The PANSY Meteor Head-echo Orbit Catalogue: Continuous Antarctic Radar Observations of Southern Meteoroid Streams

Abstract: Meteor head echoes from high-power, large-aperture radars provide pulse-resolved positions and velocities for individual micrometeoroids in a submillimeter-radius range that contribute significantly to the mass influx to Earth. We present the first meteor head-echo orbit catalogue from the Antarctic Syowa Mesosphere--Stratosphere--Troposphere/Incoherent Scatter radar (PANSY). The catalogue contains two million meteors and 50 million pulse-resolved measurements. The observed radiant distribution contains the helion, antihelion, south toroidal and apex sources, and provides unprecedented head-echo coverage of southern ecliptic latitudes. The initial detection-height distribution is found to be double-banded, with both bands increasing in height with meteor speed and exhibiting distinct radiant distributions, consistent with a mixture of meteoroid-size differences and differential ablation. Estimation of dynamic mass indicates that the survey is sensitive to initial radii of approximately 100 micrometers. As part of an initial exploration of the catalogue, we investigate the nighttime alpha Capricornids (CAP) and the extended Daytime Capricornids-Sagittariids (DCS) radiant. Their similar activity durations at opposite nodes are consistent with membership in the CAP--169P/NEAT complex. The radiant distribution suggests a new meteor shower candidate, with peak flux near solar longitude 110 degrees, mean Sun-centered ecliptic radiant longitude 291.0 degrees, latitude -48.2 degrees, and geocentric speed 48.7 km s^{-1}. During catalogue production, event-level raw voltage cuts are retained temporarily to support improvements in data analysis. This first catalogue release, covering 2025 January 26 to 2026 July 26, will be useful for further studies of the southern-hemisphere Earth-crossing meteoroid population.

• The paper presents the first continuous Antarctic HPLA radar catalogue, containing approximately 2.1 million fitted meteor trajectories and 54 million position and Doppler measurements from Syowa.
• The analysis reveals a double-banded meteor detection-height distribution, distinct radiant populations, a narrow apex source, and corrected sporadic-source rates dominated by helion and antihelion meteoroids.
• The paper identifies the provisional July Fornax–Eridanus shower candidate, detected in two years with 138 meteors and a rate about five times above the local background, while highlighting major calibration and debiasing limitations.

The PANSY Meteor Head-echo Orbit Catalogue is the first continuous high-power large-aperture (HPLA) radar survey of meteor head echoes from the Antarctic. Produced with the 47 MHz Syowa Mesosphere–Stratosphere–Troposphere/Incoherent Scatter radar at approximately $69^\circ$ S, the catalogue contains roughly $2.1\times10^6$ fitted meteor trajectories and $54\times10^6$ pulse-resolved position and radial-Doppler measurements over the interval 2025 January 26 to 2026 July 26 (2608.17589). Its defining contribution is geometric: because some southern ecliptic latitudes remain above the horizon for 24 hours per day from Syowa, the catalogue provides head-echo coverage of southern radiants that northern surveys such as MAARSY cannot supply.

Instrument configuration and observing mode

Meteor observations exploit an eight-channel receiver installed in January 2025 that provides interferometric angle-of-arrival capability without disrupting routine atmospheric operations. The mesospheric mode uses a 1.6 ms interpulse period, a 128 $\mu$s 16-bit complementary coded waveform, and five transmit pointings (zenith plus four at $10^\circ$ off-axis), revisiting each direction every 8 ms. The mode delivers about 9 hours of observing per day, corresponding to a 39% duty cycle within its windows. Seven receiver modules of 19 Yagi antennas each are digitized independently for interferometry; digitizing the full 55-module array would have exceeded data-rate, processing, and power budgets for continuous operation. The nominal single-pulse minimum detectable radar cross section at 100 km is $6.7\times10^{-5}$ m$^2$ ($-41.7$ dBsm), although in practice many detections occur in transmit sidelobes where 10–20 dB larger cross sections are required. System noise temperature varies between roughly 5000 and 22500 K with galactic-plane position, producing a sidereal sensitivity modulation of 2–6.5 dB depending on beam — a selection effect the authors document but do not fully correct.

