BirdNET Dawn Chorus Urban-Rural Study

ISEF Category: Earth and Environmental Sciences

Subcategory: Environmental Effects on Ecosystems  ·  Difficulty: Intermediate  ·  Setup: Home Setup  ·  Time: 1 to 2 Months

The Hook

Birds do not sing the same way everywhere. A loud road can make a dawn chorus look quieter, even when the birds are still there. You can test how much city noise changes what BirdNET detects along an urban-to-rural line.

What Is It?

This project studies dawn chorus, the burst of bird vocal activity that happens around sunrise. Think of it like a morning announcement system for birds. If you record at different sites, then compare the number, timing, and mix of bird calls, you can see how bird communication changes with habitat.

The cool part is that you do not need a fancy field station. A cheap USB microphone, a phone or laptop, and BirdNET can turn audio into data. BirdNET is software that tries to identify bird species from sound. You can pair that with noise measurements, so you are not just asking where birds sing, but how traffic noise may change what you can detect.

The real science question sits at the overlap of ecology and sound. Urban areas often have more low-frequency traffic noise, and that noise can mask bird songs, especially if the songs sit in a similar frequency range. Your project can compare places with different noise levels and ask whether the dawn chorus shifts in timing, richness, or loudness as human activity changes.

Why This Is a Good Topic

This is a strong science fair topic because you can collect your own field data, define clear variables, and compare real environments. You can measure bird activity, noise spectra, and site differences without needing a university lab. The project connects to habitat quality, urban ecology, and how human noise affects wildlife. You also learn field sampling, audio analysis, and basic statistics, all skills judges like to see.

Research Questions

• How does traffic noise level affect the number of bird species BirdNET detects at dawn?
• What is the effect of urbanization level on the start time of peak dawn-chorus activity?
• Does the frequency profile of traffic noise predict which bird calls are masked most often?
• To what extent do BirdNET detections differ between quiet green spaces and noisy roadside sites?
• Which habitat features best explain dawn-chorus richness along an urban-to-rural gradient?
• How does the time of sunrise relate to the timing of first bird detections across sites?

Basic Materials

• Cheap USB microphone with known sensitivity
• Laptop or smartphone for recording and analysis
• Tripod or stable mount for the microphone
• Headphones for checking recordings
• GPS app or map to log site locations
• Notebook or spreadsheet for field notes
• Sound level meter app or decibel meter for relative noise comparison
• Portable power bank if you record in the field for long periods.

Advanced Materials

• Calibrated omnidirectional field microphone
• Audio interface or recorder with flat frequency response
• External battery pack for long field sessions
• Wind screen for the microphone
• Handheld sound level meter with octave-band or spectra logging
• Laptop with BirdNET Analyzer installed
• Software for spectral analysis such as Audacity or Raven Lite
• GPS unit for precise site coordinates
• Weather meter for temperature, wind, and humidity.

Software & Tools

• BirdNET Analyzer: Identifies bird vocalizations from recordings and helps you count detections by species.
• Audacity: Lets you inspect recording quality, trim files, and compare background noise across sites.
• R or RStudio: Helps you graph detection counts, compare sites, and test relationships with noise or habitat.
• ImageJ: Can measure spectrogram features if you export images for simple visual comparisons.
• QGIS: Maps your sampling sites and helps you relate bird activity to urban and rural location patterns.

Experiment Steps

1. Define your gradient, then choose sites that differ clearly in urbanization, traffic exposure, and green space.
2. Decide your main response variable, such as species richness, first detection time, or total BirdNET detections.
3. Plan how you will measure noise, including a way to compare low-frequency traffic spectra across sites.
4. Build a sampling schedule that matches sunrise and keeps recording conditions as similar as possible.
5. Set up controls that separate habitat effects from noise effects, such as repeated sampling at multiple sites with similar vegetation.
6. Choose your analysis plan before you collect data, including how you will compare sites and handle outliers or false positives.

Common Pitfalls

• Recording too close to roads at some sites and deep inside quiet parks at others, which mixes habitat choice with noise exposure.
• Letting microphone position change between sessions, which makes BirdNET detections hard to compare.
• Ignoring wind, rain, or insects, which can flood recordings with non-bird sound.
• Trusting every BirdNET ID without checking uncertain calls, which can inflate species counts.
• Comparing raw bird counts without accounting for traffic frequency bands, which misses the masking effect you set out to test.

What Makes This Competitive

A strong version of this project goes past simple bird counts. You can separate total noise from noise frequency, which matters because masking depends on spectrum, not just volume. You can also compare multiple habitat metrics, use repeated sampling, and test whether vocal activity changes differently for different species groups. If you pair clean field methods with careful statistics, your project starts to look like real urban ecology research.

Project Variations

• Compare dawn-chorus patterns in schoolyards, parks, and roadside strips instead of a full rural-to-urban transect.
• Focus on one bird group, such as songbirds or cavity nesters, and test whether they respond differently to traffic noise.
• Use a smartphone recorder plus a calibrated app-based noise measure to see how well low-cost tools match a USB microphone setup.

Learn More

• BirdNET project pages: Search for the official BirdNET site and its documentation on species detection from audio.
• Cornell Lab of Ornithology Macaulay Library: Search the Macaulay Library for bird calls, spectrogram examples, and recording tips.
• USGS Bird Banding Laboratory: Use it for background on bird ecology and seasonal movement patterns.
• NOAA Sound and Noise resources: Search NOAA for background on noise, acoustic environments, and environmental sound measurement.
• NIH PubMed: Search for review articles on urban noise, bird communication, and acoustic masking.
• MIT OpenCourseWare ecology materials: Look for free course notes on population ecology, field sampling, and experimental design.