Abstract
We present the first very long baseline interferometric (VLBI) observations of the blazar OJ 287 carried out jointly with the Global Millimeter VLBI Array (GMVA) and the phased Atacama Large Millimeter/submillimeter Array (ALMA) at 3.5 mm on 2017 April 2. The participation of phased ALMA has not only improved the GMVA north–south resolution by a factor of ∼3, but has also enabled fringe detections with signal-to-noise ratios up to 300 at baselines longer than 2 G
Export citation and abstract BibTeX RIS
Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
1. Introduction
The BL Lac–type object OJ 287 (z = 0.306; Stickel et al. 1989) is a well-studied low synchrotron peaked BL Lac object that has attracted great interest, as it shows quasiperiodic optical outbursts with a cycle of about 12 yr. These outbursts appear to come in pairs, with separations of one to two years, and they have been suggested to originate due to the presence of a supermassive binary black hole (SMBBH) system at its center (e.g., Sillanpää et al. 1988; Lehto & Valtonen 1996). According to this model, the observed quasiperiodic double-peaked optical outbursts are triggered when the secondary supermassive black hole (SMBH) impacts the accretion disk of the primary in its orbit. Further advances of this model have accounted for general relativistic effects, and the parameters have also been further constrained with follow-up observations (e.g., Valtonen et al. 2008, 2011; Dey et al. 2018). The model requires a compact binary with a major axis of the orbit of 0.112 pc (corresponding to an angular scale of ∼26
Another observational signature of OJ 287 is that the position angle (PA) of the parsec-scale jet was found to be "wobbling" by previous very long baseline interferometric (VLBI) observations (e.g., Tateyama & Kingham 2004; Agudo et al. 2012; Cohen 2017; Britzen et al. 2018). Such changes in the inner jet PA could also be explained by the SMBBH model (e.g., Dey et al. 2021), but alternative models cannot be fully ruled out. For instance, Agudo et al. (2012) suggest instabilities coupled to the accretion disk as the likely origin of the nonperiodic changes in the inner jet orientation. Britzen et al. (2018) suggest that the flux variation could be explained by viewing angle changes and the Doppler beaming effects of a precessing jet. The precession could be driven by either the binary motion (e.g., Dey et al. 2021) or the Lense–Thirring effect due to the misalignment between the BH spin and the accretion disk (e.g., Liska et al. 2018; Chatterjee et al. 2020).
The massive central BH, the relatively low redshift, and the bright relativistic jet close to the line of sight also make OJ 287 one of the nearest high-luminosity active galactic nuclei in which the magnetic launching and acceleration of jets can be studied through high-resolution VLBI observations. Two competing scenarios have been proposed for the formation of relativistic jets. The main difference between them is whether the magnetic fields are twisted by the rotational energy of the BH (Blandford & Znajek 1977) or its accretion disk (Blandford & Payne 1982). It is also possible that both mechanisms are at work (e.g., Chiaberge et al. 2000). In the innermost region of the jet, the plasma flow is accelerated and collimated in the presence of a spiral magnetic field, while the jet expands in width and propagates downstream into the interstellar space. The disruption of the accretion flow and the interaction with the ambient medium often result in the formation of moving and standing shocks. The detailed process of jet formation, acceleration, and collimation remains unclear, as extremely high angular resolution is required to probe into the innermost region in the vicinity of the central BH.
High-resolution VLBI observations are ideal for probing the compact structure near the central engine. Previous VLBI observations of OJ 287 have provided key information on the parsec-scale structure and the dynamics of the jet (e.g., Cohen 2017; Hodgson et al. 2017; Britzen et al. 2018). In particular, Gómez et al. (2022) recently presented 22 GHz images of OJ 287 with unprecedented angular resolution for the source, obtained with RadioAstron space–ground VLBI observations. The images revealed a progressive bending of the inner jet with increasing angular resolution, by comparison with multiband ground-based VLBI images. The inner jet components show high brightness temperatures that exceed the inverse Compton limit, indicating strong Doppler boosting in the jet. The polarized images show electric vector position angles (EVPAs) aligned with the jet axis, which indicates that the jet has a predominantly toroidal magnetic field. Multifrequency analysis shows hints of a rotation measure gradient across the jet, which suggests that the VLBI core is threaded by a helical magnetic field.
