Speaker
Description
Data reduction pipelines are the cornerstone of scientific output. Catching mistakes as early as possible is paramount for more complex pipelines to succeed. Over the past decade our understanding of the LOFAR instrument and data reduction workflows has grown substantially and now encompasses at least three major pipelines, each tackling more complex or harder problems. In this talk I will discuss the recent addition of self-calibration to the LINC pipeline, added to improve the calibration on difficult fields such as the galactic plane or those containing bright compact sources. Based on this I will show examples of poor calibration, how things can go wrong early on and emphasize the need for (automated) quality control checks as early on as possible. Now that complex tools and pipelines are readily available, it is time to make our pipelines stricter in terms of what we accept for processing.
