A SETI Survey of the Vela Region using the Murchison Widefield Array: Orders of Magnitude Expansion in Search Space
PPublications of the Astronomical Society of Australia (PASA)doi: 10.1017/pas.2020.xxx.
A SETI Survey of the Vela Region using the MurchisonWidefield Array: Orders of Magnitude Expansion inSearch Space
Tremblay, C.D. & Tingay, S.J. CSIRO Astronomy and Space Science, PO Box 1130, Bentley WA 6102, Australia International Centre for Radio Astronomy Research, Curtin University, Bentley, WA 6102, Australia
Abstract
Following the results of our previous low frequency searches for extraterrestrial intelligence (SETI) usingthe Murchison Widefield Array (MWA), directed toward the Galactic Centre and the Orion MolecularCloud (Galactic Anticentre), we report a new large-scale survey toward the Vela region with the lowestupper limits thus far obtained with the MWA. Using the MWA in the frequency range 98-128 MHzover a 17 hour period, a 400 deg field centered on the Vela Supernova Remnant was observed witha frequency resolution of 10 kHz. Within this field there are six known exoplanets. At the positionsof these exoplanets, we searched for narrow band signals consistent with radio transmissions fromintelligent civilizations. No unknown signals were found with a 5 σ detection threshold. In total, acrossthis work plus our two previous surveys, we have now examined 75 known exoplanets at low frequencies.In addition to the known exoplanets, we have included in our analysis the calculation of the EffectiveIsotropic Radiated Power (EIRP) upper limits toward over 10 million stellar sources in the Vela fieldwith known distances from Gaia (assuming a 10 kHz transmission bandwidth). Using the methods ofWright et al. (2018) to describe an eight dimensional parameter space for SETI searches, our surveyachieves the largest search fraction yet, two orders of magnitude higher than the previous highest (ourMWA Galctic Anticentre survey), reaching a search fraction of ∼ × − . We also compare our resultsto previous SETI programs in the context of the EIRP min - Transmitter Rate plane. Our results clearlycontinue to demonstrate that SETI has a long way to go. But, encouragingly, the MWA SETI surveysalso demonstrate that large-scale SETI surveys, in particular for telescopes with a large field-of-view,can be performed commensally with observations designed primarily for astrophysical purposes. Keywords: planets and satellites: detection – radio lines: planetary systems – instrumentation: interferom-eters – techniques: spectroscopic
In this paper, we continue to report on our programto utilise the Murchison Widefield Array (MWA: Tin-gay et al. (2013); Wayth et al. (2018)) in a Search forExtraterrestrial Intelligence (SETI) at low radio frequen-cies, over extremely wide fields of view.In previous work, we have examined two survey fields,encompassing 400 deg toward the Galactic Centre in thefrequency range 103 - 133 MHz (Tingay et al., 2016) and625 deg toward the Galactic Anticentre direction in thefrequency range of 99 −
122 MHz (Tingay et al., 2018).In these two survey fields, 45 and 22 exoplanets wereknown at the times of observation, respectively, and nocandidate signals were detected above the observationaldetection limits, which were approximately 4 × W and 1 × W for the closest exoplanets in thefields, respectively (assuming isotropic transmitters anda 10 kHz transmission bandwidth to calculate EffectiveIsotropic Radiated Power: EIRP).A general improvement in our data processing tech-niques between the two sets of observations, and thefact that on average the known exoplanets toward theGalactic Anticentre are closer than those known towardthe Galactic Centre, means that our upper limits on theEIRP for exoplanets toward the Galactic Anticentre arelower, in general. Tingay et al. (2018) placed our resultsto that point in the context of the overall SETI endeavorand we refer the reader to that discussion and referencestherein for this context.The MWA provides a unique facility to search for1 a r X i v : . [ a s t r o - ph . I M ] S e p Tremblay & Tingay technosignatures at low radio frequencies, being highlysensitive, located at the radio-quiet Murchison Radio-astronomy Observatory (MRO), and having a very widefield-of-view (the surveyed areas noted above