Display of the results

Pieflag displays its results to the user for iterative adjustment of the parameters (Figure 1). The visibility amplitudes of one baseline and one channel are displayed, either from all pointings at once, or from one pointing only. The latter is more instructive when attempting to understand why data have been flagged. In the single source mode, $x_{\rm b,p}$, $x_{\rm b,p}\pm ny$, and $x_{\rm b,p}\pm 2ny$ are indicated on the right of the plot. Although the plotting stage allows interactive flagging and unflagging using mouse and keyboard controls, its primary intend is to allow one to inspect the results. Consequently, once Pieflag is tuned to fit one's data, it can be run without plotting.

Figure 1: An example of how Pieflag displays its results. The labels are very small in print, but are large enough to read on a computer screen for which they are intended. Amplitudes of one pointing, measured in one frequency channel on one baseline are displayed. The plot illustrates the two flagging algorithms. All data observed after 11:00h have amplitudes higher than $x_{\rm b,p}+7y$ or $x_{\rm b,p}+14y$, indicated on the right of the plot. These points have been flagged by amplitude-based flagging, followed by postprocessing. Several scans between 6:00h and 11:00h have amplitudes which are below the threshold of amplitude-based flagging, but have been detected by rms-based flagging. Other scans between 6:00h and 11:00h have not been detected, but were flagged in postprocessing, because data from adjacent pointings were detected and flagged.