Mass Production of 2023 KMTNet Microlensing Planets I: Low Mass Ratio

This paper presents the first systematic search for low-mass-ratio planets ($q<2\times 10^{-4}$) in the re-reduced 2023 KMTNet microlensing data, identifying three strong planet candidates including KMT-2023-BLG-0164, whose host system was characterized via spectroscopy despite being projected on a bright foreground star.

The Gist

Imagine the night sky as a vast, crowded dance floor. Most of the time, the stars are just dancing in the background, but occasionally, a star passes directly in front of another, acting like a cosmic magnifying glass. This phenomenon is called gravitational microlensing.

This paper is a report from a team of astronomers (the KMTNet Collaboration) who are acting like cosmic detectives. They are using a network of three powerful telescopes in Chile, South Africa, and Australia to scan the "bulge" of our galaxy (the crowded center) for these magnifying glass events. Their specific goal? To find exoplanets (planets outside our solar system) that are very small compared to their host stars.

Here is a breakdown of their findings using simple analogies:

1. The Detective Work: "The Anomaly Finder"

The team uses a computer program called AnomalyFinder. Think of this program as a super-advanced spell-checker for starlight.

• The Normal Spell: When a star passes in front of another, the light curve (a graph of brightness over time) looks like a smooth, symmetrical hill.
• The Glitch: If a planet is orbiting the front star, it creates a tiny "glitch" or "bump" in that smooth hill.
• The Mission: The team looked at data from 2023 and found 11 "glitches" that looked like they might be caused by very small planets (tiny mass ratios).

2. The Filter: Sorting the Real from the Fake

Out of the 11 suspicious glitches, the team had to do some detective work to see which ones were real planets and which were just "glitches" in the camera or data.

• 6 were fakes: They turned out to be image artifacts (like a smudge on a camera lens) or other non-planetary issues.
• 2 were ambiguous: These looked like planets, but they could also be explained by two stars passing in front of each other (a "binary source"). It's like hearing a noise and not knowing if it's a dog barking or a car backfiring. The team analyzed them but decided they are too confusing to include in their official "planet count" just yet.
• 3 were confirmed: These are the stars of the show. They found three solid, low-mass planets.

3. The Three New Planets

The paper highlights three specific discoveries, each with a unique story:

• KMT-2023-BLG-0164 (The Bright Neighbor):

  • The Story: This planet was found near a very bright star. The light from the planet's host star was mixed up with the light from this bright neighbor, making it hard to tell who was who.
  • The Clue: The team got a "spectrum" (a chemical fingerprint) of the bright star. It turned out to be a star very similar to our Sun.
  • The Twist: The planet's host star is likely a dimmer companion to this bright Sun-like star. It's like finding a tiny mouse hiding behind a giant elephant; you see the elephant, but the mouse is the one with the baby.
  • Future: To see the host star clearly, they will need to wait until the stars drift apart and use a future giant telescope (the EELT) to take a high-resolution photo.

• KMT-2023-BLG-1286 (The Sharp Peak):

  • The Story: This planet caused a very sharp, quick spike in brightness.
  • The Result: It's a Neptune-sized planet orbiting a small, red dwarf star (a "middle M dwarf") located in our galactic neighborhood. It's a classic, clean detection.

• KMT-2023-BLG-1746 (The Long Mystery):

  • The Story: This event lasted a long time, but the data was a bit sparse (like a movie with missing frames).
  • The Result: There are two possible explanations for this one. Either the planet is very close to its star, or it's a bit further away. The team can't decide which yet without better data. It's like hearing a sound and not knowing if the source is a whisper nearby or a shout far away. Future telescopes will help solve this.

4. Why This Matters: The "Mass Function"

Why are they so obsessed with counting these tiny planets?

