This figures shows the changes in amplitudes and coherences on the phase calibrator 3C84 after applying the water line phase corrections. The left hand side of the figure shows the observed astronomical amplitudes and coherences as a function of the projected baseline length before applying any water line corrections. The amplitudes are distributed between 0 and 4 Jy, with the largest scatter occurring on longer baselines. Ideally the amplitudes should be constant as a function of baseline length since the observed source is point-like. The observed scatter is of course due to de-coherence during the 3 minute astronomy integration.

The right hand side of the above figure shows the corresponding plots after applying the water line corrections. The amplitudes in general increase on all baseline lengths and show a much less scatter than without the water line corrections. Further, the coherences are generally above approximately 0.8 over all baselines. (Note: The half of dozen or so integrations that were not significantly improved by the WLM technique have been traced to a hardware problem in the WLM boxes during a single integration near the end of the track.)