GN-z11-flash was a signal from a man-made satellite not a gamma-ray burst at redshift 11
Micha? J. Micha?owski, Krzysztof Kami?ski, Monika K. Kami?ska, Edwin Wnuk
GGN-z11-flash was a signal from a man-made satellite nota gamma-ray burst at redshift 11
Michał J. Michałowski , Krzysztof Kami´nski , Monika K. Kami´nska & Edwin Wnuk Astronomical Observatory Institute, Faculty of Physics, Adam Mickiewicz University,ul. Słoneczna 36, 60-286 Pozna ´n, Poland, [email protected] , [email protected] Long gamma-ray bursts (GRB), explosions of very massive stars, provide crucial informa-tion on stellar and galaxy evolution, even at redshifts z ∼ − . , when the Universe wasonly 500–600 million years old . Recently, during observations of a galaxy at a redshiftof z ∼ (400 million years after the Big Bang)
5, 6 , a bright signal, named GN-z11-flash,shorter than 245 s was detected and interpreted as an ultraviolet flash associated with a GRBin this galaxy , or a shock-breakout in a Population III supernova . Its resulting luminositywould be consistent with that of other GRBs , but a discussion based on probability argu-ments started on whether this is instead a signal from a man-made satellite or a Solar Systemobject . Here we show a conclusive association of GN-z11-flash with Breeze-M upperstage of a Russian Proton rocket on a highly elliptical orbit. This rules out GN-z11-flash asthe most distant GRB ever detected. It also implies that monitoring of a larger sample ofvery high redshift galaxies is needed to detect such distant GRBs. This also highlights theimportance of a complete database of Earth satellites and debris, which can allow properinterpretation of astronomical observations. a r X i v : . [ a s t r o - ph . H E ] F e b e searched Space-Track, the largest publicly available database of Earth satellites and spacedebris ( ) for an object close to the position of GN-z11-flash at thetime of observations. We found the Breeze-M space debris (North American Aerospace DefenseCommand (NORAD) number 40386; Committee on Space Research (COSPAR) number 2015-005C), with a trajectory consistent with the position of GN-z11 at the time of the flash. It has ahighly elliptical orbit with the semi-major axis of 13,946 km, the perigee distance to the Earth’ssurface of 394.8 km and the apogee distance of 14,757.2 km. Its orbit is inclined by 50.7 ◦ to theEarth’s equator and the orbital period is 273.2 minutes.Using orbital elements from the database and our Satellite Trail Predictor software package(see Methods), we found that the closest angular separation between the satellite and the positionof GN-z11 was (18 ± (cid:48)(cid:48) , consistent with zero (the offset from the edge of the slit used for theflash detection was (7 ± (cid:48)(cid:48) ). This was at the UT time of 2017-04-07 08:07:50.80, within theintegration of the frame with the flash detection, which started at 08:07:19.86 and lasted 179 sec .Breeze-M debris was moving at a position angle of 104.24 ◦ with the angular velocity of 35.9 (cid:48)(cid:48) persec. At that time its distance from the Earth’s surface was 13 758 km, whereas the distance to theobserver in Hawai‘i was 15 186 km. The satellite was also outside the Earth’s shadow.Ref. 7 presented several arguments against GN-z11-flash being a signal from a satellite. Wedemonstrate that none of these concerns applies to Breeze-M debris, because it falls in the narrowregions of the parameter space not ruled out by the authors. First, ref. 7 estimated that a high-Earth orbit satellite would need to have an orbit inclined by more than ◦ to the Earth’s equator.2ndeed, the inclination of Breeze-M debris is 50.7 ◦ . Ref. 7 stated that “the compact spectral shapealready suggests that the trajectory of any hypothetical moving object must be nearly perpendicularto the slit, or else the spectrum would be broader.” This is indeed the case for Breeze-M debris,as the angle between the slit and the satellite’s trajectory is 118.74 ◦ , so an extended spectrum isnot expected above the measured width in the spatial direction, which is 20–25% broader than thepoint spread function (Extended Data Fig. 1 of ref. 7).Second, the estimated V -band magnitude of GN-z11-flash of 19.2 mag was used to implythe actual visual magnitude of . mag . However, this calculation assumed an angular velocityof 20–30 (cid:48) per sec with respect to the observer, corresponding to a low-Earth orbit. This is 33–50times faster than the actual velocity of Breeze-M debris, so the satellite should be much fainterthan 5.5 mag.Indeed, our archival observations carried out with the Roman Baranowski Telescope/Pozna ´nSpectroscopic Telescope 2 (RBT/PST2; see Fig. 1 and Methods for the description) revealed thebrightness of Breeze-M debris with an infrared cut-off filter of ∼ . mag at the distance of ∼ , km from the observer (the distance to the observer during the flash detection). Taking intoaccount the phase angle change, the estimated V magnitude