Astronomical objects that change rapidly give us insight into extreme environments, allowing us to identify new phenomena, test fundamental physics, and probe the Universe on all scales. Transient and variable radio sources range from the cosmological, such as gamma-ray bursts, to much more local events, such as massive flares from stars in our Galactic neighbourhood. The capability to observe the sky repeatedly, over many frequencies and timescales, has allowed us to explore and understand dynamic phenomena in a way that has not been previously possible. In the past decade, there have been great strides forward as we prepared for the revolution in time domain radio astronomy that is being enabled by the SKA Observatory telescopes, the SKAO pathfinders and precursors, and other ‘next generation’ radio telescopes. Hence it is timely to review the current status of the field, and summarise the developments that have happened to get to our current point. This review focuses on image domain (or ‘slow’) transients, on timescales of seconds to years. We discuss the physical mechanisms that cause radio variability, and the classes of radio transients that result. We then outline what an ideal image domain radio transients survey would look like, and summarise the history of the field, from targeted observations to surveys with existing radio telescopes. We discuss methods and approaches for transient discovery and classification, and identify some of the challenges in scaling up current methods for future telescopes. Finally, we present our current understanding of the dynamic radio sky, in terms of source populations and transient rates, and look at what we can expect from surveys on future radio telescopes.

Impulsive radio signals such as fast radio bursts (FRBs) are imprinted with the signatures of multi-path propagation through ionised media in the form of frequency-dependent temporal broadening of the pulse profile, commonly referred to as scattering. The dominant source of scattering for most FRBs is expected to be within their host galaxies, an assumption which can be tested by examining potential correlations between the scattering properties of the FRBs and global properties of their hosts. Using results from the Commensal Real-time ASKAP Fast Transient (CRAFT) survey, we investigate correlations across a range of host galaxy properties against attributes of the FRB that encode propagation effects: scattering timescale $\tau$, polarisation fractions, and absolute Faraday rotation measure. From 21 host galaxy properties considered, we find three that are correlated with $\tau$, including the stellar surface density (or compactness; Pearson’s p-value p = 0.002 and Spearman’s p = 0.010), the mass-weighted age (Spearman’s p-value p = 0.009), and a weaker correlation with the gas-phase metallicity (Spearman’s $p = 0.017$). Weakly significant correlations are also found with $H\alpha$ equivalent widths and stellar gravitational potential. From 10 000 trials of reshuffled datasets, we expect two strong Spearman’s correlations only 2% of the time and three weaker correlations in 6.6% of cases. Compact host galaxies may have more ionised content which scatters the FRB further. Compact galaxies were also found to correlate with gas-phase metallicity in our sample, while H ii regions along the line-of-sight are also a potential contributing factor. No correlation is seen with host galaxy inclination, which weakens the case for an inclination bias, as previously suggested for samples of localised FRBs. A strong ($p = 0.002$) correlation is found for absolute rotation measure with optical disc axis ratio b/a; greater rotation measures are seen for edge-on host galaxies. Further high-time resolution FRB detections, coupled with localisation and detailed follow-up on their host galaxies, are necessary to corroborate these initial findings and shed further light into the FRB mechanism.

