Simultaneous radar and video meteors—I: Metric comparisons

Simultaneous radar and video measurements were made using the Canadian Meteor Orbit Radar (CMOR) and several Gen-III image-intensified CCD cameras to observationally validate metric instrument errors determined through Monte Carlo modelling. We find that our radar interferometry accuracy is ∼0.8°∼0.8° using multiple independent techniques validated with video data. Our average radar–video radiant difference is 3.4°, suggesting that radiant errors for CMOR are dominated by errors in time-of-flight lag time determinations. Our video speeds were found to be consistently lower for slower speed meteors, with our modelled video speed errors following the relation log10δv=−1.64+0.02v. Our modelled video radiant errors had error distributions of 0.73°±0.51°. Errors in the fiducial picks for video meteors were found to be anisotropic, with errors along the meteor trail being larger than those perpendicular to the trail, primarily affecting the fit speed. We also find that the majority of our radar detections occur near the end of the observed video height interval. Our average video speeds are higher than our radar speeds, consistent with decelerations and specular reflections occurring preferentially near the end of trails. Range comparisons show our radar determined specular ranges to be systematically +0.32 km farther in range, although this is smaller than the statistical spread. We find 7%±3% of our video events are simultaneously detected by our radar system. This is above the expected 2%–5% range determined through modelling, suggesting our observations are biased towards larger, non-fragmenting meteoroids.

► We compare a dataset of simultaneously observed radar and video meteors to validate numerical error models.
► We find simultaneous observations are biased towards larger, non-fragmenting meteoroids.
► We find our radar detections to occur near the end of the corresponding video height interval.
► Our radar determined speeds should be subject to larger deceleration corrections than our measured video speeds.
► In our study, we conclude our radar interferometry errors are known to 0.8°.