Speaker
Description
Over the past decade, significant efforts has gone into developing calibration and imaging pipelines to automatically process data from the Dutch high-band antennas of the Low Frequency Array (LOFAR), which observe the Universe at about 150 MHz and reach with the Dutch array a 6" resolution (e.g. Shimwell et al. 2017; 2019). Recently, we have extended this work by developing a pipeline to calibrate and image data taken with all European LOFAR stations, reaching sub-arcsecond resolutions across a 2.5x2.5 degrees field of view (Morabito et al. 2022; Sweijen et al. 2022; de Jong et al. 2024). Since the computational costs for producing wide-field high-resolution images represent a major bottleneck, we have focused on reducing these costs by up to a factor of 15, through the development of advanced techniques and refined calibration strategies to mitigate ionospheric distortions and improve image fidelity as well (de Jong et al. 2025; de Jong et al., submitted). These advancements have led to the production of the deepest radio image to date with a field of view of 2.5x2.5 degrees and reaching an RMS near the pointing centre of about 6 μJy/beam at about 0.3″ resolution, using 200 hours of LOFAR data (de Jong et al. in prep.).
In this talk, I will present the latest developments in the LOFAR VLBI pipeline for wide-field imaging and highlight the recent enhancements that enabled this record-breaking image. I will also briefly highlight future directions, including the work on improved automated decision-making, opportunities for the use of AI, and data volume management. These developments will make the pipeline more robust and pave the way for high-resolution (and ultra-deep) LOFAR surveys at 150 MHz, enabling us to study the low-frequency universe across the northern sky at the smallest angular scales.
