Fast Radio Transient searches with UTMOST at 843 MHz

Introduction

High time resolution astronomy over the last decade has led to the discovery of new classes of radio sources such as the Rotating Radio Transients (RRATs) (McLaughlin et al. 2006) and Fast Radio Bursts (FRBs) (Lorimer et al. 2007 Lorimer et al. 2007; Thornton et al. 2013 Thornton et al. 2013; Spitler et al. 2014 Spitler et al. 2014; Burke-Spolaor & Bannister 2014 Burke-Spolaor & Bannister 2014; Petroff et al. 2015 Petroff et al. 2015; Ravi et al. 2015 Ravi et al. 2015; Champion et al. 2015 Champion et al. 2015; Masui et al. 2015 Masui et al. 2015 ; Keane et al. in prep). The majority of the RRATs and all the FRBs are characterized by millisecond duration pulses implying coherent physical processes in their origin if they are compact sources. RRATs have been found to repeat on timescales of a few pulses an hour to a few pulses a day while FRBs have not yet been seen to repeat. These elusive bursts have $\sim$ Jy peak flux densities and dispersion measures that well exceed the contribution from the Milky Way along the line of sight indicative of a possible a cosmological origin. Only a handful of FRBs are known, and to date no transient event or afterglow has been seen at any other wavelength despite major efforts to do so (Petroff et al. 2015). Several cosmological and non-cosmological models for the origin of FRBs have been suggested, including radio emission from pulsars (Cordes & Wasserman 2015; Connor et al. 2015), collapsing gravitationally unstable black holes (Falcke & Rezzolla 2014), hyper flares from magnetars (Lyubarsky 2014) and dark matter induced collapse of neutron stars (Fuller & Ott 2015).

There is clearly a need to discover FRBs more efficiently as the present discovery rate is only of order 1 per $\sim$ 12 days on sky at Parkes. The 50 year old Molonglo Observatory Synthesis Telescope in Australia is currently being refurbished with a new digital backend system and increased bandwidth as part of an upgrade to transform it into a burst finding machine. This instrument being an interferometer will help discern if FRBs are truly a celestial population by measuring a parallax to the sources. In this paper we introduce the Molonglo observatory synthesis telescope and discuss its single pulse sensitivity in Section 2. The first FRB survey at 843 MHz using the UTMOST instrument and limits on the detectability of FRBs is discussed in Section 3. We make estimates of the FRB rates we can expect with the UTMOST instrument, showing that at full sensitivity it is considerably more effective than Parkes for doing FRB surveys due to its large field-of-view and high observing duty cycle (Section 4) under conservative assumptions for the FRB spectral index. At full sensitivity we expect to detect an event every few days. We constrain the FRB event rate and mean spectral index based on the non-detection of FRBs in these pilot surveys and draw our conclusions in Section 5.

The Molonglo Observatory Synthesis Telescope (MOST)

The Molonglo telescope was originally a “Mills Cross” design, completed in 1967 (Mills 1981) and operating as a transit instrument at 408 MHz. It is located about 300 km south-west of Sydney, near Canberra, and is a field station of the University of Sydney. It played a crucial role in radio astronomy with the discovery of the Vela pulsar (Large et al. 1968) and 155 new pulsars in the second Molonglo pulsar survey (Manchester et al. 1978). It was substantially modified in the early 1980s to make the East-West (E-W) arms fully steerable and increase the operating frequency to 843 MHz and a 3 MHz bandwidth (Robertson 1991). The telescope was renamed the Molonglo Observatory Synthesis Telescope (MOST); it has a collecting area of 18000 m2, the largest in the Southern hemisphere.

The E-W arm consists of two collinear cylindrical paraboloids, each 11.6 m wide and 778 m long, separated by a 15 m gap (Bock et al. 1999). Each paraboloid is divided into smaller sections called “modules”, each with a beam of order $4.64^{\circ}\times 2.14^{\circ}$ (EW-NS). Four such modules were linked together digitally to form a “bay” and 44 such bays constitute one arm. A line feed system of 7744 right circularly polarised dipoles (22 per module), in 352 resonant chambers each feeding a Low Noise Amplifier (LNA) means that the telescope is effectively an array of 352 receivers operating at a system temperature of $\sim$ 70 K (Campbell-Wilson et al. 1997). The E-W arms can be tilted North-South (N-S), while E-W pointing is attained by differential rotation of the ring antennae (spaced at 0.54 $\lambda$) on the line feed. The telescope can access the whole sky south of $\delta=+18^{\circ}$, although hour angle coverage is limited to an E-W tilt of $\pm 60^{\circ}$.