• The Goal: They want to build a "census" of planets. They want to know: Are small, Earth-like planets common, or are giant Jupiter-like planets more common?
• The Challenge: Usually, it's hard to know the exact mass of a planet found this way because you don't know the mass of the star it orbits.
• The Future Solution: The paper discusses a "time machine" strategy. They will wait for the host star and the background star to drift apart (which takes years). Then, they will use the European Extremely Large Telescope (EELT)—a telescope so powerful it will be like switching from a standard camera to a 4K super-camera—to take a picture of the host star. This will finally reveal the true mass of the planet.

Summary

In short, this paper is the first batch of results from a "mass production" line of planet hunting in 2023. They successfully filtered out the noise, found three new, very small planets, and identified two more that need more investigation. They are laying the groundwork for a future where we can not only find these tiny worlds but also weigh them precisely, helping us understand how common Earth-like worlds really are in the universe.

Technical Summary

Here is a detailed technical summary of the paper "Mass Production of 2023 KMTNet Microlensing Planets I: Low Mass Ratio" by Ryu et al.

1. Problem and Motivation

The Korea Microlensing Telescope Network (KMTNet) aims to construct a statistical sample of exoplanets discovered via gravitational microlensing to measure the mass-ratio function ($q$, the ratio of planet mass to host mass) and eventually the planet mass function.

• The Challenge: Previous studies (2016–2019) suggested two distinct populations of planets (gas giants vs. solid-dominated planets), but these were based on a small sample (63 planets). To confirm these findings, a larger, systematic sample is required.
• The Specific Focus: This paper targets "low mass-ratio" planets ($q < 2 \times 10^{-4}$), which are difficult to detect due to subtle anomalies in light curves.
• Methodological Shift: Unlike previous seasons where planetary anomalies were found in standard pipeline reductions and then re-analyzed ("Tender Loving Care" or TLC), the 2023 season is the first where all event light curves were re-reduced using an optimized pipeline (Yang et al. 2024) before the search for anomalies. This aims to increase sensitivity to subtle planetary signals.
• Long-term Goal: The ultimate goal is to measure host masses to convert mass ratios into absolute planet masses. While this currently requires waiting decades for high-resolution imaging (to resolve the host from the source), the upcoming 39m European Extremely Large Telescope (EELT) will drastically shorten this wait time. This paper initiates the "mass production" of candidates to prepare for that era.

2. Methodology

The authors analyzed 2023 KMTNet data using the AnomalyFinder algorithm, which searches for deviations from standard single-lens, single-source (1L1S) Paczyński light curves.

• Data Reduction: All $\sim12,000$ light curves from the 2023 season were re-reduced using the TLC pySIS pipeline (Yang et al. 2024), which optimizes difference image analysis (DIA) for end-of-season analysis.
• Candidate Selection: The team focused on 11 candidates with preliminary mass-ratio estimates of $\log q < -3.7$ (i.e., $q < 2 \times 10^{-4}$).
• Modeling:
  • Static Models: Grid searches were performed on the separation ($s$) and mass ratio ($q$) planes.
  • Parallax & Orbital Motion: For long-duration events, microlens parallax ($\pi_E$) and orbital motion ($\gamma$) were fitted to break degeneracies.
  • Alternative Models: Candidates were rigorously tested against non-planetary interpretations, specifically Binary Source (1L2S) models, which can mimic planetary bumps.
• Source Characterization: Source properties (angular radius $\theta_*$) were derived using Color-Magnitude Diagrams (CMDs) and the red clump method.
• Bayesian Analysis: For secure candidates, Galactic model priors were applied to estimate host mass ($M_{host}$), distance ($D_L$), and planet mass ($M_{planet}$).
• Spectroscopy: For the most promising candidate, high-resolution spectroscopy was obtained to characterize the blended light (foreground star).

3. Key Contributions

• First Systematic Low-$q$ Search: This is the first systematic search for low-mass-ratio planets in a season where the entire dataset was pre-reduced with an optimized pipeline, potentially increasing detection sensitivity.
• Validation of Pipeline: The study validates the efficacy of the Yang et al. (2024) pipeline in recovering planetary signals that might be missed in standard real-time reductions.
• Spectroscopic Breakthrough: The paper presents a rare case where spectroscopy of the blended light (foreground star) was used to constrain the lens distance and mass, bypassing the need to wait for future high-resolution imaging for a preliminary mass estimate.
• Comprehensive Candidate Analysis: The authors provide a "mass production" analysis of all low-$q$ candidates, distinguishing between secure planetary events and ambiguous cases, rather than cherry-picking only the best results.