of Breeze-M debris— at the time of theGN-z11-flash observation is ∼ . mag. The detection of GN-z11-flash was made with the Multi-Object Spectrometer For Infra-Red Exploration (MOSFIRE) mounted at the Keck telescope,with the slit width of . (cid:48)(cid:48) and the position angle of 345.5 ◦ . The angular speed of Breeze-M debrisimplies that it was visible in the slit for around . sec. Each MOSIFRE frame was integrated3or 179 sec, so this means that the satellite appeared a factor of / . ∼ times (or ∼ . mag) fainter than it really was. This implies the expected magnitude of 19.2 mag, closeto the one measured from the MOSFIRE data. Given the uncertainties in the Breeze-M debrisbrightness resulting from the irregular shape, rotation, albedo and changes of phase angle, thebrightness recorded by MOSFIRE is consistent with our measurements for this satellite.Third, the MOSFIRE observations were conducted simultaneously for 21 slits, so a movingobject should produce a similar flash in one or more of other slits “unless its trajectory was nearlyperpendicular to the slit ( ∼ ◦ ± ◦ ) ” . Indeed, the trajectory of Breeze-M debris was inclinedby 118.74 ◦ with respect to the slit, just outside the interval quoted by ref. 7. Hence, the fact that theflash was not visible through other slits was just because Breeze-M debris was unlucky to miss allof them. The time gap between frames was 33 sec, during which the satellite moved 23 (cid:48) , i.e. morethan the field of view of MOSFIRE (6 (cid:48) ). Hence, the satellite was not visible in the previous or nextframe.In order to demonstrate that Breeze-M debris missed other slits, on Fig. 2 we show its trajec-tory together with the positions of the MOSFIRE slits obtained from the Keck Observatory Archive(KOA; Program ID: S324, PI: Kashikawa, frame name: MF.20170407.29239). We show the bestorbit with the , , and σ confidence intervals (see Methods). In order to show systematic uncer-tainties we also show trajectories resulting from the orbital elements available for epochs betweenone week before and after the MOSFIRE observations. Trajectories crossing the slit correspond-ing to the GN-z11-flash position and missing all other slits are consistent with the best orbit at a4evel below σ . Moreover, if this satellite was not associated with GN-z11-flash, then it wouldmost probably be observed at another slit at this time, as there are very few trajectories within theconfidence interval that do not cross any slit.Hence, we conclude that GN-z11-flash was the signal from Breeze-M debris, which had aposition consistent with GN-z11 at the time of observations and all other properties consistent withobservational results. Hence, we confirm the hypothesis of ref. 10 and ref. 12 that this was a man-made signal. This also highlights the importance of a complete database of Earth satellites anddebris, which allows the interpretation of astronomical observations. MethodsOrbit analysis software.
The trajectory of the satellite was calculated using our Satellite TrailPredictor software package, which is based on the Simplified General Perturbation (SGP) orbitalmodels . This software has recently been developed by K.K. and M.K.K. specifically for thepurpose of predicting the position, orientation and brightness of satellite trails in astronomicalimages and will soon be released as a publicly available on-line service ( ).In order to verify these orbit calculations we also used the following publicly availablesoftware packages: JPL Horizons
15, 16 ( ssd.jpl.nasa.gov/horizons.cgi ), OREKIT ( ), and SkyField ( rhodesmill.org/skyfield/ ). The resulting orbitsagree with our calculations within 0.1 km (1.5 (cid:48)(cid:48) ).5 BT data. (cid:48)(cid:48) with a standard deviation of 30 (cid:48)(cid:48) . We adopt the latter value as the positional uncertainty.This uncertainty is consistent with that typically found for orbital models. Typical errors ofthe SGP orbital models are 1 km and the orbital data errors perpendicular to the orbit are 1.1 kmfor highly elliptical orbit satellites . This corresponds to a total error of the trajectory of 1.5 km,or 21 (cid:48)(cid:48) at the distance of 15 186 km. Supplementary Information cknowledgements M.J.M. acknowledges the support of the National Science Centre, Poland throughthe SONATA BIS grant 2018/30/E/ST9/00208. We acknowledge the use of the Space-Track database,JPL Horizons, OREKIT, and SkyField packages. This research has made use of the Keck ObservatoryArchive (KOA), which is operated by the W. M. Keck Observatory and the NASA Exoplanet Science Insti-tute (NExScI), under contract with the National Aeronautics and Space Administration.
Author Contributions
M.J.M. designed this project and led the manuscript writing, K.K. and M.K.K. searchedfor the satellite, calculated its properties, planned and analysed the RBT observations, made the figures, andcontributed to the manuscript writing. E.W. also calculated the trajectory of the satellite.
