Urban Animal Genetics and City Heat Islands

ISEF Category: Computational Biology and Bioinformatics

Subcategory: Computational Evolutionary Biology  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

Cities are not just concrete. They are tiny heat islands that can change how animals move, breed, and spread. That means rats, pigeons, and cockroaches may not be mixing genes the same way in every neighborhood. You can test that with public data instead of waiting for a lab bench.

What Is It?

Phylogeography asks a simple question, where do populations live, and how are they related? Think of it like a family tree with a map attached. If two groups live in different places and rarely mix, their DNA can start to separate over time.

Your project uses public mitochondrial DNA sequences, which are small pieces of DNA passed down through the maternal line, plus location data from iNaturalist and other public records. You then ask whether urban heat islands, the warmer parts of cities caused by buildings, roads, and low tree cover, line up with genetic differences. In plain terms, you are checking whether city conditions might act like a barrier or a filter for movement and breeding.

This topic sits at the edge of ecology, evolution, and data science. You do not need to collect your own DNA. You need to ask a sharp question, gather clean public data, and compare patterns across neighborhoods or cities.

Why This Is a Good Topic

This makes a strong science fair topic because you can test a real evolutionary idea with data you can access now. The question is measurable, the inputs are public, and the analysis can be as simple or as deep as your skill allows. It also connects to a real problem, urban change can reshape animal movement, pest spread, and disease risk. You can learn how to clean datasets, map coordinates, compare genetic lineages, and test whether observed patterns are stronger than chance.

Research Questions

• How does urban heat island intensity relate to mitochondrial haplotype diversity in city rat populations?
• What is the effect of neighborhood tree cover on genetic distance among pigeon samples collected in the same metro area?
• Does sample location within a city predict mitochondrial lineage clustering more strongly than random spatial expectation?
• To what extent do cockroach sequences from warmer urban cores differ from those in cooler suburban sites?
• Which city features, such as impervious surface, elevation, or green space, best explain phylogeographic clustering in urban animals?
• How does the amount of public sample metadata affect the strength of a phylogeography model?
• What is the effect of combining iNaturalist coordinates with GenBank sequence records on geographic resolution?

Basic Materials

• Laptop or desktop computer with internet access.
• Spreadsheet software, such as Google Sheets or Excel.
• Free GIS or mapping software, such as QGIS.
• Sequence database access, such as GenBank through NCBI.
• Public observation database access, such as iNaturalist.
• Basic statistics calculator or spreadsheet functions.
• Notebook for tracking data-cleaning decisions.

Advanced Materials

• Computer with enough memory to handle sequence alignments.
• R with phylogenetics and spatial analysis packages.
• Python with pandas, geopandas, and Biopython.
• MEGA or another sequence alignment and tree-building tool.
• QGIS for spatial layers and urban heat island mapping.
• Access to climate or land-cover rasters from NOAA, NASA, or USGS.
• Optional high-performance computing access for larger datasets.

Software & Tools

• QGIS: Maps sample locations, urban heat layers, and city land-cover data.
• R: Runs statistical tests, spatial models, and data visualizations.
• Python: Cleans metadata, merges datasets, and automates sequence workflows.
• MEGA: Aligns mitochondrial sequences and builds simple phylogenetic trees.
• Google Sheets: Helps you screen, organize, and quality-check public records.

Experiment Steps

1. Define one animal group, one geographic scale, and one urban heat metric so your question stays narrow.
2. Gather public sequence records and observation coordinates, then decide which metadata fields you will trust and which you will exclude.
3. Clean the dataset so each sample has a usable location, a sequence ID, and a city context you can compare across sites.
4. Build a simple genetic summary, then decide whether you need haplotypes, pairwise distances, or a phylogenetic tree.
5. Add urban environment layers, then plan how you will compare genetic clustering against heat, green space, or built-up cover.
6. Choose a test that matches your data size, then define controls or null models that rule out random spatial clustering.

Common Pitfalls

• Mixing records from different species or subspecies, which can make the genetic pattern look stronger or weaker than it really is.
• Using public coordinates without checking accuracy, which can place a sample in the wrong neighborhood or even the wrong city.
• Comparing sequences of different gene regions, which breaks direct distance comparisons.
• Treating missing metadata as real biology, which can turn a data gap into a false trend.
• Ignoring uneven sampling density, which makes heavily sampled cities look more diverse just because more records exist there.

What Makes This Competitive

A strong version of this project does more than map dots on a city map. It tests a clear hypothesis with a null model, so you can tell real structure from sampling noise. It also checks whether the result holds across more than one species or more than one city, which makes the pattern more convincing. The best entries use careful data cleaning, a thoughtful spatial comparison, and a strong statistical test instead of a simple visual guess.

Project Variations

• Focus on rats only and compare downtown, park-edge, and suburban samples within one metro area.
• Swap mitochondrial DNA for another public marker, then test whether nuclear data tell the same geographic story.
• Compare two cities with different heat island intensity to see whether stronger urban warming matches stronger genetic clustering.

Learn More

• NCBI GenBank: Search public DNA sequences and metadata for your target species.
• NCBI Bookshelf: Find free chapters on population genetics and phylogeography.
• iNaturalist: Search observations and coordinates for urban wildlife and pest species.
• QGIS Documentation: Learn how to map sample points and overlay land-cover layers.
• NOAA Climate Data Online: Find temperature and climate records for urban heat comparisons.
• USGS EarthExplorer: Download land-cover and satellite data for city surface analysis.