representsingle MWA pointings). The Galactic Centre field surveywe previously reported was placed in the context of pastSETI surveys by Gray & Mooley (2017), who show thatthe limits we achieved are highly competitive. In theiranalysis of “How Much SETI Has Been Done? FindingNeedles in the n-dimensional Cosmic Haystack”, Wrightet al. (2018) examine an eight dimensional parameterspace for radio SETI and find our two previous surveysto have the highest searched fractions for this parameterspace for single surveys, factors of approximately two andten greater than the next highest, respectively. However,the highest search fractions still sit at an order of 10 − ,indicating that only a vanishingly small fraction of theSETI parameter space has been covered thus far.While this conclusion may appear discouraging, causefor encouragement comes from the fact that SETI sur-veys can increasingly be performed effectively as com-mensal science in parallel with primary astrophysicalinvestigations. This has been our approach using theMWA, whereby we utilise data collected and processedin wide field searches for low frequency spectral lines(e.g. Tremblay et al. 2018). The F AST collaborationintends to complete commensal and dedicated SETI ex-periments, using a real-time data processing pipelineoriginally developed for the
SET I @ Home platform tosearch for technosignatures from 1-1.5 GHz during nor-mal science operations (Zhang et al., 2020; Li et al.,2020). Similar ideas exist for commensal searches beingplanned with MeerKAT (Gajjar et al., 2019).We continue this approach with the MWA here, addinga survey field centred on the Vela Supernova Remnant.In §2 we describe the observations and data processing,including increments in the quality of the data processingthat lead to almost an order of magnitude improvementin our flux density sensitivity with commensurate im-provements in our EIRP detection limits (for a fixeddistance). In §3 we describe our results, examining thesix known exoplanets in the Vela field as well as thefull population of stellar systems in the field (millions ofsystems). In §4 we discuss our results and conclusions.
The Murchison Widefield Array (MWA; Tingay et al.2013) is a low-frequency interferometer operating be-tween 70 and 300 MHz at the Murchison Radio-astronomy Observatory in Western Australia. In 2018,the telescope was upgraded to the “Phase II” array(Wayth et al., 2018), doubling the number of aperturearray tiles from 128 to 256 and approximately doublingthe maximum baseline from 3 km to 5.5 km.
Figure 1.
Summary of the data processing pipeline from Figure 2of Tremblay et al. (2017) used to create integrated spectral cubeswith the MWA.
Observations of the Vela region took place between05 January 2018 and 23 January 2018 for a total of30 hours, the details of which are summarised in Table1. These observations were taken during the buildingand commissioning of the Phase II array and included91 of the new 128 tiles. Of the 30 hours of observation,17 hours were free from imaging artifacts likely causeddue to the instrument being actively worked on duringthe day, while the observations were taken at night.The MWA has an instantaneous bandwidth of30.72 MHz that is distributed between 3072 ×
10 kHzfine frequency channels. Our data were processed follow-ing the procedure detailed in Tremblay et al. (2017) andTremblay (2018) but a summary is provided here and inFigure 1. The bandpass and phase solutions were derivedeach night from a two-minute observation of Hydra A (aLINER galaxy with a flux density of 243 Jy at 160 MHz(Kühr et al. 1981)). The solutions were further refinedby using self-calibration before they were applied to each5-minute observation of the Vela region field.For each of the 5-minute observations, the fine fre-quency channels (10 kHz) are imaged at a rate of 100fine channels per every 1.28 MHz coarse channel to avoidchannels affected by aliasing. This means only 78 per centof the band is imaged. The Phase-II configuration of theMWA used in these observations removed the compactcore and had shortest baselines of 1.5 km. In order toobtain as much sensitivity to diffuse emission as possi-ble, all images were created using a Briggs weightingof 0.5. This produced a field-of-view of 400 deg and a ETI with the MWA Figure 2.
