Ionospheric Calibration of Low Frequency Radio Interferometric Observations using the Peeling Scheme II. Method Extensions & Application to a Larger Array

In a previous article, we presented a description and first results of SPAM (Source Peeling and Atmospheric Modeling), a method for calibration of direction-dependent ionospheric phase errors in low-frequency radio interferometric observations. By assuming a time-constant instrumental phase offset per antenna, and by representing the ionosphere with a single 2-dimensional phase screen model at fixed height, SPAM was able to improve the calibration accuracy of 74 MHz observations from the VLA in relatively compact configurations (. 20 km baselines) as compared to other existing calibration methods. In this paper, we present two extensions of the SPAM method: a multi-layer ionosphere model to represent vertical ionospheric structure, and an estimator for slow instrumental phase drifts in the presence of ionospheric phase fluctuations. We describe the data reduction steps while applying SPAM on archival 74 MHz observational data from the Very Large Array in its most extended A-configuration (up to 35 km baselines) during quiet ionospheric conditions. Detection and removal of phase drifts significantly reduces the RMS residual phase when fitting ionospheric phase models. The two target field data sets are calibrated and imaged using both the single-layer and multi-layer ionosphere model. Image analysis shows an equal performance of the single- and multi-layer models in terms of background noise (� 30 mJy beam 1 ) and source peak fluxes, but a slight (5–10 percent) improvement in the overall astrometric accuracy of the multi-layer model images. This outcome is consistent with the concept of a smooth bulk ionosphere that is poorly represented by a single-layer model.