THE PECULIAR RADIO SOURCE M17 JVLA 35

In a study made with the Karl G. Jansky Very Large Array (JVLA) at 4.96 and 8.46 GHz toward the classic H ii region M17, Rodríguez et al. (2012) detected 38 compact radio sources. Of these sources, 19 were found to have stellar counterparts detected in the infrared, optical, or X-rays. Among the sources without previously known counterparts, these authors noted that M17 JVLA 35 (Figure 1) was remarkable because of its steep positive spectral index α ⩾ 2.9 ± 0.6 (S_ν∝ν^α). Sources with such a spectral index at these frequencies are rare. For example, Dzib et al. (2013) determined the spectral index between 4.5 and 7.5 GHz for 165 sources in the Ophiuchus Complex. Only two of them have spectral indices comparable to that of JVLA 35.

Zoom In Zoom Out Reset image size

Counterparts to JVLA 35 were searched unsuccessfully in SIMBAD as well as in the GLIMPSE catalog of candidate young stellar objects with a high probability of association with the M17 complex (Povich et al. 2009). Its unusual radio spectrum deserves further study.

In this paper we report new VLA observations of M17 JVLA 35 that extend its spectrum to the high frequencies: from 18.5 to 36.5 GHz. Our main goal was to obtain a spectrum extending sufficiently in frequency to attempt a better classification of the source.

Observations of M17 were made with the Karl G. Jansky Very Large Array of the National Radio Astronomy Observatory (NRAO).⁵ We observed the continuum emission in the K and Ka bands in C configuration during 2013 June 22 (project code: 13A-125). For each band, we observed a total continuum bandwidth of 8 GHz covering the frequency ranges 18–26 GHz in the K band, and 29–37 GHz in the Ka band. Each band is divided into 4096 channels of 2 MHz each. Bandpass and complex gain calibration were made by observing J1825−1718 every 2 minutes. Flux calibration was achieved by observation of the standard flux calibrator 3C 48. We centered our observations at the position of M17 JVLA 35 (α(J2000) = 18^h20^m33100, β(J2000) = −16°09'4100). The on-source time in each band was ∼20 minutes. Calibration of the data were performed with the data reduction package Common Astronomy Software Applications (CASA; version 4.1.0)⁶ following standard VLA procedures for high frequency data.

We divided the calibrated data of each band in eight parts of 1 GHz each and made cleaned continuum maps of each chunk by using the task clean of CASA. In order to resolve out the extended emission from the M17 H ii region, we used only visibilities larger than 50 kλ in our imaging, thus suppressing structures larger than ∼4''. A similar imaging procedure was used by Rodríguez et al. (2012). The central frequencies, rms noises, and synthesized beams for each map are shown in Table 1. Source JVLA 35 is detected in all our 1 GHz wide maps and we obtained its flux density at each frequency by performing a Gaussian fit. Our typical angular resolution is ∼1'' and the source is unresolved in our images.

In Table 1 we show the flux density of JVLA 35 obtained from our data and in Figure 2 we show the spectral energy distribution (SED). In the SED we have included the flux density at 8.46 GHz and the upper limit at 4.96 GHz obtained by Rodríguez et al. (2012). As can be seen, the spectral index of the emission is positive at low frequencies, there is a maximum of emission at intermediate frequencies, and then the emission diminishes with frequency with a negative spectral index at high frequencies.

Zoom In Zoom Out Reset image size

We fitted the SED by assuming a synchrotron-like spectrum in the form

$$\begin{equation} S_\nu = S_{\nu _0} \left(\frac{\nu }{\nu _0} \right)^{\alpha _{{\rm thick}}} (1 - {\rm exp}(-\tau _{\nu _0} (\nu /\nu _0)^{\alpha _{{\rm thin}}-\alpha _{{\rm thick}}})), \end{equation} \tag{ 1 }$$

where $S_{\nu _0}$ and $\tau _{\nu _0}$ are the flux density and the optical depth at a reference frequency ν₀. This equation assumes that the emission is optically thick at low frequencies with a spectral index α_thick, and optically thin at high frequencies with a spectral index α_thin. We assumed the theoretically expected value of α_thick = 2.5 and obtained α_thin = −2.0 by fitting the data at the highest frequencies. The maximum of the emission takes place at ∼13.3 GHz. It should be noted that the spectral index at low frequencies is only an estimate since we only have an upper limit and one measured flux density. Additional observations are required.

In order to further constrain the observed properties of M17 JVLA 35, we extended our search for counterparts to a wider range of frequencies by analyzing several deep survey images of M17. We retrieved these from the public archives of the United Kingdom Infrared Telescope (UKIRT) Infrared Deep Sky Survey (UKIDSS; Lawrence et al. 2007) in the JHK near-infrared bands, the Spitzer Galactic Legacy Infrared Midplane Survey Extraordinaire (GLIMPSE; Benjamin et al. 2003; Churchwell et al. 2009) taken with the Infrared Array Camera (IRAC) at 3.6, 4.5, 5.8, and 8 μm, and the Herschel Infrared Galactic Plane Survey (Molinari et al. 2010) at 70, 160, 250, 350, and 500 μm. No compact infrared counterparts were found in any of these images within ±1'' of the JVLA position. Using the appropriate survey calibrations, we derived the following upper limits to the flux densities: 16 μJy at 1.2 μm, 23 μJy at 1.6 μm, 17 μJy at 2.2 μm, 40 μJy at 3.6 μm, 60 μJy at 4.5 μm, 200 μJy at 5.8 μm, 1 mJy at 8 μm, 1.2 Jy at 70 μm, and 1.8 Jy at 160 μm. At 250, 350, and 500 μm, the extended thermal dust emission associated with M17 is too strong to derive any useful limit. As an illustration, Figure 3 shows the K-band UKIDDS and the GLIMPSE 5.8 μm images of a small field centered on M17 JVLA 35. Note also that no Chandra X-ray source was listed at this position by Broos et al. (2007).

