Deer Impact on Seed Dispersal Networks

ISEF Category: Plant Sciences

Subcategory: Ecology  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: Full Year

The Hook

A forest can look healthy and still lose its future. If deer eat too many seedlings and the wrong plants stop reproducing, the whole seed network starts to thin out. You can model that shift with public forest plot data and see how overbrowsing changes plant survival and spread.

What Is It?

This project studies how seeds move through a forest when deer numbers rise above natural levels. Seed dispersal means how seeds get from one place to another, often by animals, wind, or gravity. In a healthy forest, many plant species have a chance to spread, find open space, and replace older plants.

An agent-based simulation treats each deer, plant, or seed as an acting unit, or agent. Think of it like a game board where each piece follows rules. Deer browse on seedlings, seeds land in new spots, and forest plots change over time. You can compare a natural baseline with a deer-heavy scenario and see how the network of plant spread changes. Public Eastern U.S. forest plot data gives you real-world numbers to calibrate those rules.

Why This Is a Good Topic

This is a strong science fair topic because you can test clear variables, compare two conditions, and use real ecological data. You do not need a wet lab. You need logic, coding, and careful data analysis. The project connects to forest regeneration, invasive pressure from overabundant deer, and biodiversity loss. You can learn how to build a model, validate it with real data, and interpret patterns in a messy natural system.

Research Questions

• How does deer density change the number of plant species that successfully recruit into a forest plot?
• What is the effect of deer overpopulation on the connectivity of seed-dispersal networks?
• Does a higher browsing rate reduce the spread distance of seedlings in the simulation?
• To what extent do native tree species recover under a natural baseline compared with a deer-heavy scenario?
• Which seed-dispersal pathway, animal, wind, or gravity, is most resilient to deer overbrowsing?
• How does variation in seedling survival alter long-term forest composition under different deer densities?

Basic Materials

• Laptop or desktop computer with internet access.
• Spreadsheet software such as Google Sheets or LibreOffice Calc.
• Python installed with Jupyter Notebook or another notebook editor.
• Free public forest plot data from the USDA Forest Service, the Smithsonian ForestGEO network, or state forest inventory sources.
• Notes document for model rules, assumptions, and variables.
• Basic graphing tool for charts and network diagrams.

Advanced Materials

• Laptop or desktop computer with internet access.
• Python with NumPy, pandas, NetworkX, Matplotlib, and SciPy.
• Jupyter Notebook for simulation runs and documentation.
• GIS software such as QGIS for mapping plot locations and spatial patterns.
• PubMed or Web of Science access through a school library for ecology review articles.
• Public Eastern-U.S. forest plot datasets from ForestGEO, USDA Forest Service, or NOAA-linked data portals.
• Optional cluster or cloud notebook access for repeated simulations.

Software & Tools

• Python: Runs the agent-based simulation and processes forest plot data.
• Jupyter Notebook: Lets you build, test, and document each model step in one place.
• pandas: Organizes plot data, species records, and simulation outputs.
• NetworkX: Builds and measures seed-dispersal networks.
• QGIS: Maps forest plots and helps you compare spatial patterns.

Experiment Steps

1. Define the forest system you will model, including which plants, deer effects, and dispersal routes matter most.
2. Choose the real dataset or datasets that will anchor your baseline values and explain why those plots fit your question.
3. Translate ecological rules into agent behavior, such as browsing pressure, seed arrival, and seedling survival.
4. Build a baseline scenario and a deer-heavy scenario so you can compare network structure and regeneration outcomes.
5. Decide which metrics will count as your evidence, such as species richness, connectivity, or recruitment success.
6. Plan sensitivity tests so you can see which assumptions change the results most.

Common Pitfalls

• Using deer presence as a vague idea instead of a measurable browsing pressure, which makes the model impossible to compare.
• Picking plot data without checking whether the species list, region, or sampling method matches your question.
• Building a simulation with too many rules at once, which hides the effect of deer density.
• Treating one forest plot as proof for all Eastern forests, which weakens the scope of your conclusions.
• Reporting only the final network picture without quantifying species loss, connectivity, or recruitment change.

What Makes This Competitive

A stronger project goes beyond a simple before-and-after comparison. You can make your model stronger by testing how sensitive the results are to different browsing rules, dispersal distances, or seedling survival rates. You can also compare multiple forest types or species groups instead of one averaged forest. Clear validation against public plot data and careful network metrics will make your work read like real ecology research.

Project Variations

• Focus on understory herbs instead of trees to see whether small plants respond differently to deer pressure.
• Compare wind-dispersed species with animal-dispersed species to test which group loses more network connections.
• Add habitat fragmentation to the simulation and measure whether deer effects get worse in smaller forest patches.

Learn More

• ForestGEO: Search the network’s public data resources and site descriptions for forest plot records and long-term ecology studies.
• USDA Forest Service FIA Program: Find the Forest Inventory and Analysis database for U.S. forest composition and regeneration data.
• USGS Nonindigenous Aquatic and Terrestrial Species database: Look for background on species spread and ecological disruption.
• NOAA Climate Data Online: Use climate context for your forest sites, including temperature and precipitation patterns.
• PubMed: Search for review articles on deer browsing, seed dispersal, and forest regeneration in eastern North America.
• MIT OpenCourseWare, Ecology and Evolutionary Biology materials: Use lecture notes and assignments to learn modeling and population ecology concepts.