TESS Single-Transit Exoplanet Search Project

ISEF Category: Physics and Astronomy

Subcategory: Astronomy and Cosmology  ·  Difficulty: Advanced  ·  Setup: Home Setup  ·  Time: Full Year

The Hook

Some exoplanets orbit so far from their stars that TESS may see only one transit, or even none at all. That makes them easy to miss and hard to prove. If you can find them, you are working on the same kind of problem used in real exoplanet discovery pipelines. You also get to practice the skills that matter most in astronomy, cleaning noisy data, testing false alarms, and checking whether a signal is real.

What Is It?

This project asks you to search TESS Full-Frame Images for single-transit events around quiet K and M dwarfs. A single-transit event is one dip in brightness that may come from a planet passing in front of its star. Think of it like spotting one footprint in fresh snow. You do not get the whole path, so you have to infer what kind of object made it.

The wavelet part helps you separate real transit-shaped dips from noisy variations in the light curve. A matched filter compares your data to a model shape, which helps you score how similar each dip looks to a transit. Then Gaia DR3 RUWE gives you a clue about binarity, because stars that look oddly distorted or unresolved can hide companions that fake planet-like signals. You finish by estimating completeness, which means asking how many real signals your method would catch if they were hidden in the data.

Why This Is a Good Topic

This is a strong science fair topic because the data are public, the question is real, and the answer is measurable. You can test how well your pipeline finds one-transit events, how many false positives it creates, and whether Gaia flags suspicious stars. The project connects to exoplanet discovery, signal processing, and data validation, so you can learn more than one skill at once. You do not need a telescope, but you do need patience, code, and careful thinking.

Research Questions

• How does a wavelet-based preprocessing step change single-transit detection sensitivity in TESS Full-Frame Image light curves?
• What is the effect of stellar type, K dwarf versus M dwarf, on the rate of candidate single-transit detections?
• Does filtering out high-RUWE Gaia DR3 sources reduce the false positive rate of single-transit candidates?
• To what extent does the matched-filter threshold change the balance between completeness and purity?
• Which light-curve quality metrics best predict whether a candidate survives visual vetting?
• How does injection depth affect recovery rates for outer-orbit transit signals in quiet stars?

Basic Materials

• Laptop or desktop computer with at least 16 GB RAM.
• Stable internet connection for downloading TESS and Gaia data.
• External hard drive or cloud storage for data backups.
• Spreadsheet software for tracking candidates and metadata.
• Python installed with scientific libraries for data analysis.
• Text editor or notebook environment for code and notes.

Advanced Materials

• Access to a machine with higher memory for batch processing large TESS image sets.
• Python environment with astroquery, lightkurve, numpy, scipy, astropy, and scikit-image.
• ImageJ or similar image analysis software for quick visual checks of cutouts.
• Access to GPU-enabled computing if you test machine-learning candidate ranking.
• Database software or structured CSV workflow for candidate tracking.
• Version control system such as Git for reproducible analysis.

Software & Tools

• Python: Runs your data cleaning, transit search, and completeness calculations.
• Lightkurve: Helps you work with TESS light curves and image-based photometry.
• Astropy: Provides astronomy utilities for coordinates, time handling, and catalog work.
• Astroquery: Downloads catalog data from NASA, ESA, and other online archives.
• ImageJ: Lets you inspect image cutouts and compare suspect transit frames by eye.

Experiment Steps

1. Define your search sample by choosing quiet K and M dwarfs with usable TESS coverage.
2. Build a candidate-finding pipeline that flags single dips and scores them against a transit-like template.
3. Set up a vetting plan that checks each signal against stellar variability, data gaps, and image artifacts.
4. Cross-match each candidate with Gaia DR3 RUWE and other catalog flags to spot likely binaries or blended systems.
5. Design an injection-recovery test so you can measure how often your pipeline finds planted signals.
6. Compare candidate yield, false positives, and completeness across different star types, thresholds, or preprocessing choices.

Common Pitfalls

• Confusing instrumental jumps or scattered-light artifacts for true single-transit dips.
• Mixing stars with obvious variability into the sample, which hides weak transit signals.
• Trusting one pipeline threshold too much, which can flood your list with false positives or erase real events.
• Ignoring Gaia RUWE outliers that often point to blended stars or unresolved companions.
• Skipping injection-recovery tests, which leaves you unable to say how complete your search really is.

What Makes This Competitive

A stronger project does more than find a few candidate dips. It compares multiple detection settings, quantifies completeness, and shows how your method behaves across different stellar populations. You can also tighten the analysis by using catalog cross-checks, visual vetting rules, and a clear false-positive audit. Judges respond well when you can explain why your pipeline is better, not just that it found something.

Project Variations

• Try the same search on only M dwarfs, then compare candidate rates against K dwarfs.
• Swap the matched-filter scoring rule for a simpler thresholding method and test which one recovers more injected signals.
• Add a second validation layer, such as centroid shifts or nearby star contamination checks, to see whether candidate purity improves.

Learn More

• NASA Exoplanet Archive: Search confirmed planets, candidate catalogs, and mission data summaries on NASA’s archive site.
• MAST TESS Data Products: Find TESS Full-Frame Images, light curves, and mission documentation through the Mikulski Archive for Space Telescopes.
• Gaia Archive: Query Gaia DR3 source data, including RUWE and astrometric quality fields, in ESA’s archive.
• The Astronomical Journal: Search for review articles and methods papers on TESS exoplanet detection and transit vetting through journal access or abstract databases.
• PubMed: Search for review papers on signal processing, matched filtering, and detection methods when you want background on the analysis side.