Author information [email protected] , or K.K. (email: [email protected] ). ig. 1. Brightness of Breeze-M debris as a function of distance from the observer from ourarchival observations with the RBT/PST2 telescope with an infrared cut-off filter (see Meth-ods). At the distance of around 15,000 km, corresponding to the distance for the MOSIFREobservations, the brightness was ∼ . mag, consistent with the MOSFIRE measurementtaking into account phase angle differences and the angular speed of the satellite. ig. 2. The trajectory of Breeze-M debris in the field of view of MOSFIRE/Keck during theGN-z11-flash detection. The trajectory for the orbit corresponding to the date of the obser-vation is shown as a thick black line with , , and σ confidence intervals shown as shadedregions. The trajectories for orbit elements calculated for days up to one week before and fter the MOSIFRE observations are shown as green lines. The blue rectangles correspondto the positions and sizes of the slits used during the MOSFIRE observations. The position ofGN-z11 is shown as the black cross. Within σ of the best orbital model there are solutionscrossing the slit corresponding to GN-z11-flash and missing all other slits.
1. Salvaterra, R. et al.
GRB090423 at a redshift of z˜8.1.
Nature , 1258–1260 (2009)..2. Tanvir, N. R. et al. A γ -ray burst at a redshift of z˜8.2. Nature , 1254–1257 (2009)..3. Cucchiara, A. et al.
A Photometric Redshift of z ˜ 9.4 for GRB 090429B.
ApJ , 7 (2011)..4. Jiang, L. et al.
Evidence for GN-z11 as a luminous galaxy at redshift 10.957.
Nature Astron-omy (2020). https://doi.org/10.1038/s41550-020-01275-y ,.5. Oesch, P. A. et al.
The Most Luminous z ˜9-10 Galaxy Candidates Yet Found: The LuminosityFunction, Cosmic Star-formation Rate, and the First Mass Density Estimate at 500 Myr.
ApJ , 108 (2014)..6. Oesch, P. A. et al.
A Remarkably Luminous Galaxy at z=11.1 Measured with Hubble SpaceTelescope Grism Spectroscopy.
ApJ , 129 (2016)..7. Jiang, L. et al.
A possible bright ultraviolet flash from a galaxy at redshift z ≈ NatureAstronomy (2020). https://doi.org/10.1038/s41550-020-01266-z ,.8. Padmanabhan, H. & Loeb, A. GN-z11-flash: A shock-breakout in a Population III supernovaat Cosmic Dawn? arXiv e-prints arXiv:2101.12222 (2021)..10. Kann, D. A., Blazek, M., de Ugarte Postigo, A. & Th¨one, C. C. GN-z11-flash in the Contextof Gamma-Ray Burst Afterglows.
Research Notes of the American Astronomical Society ,247 (2020)..10. Steinhardt, C. L. et al. A more probable explanation for a continuum flash in the direction ofa redshift ≈
11 galaxy. arXiv e-prints arXiv:2101.12738 (2021)..11. Jiang, L. et al. “A more probable explanation” is still impossible to explain GN-z11-flash: inresponse to Steinhardt et al. (arXiv:2101.12738). arXiv e-prints arXiv:2102.01239 (2021)..12. Nir, G., Ofek, E. O. & Gal-Yam, A. The GN-z11-Flash Event Can be a Satellite Glint. arXive-prints arXiv:2102.04466 (2021)..13. McLean, I. S., Steidel, C. C., Matthews, K., Epps, H. & Adkins, S. M. MOSFIRE: a multi-object near-infrared spectrograph and imager for the Keck Observatory. In McLean, I. S. &Casali, M. M. (eds.)
Ground-based and Airborne Instrumentation for Astronomy II , vol. 7014of
Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series , 70142Z(2008).14. Vallado, D., Crawford, P., Hujsak, R. & Kelso, T. Revisiting spacetrack report
AIAA/AAS Astrodynamics Specialist Conference and Exhibit (American Institute of Aeronau-tics and Astronautics, 2006). URL https://doi.org/10.2514/6.2006-6753 .15. Giorgini, J. D. et al.
JPL’s On-Line Solar System Data Service. In
AAS/Division for Plane-tary Sciences Meeting Abstracts , vol. 28 of
AAS/Division for Planetary Sciences MeetingAbstracts , 25.04 (1996). 116. Giorgini, J. D., Chodas, P. W. & Yeomans, D. K. Orbit Uncertainty and Close-ApproachAnalysis Capabilities of the Horizons On-Line Ephemeris System. In
AAS/Division for Plan-etary Sciences Meeting Abstracts , vol. 33 of
AAS/Division for Planetary Sciences MeetingAbstracts , 58.13 (2001).17. Maisonobe, L., Pommier, V. & Parraud, P. Orekit: an open-source library for operational flightdynamics applications. In
Proceedings of the 4th International Conference of AstrodynamicsTools and Techniques (2010).18. Rhodes, B. Skyfield: High precision research-grade positions for planets and Earth satellitesgenerator (2019)..19. Flohrer, T., Krag, H., Klinkrad, H., Bastida Virgili, B. & Fruh, C. Improving ESA’s Colli-sion Risk Estimates by an Assessment of the TLE Orbit Errors of the US SSN Catalouge.In Lacoste, H. (ed.)
Fifth European Conference on Space Debris , vol. 672 of