Robust Clock and Ionospheric Phase Delay Separation in Radio Interferometric Calibration

In radio interferometric observations, ionospheric effects become increasingly significant at low frequencies, and can severely degrade the results of subsequent imaging. Dispersive delays due to the ionosphere are often entangled with instrumental phase delays stemming from clock synchronisation errors. For calibrator observations, separating the two effects is essential before transferring instrumental phase corrections to the target field/s. However, as the likelihood of the problem may contain multiple local minima, traditional optimisation-based calibration often fails to converge to the correct solution. To address this, we propose a regression-based initialisation method that is physically motivated and exploits the distinct spectral signatures of the two effects. We test the efficacy of the method by applying it to severely corrupted Low Band Antenna observations from the Low-Frequency Array. We show that our results are in excellent agreement with the current state-of-the-art approach. Importantly, our method is significantly simpler to apply to raw data and utilises a model with far fewer degrees of freedom, thereby reducing the risk of overfitting. We have incorporated the method into the QuartiCal calibration package and provide a stimela2 recipe implementing the full calibration strategy.