🔭 astrophysics

COUNTESS I: A Uniformly Vetted Catalog of Known and New Transiting Exoplanets in the TESS Northern Continuous Viewing Zone

This paper introduces COUNTESS, a specialized transit-search pipeline for the TESS Northern Continuous Viewing Zone that combines multi-sector light curves to detect and validate longer-period exoplanets, successfully recovering 115 known TESS Objects of Interest and identifying 10 new candidates, including two statistically validated sub-Neptunes.

Original authors: Andrew Hotnisky, Rachel B. Fernandes, Kevin K. Hardegree-Ullman, Steven Giacalone, Kiersten M. Boley, Kristo Ment, Michelle Kunimoto, Galen J. Bergsten, Sakhee Bhure, Jessie L. Christiansen, Brandon R

Published 2026-06-15

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Andrew Hotnisky, Rachel B. Fernandes, Kevin K. Hardegree-Ullman, Steven Giacalone, Kiersten M. Boley, Kristo Ment, Michelle Kunimoto, Galen J. Bergsten, Sakhee Bhure, Jessie L. Christiansen, Brandon Radzom, Suvrath Mahadevan

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, dark ocean, and stars are like lighthouses scattered across it. For decades, astronomers have been trying to spot tiny "moths" (exoplanets) flying in front of these lighthouses. When a moth flies in front of a light, it dims the beam just a tiny bit. By watching for these dips in brightness, we can find new worlds.

This paper introduces a new, super-smart tool called COUNTESS (which stands for Combining Observations to Unveil New Transiting Exoplanet Systems and Statistics) designed to help us find these moths better than before.

Here is the story of what they did, explained simply:

The Problem: The "Short Attention Span" Camera

For a long time, the main space telescope looking for these planets (called TESS) had a bit of a limitation. It was like a security camera that only watched a specific neighborhood for about 27 days before moving to the next one.

The Issue: If a planet takes a long time to orbit its star (like Earth takes 365 days), the camera moves away before it can see the planet fly in front of the star twice.
The Exception: There are two special "zones" near the top and bottom of the sky (called the Continuous Viewing Zones, or CVZs) where the camera can keep watching the same stars for much longer—almost a year or more. This is like a security camera that never blinks, giving us a much better chance to spot slow-moving planets.

The Challenge: A Messy Playlist

The authors wanted to use these long-watching zones to find planets, but there was a catch. The camera changed its settings halfway through the mission.

Part 1: It took a picture every 30 minutes.
Part 2: It switched to taking a picture every 10 minutes.
The Result: Trying to stitch these two different speeds together is like trying to mix a slow jazz song with a fast techno track. The rhythm gets messy, and it's hard to hear the "beat" of a planet's orbit. Also, the stars themselves twinkle and change brightness, which can hide the tiny signal of a planet.

The Solution: COUNTESS

The team built COUNTESS, a new computer program designed to be the ultimate "music mixer" for this data.

Cleaning the Signal: First, it smooths out the "static" and the stars' natural twinkle (like noise-canceling headphones) so the tiny dips caused by planets stand out.
Matching the Rhythm: It carefully aligns the fast 10-minute photos with the slow 30-minute ones so they fit together perfectly.
The Search: It uses a super-fast algorithm (called GERBLS) to fold the data over and over, looking for a repeating pattern that says, "Hey, something is passing in front of that star!"
The Vetting: Once it finds a potential planet, it puts it through a rigorous "police lineup" (using tools called LEOVetter and triceratops) to make sure it's not a fake signal caused by a glitch in the camera or a background star.

What They Found

The team tested COUNTESS on the "Northern" viewing zone of the sky.

The Test Run: They looked at stars that already had known planets to see if COUNTESS could find them again. It successfully found 115 out of 159 known planets. This proved the tool works.
The New Discoveries: Then, they let COUNTESS look at 26,000 stars it had never seen before.
- It found 10 new planet candidates. These are worlds that look very likely to be real planets but need a little more confirmation.
- Two of these candidates were so strong that the team officially validated them as real planets: TIC 219893931 b and TIC 237254473 b.
- These new worlds are "Sub-Neptunes" (bigger than Earth but smaller than Neptune) and take between 6 and 10 days to orbit their stars. This is exciting because most planets found by TESS are very close to their stars and orbit quickly; these are "longer-period" planets, which are harder to find.

Why It Matters

Before this, we mostly knew about planets around distant stars (like the famous Kepler mission did) or very close-in planets around nearby stars.

The Bridge: COUNTESS helps bridge the gap. It allows us to study planets around nearby stars that take longer to orbit.
The Future: Because these stars are close to us, we can point powerful telescopes at them later to study their atmospheres. This paper is just the beginning; the authors plan to use COUNTESS to scan the entire sky and find even more of these long-period worlds, helping us understand how common different types of planets are in our cosmic neighborhood.

In short: They built a new, smarter filter to clean up messy telescope data, allowing them to spot planets that orbit their stars more slowly than we usually see, specifically in the part of the sky where the telescope never stops watching.