4. Key Results

The study examined 11 candidates. Six were rejected as image artifacts. The remaining five were analyzed in detail:

A. Secure Planetary Systems (3 Events)

1. KMT-2023-BLG-0164 ($q \sim 1.3 \times 10^{-4}$):

   • Features: The source is projected on a very bright ($I=16.0$) foreground star.
   • Spectroscopy: Spectra obtained with the Magellan/IMACS telescope revealed the blend is a Sun-like star ($M \sim 1.0 M_\odot$).
   • Distance: The spectroscopic data implies a distance of $D_L \sim 1.5$ kpc.
   • Interpretation: The bright star is likely a companion to the actual host (an M-dwarf), not the host itself. The host is too faint to be the spectroscopic target.
   • Future: High-resolution imaging (EELT) is needed to resolve the host from the bright companion to measure the host mass directly.

2. KMT-2023-BLG-1286 ($q \sim 1.9 \times 10^{-4}$):

   • Features: A sharp, non-caustic crossing bump.
   • Parameters: The host is likely an M-dwarf in or near the Galactic bulge. The planet is Neptune-class.
   • Degeneracy: Resolved well; the resonant solution was ruled out ($\Delta \chi^2 > 100$).

3. KMT-2023-BLG-1746 ($q \sim 8 \times 10^{-5}$):

   • Features: A very long event ($t_E \sim 113$ days) with a shallow dip.
   • Degeneracies: Suffered from severe degeneracies due to low cadence (only 3 points on the dip). There are 16 solutions divided into two groups: "inner/outer" (low $\mu_{rel}$, bulge lens) and "close/wide" (higher $\mu_{rel}$, disk lens).
   • Resolution: Future high-resolution imaging to measure proper motion ($\mu_{rel}$) is required to distinguish between these groups, which differ in mass ratio by a factor of $\sim 1.5$.

B. Ambiguous Candidates (2 Events)

1. KMT-2023-BLG-0614:

   • The light curve is well-fit by a Binary Source (1L2S) model.
   • The 1L2S solution is physically plausible (G-dwarf + M-dwarf companion).
   • Future proper motion measurements are unlikely to rule out the 1L2S interpretation; thus, it is excluded from the statistical planetary sample.

2. KMT-2023-BLG-1593:

   • Similar to 0614, a 1L2S model provides a competitive fit.
   • The binary source interpretation (G-dwarf + M-dwarf) is physically plausible.
   • Excluded from the statistical planetary sample due to ambiguity.

5. Significance

• Statistical Sample Expansion: This work adds three confirmed low-mass-ratio planets to the KMTNet statistical sample, crucial for refining the mass-ratio function at the low-$q$ end.
• Preparation for EELT: By identifying these systems and characterizing their degeneracies (e.g., the need for proper motion measurements in KMT-2023-BLG-1746), the paper lays the groundwork for the EELT era. The EELT will be able to resolve the host stars of these events much sooner than current telescopes (Keck), allowing for the first systematic measurement of the planet mass function.
• Methodological Validation: The success of the 2023 re-reduction strategy confirms that optimizing the entire dataset before anomaly searching yields robust results, setting a precedent for future "mass production" papers covering the 2016–2019 and subsequent seasons.
• Spectroscopic Utility: The successful use of spectroscopy to constrain the lens distance in KMT-2023-BLG-0164 demonstrates a powerful alternative to waiting for late-time imaging, provided the blend is bright enough.

In summary, this paper represents a major step forward in the systematic discovery of Earth-mass and super-Earth planets via microlensing, validating new data reduction techniques and preparing the scientific community for the high-precision mass measurements that the EELT will soon enable.