We report detection and analysis of the largest ever low-frequency sample of Crab giant pulses (GPs) detected in frequency band 200–231.25 MHz. In total, about $\sim$95 000 GPs were detected, which, to our knowledge is the largest low-frequency sample of Crab GPs presented in the literature. The observations were performed between 2024-12-14 and 2025-03-31 with the Engineering Development Array 2, a prototype station of the low-frequency Square Kilometre Array telescope. The fluence distribution of GPs in the entire sample is very well characterised with a single power law N(F) $\propto$ F$^\alpha$, where $\alpha =-3.17\pm0.02$ for all GPs, and $\alpha_{MP} =$ $-3.13\pm0.02$ and $\alpha_{IP} =-3.59\pm0.06$ for GPs at the phases of the main pulse and low-frequency interpulse, respectively. We do not observe flattening of the fluence distribution at the higher fluences. Although the index of the power law fluence distribution remained approximately constant over the observing period, the normalisation of the distribution was strongly anti-correlated (coefficient $\approx -0.9$) with the scatter broadening time. The timescale ($\sim$ weeks) of these variations indicates that intrinsic GP emission was modulated by the refractive scintillation as the signals propagated through the Crab Nebula and ISM. As a result, the measured fluence distribution was augmented for lower ($\tau \approx$ 2 ms) and diminished for higher ($\tau \approx$ 5 ms) scatter broadening time $\tau$ causing the GP detection rate to vary between 3 000 and 100 per hour, respectively (the correlation coefficient $\approx -0.9$). Furthermore, for the first time at low frequencies we observe indications of positive correlation (correlation coefficient $\approx$0.7) between the scatter broadening time ($\tau$) and dispersion measure. Our modelling favours the screen size $\sim10^{-5}$ pc with mean electron density $\sim 400\textit{e}^{-}$cm$^{-3}$ located within 100 pc from the pulsar (Crab Nebula or Perseus arm of the Milky Way galaxy). The observed frequency scaling of the scattering broadening time $\beta \approx -3.6\pm0.1$ (where $\tau \propto \nu^{\beta}$) is in agreement with the previous measurements. The observed maximum spectral luminosities $\sim 10^{25}$ erg/Hz/s approach those of the weakest pulses from some repeating fast radio bursts (FRBs). However, the distribution of pulse arrival times is consistent with a purely random Poisson process, and we do not find evidence of clustering. Overall, our results agree with the current views that GPs from extra-galactic Crab-like pulsars can be responsible for some very weak repeating FRBs, but cannot explain the entire FRB population. Finally, these results demonstrate an enormous transient science potential of individual SKA-Low stations, which can be unlocked by milli-second all-sky imaging.

We present microsecond-resolution, coherently dedispersed, polarimetric measurements of 35 fast radio bursts (FRBs) detected during the Commensal Real-time ASKAP Fast Transients (CRAFT) incoherent sum (ICS) survey with the Australian Square Kilometre Array Pathfinder (ASKAP). We find a wide diversity of time–frequency morphology and polarisation properties broadly consistent with those of currently known non-repeating FRBs. The high S/N and fine time-resolution of our data however reveals a wealth of new information. Key results include (i) the distribution of scattering timescales, ${{{\unicode{x03C4}}_\textrm{obs}}}$, is limited purely by instrumental effects, with no downturn at high ${{{\unicode{x03C4}}_\textrm{obs}}}$ as expected from a log-normal distribution; (ii) for the 29 FRBs with known redshift, there is no detectable correlation between ${{{\unicode{x03C4}}_\textrm{obs}}}$ and dispersion measure (DM) fluctuations about the Macquart relation, in contrast to expectations from pulsar scattering–DM relations; (iii) all FRBs probably have multiple components, and at least a large fraction have variable PA, the identification of which is limited by scattering; (iv) at least half of all FRBs exhibit PA microstructure at 200 $\mu{}$s–200 ns timescales, with behaviour most closely resembling a sub-category of Crab main pulses; (v) that there is a break in the FRB circular polarisation distribution at Stokes $V \gtrsim 20$%, which is suggestive of a discrete sub-population.