VLBI observations at wavelengths shorter than 7 mm have the potential to probe areas closer to the central engine that are optically thick at lower frequencies (see, e.g., Boccardi et al. 2017). Previous VLBI observations at 3.5 mm with the Global Millimeter VLBI Array (GMVA) show the existence of quasi-stationary components and changes in the morphology and PA in the innermost jet region (e.g., Hodgson et al. 2017). However, most of the previous GMVA observations are limited in sensitivity, due to typically shorter atmospheric coherence times, lower antenna efficiencies, and thus higher system-equivalent flux densities compared to longer wavelengths. The participation of large sensitive stations in millimeter-VLBI observations is desirable, together with further developments of the instruments and calibration methods (e.g., Rioja & Dodson 2011; Rioja et al. 2017; Zhao et al. 2018).
In this paper, we present the first VLBI observations of OJ 287 with the GMVA and phased Atacama Large Millimeter/submillimeter Array (ALMA) on 2017 April 2. These observations are accompanied by a multiwavelength campaign, including the first 1.3 mm observation of the source with the Event Horizon Telescope (EHT; Event Horizon Telescope Collaboration et al. 2019a), the results of which will be presented in a forthcoming paper. The campaign was carried out during a major outburst event of OJ 287 in 2016–17, with the largest X-ray outburst recorded so far (Komossa et al. 2017, 2021c) and the first very-high-energy (VHE) flare detection (Mukherjee & VERITAS Collaboration 2017).
We summarize the details of the GMVA + ALMA observations and the methods that we use to calibrate, image, and analyze the data in Section 2. We present our observational results, including total intensity and linear polarization images in Section 3. In Section 4, we discuss the nature of the components in the jet and possible constraints on the theoretical models, followed by a summary in Section 5.
2. Observations and Data Analysis
In this section, we describe the details of our 3.5 mm observations of OJ 287 with GMVA + ALMA, the data calibration procedure, and the methods used to obtain subparsec-scale images of OJ 287.
2.1. Observations
We carried out high-resolution VLBI observations toward OJ 287 at 3.5 mm with GMVA on 2017 April 2. These observations mark the first VLBI observations with the phased ALMA, which consists of 37 ALMA antennas and is equivalent to a 70 meter dish (Event Horizon Telescope Collaboration et al. 2019a). The participating stations also include eight Very Long Baseline Array (VLBA) stations and five European stations (Effelsberg, IRAM-30 m, Metsähovi, Onsala, Yebes-40m). The on-source time was around 375 minutes between UT 17 and UT 7 the next day (April 3).
Most stations had good or typical weather conditions during the observation, except for the VLBA Maunakea and Pie Town stations, which resulted in few fringe detections with limited signal-to-noise ratio (S/N) on the baselines to these two stations. No fringes were found on the baselines to Metsähovi, due to a faulty backend setup. All data were recorded in full polarization mode, with most stations recording on a circular polarization basis, while the ALMA data were converted from a mixed linear–circular basis to circular polarization mode using PolConvert (Martí-Vidal et al. 2016). The Yebes-40 m telescope recorded only left-handed circular polarization (LCP). The bandwidths and frequency ranges recorded are not the same for all stations. 36 Only the common frequency ranges among all participating stations are used in later processing.
2.2. Data Reduction
Data correlation was performed with the DiFX correlator (Deller et al. 2007) at the Max-Planck-Institut für Radioastronomie in Bonn, Germany. The final correlated data have a total bandwidth of 232 MHz, which was further divided into four 58 MHz intermediate frequency (IF) bands.
The postcorrelation data set was then processed with the ParselTongue (Kettenis et al. 2006) (Greisen 2003) interface for fringe fitting and a priori amplitude calibration. We first performed parallactic angle correction with the task clcor 37 , and manual phase calibration using short segments of data to remove the instrumental phase offset between the different IFs. We then performed a global fringe fitting of the data using the task fring, with a solution interval of 10 s and subintervals down to 2 s, and by integrating over the whole 232 MHz bandwidth and averaging parallel-hand polarizations (RR and LL).
