The nature of the radio source detected towards the exoplanet system 1RXS1609.1−210524

ABSTRACT

Several studies have been carried out to detect radio emissions from known exoplanets. Some of these studies have resulted in tentative detections of radio sources near the position of known exoplanets. One such planet/brown dwarf around which a radio source was detected is 1RXS1609.1−210524 (hereafter 1RX) b. A radio source near 1RX was detected with the TIFR GMRT Sky Survey (TGSS) at 150 MHz and the NRAO VLA Sky Survey (NVSS) at 1.4 GHz. However, since these surveys’ spatial resolution was low, it was not possible to ascertain whether the radio emission originated from the system or a background source. This work presents results from the 1RX field’s targeted observations at 150, 325, and 610 MHz with Giant Meterwave Radio Telescope (GMRT). These observations have a higher angular resolution as compared to TGSS and NVSS. I detected the radio source near the position of 1RX at all frequencies with GMRT. I further used the Very Large Array Sky Survey (VLASS) data at 3 GHz to determine the flux density and position at high frequency. With the targeted GMRT observations and observations from VLASS, I show that the radio emission does not originate from the 1RX b but is from a background source about ∼13 arcsec away from the host star. Further, no radio emission was detected from the position of 1RX.

Stars: individual: 1RXS1609.1−210524, planetary systems, planet–star interactions, radio continuum: planetary systems

1 INTRODUCTION

Planets in the Solar system that possess a strong magnetic field also have radio emissions associated with them (e.g. Burke & Franklin 1955; Bigg 1964; Zarka 2000). The primary mechanism for radio emission from planets in the Solar system is the electron cyclotron maser instability (ECMI) process. In this mechanism, energetic keV electrons gyrate and accelerate in the planets’ magnetic field and produce radio emission (Melrose & Dulk 1982; Dulk 1985). Similar to planets in the Solar system, exoplanets with strong magnetic fields are also thought to emit at radio wavelengths via ECMI (e.g. Zarka et al. 1997, 2001). The radio emission from exoplanets is comparable to or greater than the emission from their host stars (e.g. Zarka et al. 2001). This makes the observation of stars at radio wavelength an excellent tool to detect new exoplanets (Fennelly & Matloff 1974; Yantis, Sullivan & Erickson 1977; Grießmeier et al. 2005; Lazio 2017). Thus, extensive and deep all-sky radio surveys, such as TIFR GMRT Sky Survey (TGSS) (Intema et al. 2017) at 150 MHz, NRAO VLA Sky Survey (NVSS) (Condon et al. 1998) at 1.4 GHz, GaLactic, Extragalactic All-sky Murchison Widefield Array survey (Hurley-Walker et al. 2017) at 200 MHz, LOFAR Two-metre Sky Survey (LoTSS) (Shimwell et al. 2017) at 120–168 MHz, and Karl G. Jansky Very Large Array Sky Survey (VLASS) (Lacy et al. 2020) at 3 GHz (2–4 GHz), can be used to not only detect radio emission from known exoplanetary systems but also search for new exoplanet candidates.

There have been a few attempts at using large all-sky surveys to detect radio emission from exoplanets and brown dwarfs (Sirothia et al. 2014; Murphy et al. 2015; Lynch et al. 2018; Vedantham et al. 2020a,b). Sirothia et al. (2014) using TGSS at 150 MHz detected radio emission near the position of seven exoplanets. One of the systems that Sirothia et al. (2014) examined was 1RXS1609.1−210524 (hereafter 1RX). Sirothia et al. (2014) found a strong radio source 8 arcsec away from the position of 1RX. The NVSS map also reveals the presence of a strong radio source at 1.4 GHz (6.72 arcsec away from the position of 1RX b) with a flux density of 5 ± 0.5 mJy. However, given the poor angular resolution of TGSS and NVSS, it was not possible to ascertain the nature and origin of the radio emission.

Recently, using data from the LoTSS, there have been several detections of radio emission from M dwarf stars, some of which have characteristics similar to the radio emission expected due to star–planet interaction (SPI; Vedantham et al. 2020a; Callingham et al. 2021). Furthermore, using the LoTSS (with the LOFAR telescope), Vedantham et al. (2020b) reported the discovery of radio emission from a cold methane dwarf BDR J1750+3809. The magnetic field of the brown dwarf was estimated to be ≥25 G, which is comparable to the magnetic fields of giant planets. These discoveries highlight the possibility of using a low-frequency all-sky radio survey to detect brown dwarfs and exoplanets.

