Explaining Disjunct Plant Ranges With Biogeography

ISEF Category: Plant Sciences

Subcategory: Systematics and Evolution  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

Some plants live in places that make no sense at first glance. A genus can show up in two far-apart regions with empty space in between, like pieces of a puzzle that lost the middle. Your job is to ask how that split happened. Did the plants once grow across a wider range, or did they move in separate waves?

What Is It?

Biogeography is the study of where living things live, and why. In this project, you pick a genus with a weird range pattern, then use occurrence records, climate history, and evolutionary trees to test competing explanations.

Think of the range like a map with footprints. GBIF records give you the footprints. Paleoclimate reconstructions tell you what the environment looked like in the past. Dispersal-vicariance models help you ask whether the genus spread to new places, or whether a once-connected population got split when habitats changed.

Why This Is a Good Topic

This makes a strong science fair topic because you can test real hypotheses with public data. You do not need to grow plants in a lab, and you can still ask a serious evolutionary question. The project connects climate change, plant history, and species distribution, so your results matter beyond one genus. You also learn data cleaning, mapping, and model interpretation, which are skills judges notice.

Research Questions

• How does the current geographic split of the chosen genus compare with its estimated historical range?
• What is the effect of removing low-confidence GBIF records on the inferred distribution pattern?
• Does a dispersal-vicariance model fit the genus better than a pure dispersal model?
• To what extent do paleoclimate layers improve the explanation of present-day disjunct ranges?
• Which geographic regions act as the most likely ancestral area for the genus?
• What is the effect of using different species trees on the biogeographic reconstruction?

Basic Materials

• Laptop with enough memory to handle large occurrence files.
• Internet access for GBIF, paleoclimate data, and literature search.
• Spreadsheet software for cleaning records.
• R installed with biogeography and plotting packages.
• RASP, the free biogeography program, for ancestral area reconstruction.
• Open-source map data or a GIS viewer for visualizing range maps.

Advanced Materials

• High-performance laptop or university workstation.
• Access to a university library for phylogeny and systematics papers.
• R with phylogenetic, spatial, and statistical packages.
• GIS software such as QGIS for detailed range mapping.
• Climate raster datasets from NOAA, NASA, or WorldClim.
• Sequence alignment and tree files from published datasets or your own downloaded data.

Software & Tools

• R: Cleans GBIF data, runs analyses, and makes plots and maps.
• RASP: Reconstructs ancestral ranges with dispersal-vicariance models.
• GBIF: Provides open occurrence records for your chosen genus.
• QGIS: Maps occurrence points and compares them with climate layers.
• ImageJ: Helps measure or compare mapped image outputs when needed.

Experiment Steps

1. Choose a genus with a clear disjunct distribution and enough public occurrence records to analyze.
2. Define the geographic regions you will treat as separate areas for the biogeography model.
3. Clean the occurrence data so bad coordinates, duplicates, and outliers do not distort the range map.
4. Gather a time-calibrated tree or a published phylogeny that matches your genus and supports the model.
5. Compare alternative biogeography hypotheses, then test how sensitive the answer is to model choice and data filtering.
6. Link the reconstructed history to paleoclimate evidence so your explanation goes beyond a simple map pattern.

Common Pitfalls

• Using raw GBIF records without cleaning coordinate errors, which creates fake range edges.
• Choosing a genus with too few records or too little published phylogenetic work, which makes the model unstable.
• Defining regions too broadly, which hides the split pattern you are trying to explain.
• Treating one model output as proof, which ignores uncertainty in ancestral range reconstruction.
• Mixing occurrence data from different taxonomic concepts, which can combine records from plants that are not truly the same genus concept.

What Makes This Competitive

A strong project does more than make a pretty map. You would compare multiple models, test how record cleaning changes the result, and explain uncertainty clearly. You can also strengthen the project by comparing several genera, not just one, or by asking whether climate history explains one split better than another. Judges like projects that connect data quality, evolutionary theory, and a clear biological argument.

Project Variations

• Use a different disjunct genus, such as a fern or shrub genus, to compare whether the same biogeographic pattern repeats across plant groups.
• Swap in a different area coding scheme, then test whether finer regional boundaries change the ancestral reconstruction.
• Add a climate niche comparison between the present range and reconstructed past climate to test whether habitat tracking matches the biogeographic model.

Learn More

• GBIF: Search the Global Biodiversity Information Facility for free occurrence records and data download guides.
• NOAA Paleoclimate Data: Search NOAA for past climate reconstructions and climate archive tutorials.
• NASA Earthdata: Find climate and land-surface datasets that help you compare past and present environments.
• Systematic Biology: Search this journal for review articles and case studies on ancestral area reconstruction and biogeography.
• RASP documentation: Look up the free RASP manual and example projects for dispersal-vicariance analysis.