The (u,v)-coverage toward OJ 287 for all baselines with fringe detections is shown in Figure 1. We note that the participation of ALMA has provided an increase in the north–south resolution by a factor of ∼3 for observations of OJ 287. ALMA has also significantly improved fringe detection due to its high sensitivity (see Figure 2) with the maximum fringe S/N reaching ∼350 at baselines longer than 1.5 G
Download figure:
Standard image High-resolution imageA priori amplitude calibration was performed in with the task apcal, by multiplying the system temperatures (Tsys) and gain curves of each antenna. Opacity corrections were applied to stations that measure system temperatures with the noise diode method (VLBA and Effelsberg). For ALMA, IRAM-30 m, and Yebes-40 m, the Tsys measurements were performed using the hot/cold method, and therefore already included opacity correction. The ALMA Tsys values have also taken into account the phasing efficiencies derived during the quality assurance and PolConvert processes (e.g., Goddi et al. 2019). The cross-hand phase and delay offsets of the reference station were calibrated using the procedure vlbacpol.
After the calibration, the data were averaged in time (with an interval of 15 s) and frequency (with all channels within each IF averaged) for further processing.
2.3. Imaging and Model Fitting
We performed the imaging and self-calibration of the data independently with three different imaging softwares: DIFMAP, eht-imaging, and SMILI. DIFMAP is the software commonly used for the conventional CLEAN method for interferometric imaging (Shepherd et al. 1995). It interactively establishes a collection of point-source models from the inverse Fourier transform of the visibilities, i.e., the dirty map. CLEAN windows, which define the regions to search for CLEAN components, are used during our imaging process. Phase-only self-calibration is performed after each step of cleaning. Amplitude and phase self-calibration is performed once a good fit to the visibilities has been established through multiple steps of cleaning and phase self-calibration. We repeat the clean and self-calibration loops several times during our imaging process by gradually decreasing the solution interval of the amplitude and phase self-calibration. On the other hand, the regularized maximum likelihood (RML) methods, employed by the eht-imaging (Chael et al. 2016, 2018) and SMILI (Akiyama et al. 2017) libraries, reconstruct images by minimizing an objective function, which is a weighted combination of
After the imaging of the total intensity, we estimated the instrumental polarimetric leakage (known as D-terms) for each station, using the self-calibrated data of OJ 287. This process was carried out independently with two pipelines: the task lpcal and the eht-imaging library, each based on a particular self-calibrated data set generated during the total intensity imaging process, i.e., DIFMAP and eht-imaging, respectively. Both approaches provide consistent values of D-terms. Details of the leakage calibration are described in the Appendix. The polarization imaging of the lpcal processed data was carried out with DIFMAP. With eht-imaging, the imaging was performed iteratively with the D-terms calculation. The calibration of the absolute orientation of the EVPAs was performed through comparison with the ALMA array data (Goddi et al. 2021).
We also carried out nonimaging analysis of the data to measure the properties of the jet. We performed circular Gaussian model fitting to the SMILI self-calibrated visibility data with DIFMAP. The results indicated that the jet structure could be represented by four Gaussian components. We labeled the components following the convention described in Gómez et al. (2022). The total flux, size, and position offset with respect to the core (the component at the southeastern end of the jet; see Section 3 below) of all the components are listed in Table 1. The uncertainties of the fitted parameters are derived following the equations outlined in Nair et al. (2019).
Table 1. Model Fitting Parameters of OJ 287 with GMVA + ALMA on 2017 April 2
Comp | S | r | PA | FWHM | |
---|---|---|---|---|---|
Name | (Jy) | ( | ({°}) | ( | (1010 K) |
(1) | (2) | (3) | (4) | (5) | (6) |
C0 | 1.25 ± 0.13 | 0.0 | 0.0 | 31.1 ± 2.4 | 21.2 ± 4.1 |
C1 | 0.96 ± 0.10 | 63.9 ± 2.4 | −41.5 ± 2.2 | 28.9 ± 2.4 | 18.8 ± 3.9 |
C2a | 2.49 ± 0.25 | 172.9 ± 2.4 | −40.0 ± 0.8 | 44.4 ± 2.5 | 20.6 ± 3.1 |
C2b | 0.51 ± 0.05 | 212.7 ± 3.5 | −47.7 ± 0.9 | 42.5 ± 4.6 | 4.7 ± 1.2 |
Note. Columns from left to right: (1) Component ID; (2) Flux density; (3) Radial distance from the core component (C0); (4) Position angle; (5) Component FWHM; and (6) Observed brightness temperature.
