Research

The Canadian Meteor Orbit Radar (CMOR) is the primary instrument for large-scale meteor survey work in our group. Operating simultaneously at three radio frequencies (17.45, 29.85, and 38.15 MHz) from its central site near Tavistock, Ontario, CMOR detects meteors via specular backscatter from the ionization trail left as a meteoroid ablates in the mesosphere and lower thermosphere (roughly 70–120 km altitude).

Each detected meteor echo yields a geocentric radiant, speed, and heliocentric orbit through interferometry across five receiving antennas and three outlying forward-scatter stations. Since 2002, CMOR has accumulated more than 25 million meteor orbits, the largest single-facility dataset of its kind. An additional 150+ million single-station echoes provide high-cadence activity monitoring without orbital solutions.

The multi-frequency design enables direct cross-calibration and measurement of frequency-dependent effects: interference within trails is different at different wavelengths. Comparing events at the three frequencies helps characterize these effects and the properties of the trails.

• Continuous operation since 1999 (single-station) / 2002 (multi-station orbits)
• Transmitter power: 6 kW (12 kW on 29.85 MHz)
• Five outlying forward-scatter receivers for multi-station trajectories on 29 MHz
• All-sky sensitivity with minimal dead time
• Automated, real-time data reduction pipeline

Converting raw radar echo counts into physical meteoroid fluxes requires correction for several wavelength-dependent and speed-dependent biases. Our group has systematically characterised these effects using 8 years of simultaneous EMCCD and CMOR observations, yielding 10,503 dual-detection events that directly constrain the fraction of optical meteors missed by the radar.

Key corrections include:

• Initial radius effect — ionization column wider than the wavelength causes destructive interference; more severe at shorter wavelengths
• Finite velocity effect — The ionization in the trail spreads in the time it takes a slow meteoroid to cross the first Fresnel zone
• PRF effect — the pulse repetition frequency limits detection of very short-lived echoes from fast meteors
• Faraday rotation — Earth's magnetic field rotates the polarization plane during propagation through the ionosphere, attenuating echoes with certain geometries

Our key finding: fast showers (Perseids, Leonids) have been significantly undercorrected in the literature, while medium-speed showers (Geminids, South Delta Aquariids) have been somewhat overcorrected. The corrections depend on meteoroid density/strength, not just speed, which explains the discrepancy between 29 and 38 MHz flux measurements for the same shower.

For short-duration outburst showers such as the Draconids (October 8 each year), multi-longitude radar networks are essential since flux can change significantly within the duration of a single-site observing window.

The high-resolution CAMO tracking mirror system resolves individual meteors at metre-scale, making it possible to directly observe fragmentation events, measure differential deceleration between fragments, and model the morphology of meteor wakes.

A survey of ~400 meteors observed by CAMO between 2010 and 2014 found that ~90% show evidence of fragmentation — a result that applies equally to meteors on cometary and asteroidal orbits, contrary to expectations. Light curve shapes are predominantly symmetric, with a mean F parameter of 0.49, consistent with roughly uniform mass loss along the trail.

Transverse motion of fragments has been measured directly in nine fragmenting events, yielding fragment speeds of order 100 m/s perpendicular to the meteoroid trajectory — far exceeding aerodynamic lift predictions and requiring rotation or cometary spin-up scenarios.

Analysis of more than 7000 two-station meteors observed with the CAMO EMCCD system revealed a population of faint meteors with unusual properties: they begin ablation at altitudes below 86 km (far lower than predictions from standard ablation models for stony meteoroids), have slow geocentric speeds (<25 km/s), and come from asteroidal orbits with Tisserand parameter T_J > 3.

These properties are consistent with millimetre-sized pieces of iron-rich material. Their early-peaked light curves and the absence of a cometary counterpart at these orbital elements suggest an asteroidal origin; possibly fragments from the collisional disruption of iron meteorites or iron inclusions.

This population is conspicuous in the more sensitive EMCCD cameras (limiting magnitude +6.5) but largely absent from the tracking mirror data (limiting magnitude +4), suggesting either a very steep mass-frequency distribution at these sizes, or very compact, high-density objects that are bright only briefly at the start of ablation.