Since Vedantham et al. (2020a,b) and Callingham et al. (2021) were able to detect radio emission originating from brown dwarfs and M dwarfs (perhaps from SPI), using an all-sky low-frequency survey, I was motivated to examine the validity of the detection of the radio source near 1RX b from Sirothia et al. (2014). I used archival Giant Meterwave Radio Telescope (GMRT)¹ observations and observations from VLASS² at 2–4 GHz to study the nature of the radio source detected near the position of 1RX.

Most observations on detecting radio emission from exoplanets have focused on close-in hot Jupiter (e.g. Bastian, Dulk & Leblanc 2000; Lazio et al. 2004; Lecavelier Des Etangs et al. 2009; Smith et al. 2009; Lecavelier Des Etangs et al. 2011; Narang et al. 2021a,b; Turner et al. 2021). Observations of super-Earths and Neptunes have resulted in a few tentative detections of possible radio emission due to SPI. Lecavelier des Etangs et al. (2013) detected tentative signals of 3σ emission from HAT-P 11b whose light curve is consistent with an eclipse when the planet passed behind the star. The radio observation of Proxima Centauri by Pérez-Torres et al. (2021) has shown the emission to be variable and likely modulated at the orbital phase of the planet.

Recently, Cendes, Williams & Berger (2022) have observed five directly imaged planetary systems with VLA to search for radio emission from them. These directly imaged planets along with the 1RX sample are more similar to ultra-cool dwarfs (UCDs) than planets. Since radio emissions from several UCDs have previously been detected (e.g. Berger et al. 2001; Berger 2002; Hallinan et al. 2006; Zic et al. 2019; Vedantham et al. 2020b), it is possible that the radio emission detected from 1RX was due to the planet/brown dwarf.

This paper presents the results from the targeted observations of the 1RX field with GMRT at 150, 325, and 610 MHz. I also examine the 1RX field at high frequency, 3GHz (2–4 GHz), using the data from VLASS. In Section 2, I describe the target 1RX b. The observations are described in Section 3, while in Section 4, I present the results from the GMRT and VLASS (at 3 GHz) observations. Further, in Section 4, I argue that the radio source detected near 1RX is of extragalactic origin. In Section 5, I discuss the findings and summarize the results in Section 6.

2 TARGET

1RX (α = |$\rm 16^h09^m30.31^s$| and δ − 21°04′58.94″) is a pre-main-sequence star with a spectral type of K7 (Lafrenière, Jayawardhana & van Kerkwijk 2008) located in the Upper Scorpius OB Association of the Scorpius-Centaurus Association (Preibisch et al. 1998) at a distance of 137.37 pc (Gaia Collaboration 2021; Bailer-Jones et al. 2021). The star is estimated to be 11 Myr old (Pecaut, Mamajek & Bubar 2012). 1RX is also host to a |$14_{-3}^{+2} M_J$| planet/brown dwarf (Pecaut et al. 2012), which was detected by direct imaging (Lafrenière et al. 2008; Lafrenière, Jayawardhana & van Kerkwijk 2010). The planet 1RX b is at a projected distance ∼306 au away from the host star (Lafrenière et al. 2008, 2010).

Observation with the Spitzer Space Telescope shows infrared excess at 24 μm, indicating the presence of a debris disc (Bailey et al. 2013). Wu et al. (2015) observed the system using the MagAO on the 6.5-m Clay Telescope and derived a T_eff of 2000 ± 100 K (Spectral type L2) for 1RX b. Wu et al. (2015) also calculated an extinction A_V of 4.5 mag for 1RX b, hinting at the presence of a disc around 1RX. Wu et al. (2015), however, did not find any signature of accretion indicating that either 1RX b is accreting very slowly or 1RX b is not accreting at all.