Download table as: ASCIITypeset image
3. Results
3.1. Jet Morphology
Figure 3 shows the total intensity maps of OJ 287 obtained with our GMVA + ALMA observations, achieving the highest angular resolution to date of the source at the wavelength of 3.5 mm. The imaging results are consistent across different imaging methods (CLEAN and RML). Under the nominal resolution, the jet appears to consist of three major features, extending along the southeast to northwest direction. We denote the three features as the components C0, C1, and C2, as shown in the bottom right panel of Figure 3.
Download figure:
Standard image High-resolution imageComponent C0, which lies at the southern end of the jet, is compact and shows the highest brightness temperature (Table 1). This feature is more likely to be the VLBI core at 3.5 mm. The component C2 has the highest flux density among the three components. This feature shows complex substructures under the fine resolution of the RML images (Figure 3, top middle and top right panels). We see hints of the jet bending and extending toward the western direction downstream of C2. This bend is more obvious in the lower-frequency maps which are more sensitive to the extended lower-brightness regions despite the lower angular resolutions (e.g., Cohen 2017; Jorstad et al. 2017). The downstream jet is largely resolved out and not well constrained in our high-resolution images, because of their steep spectra and extended structure. Our higher-resolution images reveal for the first time the twisted morphology of the innermost, ultracompact jet region. The first bending occurs between C0 and C1, with the jet axis gradually changing from north to northwest (clockwise). We also see hints of a subsequent bending occurring downstream of C1, where the jet axis turns in the counterclockwise direction.
The three-component structure is also consistent with the recent 22 GHz RadioAstron space–ground VLBI observations of OJ 287 made at a similar resolution (Gómez et al. 2022). However, a PA difference of ∼50° in the inner jet can be found when comparing with the RadioAstron image obtained in 2014. Such a difference could be attributed to the variation in the PA over ∼ 3 yr. A detailed analysis of the inner jet PA variation on a yearly scale, and a comparison with theoretical predictions, will be presented in a forthcoming paper (G-Y. Zhao et al., in preparation).
In order to quantify the PA evolution along the jet, we fit the jet ridgeline on the eht-imaging map. First, we transform the image to polar coordinates centered on the jet origin, then slice it transversely. For each slice, we store the flux density peak position, then transform them back to Cartesian coordinates. Thus, we obtain a collection of positions tracing the jet axis between C0 and C2. The results are presented in Figure 4, where we also show a sketch tracing the conical structure of C2 in the figure. The jet axis near C0 extends along a PA of ∼−15°, decreases to ∼−50° at C1, and then starts to increase again near C2. A similar trend can also be found in the SMILI image.
Download figure:
Standard image High-resolution image3.2. Brightness Temperatures
We investigate the brightness temperature of the OJ 287 jet using two independent approaches. (1) We calculate the observed brightness temperature of each Gaussian component from the model fitting results, using the following equation (e.g., Tingay et al. 2002):
where S is the component flux density in Jy,
Download figure:
Standard image High-resolution imageWe estimate the intrinsic brightness temperature, , by (e.g., Gómez et al. 2016):
where
3.3. Polarization
We perform polarimetric imaging of the instrumental polarization calibrated data with CLEAN and eht-imaging independently. The corresponding images are shown in Figure 6. Our images show that the overall degree of polarization of OJ 287 is ∼8%, which is in quantitative agreement with the ALMA array results of 8.8% presented in Goddi et al. (2021). The EVPAs extend mostly along the mean jet axis, which suggests that the magnetic field in the jet has a predominant toroidal component. Again, the image reconstructed by eht-imaging shows a fine structure, because of the super-resolution that is naturally achieved by the forward-modeling method. However, even in the CLEAN image, which is convolved with the nominal beam, we see a remarkable polarimetric structure in the inner jet. The overall structure is consistent between the two images that have been reconstructed independently with different approaches. We notice that the apparent difference in the fractional polarization between the two maps is due to the fact that the CLEAN images are convolved with the nominal beam. The overall degree of polarization and the EVPA distributions agree well between the two images.