Signal processing and orbit determination

Processing proceeds through range–Doppler matched filtering of each transmitted pulse against the measured code, followed by event formation by linking detections whose range and Doppler evolve smoothly. Events require an 8.5 dB SNR threshold, at least ten thresholded pulses, and candidate positions below 180 km altitude. Interferometric angle of arrival is obtained from the phase coherence across 21 receiver-module baselines; because the sparse receive geometry produces grating aliases, all plausible coherence maxima are carried forward and resolved using time continuity of the full trajectory rather than single-pulse maxima. The selected path is fitted with a single-body shrinking-radius drag–ablation model using fixed material constants ($\rho_m = 3000$ kg m$^{-3}$, $2.1\times10^6$0 kg J$2.1\times10^6$1) and NRLMSISE-00 densities. Geocentric states are propagated with REBOUND/IAS15 to the Sun–Earth Hill sphere to remove zenithal attraction, yielding asymptotic geocentric velocities, radiants, and osculating Keplerian elements with uncertainties estimated from ten perturbed state realizations per meteor.

A representative example demonstrates the achievable precision: position residuals of 180, 130, and 32 m in the east–west, north–south, and vertical components, and a Doppler RMS residual of 530 m s$2.1\times10^6$2. The quality selection retains events with initial-state position uncertainty $2.1\times10^6$3 m and radiant-angle uncertainty $2.1\times10^6$4; this threshold keeps 61.6% of trajectory fits. The $2.1\times10^6$5 Aquariid radiant histogram confirms angular resolution consistent with the estimated uncertainties.

Radiant morphology and the double-banded height distribution

The debiased Sun-centered ecliptic radiant distribution reproduces the helion, antihelion, apex, and southern toroidal sources. Two levels of empirical correction are applied: a zenith-angle weight with exponent $2.1\times10^6$6 fitted by matching mirrored northern and southern apex rates, and a first-order velocity weight proportional to $2.1\times10^6$7 approximating the speed dependence of ablation rate and head-echo plasma production. The authors emphasize these are not calibrated fluxes; a complete correction would require radar-cross-section modeling and physical ablation models, which were intractable because initial radii are poorly constrained.

Two results stand out. First, the initial detection-height distribution is distinctly double-banded, with both bands rising in height as meteor speed increases. The two bands exhibit different radiant distributions: the upper band shows more structured helion and antihelion sources and a stronger $2.1\times10^6$8 Aquariid signature, while the narrow apex feature appears only in the lower band. The authors interpret this qualitatively as a mixture of meteoroid-size differences and differential ablation — volatile alkalis and organic pyrolysis releasing mass at higher altitudes than refractory metals — while conceding that size and compositional effects cannot be disentangled with the present data and that onset height depends on sensitivity, geometry, mass, velocity, and composition jointly.

Second, the narrow apex source — previously reported only at Jicamarca and absent from specular meteor-radar surveys — appears as a compact meridionally extended structure peaking near $2.1\times10^6$9, isolated via spherical-harmonic low-pass decomposition. Its morphology resembles the central apex condensations modeled by Wiegert et al. as old ($54\times10^6$0 yr), small ($54\times10^6$1m) retrograde meteoroids from 55P/Tempel–Tuttle or a dynamically similar Halley-type parent. Its confinement to the lower height band is tentatively attributed to volatile depletion through long-term space weathering, though the paper presents this as a qualitative consistency rather than a demonstrated causal link.

Sporadic-source rates

Aperture-integrated source fractions show how strongly observational bias shapes raw counts. Apex-region detections dominate raw counts at 72.9%, but after exposure, zenith-angle, and $54\times10^6$2 corrections the helion and antihelion apertures together account for 78.0% of the debiased rate — consistent with dynamical models attributing ~85% of terrestrial dust mass influx to Jupiter-family-comet particles. The velocity-weighted antihelion-to-helion ratio is approximately 1.20, agreeing with the size-dependent asymmetry predicted by Wiegert et al.'s dynamical model for small radar meteoroids. The authors caution explicitly that these fractions are averages over this catalogue's particular temporal sampling, that compact shower foregrounds contaminate the broad apertures, and that seasonal visibility from a high-latitude site can bias low-ecliptic-latitude source ratios.

