Park Soundscape Biodiversity Mapping
ISEF Category: Animal Sciences
Ready to Turn This Idea Into a Real Project?
This guide was put together with the help of AI research tools to give you a solid starting point. But a competitive science fair project lives in the details: refining your research question, fine-tuning your variables, analyzing your data, and presenting your findings like a seasoned scientist.
For next steps tailored to your interests, skill level, and timeline, work one-on-one with a MehtA+ mentor. Learn more about MehtA+ Science & Engineering Research Mentorship →
Subcategory: Ecology and Agriculture · Difficulty: Intermediate · Setup: Home Setup · Time: 1 to 2 Months
The Hook
Two parks can look almost the same on a map, but sound totally different. A tiny Raspberry Pi can record those differences like an audio fingerprint. When you match that sound with satellite data, you can test which habitats support more bird activity.
What Is It?
This project turns park audio into a biodiversity score. BirdNET is a tool that listens to bird calls and guesses which species are present. You are not counting every bird by sight. You are measuring the soundscape, which means the mix of natural sounds in one place.
Think of it like comparing two playlists. One park might have more species, more calls, and more variety. Another might have fewer bird sounds because of roads, pavement, or less tree cover. Satellite imagery helps you measure the habitat around each recording spot, so you can compare what you hear with what the land looks like from above.
The big idea is simple. If birds prefer certain habitats, the audio score should change with tree cover, water, grass, or nearby buildings. That makes the project part ecology, part data science, and part remote sensing.
Why This Is a Good Topic
This is a strong science fair topic because you can measure both the soundscape and the habitat around it. You do not need a university lab to collect useful data, but you still get to ask a real ecology question. The project connects bird activity, urban green space, and land-use change, which gives it clear real-world value. You can also learn how to handle field data, map sites, and compare variables with statistics.
Research Questions
- How does tree cover percentage relate to BirdNET species counts in different park locations? ?
- What is the effect of distance from roads on the audio biodiversity index? ?
- Does the audio biodiversity index differ between park edges and park interior sites? ?
- To what extent does nearby water predict higher BirdNET call diversity? ?
- Which land-cover variable, tree canopy, grass cover, pavement, or water, best predicts the soundscape score? ?
- How does time of day affect the audio biodiversity index at the same site? ?
Basic Materials
- Raspberry Pi with power supply and microSD card.
- USB microphone or USB audio recorder.
- Weather-resistant case with an acoustic opening.
- Portable battery pack.
- Smartphone with GPS and camera.
- Laptop with internet access.
- BirdNET Analyzer or BirdNET app.
- Google Earth Pro or another free satellite imagery source.
- Spreadsheet software for data logging.
- Measuring tape or park map for site selection.
Advanced Materials
- Higher-quality field recorder with a calibrated microphone.
- Directional microphone or small mic array for comparison tests.
- External GPS logger for exact site coordinates.
- GIS software with land-cover layers, such as QGIS.
- Statistical software for mixed-effects models.
- Access to high-resolution satellite imagery or local land-cover datasets.
- Weather station data for wind, rain, and temperature matching.
- Acoustic annotation software for spot-checking BirdNET calls.
Software & Tools
- BirdNET Analyzer: Identifies bird calls from audio clips and gives you a species-level output for each site.
- QGIS: Maps sampling points and calculates land-cover features around each recording location.
- Google Earth Pro: Helps you inspect tree cover, paths, water, and pavement near your sites.
- R: Fits statistical models that compare acoustic diversity with land-cover variables.
- Python: Automates file naming, batch processing, and cleanup of recording folders.
Experiment Steps
- Define the park sites, the habitat features you want to compare, and the one biodiversity metric you will use.
- Match each recording site with a land-cover snapshot from satellite imagery so your audio and habitat data line up.
- Build a naming system that keeps every clip linked to place, date, and site type.
- Set controls for weather, time of day, and background noise so habitat is the main difference.
- Choose the analysis method that turns BirdNET output into a site score and tests its relationship with land cover.
- Predefine how you will handle outliers, missed detections, and repeated visits to the same location.
Common Pitfalls
- Recording near busy trails, which turns human noise into a fake drop in biodiversity.
- Mixing sites from different seasons without accounting for bird migration, which makes habitat effects messy.
- Using raw BirdNET detections without one consistent site score, which makes comparisons unfair.
- Measuring land cover over an area much larger than the mic actually sampled, which blurs the habitat signal.
- Changing microphone position or settings between sessions, which makes hardware drift look like ecology.
What Makes This Competitive
A class-level version of this project can stop at basic species counts, but a stronger entry goes deeper. You can compare park edge and interior sites, test multiple habitat variables, and use one clear statistical model instead of simple averages. You can also validate a sample of BirdNET IDs by hand and report confidence, not just raw counts. That shows you understand both the ecology and the limits of the data.
Project Variations
- Compare urban parks, school campuses, and roadside greenbelts to see how the audio biodiversity index changes with human disturbance.
- Swap BirdNET species counts for call activity rate and test whether a simpler sound metric tracks land cover just as well.
- Use two satellite variables, such as canopy cover and impervious surface, to see which one predicts the acoustic score better.
Learn More
- BirdNET: Free bird sound identification tools and documentation on the BirdNET project website.
- QGIS Documentation: Free GIS guides for mapping sites and measuring land-cover variables.
- NASA Earthdata: Free satellite imagery, tutorials, and remote-sensing resources from NASA.
- USGS Land Cover Data: Free land-cover datasets and background information from the USGS website.
- NOAA Climate Data Online: Free weather records you can use to match recording conditions.
- PubMed: Search for review articles on soundscape ecology, bioacoustics, and acoustic biodiversity.