Download figure:
Standard image High-resolution imageOf the several jet components, C0 shows the lowest fractional polarization of ∼5%, as measured from the eht-imaging map. This further supports this component being the jet core, which is usually depolarized (e.g., Lister & Homan 2005). C1 exhibits a high level of polarization (∼16%), which indicates that the magnetic field is more ordered in this region. C2 shows a conspicuous polarimetric structure, which can be further divided into two subcomponents, with the EVPAs lying perpendicular to each other. The EVPAs in the upper subcomponent also lie perpendicular to the direction along which the brightness extends, while in the bottom subcomponent, they lie nearly in parallel. The degrees of polarization are ∼7% and ∼13% in the upper and lower subcomponents, respectively. These substructures are clearly seen in both the eht-imaging and CLEAN images.
4. Discussion
4.1. Nature of the C2 Region
Our GMVA + ALMA observations have revealed a remarkable structure of the inner jet of OJ 287 because of the improved (u,v)-coverage and high sensitivity. In particular, the C2 component shows a complex conical structure in both total and linearly polarized intensity, and a bimodal distribution in the EVPAs. Previous multi-epoch observations have shown that this component is nearly stationary (e.g., Hodgson et al. 2017; Jorstad et al. 2017; Lico et al. 2022). In the following, we discuss the possible nature of this component.
Oblique shocks could result from the jet striking a cloud of interstellar media. Under the precessing jet model, this would naturally occur for some period, as the jet sweeps through the ambient material. Since the location of C2 coincides with where the jet bends, the northeastern section of C2 could be interpreted as an oblique shock on one side of the jet. The oblique shock is in a plane making a small angle to the jet boundary on the northeast side. The flow is then bent by the shock toward the west. The magnetic field could become compressed to strengthen the component nearly parallel to the jet. Therefore, the EVPAs on the northeast side are roughly perpendicular to the jet. The southwestern section of C2 could then just be the main jet after the bend, with the magnetic field transverse to the jet direction at that point, as usual for a BL Lac object.
Conical shock waves can be formed when there is a pressure imbalance between the jet plasma and the ambient medium. The properties of shocks in relativistic jets have been explored by numerical and semidynamical simulations. Gomez et al. (1995) carried out relativistic hydrodynamics simulations of a parsec-scale jet surrounded by an ambient medium with constant or decreasing pressure. The simulations confirmed the existence of stationary components associated with recollimation shocks. Gómez et al. (1997) simulated the interactions of standing shocks and relativistically moving perturbations propagating down the stable jet, and found that the shock could enhance the emission of the moving feature and the stationary component could be temporarily "dragged" downstream. Further simulations of the interactions between recollimation shocks and traveling shocks are presented in Fromm et al. (2016), based on the observations presented in Fromm et al. (2011, 2013a, 2013b, 2015), for the particular case of CTA 102. Various configurations of the upstream magnetic field components have also been included in subsequent numerical simulations (e.g., Broderick & McKinney 2010; Porth et al. 2011; Fuentes et al. 2018). In particular, Mizuno et al. (2015) studied kinematically dominated jets with different magnetic field configurations, including axial, toroidal, and helical, based on a relativistic magnetohydrodynamics (RMHD) simulation code. Fuentes et al. (2018) characterized the properties of recollimation shocks in RMHD simulations of jets at the parsec scale as a function of the dominant type of energy: internal, kinetic, or magnetic. By solving the radiative transfer equations for synchrothron radiation, using these simulations as inputs, they analyzed the total intensity and linear polarization signatures imprinted in the stationary components associated with these shocks. Fuentes et al. (2021) extended the analysis to RMHD jet models threaded by helical magnetic fields with larger magnetic pitch angles, and also explored the effect of different nonthermal particle populations on the polarimetric properties of stationary features and the overall observed synchrotron radiation.