MWA spectrum for a data cube, with a total integration time of 17 hours, used within this survey at the position of HD75289 b. Some of the channels in the lower end of the band are affected by narrow-band RFI. The vertical shaded areas mark regions ofknown narrow-band RFI and the green region of the spectrum shows the top end of the FM band. Flagged channels are blanked out inthe spectrum. The horizontal grey shaded region represents the ± σ RMS value used in Table 2. synthesised beam-width of 1 .In previous SETI surveys completed with the MWAtoward the Galactic Centre and the Orion Molecularcloud, only 4 hours and 3 hours, respectively, of ob-servations were obtained. In this survey, 17 hours oftotal integration time is used to provide our deepestlow-frequency survey, producing a mean spectral RMS(root mean squared) of 0.05 Jy beam − across much ofthe field, in comparison to the previous 0.35 Jy beam − RMS.The MWA is situated in an RFI-protected environ-ment but occasional intermittent interference occurs(Offringa et al., 2015; Sokolowski et al., 2017). Each5-minute observation was flagged using
AO F l ag g e r (Offringa et al., 2015) to remove strong radio frequencyinterference (RFI) signals from the raw visibilities basedon statistical methods. This is not expected to impactour science goals, as the chance of a real astronomical orsignal from an ETI being strong enough to be flagged in asingle 5-minute observation is very small. It is estimatedthat this process removes less than 5–20 per cent of thetotal visibilities, having little impact on an observation’ssensitivity. For these observations, after integrating eachof the snapshot images, significant narrow-band RFI wasdetected in the commercial FM radio bands between 98MHz and 108 MHz. This left 64% of the band availablefor narrow band signal searches. These data are comensally searched for spectral linesignals of an astrophysical nature, which will be re-ported in a separate publication (Tremblay et al. ApJsubmitted). An example of a typical spectrum with nosignificant signal, as seen toward HD 75289 b, is shownin Figure 2.The source finding software
A e g e a n (Hancock et al.,2018) is used to search each of the 2400 (10 kHz) finefrequency channels independently for signals over a 5 σ limit. A e g e a n works by fitting Gaussians to the pixeldata and applies a correction for the background tocalculate the flux density for potential sources. Anypotential source is further evaluated based on variousquality control checks, including but not limited to, en-suring the signal is greater than 5 σ in both the spectraland image plane. Any remaining signals are cross refer-enced to a combination of chemical databases and newchemical modeling reported in a future publication. Fol-lowing this search, we found no signals of an unknownnature. Utilising the processed data described in §2, we haveaccess to the spectrum across our bandwidth at 10 kHz The background is defined by the 50 th percentile of fluxdistribution in a zone 30 times the size of the synthesised beam. Tremblay & Tingay
Table 1
MWA Observing ParametersParameter ValueCentral frequency 113.28 MHzTotal bandwidth 30.72 MHzNumber of imaged channels 2400Channel separation 10 kHzSynthesized beam FWHM ∗ Primary beam FWHM ∗
30 degreesPhase center of image (J2000) 08h35m27s –45d12m19sTotal Integration Time 17 hours ∗ Full Width at Half Maximum (FWHM) resolution, for every pixel in our 400 deg field-of-view.Thus, as per our previous work (Tingay et al., 2018,2016), we can examine the spectrum at the locations ofstellar systems with known exoplanets and search for nar-row band signals that may constitute technosignatures.We undertake this examination below in §3.1. Further,given that only a tiny fraction of the exoplanets in thisfield are known, we can also undertake a blind survey ofall stellar systems in the field with known distances from Gaia (Bailer-Jones et al., 2018) and undertake the samesearch. We undertake this examination in §3.2, below.To calculate the upper limit on the total EquivalentIsotropic Radiated Power (EIRP) we use the equation:EIRP(W) < . × S rms R , (1)where S rms is the RMS intensity value in Jy beam − and R is the distance to the stellar system in pc. Thisassumes that the transmission bandwidth is matched tothe MWA fine channel bandwidth of 10 kHz. For trans-mission bandwidths less than 10 kHz, the maximumEIRP estimates are increased by kHz ∆ ν t , where ∆ ν t isthe transmission bandwidth. For example, a 10 Hz trans-mission bandwidth would cause our EIRP upper limitsto be raised by a factor of one thousand. A search of our field-of-view in the Exosolar PlanetsEncyclopedia Catalog (as of March 2020) returns sixexoplanets hosted by five stellar systems. These exoplan-ets are listed in Table 2, including basic parameters ofthe exoplanets, their stellar host, and the radio observa-tions. In no case were any narrow band signals detectedtoward these objects in our observing band at or abovea level of 5 σ . As in our previous work, we assign upperlimits to the EIRP based on the RMS of the measuredspectrum, listed in Table 2. http://exoplanet.eu/catalog Figure 3.