The telescope is currently being upgraded both in the backend receivers and with the installation of a new graphics processing unit (GPU) based correlator, in a collaboration between Sydney and Swinburne Universities. The installation of high-performance GPUs at MOST has transformed it into a powerful instrument, the Swinburne University of Technology upgrade for the MOST (UTMOST ; Bailes et al. in prep) and enlarged the field of view to twice that of the Sydney University Molonglo Sky Survey (Bock et al. 1999, SUMSS ;) due to processing data from each ‘module’ rather than each ‘bay’.

The sensitivity of UTMOST to FRB events (i.e. single pulse events) can be calculated using the radiometer equation,

The main properties of UTMOST and Parkes from the point of view of discovering FRBs are shown in Table 1.

FRB surveys at UTMOST

Two FRB searches have been performed at UTMOST at different fractional sensitivities during the ongoing upgrade. These two surveys are called V1.0 and V2.0. The antennae are aligned and fringe stopped to maintain stable and flat phases and then combined into a tied-array beam, centered on the primary beam boresight. This beam is then re-steered into 352 tied array beams called “fan beams” that are “tiled” across the 4 degree East-West axis of the primary beam. Time series from each fan beam are detected and integrated from 1.28 to 655.36 $\upmu$s sampling and also requantised to 8-bits/sample.

The total data rate to the backend is 11 GBps, and the resulting output data rate from the 352 beams is approximately 10 MBps for both surveys. The input stream at UTMOST in FRB search mode is 16 MHz for survey V1.0 and 31.25 MHz for V2.0, of single polarisation baseband data from 352 antennae, in 20 frequency channels produced in a polyphase filterbank (PFB). We upgraded to 40 coarse channels in FRB survey V2.0. After completion of V1.0, the rest of the GPUs were installed onsite (May 2015) so that the full 31.25 MHz could be processed for V2.0. As a consequence it is clear that roll-off at the edge of the bandpass is quite pronounced so that the extra bandwidth is not usable. We conservatively assume 16 MHz of final effective bandpass for all the results in this paper.

2 FRB Survey V2.0

Estimates of FRB rates at UTMOST

Hassall et al. 2013 have estimated FRB rates that might be seen at a wide range of radio telescopes operating over a wide range of frequencies. They assumed the bursts to be standard candles, to have a constant spectral index and a constant co-moving space density. They estimated a detection rate of $\sim$ 3 per day at MOST, but this is an overestimate for the present system being installed. The MOST telescope specifications they adopt from Green et al. 2012 are for a more ambitious upgrade path than the current UTMOST design, which has a bandwidth a factor of 6 smaller and a field of view smaller by 60$\%$ (Bailes et al. in prep). We now estimate the rate of FRBs for the current upgrade at UTMOST using two methods, in both cases scaling from the event rate at Parkes.

2 Empirical scaling from events at Parkes

3 Event rate based on surveyed volume

Discussion and Conclusions

The discovery of FRBs has opened up numerous exciting possibilities for the exploration of the extragalactic Universe. However their extragalactic/celestial origin is yet to be decisively established. With the newly upgraded UTMOST array, we will be able to affirm if these sources are truly extraterrestrial when we detect one, as the array’s Fresnel zone is at $\sim$ 14,000 km. FRB searches at two different fractional sensitivities (7$\%$ and 14$\%$) were performed as part of commissioning science with the telescope parked at at $\delta=-46^{\circ}$. The chosen declination was to allow the diurnal passage of bright southern pulsars Vela and PSR J1644$-$4559 and a bright calibrator, radio galaxy J1935$-$4620, so that the system performance, phases and delays could be monitored. No FRBs were detected down to fluence limits of 23 Jy ms and 11 Jy ms after spending 467 and 225 hours on sky respectively.

Acknowledgements

The authors would like to thank the staff at the Molonglo Observatory for the exceptional support provided. The Molonglo Observatory is owned and operated by the University of Sydney with support from the Australian Research Council (ARC) and the Science Foundation within the School of Physics. The upgrade to the telescope has been supported by the University of Sydney, Swinburne University of Technology and the ARC, including via CAASTRO. Parts of this research were conducted by the Australian Research Council Centre for All-Sky Astrophysics (CAASTRO), through project number CE110001020.

References