On the other hand, Cawthorne & Cobb (1990) established a semidynamical model, assuming only that the shock front is emitting, and found that conical shock waves could result in polarization angles that were either parallel or perpendicular to the jet axis. This model also only considered random magnetic fields in the upstream jet. In Cawthorne (2006), a poloidal magnetic field component was added to the model, and the results can well explain the observed polarization of the knot K1 in 3C 380. Furthermore, Cawthorne et al. (2013) extended this model to include a paired collimating and decollimating shock, and the predicted EVPAs could successfully describe the observational results of the BL Lac object 1803 + 784.
Comparing our observational results of C2 with the numerical and semidynamical studies, we find that the conical shape of the emitting region is quite consistent between our observations and the simulations. Numerical simulations predict a series of stationary shocks along the jet that can be triggered by a pressure imbalance between the jet and the external medium. The reason that we find only one conical-shaped component is most likely the adiabatic expansion of the jet. As also shown in Gomez et al. (1995), with decreasing pressure downstream of the jet, the intensity of the stationary components gradually decreases and the separation between the components increases, so the downstream shocks may be too faint and become undetectable at our observing frequency. Regarding the polarized emission, the semidynamical simulations show different EVPA distributions across the cone. However, the EVPA pattern is more symmetric with respect to the cone axis. The numerical simulations also show that the EVPA pattern will depend on the upstream magnetic field configuration and the viewing angle (e.g., Mizuno et al. 2015; Gómez et al. 2016; Fuentes et al. 2021). Fuentes et al. (2021) pointed out that jets with a large magnetic pitch angle, i.e., threaded by a helical magnetic field dominated by its toroidal component, can exhibit a bimodal EVPA distribution around recollimation shocks for small viewing angles. This EVPA configuration could imply a sign flip of the Stokes parameter that leads to an EVPA flip, which then results in a dip in the linearly polarized emission, as we observe in the C2 component from the reconstructed polarimetric images.
Alternative to the standing shock scenario, the observed properties of the C2 component could be a result of geometric effects due to the bending of the jet axis toward the line of sight. With a decreasing viewing angle, the enhanced Doppler boosting could amplify the emission in this region and make C2 the brightest component in the inner jet. If the viewing angle becomes smaller than the jet opening angle, the bimodal distribution of the EVPAs could be produced by the existence of helical magnetic fields in the jet, as the direction of the projected magnetic field is different across the component (Fuentes et al. 2021). This scenario is supported by previous observations that revealed the existence of bending around C2 (e.g., Hodgson et al. 2017; Jorstad et al. 2017; Gómez et al. 2022). However, it is difficult to explain the conical shape of the emission region with this assumption.
Moreover, by means of multi-epoch GMVA observations, Lico et al. (2022) identified a new jet feature in the region of C2, in a quasi-concurrent GMVA observing epoch. The authors argued that the passage of this new jet component through the stationary feature at 0.1 mas core separation (i.e., C1) triggered the high-energy outburst during 2016–2017 (Komossa et al. 2017, 2021a), including the faint VHE flare detected in February 2017 (Mukherjee & VERITAS Collaboration 2017)
38
and then moved down to the C2 jet region at the time of these observations. In this scenario, the C2 component in our observations could correspond to the blending of the new feature and the standing shock. The observed bimodal distribution of the EVPAs could be due to the different polarimetric properties of the two components. A similar case was found in the core region of PKS 1510-089 during a
4.2. Testing the SMBBH Model
OJ 287 is one of the most promising candidates for harboring an SMBBH system at the center. In fact, OJ 287 is among the candidates for hosting a nano-Hz gravitational wave–emitting SMBBH system (Valtonen et al. 2021). The binary model has been successful in explaining the periodic light curves and predicting upcoming impact flares, which have been confirmed by observations within a few hours (e.g., Laine et al. 2020). The direction of the jet axis was also found to vary with time, and this could also be related to the orbital motion of the BHs (Dey et al. 2021). Models that do not require a secondary BH to explain the observed variability have also been proposed. For instance, the flux variation could be explained by viewing angle changes and the Doppler beaming effects of a precessing jet. The precession could be driven by the Lense–Thirring effect, due to misalignment between the BH spin and the accretion disk (e.g., Britzen et al. 2018; Liska et al. 2018; Chatterjee et al. 2020; Liska et al. 2021). It would also be possible for MHD instabilities (current-driven or Kelvin–Helmholtz) to produce a helical distorted jet structure (e.g., Mizuno et al. 2012; Perucho et al. 2012; Vega-García et al. 2019).