Shower science: CAP–DCS and the July Fornax–Eridanus candidate

For the $54\times10^6$3 Capricornids (CAP), PANSY measures activity spanning $54\times10^6$4–$54\times10^6$5 with mean parameters ($54\times10^6$6 relative to comet 169P/NEAT) consistent with CMOR and Atlas values. The extended Daytime Capricornids-Sagittariids (DCS) radiant spans $54\times10^6$7–$54\times10^6$8 with $54\times10^6$9. The nearly equal durations (~34° and ~33°) at opposite nodes support membership in the CAP–169P/NEAT complex, though the authors note that orbital similarity alone does not demonstrate common origin; the largest element differences lie in $\mu$0 and $\mu$1, as expected for millennia-old ejecta under planetary precession. A localized DCS enhancement near $\mu$2 is reported but treated as possible substructure within one extended radiant, since incomplete exposure prevents robust measurement of the neighboring minimum.

The most striking new result is the July Fornax–Eridanus (JFE) shower candidate: a compact radiant near $\mu$3, $\mu$4, $\mu$5 km s$\mu$6, detected in both 2025 and 2026 with 138 meteors. Against adjacent solar-longitude intervals, the shower rate of $\mu$7 h$\mu$8 exceeds the combined local background of $\mu$9 h$10^\circ$0 by about a factor of five. Comparison against all 1339 parameter sets in the IAU Meteor Data Center Working List finds no convincing identification — the nearest entry, M2024-N1, differs by $10^\circ$1 in radiant separation and $10^\circ$2 — and a Minor Planet Center search identifies no plausible parent body. The status remains provisional pending independent recovery.

Pulse-to-pulse Doppler processing and mass constraints

The production catalogue deliberately omits radius and mass estimates because within-pulse FFT Doppler scatter is too large to separate geometric deceleration from drag-induced deceleration. As a demonstration, the authors implement a three-pulse complex-voltage fit exploiting inter-pulse phase progression, resolving discrete velocity branches separated by $10^\circ$3 m s$10^\circ$4 at the 8 ms same-beam spacing. For one example event, this reduces the Doppler residual standard deviation from 527 m s$10^\circ$5 to 14 m s$10^\circ$6 and converts an unbounded high-mass-tail radius profile into a bounded 95% interval of $10^\circ$7–$10^\circ$8m. Applied to 50,000 randomly chosen events (9,157 with paths exceeding 15 km yielded constraints), the method indicates median bounds of order $10^\circ$9m for slow meteors and $6.7\times10^{-5}$0–$6.7\times10^{-5}$1m for fast ones — establishing that the survey is sensitive to initial radii of approximately 100 $6.7\times10^{-5}$2m, with mass uncertainty still near one order of magnitude. Velocity side bands from the phase ambiguity are not always rejected automatically, so this remains a demonstration rather than a catalogue-scale pipeline.

Limitations and open questions

Several limitations bear directly on interpretation. Raw detection rates are not absolute fluxes: they are modulated by sky-noime-dependent sensitivity, sparse-array sidelobe illumination, range, viewing geometry, and speed-dependent detection probability, and the debiasing weights are explicitly first-order. The sporadic-source fractions assume temporal stationarity that the data themselves contradict. The double-band height interpretation remains qualitative, and whether differential ablation models can reproduce the bands is left as an open question requiring dedicated ablation modeling. Dynamic masses depend on fixed density and ablation-coefficient assumptions that are not estimated per meteor. No systematic minor-shower search has yet been performed; several visually identified compact enhancements beyond JFE await formal analysis. Finally, the level-1 raw voltage cuts (~10 TB) are retained only temporarily, so the reprocessing capability underpinning future Doppler improvements is not permanently preserved.

Conclusion

This catalogue establishes PANSY as a stable, nearly continuously sampled southern-hemisphere HPLA meteor survey covering all solar longitudes, with two million orbits, unprecedented coverage of southern ecliptic radiants, and demonstrated resolution of both established streams (CAP, DCS, $6.7\times10^{-5}$3 Aquariids) and a new shower candidate (JFE). The double-banded initial-height distribution and its association with distinct radiant populations offer a new observable for meteoroid composition studies, contingent on ablation modeling that remains to be done. Operations continue until September 2027, when the radar is scheduled to cease; linear extrapolation predicts approximately 3.7 million fitted head echoes by then, and a joint analysis with the comparable Arctic MAARSY catalogue would provide nearly global radiant-latitude coverage with closely matched instrumentation.