Histogram of EIRP upper limits based on the distribu-tion of stellar distances (set to a maximum of 6350 pc) from the
Gaia catalog.
Given the small number of known exoplanets associatedwith the stellar systems in this field, it is likely that avast number of exoplanets remain unknown. Thus, weexamine the general limits we can derive for stars in thisfield. We do this by examining the
Gaia catalog, extract-ing the distances of stars within the field determined bytheir parallax measurements (Bailer-Jones et al., 2018).There are 10,355,066 such stars within the field-of-viewfor this survey. As no detections were made in the searchdiscussed in §2, we utilise the RMS value as a func-tion of position across our field and the coordinatesand distances of the stars, to derive the EIRP upperlimit histogram for all 10,355,066 stars with a distancesmaller than 6350 pc in Figure 3. As the distribution ofdistances is dominated by stars within the spiral armof the Galaxy at 1.5–2.5 kpc, the EIRP upper limitsare also very large, orders of magnitude larger than thelowest upper limits from our previous work.To examine the most interesting part of this distribu-tion, at low values of EIRP upper limit, we show theEIRP upper limit histogram for those stars within 30and 50 pc in Fig 4. For the ten closest stars, Table 3lists the RMS and derived EIRP limits in more detail(excluding those systems referenced in Table 2).
The median distance for the six known exoplanet systemsin the Vela field is 28.8 pc (treating HD 73526 b andc as a single system), compared to 50 pc for the 22exoplanets examined by Tingay et al. (2018) and ≈ ETI with the MWA Table 2
Known exoplanets in the survey field, from the exoplanet catalog: http://exoplanet.eu/
Designation RA (J2000) Dec (J2000) Distance MSin(i) Period Spectral Detection RMS EIRP d hh:mm:ss dd:mm:ss (pc) (M J a ) (days) type b method c (Jy/beam) (10 W)HD 75289 b 08:47:40.0 − < ± × − HD 73526 b 08:37:16.0 − ± ± < − ± ± < − ±
400 G5 IV-V RV 0.039 < − ± − ± − I 0.044 < − ±
19 0.91 ± − PT 0.052 < ± × − a Mass of planet times the sine of orbit inclination, in Jupiter masses b Spectral type of host star c RV= Radial Velocity; I=Imaging; PT=Primary Transit d Equivalent Isotropic Radiated Power
Table 3
Gaia stellar systems in the survey field, from the Gaia DR2 release
Designation RA (J2000) Dec (J2000) Est. Distance RMS EIRP a hh:mm:ss dd:mm:ss (pc) (Jy/beam) 10 (W)5534076974490020000 08:00:39.6 − ± < − ± < − ± < − ± < − ± < − ± < − ± < − ± < − ± < − ± < a Equivalent Isotropic Radiated Power
Tremblay & Tingay
Figure 4.