Dey et al. (2021) established a model to explain the parsec-scale jet direction variations at different frequencies, in which the jet precession is powered by the SMBBH with parameters constrained by optical observations. This model predicts that the 86 GHz jet axis should be ∼−37° around 2017 April, assuming a disk model. The PAs of the inner jet components (e.g., C1 and C2a) measured in our GMVA + ALMA observations agree well with this prediction (see Table 1). However, we note that this agreement is partially due to the observing epoch being not far apart from the 86 GHz GMVA data used to constrain the model. Furthermore, this agreement will not rule out other possible scenarios. For example, the tilted accretion could also result in the precession of the inner jet. Britzen et al. (2018) argue that the PA change observed at 15 GHz can be modeled by a jet precession combined with a nutation of the axis. The precession could be a result of Lense–Thirring effects, and a secondary BH is not always required. Furthermore, our RML images also revealed a twisted pattern of the innermost jet, which resembles a precessing jet in projection.
Future kinematic studies with multi-epoch GMVA and EHT observations will hopefully provide further insights to distinguish among different theoretical models regarding the underlying nature of the source.
Dey et al. (2021) also explored the possibility of the existence of a jet from the secondary SMBH based on the SMBBH model. With the high sensitivity and improved north–south resolution resulting from the participation of ALMA, we found no evidence for a secondary jet, even in the eht-imaging and SMILI images with super-resolution. There could be several possible reasons for such a nondetection. First, the jet is likely to be short-lived, as commented upon in Dey et al. (2021). Since the projected separation of the two SMBHs in 2017 April is ∼10
We further note that the GMVA + ALMA observations presented in this work are part of a multiwavelength observing campaign of OJ 287. Close-in-time observations with the EHT at 230 GHz (on 2017 April 4 and 9) and with the RadioAstron space-VLBI mission at 22 GHz (on 2017 March 7) could provide even higher angular resolutions and probe slightly different regions of the inner jet. Together with the observations at X-ray and optical bands (e.g., Komossa et al. 2017, 2020, 2021a, 2021b, 2021c), we will be able to test or obtain constraints on the physical parameters of the possible jet associated with the secondary SMBH.
5. Summary
We carried out GMVA + ALMA observations of OJ 287 on 2017 April 2, which are the first VLBI observations with the phased ALMA. The improved north–south resolution and array sensitivity, together with the newly developed RML methods, have enabled us to obtain high-fidelity super-resolved images of the OJ 287 jet with unprecedentedly high angular resolution. The convolved RML images also agree with the CLEAN reconstruction. The images have revealed a twisted structure in the innermost region of the jet. Our result suggests that the C0 component lying at the southeastern end of the jet is more likely the VLBI core, as it is bright, compact, and relatively depolarized. The C2 component located at ∼200
The work at the IAA-CSIC is supported in part by the Spanish Ministerio de Economía y Competitividad (grants AYA2016-80889-P and PID2019-108995GB-C21), the Consejería de Economía, Conocimiento, Empresas y Universidad of the Junta de Andalucía (grant P18-FR-1769), the Consejo Superior de Investigaciones Científicas (grant 2019AEP112), and the State Agency for Research of the Spanish MCIU through the "Center of Excellence Severo Ochoa" award to the Instituto de Astrofísica de Andalucía (SEV-2017-0709). This publication acknowledges the project M2FINDERS, which is funded by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No. 101018682). L.L. acknowledges the support of the DGAPA/PAPIIT grants IN112417 and IN112820, the CONACyT-AEM grant 275201, and the CONACyT-CF grant 263356. T.S. was supported by the Academy of Finland projects 274477, 284495, 312496, and 315721. Y.K. was supported in the framework of the State project "Science" by the Ministry of Science and Higher Education of the Russian Federation, under the contract 075-15-2020-778. J.Y.K acknowledges support from the National Research Foundation of Korea (grant no. 2022R1C1C1005255). R.-S. L is supported by the Max Planck Partner Group of the MPG and the CAS, the Key Program of the National Natural Science Foundation of China (grant No. 11933007), the Key Research Program of Frontier Sciences, CAS (grant No. ZDBS-LY-SLH011), and the Shanghai Pilot Program for Basic Research – Chinese Academy of Science, Shanghai Branch (JCYJ-SHFY-2022-013).