Zoom In Zoom Out Reset image size

In this section we discuss continuum emission mechanisms that could explain the spectrum observed for M17 JVLA 35. In addition to the mechanisms discussed below, we note that an explanation in terms of a galactic background massive star seems unlikely. These stars have ionized winds that at high frequencies show a spectral index of +0.6 (i.e., Abbott et al. 1985), in contrast to the negative spectral index shown by JVLA 35.

4.1. Self-absorbed Synchrotron Source

The observed spectrum is similar to those seen in the high frequency peakers (HFPs). These are compact, powerful extragalactic radio sources with well-defined peaks in their radio spectra above 5 GHz, with most of them being high redshift quasars (Dallacasa et al. 2000). A possible explanation for the HFP radio sources is that we are observing self-absorbed synchrotron emission. An estimate for the magnetic field B in the region can be obtained from (Kellermann & Pauliny-Toth 1981)

$$\begin{equation*} \left[{{B} \over {{\rm G}}} \right] \simeq 9.1 \times 10^{-2} \left[{{\nu _{to}} \over {5 \,{\rm GHz}}} \right]^5 \left[{{S_{to}} \over {{\rm Jy}}} \right]^{-2} \left[{{\theta _{to}} \over {{\rm mas}}} \right]^{4}, \end{equation*}$$

where ν_to, S_to, and θ_to are the frequency, flux density, and angular size of the source at the turnover. For an HFP radio source, the terms in square brackets on the right-hand side of the equation are of order unity and a magnetic field of order 100 mG is obtained (e.g., Orienti & Dallacasa 2014). However, M17 JVLA 35 is clearly resolved at 8.46 GHz (with deconvolved dimensions of 065 ± 009 × 037 ± 006;  P.A.  =  19° ± 12°; Rodríguez et al. 2012). We will adopt as the characteristic angular size the geometric mean of the major and minor axes, ∼05. For this angular dimension the resulting magnetic field is unrealistically large for any type of astronomical object. We then consider alternative explanations for the nature of JVLA 35.

4.2. Optically Thin Synchrotron with Free–Free Absorption by M17

M17 JVLA 35 could be a source with a single power-law spectrum, with the turnover observed at ∼13 GHz due to free–free absorption in the low frequencies produced by the M17 H ii region. To test this hypothesis, we calibrated VLA archive observations made at 1.42 GHz in the DnC configuration. These observations were made on 1991 February 12 under project AM279. We calibrated the data using the usual NRAO procedures and made images from the average of the pure continuum (that is, without detectable H i absorption) channels. The data was self-calibrated in phase and the final image is shown in Figure 4. The total flux density from the H ii region in this image is ∼370 Jy. This result indicates that we are detecting most of the extended emission from the region since the single dish measurement of Altenhoff et al. (1970) at the same frequency gives a total flux density of 500 ± 50 Jy. The position of JVLA 35 is indicated with an × symbol.

Zoom In Zoom Out Reset image size

The flux density per beam at the position of JVLA 35 is 2.25 Jy beam⁻¹. In contrast, the flux density per beam at the position of peak emission in the image (α = 18^h20^m2505; δ = −16°11'347) is 9.09 Jy beam⁻¹. These flux densities per beam imply brightness temperatures of ∼1500 K for the line of sight to JVLA 35 and of ∼6200 K for the position of peak emission. Assuming that the emission in the latter position traces optically thick emission, we assume that the electron temperature of the H ii region is T_e ≃ 6200 K. Then, using

$$\begin{equation*} T_B = T_e[1 - {\rm exp}(-\tau _\nu)], \end{equation*}$$

we derive a free–free optical depth of τ(1.42 GHz) ≃ 0.3 for the emission in the line of sight to JVLA 35. Since the free–free optical depth goes as ν^−2.1, we expect τ(13 GHz) ≃ 0.002. We then conclude that the turnover at 13 GHz cannot be explained by free–free absorption from M17.

4.3. Optically Thin Synchrotron with Associated Free–Free Absorption

4.4. Self-absorbed Free–Free

4.5. Dust Emission

4.6. Spinning Dust Emission

4.7. Self-absorbed Synchrotron Source with Line-of-sight Plasma Scattering

We discussed several possibilities to explain the observed spectrum and angular size of the source M17 JVLA 35. We favor that it is an HFP whose angular size is plasma-broadened as the radiation travels across M17. This hypothesis can be tested by searching the expected ν⁻² angular size dependence in the centimeter range.

This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France. C.C.G. and L.F.R. are grateful to CONACyT, Mexico and DGAPA, UNAM for their financial support. The UKIDSS project is defined in Lawrence et al. (2007). UKIDSS uses the UKIRT Wide Field Camera (WFCAM; Casali et al. 2007) and a photometric system described in Hewett et al. (2006). The pipeline processing and science archive are described in Hambly et al. (2008). The Spitzer Space Telescope is operated by the Jet Propulsion Laboratory, California Institute of Technology (CIT) under National Aeronautics and Space Administration (NASA) contract 1407. Herschel is an ESA space observatory with science instruments provided by European-led Principal Investigator consortia and with important participation from NASA.

Facilities: VLA - Very Large Array, UKIRT - United Kingdom Infrared Telescope, Spitzer - Spitzer Space Telescope satellite, Herschel - European Space Agency's Herschel space observatory