3 OBSERVATIONS AND DATA REDUCTION

To determine the nature of emission from 1RX b, I first re-examined the radio maps of the 1RX field. The radio source at the position of 1RX b was detected with TGSS alternative data release (ADR1) by Intema et al. (2017) at 150 MHz with an integrated flux density of 26.1 ± 7.5 mJy (a ∼3.5σ detection) (Fig. 1a). The radio source was also detected with the NRAO VLA Sky Survey at 1.4 GHz (Condon et al. 1998) with an integrated flux density of 5.3 ± 0.5 mJy (a ∼11σ detection) (Fig. 1b). At 150 MHz, the radio source is 13.6 arcsec away from 1RX, and at 1.4 GHz, the radio source is 6.8 arcsec from 1RX. Since the spatial resolution of TGSS (32 arcsec at 150 MHz) and NVSS (45 arcsec at 1.4 GHz) is poor, we cannot be certain about the origin of the radio emission. Hence, the 1RX field was re-observed with GMRT at multiple frequencies to better understand the radio source and ascertain whether the radio emission is from the planet/host star or a background object. The proposal id for these observations is |$23\_059$|⁠. One of the key goals of this proposal was to observe and characterize the radio emission detached close to 1RX in the TGSS survey. Two other targets, HATP-11 and 55 Cnc, were also observed as part of this proposal. The 55 Cnc observations were published as part of (Narang et al. 2021a).

Figure 1.

(a) The TGSS image contours at 150 MHz and (b) NVSS image contours at 1.4 GHz overlaid on the 2MASS J-band (1.2 μm) image of the 1RX field. The contour levels are 3σ at 150 MHz and 3σ and 5σ at 1.4 GHz. The position of the host star is marked as the red cross. On the bottom left corner, the beam size is shown as a blue ellipse.

The 1RX field was observed at 150 MHz for ∼4 h on 2013 February 23rd. 3C 286 and 3C 48 were used as the flux calibrators, and J1714−252 was used as the band-pass and phase calibrator. The flux calibrators were observed twice, once at the beginning (3C 286 for 15 min) and once at the end of observations (3C 48 for 25 min). The target 1RX and band-pass and phase calibrator were observed in a loop with 52 min on the science target and 6 min on the band-pass and phase calibrator.

At 325 MHz, the 1RX system was observed on 2013 February 15th for ∼2.5 h. 3C 286 was used as the flux calibrator and was observed twice, once at the beginning and once at the end of observations for 15 min each time. J1714−252 was used as the band-pass, and the phase calibrator was observed for 6 min in a loop with the science target similar to the previous observations.

On 2013 February 24th, 1RX was observed at 610 MHz for 3.5 h. 3C 286 and 3C 48 were used as the flux calibrators, and J1714−252 was used as the band-pass and phase calibrator. The flux calibrators were observed at the beginning (3C 286 for 16 min) and at the end of observations (3C 48 for 18 min). The science observations and the phase and band-pass were again observed in a loop.

I reduced the observations using the source peeling and atmospheric modelling (SPAM) pipeline (Intema 2014a,b). SPAM is a set of AIPS scripts in python that can reduce and image the legacy GMRT data at 150, 325, and 610 MHz. SPAM includes direction-dependent (ionospheric) calibration and imaging and RFI and bad data mitigation methods.

4 RESULTS

The radio source near the position of 1RX (radio source from hereafter) is detected in all the three targeted observations (see Fig. 2). At 150 MHz, the integrated flux density from the radio source is 27.6 ± 3.4 mJy (Fig. 2a). The flux density from TGSS at 150 MHz was 26.1 ± 7.5 mJy. At 325 MHz, the integrated flux density from the source is 34.23 ± 0.55 mJy. While at 610 MHz, the integrated flux density is 19.28 ± 0.14 mJy. The radio source is also detected at 3 GHz with the Karl G. Jansky VLASS (see Fig. 3). At 3 GHz, the integrated flux density from the source is 2.9 ± 0.5 mJy. No radio emissions were detected from the location of 1RX.

Figure 2.

The GMRT observations of the 1RX field at (a) 150 MHz, (b) 325 MHz, and (c) 610 MHz. The contours (black solid lines) are overlaid on the 2MASS J-band (1.6 μm) image of the 1RX field. At 150 MHz, the contours are 3σ, 5σ, and 10σ, while the contours at 325 and 610 MHz are 3σ, 5σ, 10σ, 30σ, 100σ, and 200σ. The position of the host star is marked as the red cross. On the bottom left corner, the beam size is shown as a blue ellipse.

Figure 3.

