Speaker
Description
LOTAAS, the LOFAR Tied-Array All-Sky Survey, offers an extensive dataset for discovering low-frequency radio transients. However, the original single-pulse search was not optimized to detect longer-duration or heavily scattered signals, and was limited by high false positive rates arising from instrumental variability and RFI. Since then, our understanding of FRBs at low frequencies has significantly advanced. In particular, two repeating sources have now been detected with LOFAR, demonstrating that low-frequency emission is indeed produced in some FRBs and can escape their local environments. Meanwhile, surveys at higher frequencies continue to reveal a population of highly scattered FRBs and long-period transient sources, raising the possibility that similar signals remain undetected in the LOTAAS data.
In this talk, I present the reprocessing of the LOTAAS survey using updated methods tailored to this new landscape. A central component of this work is “flat-fielding,” a normalization technique we introduce to suppress beam-to-beam gain variations and reduce the impact of persistent RFI across the array. This is done by dividing each beam by the average of the central beams within a sub array pointing, effectively flattening the instrumental response across frequency and time while preserving astrophysical signal morphology. This significantly lowers the false positive rate—by several orders of magnitude—and improves the reliability of single-pulse detection. Additionally, the pipeline improves sensitivity to wide and scattered bursts and is designed to process data efficiently at scale. I’ll highlight early results from reprocessed beams and discuss how this method provides a path forward for scalable transient detection in future datasets, specifically through the EuroFlash cluster for LOFAR 2.0.
