Qual Reference: Transient Rate Measurements

2010 August 2

There are many measurements of radio transient rates out there in the literature. Unfortunately, there are many different ways to quantify transient rates and they tend to be very hard to intercompare. In lieu of trying to standardize everything, I’ll just report some of the values that I’ve seen and the way in which they’re measured:

• Becker+ 2010. The headline number is 1.6 Galactic sources per square degree which vary by more than 50% on a timescale of <~ years. They ignored scintillation. This is based on guessing how many XG sources were in the background. (Found ~4:1 G:XG.) They had to work around the fact that their two comparison images were made in different VLA configurations and so fluxes are not as easily comparable as you’d like. Found a strongly increasing variable fraction as they got closer to the GC, suggesting that the areal variable density there would be quite high; likely ISS.

• Bower+ 07. By far the most thorough job of thinking about / reporting transient rates. For short timescale, S > 370 uJy, 5 & 8.4 GHz, rate is between 0.07 and 40 per sqdeg per yr. (Timescale is between 20 min and 7 d) Two-epoch survey should get 1.5 ± 0.4 transients per sqdeg for S > 370 uJy. They have tables of effective area searched as a function of flux limit for various timescales. Based on two transients with galaxy hosts, compute volumetric rates of 3e5 / Gpc^3 / yr (t_char = 7d) and 4e3 / Gpc^3 / yr (t_char = 60d), or joint rate of (~2 ± 2)e5 / Gpc^3 / yr. Reported SN Ib/c rate is 4.8e4. Figure 9 summarizes their multi-depth results and those of other surveys:

  

• Hyman+ 2008 (GMRT). No effort to compute rates here. For the GMRT work, 1 transient found in 66 hours of observing (22 epochs) with a typical limiting flux of ~20 mJy at 0.235 GHz. Snapshot FWHM of 2 deg gives you an equivalent 2-epoch area of 66 sqdeg. This is of course looking at the GC.

• Gregory & Taylor 1986. Nothing explicitly given on rates here, either. 500 sqdeg surveyed at 4.76 GHz. 32 variables and 27 “possible variables” detected, about 11 of which are false positives, with 64% showing “short-term” variability. (Most sources are faint so variability implies changes in S of order 50%.) 1274 sources total -> ~3% variable fraction. Estimate >90% are XG. Some comparisons to early 80’s work that I’m going to ignore.

• Carilli+ 03. 19d and 17m timescales probed at 1.4 GHz. Find rate < 18 / sqdeg for S > 100 uJy.

• Frail+ 03. Find four variables, no transients in GRB afterglow searches. Find R_variable < 6 / sqdeg.

• Niinuma+ 07, Kuniyoshi+ 07, Matsumura+ 07 (see Bower+ 08 for refs). Bright transients in drift scans; three >1 Jy lasting < 1d, one > 3Jy lasting ~3d; not widely believed.

• Levinson+ 02, Gal-Yam+ 06. FIRST/NVSS comparisons. 2400 sqdeg at S > 6 mJy and 1.4 GHz. Limit that there are < 65 RTs in entire sky above 6 mJy at once (95% CL). Limit volumetric rate < 1000 / Gpc^3 / yr. (Typo in abstract: Gpc^1 not Gpc^-3.) Predict ~1000 in similar 1/17-sky survey if S > 1 mJy.