Histogram of EIRP upper limits based on the distribu-tion of distances from the
Gaia catalog, with a focus on sourceswith distances <
50 and <
30 pc. median EIRP upper limit from this work is therefore anorder of magnitude better than the median upper limitfrom Tingay et al. (2018). This represents continuedimprovement in our techniques and general upper limits.Our best upper limit, from Table 3 of 6 . × W (notfor a known exoplanet) approaches a 50% improvementon our best upper limit from Tingay et al. (2018), stillnoting that an EIRP of 10 W is high compared to thehighest power transmitters on Earth at these frequencies(see Tingay et al. (2016) for a discussion).Sheikh et al. (2020) recently completed an in-depthanalysis of 20 stars within the Earth transit zone be-tween 3.95–8 GHz with the NRAO Green Bank Tele-scope (GBT). They determined an EIRP of detectablenarrow-band signals that ranged from 47–17590 × Wfor stellar distances between 7–143 pc. These are simi-lar distances to the stars we present in Table 3 for thenearby stars from
Gaia in the Vela field, but our samplerepresents a much larger population of on-average closersources. Sheikh et al. (2020) also convert their EIRPvalues to the fraction of signal capacity for the AriceboTransmitter (L A ). Their value of 0.033 for a star at 27 pcis a factor of two lower than for a star in our survey atthe same distance (0.068), recognising the difference infrequency between the GBT and the MWA and the factthat the GBT has better frequency resolution than theMWA.For the first time, we obtain simultaneous upper limitson EIRP for in excess of 10 million stellar systems with-out known exoplanets, although the EIRP limits for themajority of distant systems are well above 10 W. Forany future exoplanet discoveries for systems in this field,the low frequency EIRP upper limits are immediatelyavailable from our data.Seto & Kashiyama (2020) completed an astrometricstudy of F-,G- and K-type stars in
Gaia
Data Release2 for interstellar communications, from the view point of the sender, and concluded that surveys like
Gaia willbe necessary to target these potential signals. Petiguraet al. (2013) suggest that approximately 20 per cent ofGalactic sun-like stars could have Earth-sized planets intheir habitable zones and Kipping (2020) suggests thatsearching for technosignatures from stars with stellartypes much earlier than our Sun may not be necessary, aslife is unlikely to ever evolve. This means that matchingSETI survey data to the
Gaia survey is going to be animportant approach for the future of SETI.We compare our results to Figure 5 of Price et al.(2020), in which the results of previous surveys are pre-sented in a plane defined by minimum EIRP (EIRP min )at the maximum stellar distance and Transmitter Rate,(N star ( ν c ν tot )) − , where N star is the total number of starssearched and ν c and ν tot are the central frequencyof the band (113.28 MHz) and the total bandwidth(30.72 MHz), respectively. For our survey using the Gaia catalog, we see that our results sit below the most con-straining limits set by prior work within this particularparameter space when using a distance of 1.7 kpc and achannel bandwidth of 10 kHz.We also consider a different metric, utilising themethod of Wright et al. (2018) in order to calculatethe “haystack fraction” accessible to our observationsof the Vela region. This metric takes into account theobservational parameters without significant assump-tions. We find the haystack fraction to be ∼ × − ,which is almost two orders of magnitude higher than thehighest previous fraction listed in Wright et al. (2018),which was for our previous observations of the Orionmolecular cloud (Galactic Anticentre) field. Our newresult yields an haystack fraction almost three ordersof magnitude higher than the largest non-MWA surveylisted by Wright et al. (2018).Recently, Westby & Conselice (2020) described theso-called Strong Astrobiological Copernican scenario, inwhich life must arise in a system on timescales compa-rable to those experienced on Earth (4.5 − ± civilisations ca-pable of generating technosignatures exist in our Galaxy.The closest system to Earth would be 17000 ± lt-yrdistant. The numbers are not large and represent a verysmall part of the haystack fraction parameter space,throwing into focus that SETI experiments will need toenter the statistical domain of Gaia -sized samples.Overall, our MWA surveys show the rapid progressthat can currently be made in SETI at radio frequencies,using wide field and sensitive facilities, but also showthat SETI surveys have a long way to go. The continueduse of the MWA, and the future similar use of the SKAat much higher sensitivities, offers a mechanism to makesignificant cuts into the haystack fraction of Wright et al.(2018), while maintaining a primary focus on astrophys-ical investigations, making excellent commensal use ofthese large-scale facilities.
ETI with the MWA Figure 5.
