.

Project Variations

• Model shrubland, grassland, or forest recovery to see whether plant type changes the invasion outcome.
• Swap rainfall for soil moisture retention or erosion risk to study another part of land reclamation.
• Compare NetLogo and Mesa outputs to see whether platform choice changes model behavior or interpretation.

Learn More

• NetLogo Models Library: Search the NetLogo site for ecology and population dynamics models to study how agent rules are structured.
• Mesa Documentation: Read the official Mesa docs for Python-based agent modeling examples and data collection tools.
• NOAA Climate Data Online: Find rainfall records and climate normals to ground your scenarios in real weather patterns.
• USGS Ecosystems Program: Search for reports on invasive species, fire recovery, and vegetation change after disturbance.
• Ecological Modelling: Search the journal in PubMed, Google Scholar, or your school library for agent-based plant competition papers.
• NASA Earthdata: Look for burned area, vegetation, and climate datasets that can help you connect simulation to real landscapes.