The 3 GHz VLASS image of the 1RX region. The 3σ, 5σ, and 10σ contours (black solid lines) are overlaid on the 2MASS J-band (1.6 μm) image of the 1RX field. The position of the host star is marked as the red cross. On the bottom left corner, the beam size is shown as a blue ellipse.

These observations (from GMRT and VLA) have a higher sensitivity and better resolution than the TGSS and NVSS observations (see Table 1). Hence, they can separate out the emission from a background radio source and emission from the 1RX system. At various wavebands, the radio source has an average positional offset of 13 arcsec (see Table 1) away from the position of 1RX from Gaia EDR3 (Gaia Collaboration 2021). The separation between the star and the planet is ∼2.2 arcsec (Lafrenière et al. 2008, 2010). The maximum beam size at 610 MHz is ∼7.4 arcsec, and the maximum beam size at 3 GHz is ∼2.8 arcsec. The radio source is at least ∼2–4 beams away from the Gaia position of the host star and planet. This means that the radio source is not associated with 1RX or 1RX b.

Table 1.

The summary of the targeted observation of 1RX b with GMRT along with the 3 Hz observations with VLASS. For each of the frequencies, I have calculated the integrated flux density from the object, the peak flux density, the rms flux at the position of 1RX, the beam size [minimum (B_min) and maximum (B_max)] and the offset between the centre of the radio source and the position of 1RX (Offset).

Frequency	Date	Integrated flux density (mJy)	Peak flux density (mJy beam−1)	rms	Bmin (arcsec )	Bmax (arcsec)	Offset (arcsec)
150 MHz	2013 Feb 23th	27.6 ± 3.4	25.9 ± 1.8	2 mJy	15	28	12.56
325 MHz	2013 Feb 15th	34.2 ± 0.55	33.9 ± 0.3	150 μ Jy	6.4	13.3	12.40
610 MHz	2013 Feb 24th	19.3 ± 0.14	19.8 ± 0.08	70 μ Jy	3.5	7.4	13.98
3 GHz	2019 June 29th	2.9 ± 0.5	1.9 ± 0.2	160 μ Jy	1.7	2.8	13.4

In Fig. 4, the SED of the radio source is shown. I have included the integrated flux densities from the targeted observations with GMRT (150, 325, and 610 MHz), the NVSS detection at 1.4 GHz, and the VLASS detection at 3 GHz. From the SED (Fig. 4) of the radio source, I find that the emission rises between 150 and 610 MHz, with the peak emission being somewhere between 325 and 610 MHz. After 610 MHz, the emission from the source falls. I calculate the power-law index α (S_ν ∼ ν^−α), which is +1.3 between 150 and 325 MHz. While α is ∼−0.56 between 610 MHz and 3 GHz. From the analysis of the SED, I conclude that the radio source is a gigahertz-peaked spectrum or GPS source. These sources are characterized by a convex radio spectrum with steep spectral indices at high frequencies and spectral turnovers near 1 GHz (Gopal-Krishna, Patnaik & Steppe 1983; O’Dea, Baum & Stanghellini 1991; O’Dea 1998). I further searched for an optical counterpart for the radio source using the Six-degree Field Galaxy Survey (6dFGS), (Jones et al. 2004, 2009), UKIRT Infrared Deep Sky Survey (UKIDSS)-DR9 (Lawrence et al. 2007), and Pan-STARRS (Chambers et al. 2016). No optical/NIR source was detected within 3 arcsec of the radio source.

Figure 4.

The spectral energy distribution of the radio source detected near 1RX.

5 DISCUSSION

There could be several reasons why no emission was detected from the position of 1RX. Although I have some of the deepest observations for an exoplanet field (see Griessmeier 2017; Lazio 2017; Narang et al. 2021b and Turner et al. 2021), no emission was detected from the position of 1RX at any of the frequencies. One of the characteristics of ECMI is that the radio emission peaks at the maximum cyclotron frequency. If the maximum cyclotron frequency were higher or lower than the frequency range I examined, the observations would have missed the emission. The maximum cyclotron frequency ν_c in terms of the magnetic field of the planet (B_P) is

$$\begin{eqnarray*} \nu _\mathrm{c}~=~2.8\,{\rm MHz}\,\,[B_{\mathrm{P}}/\mathrm{G}]. \end{eqnarray*}$$

(1)