We present the first results from a new backend on the Australian Square Kilometre Array Pathfinder, the Commensal Realtime ASKAP Fast Transient COherent (CRACO) upgrade. CRACO records millisecond time resolution visibility data, and searches for dispersed fast transient signals including fast radio bursts (FRB), pulsars, and ultra-long period objects (ULPO). With the visibility data, CRACO can localise the transient events to arcsecond-level precision after the detection. Here, we describe the CRACO system and report the result from a sky survey carried out by CRACO at 110-ms resolution during its commissioning phase. During the survey, CRACO detected two FRBs (including one discovered solely with CRACO, FRB 20231027A), reported more precise localisations for four pulsars, discovered two new RRATs, and detected one known ULPO, GPM J1839 $-$10, through its sub-pulse structure. We present a sensitivity calibration of CRACO, finding that it achieves the expected sensitivity of 11.6 Jy ms to bursts of 110 ms duration or less. CRACO is currently running at a 13.8 ms time resolution and aims at a 1.7 ms time resolution before the end of 2024. The planned CRACO has an expected sensitivity of 1.5 Jy ms to bursts of 1.7 ms duration or less and can detect $10\times$ more FRBs than the current CRAFT incoherent sum system (i.e. 0.5 $-$2 localised FRBs per day), enabling us to better constrain the models for FRBs and use them as cosmological probes.

The Variables and Slow Transients Survey (VAST) on the Australian Square Kilometre Array Pathfinder (ASKAP) is designed to detect highly variable and transient radio sources on timescales from 5 s to $\sim\!5$ yr. In this paper, we present the survey description, observation strategy and initial results from the VAST Phase I Pilot Survey. This pilot survey consists of $\sim\!162$ h of observations conducted at a central frequency of 888 MHz between 2019 August and 2020 August, with a typical rms sensitivity of $0.24\ \mathrm{mJy\ beam}^{-1}$ and angular resolution of $12-20$ arcseconds. There are 113 fields, each of which was observed for 12 min integration time, with between 5 and 13 repeats, with cadences between 1 day and 8 months. The total area of the pilot survey footprint is 5 131 square degrees, covering six distinct regions of the sky. An initial search of two of these regions, totalling 1 646 square degrees, revealed 28 highly variable and/or transient sources. Seven of these are known pulsars, including the millisecond pulsar J2039–5617. Another seven are stars, four of which have no previously reported radio detection (SCR J0533–4257, LEHPM 2-783, UCAC3 89–412162 and 2MASS J22414436–6119311). Of the remaining 14 sources, two are active galactic nuclei, six are associated with galaxies and the other six have no multi-wavelength counterparts and are yet to be identified.

The recent increase in well-localised fast radio bursts (FRBs) has facilitated in-depth studies of global FRB host properties, the source circumburst medium, and the potential impacts of these environments on the burst properties. The Australian Square Kilometre Array Pathfinder (ASKAP) has localised 11 FRBs with sub-arcsecond to arcsecond precision, leading to sub-galaxy localisation regions in some cases and those covering much of the host galaxy in others. The method used to astrometrically register the FRB image frame for ASKAP, in order to align it with images taken at other wavelengths, is currently limited by the brightness of continuum sources detected in the short-duration (‘snapshot’) voltage data captured by the Commensal Real-Time ASKAP Fast Transients (CRAFT) software correlator, which are used to correct for any frame offsets due to imperfect calibration solutions and estimate the accuracy of any required correction. In this paper, we use dedicated observations of bright, compact radio sources in the low- and mid-frequency bands observable by ASKAP to investigate the typical astrometric accuracy of the positions obtained using this so-called ‘snapshot’ technique. Having captured these data with both the CRAFT software and ASKAP hardware correlators, we also compare the offset distributions obtained from both data products to estimate a typical offset between the image frames resulting from the differing processing paths, laying the groundwork for future use of the longer duration, higher signal-to-noise ratio (S/N) data recorded by the hardware correlator. We find typical offsets between the two frames of ${\sim}0.6$ and ${\sim}0.3$ arcsec in the low- and mid-band data, respectively, for both RA and Dec. We also find reasonable agreement between our offset distributions and those of the published FRBs. We detect only a weak dependence in positional offset on the relative separation in time and elevation between target and calibrator scans, with the trends being more pronounced in the low-band data and in Dec. Conversely, the offsets show a clear dependence on frequency in the low band, which we compare to the frequency-dependent Dec. offsets found in FRB 200430. In addition, we present a refined methodology for estimating the overall astrometric accuracy of CRAFT FRBs.