This paper makes use of the following ALMA data: ADS/JAO.ALMA#2016.1.01116.V. ALMA is a partnership of ESO (representing its member states), NSF (USA), and NINS (Japan), together with NRC (Canada), MOST and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO, and NAOJ. The ALMA data required nonstandard processing by the VLBI QA2 team (C. Goddi, I. Martí-Vidal, G. B. Crew, H. Rottmann, and H. Messias). This work is partially based on observations with the 100-m telescope of the MPIfR (Max-Planck-Institut für Radioastronomie) at Effelsberg. This research has made use of data obtained with the Global Millimeter VLBI Array (GMVA), which consists of telescopes operated by the MPIfR, IRAM, Onsala, Metsähovi, Yebes, the Korean VLBI Network, the Greenland Telescope, the Green Bank Observatory (GBT), and the Very Long Baseline Array (VLBA). The VLBA and the GBT are facilities of the National Science Foundation, operated under a cooperative agreement by Associated Universities, Inc. The data were correlated at the correlator of the MPIfR in Bonn, Germany.
Facilities: ALMA - Atacama Large Millimeter Array, GMVA - , VLBA - , GBT - , EVN - , Effelsberg - , Metsähovi Radio Observatory - , Onsala Space Observatory - , IRAM-30 m RT - , Yebes-40 m RT. -
Software: (Greisen 2003), DiFX (Deller et al. 2007), DIFMAP (Shepherd et al. 1995), eht-imaging (Chael et al. 2018), ParselTongue (Kettenis et al. 2006), SMILI (Akiyama et al. 2017), Altair (VanderPlas et al. 2018), Vega-Lite (Satyanarayan et al. 2017).
Appendix: Calibration of the Instrumental Polarization
Calibration of the instrumental polarization leakage (also known as the D-terms) is required to obtain reliable polarimetric maps of the target. Each of the two pipelines that we used to perform polarimetric imaging (see Section 2) has independently implemented this calibration step.
The lpcal pipeline loads the self-calibrated visibility data and the CLEAN Stokes I image of OJ 287 produced by DIFMAP and runs the task lpcal to solve for the D-terms. lpcal assumes that the source can be divided into a few subcomponents, each with a constant fractional polarization. lpcal solves the D-terms for each IF independently; the results are shown in the top left panel of Figure 7. We have flagged the stations that only have data for one circular polarization (Yebes-40m) and stations that show low S/N on cross-hands (RL and LR) polarization data (VLBA North Liberty and Onsala).
Download figure:
Standard image High-resolution imageOn the other hand, the eht-imaging pipeline performs the instrumental polarization calibration in parallel with the imaging of the polarimetric data products. The pipeline computes the leakage terms by minimizing the difference between the self-calibrated data and the sampled data from the corrupted reconstructions. For details of the polarimetric imaging with eht-imaging, refer to Chael et al. (2016). The eht-imaging software by default averages the data at different IFs, so we have flagged the stations that show large differences in D-terms across IFs (VLBA BR and OV) in our polarimetric analysis. The eht-imaging results are shown in the top right panel of Figure 7.
Despite the different approaches to solving instrumental leakages, the two pipelines provide very consistent results in terms of the D-terms estimation, as shown in the bottom panels of Figure 7, which validate our polarization calibration. The absolute calibration of the EVPA was obtained by comparison with the ALMA observations of OJ 287 at the same frequency performed during the same observation campaign (Goddi et al. 2021).
Footnotes
- 36
The recorded bandwidth for each station is as follows: ALMA 32 × 62.5MHz, VLBA 2 × 128MHz, and most European stations 1 × 512MHz.
- 37
We note that the mount types for IRAM-30 m (Nasmyth-Left) and Yebes-40 m (Nasmyth-Right) are different from the rest of the antennas in the array (altitude–azimuth). The Yebes-40 m data were not used for polarimetric analysis, as they were only recorded in LCP.
- 38
In fact, the high X-ray flux of OJ 287 detected during the Swift MOMO program triggered the VHE observations, which led to the first VHE detection.