Further, the maximum cyclotron frequency ν_c can be expressed in terms of the planet mass M_P, rotation rate ω_P, and radius R_P, as (Narang et al. 2021b)

$$\begin{eqnarray*} \nu _\mathrm{c}~=~2.8\,{\rm MHz}\,\,[B_{\mathrm{P}}/\mathrm{G}]~=~ 23.5\,{\rm MHz} \, \bigg (\frac{\omega _\mathrm{P}}{\omega _\mathrm{J}}\bigg) \bigg (\frac{M_\mathrm{P}}{M_\mathrm{J}}\bigg) \, \bigg (\frac{R_\mathrm{J}}{R_\mathrm{P}}\bigg). \, \end{eqnarray*}$$

(2)

The maximum cyclotron frequency using equation (2) and assuming that ω_P = ω_J and R_P = R_J is 329 MHz. A more sophisticated method of calculating the planet’s magnetic field and hence the cyclotron frequency is given by Reiners & Christensen (2010). Using results from Reiners & Christensen (2010), I estimate the magnetic field for 1RX b to be ∼1 kG. This gives the cyclotron frequency ν_c as 2800 MHz.

Based on the two estimates of the magnetic fields of 1RX, the emission peaks either in band 3 of GMRT (325 MHz) or in the VLASS survey (2–4 GHz). However, no emission was detected at these frequencies. I put a 3σ upper limit of 0.45 mJy at 325 MHz and 0.48 mJy at 3 GHz (assuming a steady-state component to the emission beamed towards the Earth). However, if the emission is not beamed towards the Earth or if the emission is highly variable, in such cases, the emission will not be detected.

Furthermore, if the emission from 1RX is inherently weak, the emission would not be detected. From observations of planets in the Solar system, it has been established that the radio power from the planets is proportional to the input power supplied to the electrons (e.g. Zarka et al. 1997). Most of the planets that have been observed at radio wavelengths are close-in hot Jupiters. Since these planets are so close to the host star, the input power to the electrons due to the interaction of the planet and stellar wind would be large. Due to this, the expected flux density from these planets can be of the order of a few mJy. 1RX b, on the other hand, is at 306 au away from the host star. Due to this, the input power to the electrons due to the host star’s stellar wind would be negligible (order of 10⁻⁷ smaller than that of hot Jupiters). Due to this, the expected flux from 1RX b would be undetectable with the current generation of radio telescopes. However, if there was a large local source of keV electrons, perhaps from the accretion disc around the planet or due to the presence of a satellite, there could be detectable radio emission from the planet (Nichols 2011, 2012).

6 SUMMARY

This paper analysed the 1RX field at radio wavelengths, covering the frequency range between 150 MHz and 3 GHz. I show that the radio source detected near 1RX from Sirothia et al. (2014) is a background source ∼13 arcsec from the position of 1RX. Even though I have carried out one of the deepest observations to detect radio emission from an exoplanet, no emission was detected from the 1RX system. There could be several reasons why I did not detect radio emission from the 1RX b. However, the primary reason why no emission was detected from 1RX b is that the planet is 306 au away from the host star. Due to the considerable distance between the host star and planet, the standard models of SPI will not produce detectable radio emission.

ACKNOWLEDGEMENTS

I thank the referee for their insightful comments and suggestions that have resulted in a significant improvement of the manuscript. This work is based on observations made with the Giant Metrewave Radio Telescope, operated by the National Centre for Radio Astrophysics of the Tata Institute of Fundamental Research and located at Khodad, Maharashtra, India. I thank the GMRT staff for the efficient support of these observations. This research has used the NASA Exoplanet Archive, operated by the California Institute of Technology, under contract with the National Aeronautics and Space Administration under the Exoplanet Exploration Program. This research has also used NASA’s Astrophysics Data System Abstract Service and the SIMBAD data base, operated at CDS, Strasbourg, France. I acknowledge the support of the Department of Atomic Energy, Government of India, under project identification no. RTI4002.

DATA AVAILABILITY

The data presented in this article are available on the GMRT and VLASS archives. The GMRT data can be accessed from https://naps.ncra.tifr.res.in/goa/, with proposal id |$23\_059$|⁠.

Footnotes

1

https://naps.ncra.tifr.res.in/goa/data/search

2

https://archive-new.nrao.edu/vlass/