Technical Summary: COUNTESS – A Uniformly Vetted Catalog of Known and New Transiting Exoplanets in the TESS Northern Continuous Viewing Zone

Problem Statement
The Transiting Exoplanet Survey Satellite (TESS) has revolutionized the study of nearby exoplanetary systems. However, its nominal observing strategy provides only $\sim$ 27 days of continuous coverage per sector for most of the sky. This short baseline biases standard transit searches toward planets with orbital periods $\lesssim$ 10 days, as at least two transits are required for detection. While the TESS Continuous Viewing Zones (CVZs) near the ecliptic poles offer extended temporal baselines (up to 351 days in the Primary Mission), they introduce significant analytical challenges. Specifically, the photometric cadence changes from 30 minutes in the Primary Mission (PM) to 10 minutes in the first Extended Mission (EM1). This heterogeneity in cadence and noise properties complicates the coherent detection of transit signals across multiple sectors and mission phases. Existing TESS transit-search pipelines lack a homogeneous, end-to-end framework specifically designed to address these CVZ-specific and multi-mission challenges, hindering the robust characterization of planet occurrence rates and demographics for long-period planets around nearby stars.

Methodology
To address these limitations, the authors present COUNTESS (Combining Observations to Unveil New Transiting Exoplanet Systems and Statistics), a unified transit-search pipeline optimized for long-baseline, multi-cadence TESS CVZ light curves. The methodology proceeds through the following stages:

Stellar Sample Construction: The authors constructed a homogeneously characterized catalog of 391,059 FGKM stars in the TESS northern CVZ. Using Gaia DR3 and 2MASS photometry, they derived stellar effective temperatures, luminosities, radii, and masses using a methodology consistent with K. K. Hardegree-Ullman et al. (2025) to ensure compatibility with Kepler and K2 datasets. The sample was filtered for single stars (Gaia RUWE < 1.4) and distance uncertainties (<10%), resulting in a final sample of 391,059 stars for the stellar catalog, with 26,114 stars selected for the transit search based on TESS-SPOC light curve availability.
Data Processing and Detrending: The pipeline combines TESS-SPOC Full Frame Image (FFI) light curves from the PM (30-min cadence) and EM1 (10-min cadence). To mitigate heterogeneous sampling, EM1 data were binned to 30 minutes to match the PM cadence. Stellar variability was removed using a biweight filter implemented in the Wōtan package. A two-stage window-length approach was adopted: a 0.5-day window for periods $\le$ 100 days and an extended 1.75-day window for periods >100 days to prevent over-detrending of long-duration transit signals.
Transit Search and Fitting: Periodic signals were identified using GERBLS (Greatly Expedited Robust Box Least Squares), a fast-folding algorithm that accelerates period detection without requiring GPU acceleration. Threshold Crossing Events (TCEs) with a Signal-to-Noise Ratio (SNR) > 6 were selected. These candidates underwent transit fitting using EXOTIC, which employs Bayesian nested sampling to derive physically meaningful parameters (radius ratio, impact parameter, inclination, limb darkening).
Vetting and Validation: Candidates were vetted using LEOVetter, which applies 12 false alarm (FA) tests and 5 false positive (FP) tests to distinguish instrumental artifacts and astrophysical false positives (e.g., eclipsing binaries) from planetary signals. Pixel-level vetting ensured the signal originated from the target star. Finally, triceratops was used for statistical validation, calculating False Positive Probabilities (FPP) and Nearby False Positive Probabilities (NFPP) to classify candidates.

Key Contributions and Results
As a proof of concept, the authors applied COUNTESS to the TESS northern CVZ using PM and EM1 data:

Recovery of Known Objects: The pipeline was tested against 159 known TESS Objects of Interest (TOIs) with periods between 0.85 and 124.72 days. COUNTESS successfully recovered 115 of these (72%), including 12 secondary signals in multi-planet systems. The 44 missed detections were primarily attributed to gaps in the PM/EM1 baseline or low SNR. The authors note that while recovered periods matched literature values, recovered radii were systematically smaller, likely due to differences in fitting methodologies and stellar parameters compared to previous catalogs.
Blind Search and New Discoveries: A blind search was conducted on 26,114 stars, yielding 38 new planetary signals. After rigorous vetting (pixel-level and statistical), 10 candidates were classified as "likely planets" (FPP < 0.5, NFPP < $10^{-1}$ ).
Statistically Validated Planets: Two candidates met the strict statistical validation thresholds (FPP < 0.015, NFPP < $10^{-3}$ $1 0^{- 3}$ ) and were designated as new planets:
- TIC 219893931 b: A sub-Neptune with $P \approx 6.41$ days and $R_p \approx 2.92 R_\oplus$ .
- TIC 237254473 b: A sub-Neptune with $P \approx 10.12$ days and $R_p \approx 3.30 R_\oplus$ .
- The authors also identified TIC 219893931.02 ( $P \approx 34.62$ days, $R_p \approx 4.19 R_\oplus$ ) as a candidate, noting it represents a detection at a period significantly longer than what is robustly identifiable in a single TESS sector.
Follow-up Metrics: Estimated radial velocity semi-amplitudes for the two validated planets are $\sim$ 3.77 m/s and $\sim$ 3.00 m/s, respectively, suggesting they are viable targets for future mass and atmospheric characterization.

Significance
The paper establishes COUNTESS as a critical tool for extending TESS's scientific reach beyond the standard short-period bias. By successfully combining heterogeneous cadences and long baselines, the pipeline enables the detection of planets with periods $>10$ days around nearby stars. This capability bridges the gap between the distant, well-characterized Kepler population and the nearby planetary systems accessible to detailed follow-up. The authors assert that this work lays the foundation for robust exoplanet demographic studies in the solar neighborhood, allowing for direct comparisons with Kepler and K2 trends (such as the radius valley and hot Neptune desert) in a regime previously inaccessible to TESS. The discovery of two statistically validated sub-Neptunes and several long-period candidates demonstrates the pipeline's efficacy and highlights the potential for COUNTESS to scale to the full TESS CVZ sample, ultimately pushing sensitivity to orbital periods beyond 100 days.