New methods to constrain the radio transient rate: results from a survey of four fields with LOFAR

Abstract

We report on the results of a search for radio transients between 115 and 190 MHz with the LOw-Frequency ARray (LOFAR). Four fields have been monitored with cadences between 15 min and several months. A total of 151 images were obtained, giving a total survey area of 2275 deg. We analysed our data using standard LOFAR tools and searched for radio transients using the LOFAR Transients Pipeline. No credible radio transient candidate has been detected; however, we are able to set upper limits on the surface density of radio transient sources at low radio frequencies. We also show that low-frequency radio surveys are more sensitive to steep-spectrum coherent transient sources than GHz radio surveys. We used two new statistical methods to determine the upper limits on the transient surface density. One is free of assumptions on the flux distribution of the sources, while the other assumes a power-law distribution in flux and sets more stringent constraints on the transient surface density. Both of these methods provide better constraints than the approach used in previous works. The best value for the upper limit we can set for the transient surface density, using the method assuming a power-law flux distribution, is 1.3 × 10 deg for transients brighter than 0.3 Jy with a time-scale of 15 min, at a frequency of 150 MHz. We also calculated for the first time upper limits for the transient surface density for transients of different time-scales. We find that the results can differ by orders of magnitude from previously reported, simplified estimates.

New Limits on the Low-frequency Radio Transient Sky Using 31 hr of All-sky Data with the OVRO–LWA

“…Because the transient search is conducted over nearly the entire primary beam FOV, each epoch of our survey has nonuniform sensitivity, with the best and worst sensitivities achieved at zenith and 10 • elevation angle, respectively. However, for a transient population described by a luminosity function with power law γ, varying sensitivity corresponds to probing different parts of the luminosity distribution, and therefore the number of transients we can expect to detect within the volume probed by our survey is a function of sensitivity and the corresponding fraction of the sky covered by each sensitivity (see, e.g., the methods presented in Carbone et al 2016). Using Equation 3, we can write…”

“…For reference, the surface density for the transient detected byStewart et al 2016, at a reported flux density of between 15 and 25 Jy, and at the flux density that same event would appear at in our survey for a maximum intrinsic bandwidth of 195 kHz (the bandwidth of the survey byStewart et al 2016). See alsoFigure 4ofCarbone et al 2016 for constraining transient surface density as a function of flux density for different power-law luminosity distributions.…”

A Matched Filter Technique for Slow Radio Transient Detection and First Demonstration with the Murchison Widefield Array

“…R16 denotes Rowlinson et al (2016). Jaeger12 (Jaeger et al 2012), Hyman09 (Hyman et al 2009), and Stewart16 (Stewart et al 2016) reported detections; the rest reported upper limits (Bell et al 2014;Cendes et al 2014;Carbone et al 2016;Polisensky et al 2016;Rowlinson et al 2016;Murphy et al 2017). We only plotted two points for Polisensky16: the limit at the lowest flux density, and the best limit reported for 10-minute transients.…”

“…However, recent blind surveys with new widefield radio interferometers, such as the Murchison Widefield Array 21 (MWA; Lonsdale et al 2009;Tingay et al 2013) and the Low-Frequency Array 22 (LOFAR; van Haarlem et al 2013), are setting increasingly stringent limits on the surface density of radio transients at different sensitivities and over a wide range of timescales. In particular, Bell et al (2014) reported a limit of <´-7.5 10 5 deg −2 above 5.5 Jy at 154 MHz for minute-to-hour transients, Carbone et al (2016) reported a limit of <´-1.0 10 3 deg −2 above 500 mJy at 152 MHz for minute-to-hour transients, Cendes et al (2014) reported a limit of <´-2.2 10 2 deg −2 above 500 mJy at 149 MHz for 11-minute transients, Rowlinson et al (2016) reported a limit of <´-6.4 10 7 deg −2 above 210 mJy at 182 MHz for 30 s transients up to <´-6.6 10 3 deg −2 for 1 yr transients, Polisensky et al (2016) reported a range of limits (~-10 4 deg −2 above 100 mJy) for 10-minute and 6 hr transients at 340 MHz, and Murphy et al (2017) reported a limit of <´-1.8 10 4 deg −2 above 100 mJy between 150 and 200 MHz for transients that lasted 1-3 yr. Jaeger et al (2012) reported a transient detection that corresponded to a surface density of 0.12 deg −2 above 2.1 mJy over 12 hr at 325 MHz, while Stewart et al (2016) reported a transient detection that implied a surface density of´-1.5 10 5 deg −2 above 7.9 Jy for ∼10-minute transients at 60 MHz.…”

Calculating transient rates from surveys

“…When one also invokes a certain timescale of occurrence, this can be converted into a transient rate. While the transient surface density is straightforward to calculate, since the total area of a given survey is typically well known, determining the transient rate is not trivial because it involves both the timescales of the survey and the duration of the transients (see, e.g., Carbone et al 2016) The transient surface density, as well as the transient rate, can be calculated as a function of various parameter of the transients, e.g., their flux and duration. If a survey results in non-detections, the common way to calculate upper limits for the transient surface density as a function of flux is to assume a Poisson distribution for transients, and use all the independent pairs of observations within the survey to calculate the total amount of surveyed area.…”