Insect monitoring radar: analytical time-domain algorithm for retrieving trajectory and target parameters

Abstract

Automated radars employing the zenith-pointing linear-polarized conical-scan (ZLC) configuration are proving effective for long-term monitoring observations of the migratory activity of a number of insect pests of agriculture. An insect passing over one of these radars produces a signal with a rather complex time variation, which is recorded and subsequently analyzed to retrieve parameters indicative of the target’s trajectory (speed, direction), its orientation, and its character (size, shape). As an alternative to earlier algorithms for retrieving these parameters, we present here a relatively straightforward analytical method that uses the form of the ZLC scan to isolate signal components in the time-domain and that allows the parameters to be estimated by a sequence of least-squares fits. Validation tests that indicate the algorithm’s precision and accuracy are also described. While the algorithm can be used to determine a starting point for a subsequent overall fit of all parameters, its retrievals are of sufficient quality for this final stage to be unnecessary in the envisaged applications. The algorithm has been adapted successfully for routine processing of data from two insect monitoring radars (IMRs) deployed in inland eastern Australia; implemented in the C++ language and running on a modern microcomputer, it processes the ∼50,000 echoes recorded on the nights of most intense migration in ∼4 min.