Figure 5 from Price et al. (2020) with our results for the
Gaia catalog survey (Section 3.2) shown for comparison. We reportthe results using a maximum RMS of 0.06 Jy beam − and when we limit the distances to less than 50 pc, as per Figure 4 and for all starsless than 1.7 kpc. The value of 1.7 kpc is chosen as it is the distance to the Vela Molecular Cloud complex, a stellar rich environmenttoward the Carina-Sagittarius spiral arm. The black ovals are the EIRP values per Equation 1 assuming a 10 kHz channel bandwidthand the red ovals are using the EIRP values assuming a transmission bandwidth of 10 Hz. These results span the diagonal grey linerepresenting a fit between the previous most constraining data points for Transmitter Rate and EIRP min at the most distant star whenusing the MWA channel bandwidth. The solid and dashed vertical lines represent the EIRP of the Arecibo planetary radar, and thetotal power from the Sun incident on the Earth, respectively. We would like to thank Daniel Price (Swinburne Uni-versity) for his insight and comments on this paper andNatasha Hurley-Walker for help with the continuumsubtraction. We would also like to thank the anonymousreviewer for their helpful comments that significantlyimproved our manuscript.
This scientific work makes use of the Murchison Radio-astronomy Observatory, operated by CSIRO. We ac-knowledge the Wajarri Yamatji people as the traditionalowners of the Observatory site. Support for the operationof the MWA is provided by the Australian Government(NCRIS), under a contract to Curtin University adminis-tered by Astronomy Australia Limited. Establishment ofASKAP, the Murchison Radio-astronomy Observatoryand the Pawsey Supercomputing Centre are initiativesof the Australian Government, with support from theGovernment of Western Australia and the Science andIndustry Endowment Fund.
We acknowledge the Pawsey Supercomputing Centrewhich is supported by the Western Australian and Aus- tralian Governments. Access to Pawsey Data StorageServices is governed by a Data Storage and Manage-ment Policy (DSMP). The All-Sky Virtual Observatory(ASVO) has received funding from the Australian Com-monwealth Government through the National eResearchCollaboration Tools and Resources (NeCTAR) Project,the Australian National Data Service (ANDS), and theNational Collaborative Research Infrastructure Strategy.This research has made use of NASAâĂŹs AstrophysicsData System Bibliographic Services.
The following software was used in the creation of thedata cubes: • ao f l ag g e r and c o t t e r – Offringa et al.(2015) • W S C l e a n – Offringa et al. (2014); Offringa &Smirnov (2017) • A e g e a n – Hancock et al. (2018) • m i r i a d – Sault et al. (1995) • T O P C AT – Taylor (2005)
REFERENCES
Bailer-Jones C. A. L., Rybizki J., Fouesneau M., Man-telet G., Andrae R., 2018, AJ, 156, 58
Tremblay & Tingay
Gajjar V., Siemion A., Croft S., Brzycki B., Bur-gay M., Carozzi T., et al., 2019, in BAAS. p. 223( arXiv:1907.05519 )Gray R. H., Mooley K., 2017, AJ, 153, 110Hancock P. J., Trott C. M., Hurley-Walker N., 2018,PASA, 35, e011Kipping D., 2020, in American Astronomical SocietyMeeting Abstracts. p. 454.03Kühr H., Witzel A., Pauliny-Toth I. I. K., Nauber U.,1981, A&AS, 45, 367Li D., Gajjar V., Wang P., Siemion A., ZhangZ., FAST Collaboration 2020, arXiv e-prints, p.arXiv:2003.09639Offringa A. R., Smirnov O., 2017, MNRAS, 471, 301Offringa A. R., McKinley B., Hurley-Walker et al., 2014,MNRAS, 444, 606Offringa A. R., Wayth R. B., Hurley-Walker N., KaplanD. L., Barry N., et al., 2015, PASA, 32, e008Petigura E., Marcy G. W., Howard A., 2013, in Ameri-can Astronomical Society Meeting Abstracts arXiv:astro-ph/0612759arXiv:astro-ph/0612759