Quote from: eeergo on 03/06/2020 10:50 am- on the one hand, it misses some of the most critical aspects of these endeavors: it ignores from the get-go the effect of satellites in deployment/decommissioning/maintenance low orbits. This is fine if you want to address the steady-state ideal system, but it will never be like that in reality. It's like modeling traffic in a city with cars never stopping anywhere - it's useful for some basic understanding, but you'll never approach a realistic representation like that. In fact, it is reasonable to believe at least 5-10% of satellites will be in that low-orbit, low-drag situation where they are extremely visible and illuminated by barely less time than in the operational config, at any given time. Crudely extrapolating total/visible satellites for the steady case, this would mean between 1000-3000 satellites in total, of which 50-150 would be *very* visible in the sky at once, in the best case of homogeneous dispersion (obviously many more if they're still close together after deployment). Multiply that by 2-4 for the case of full-scale constellations.I'm not sure why satellites in deployment/decommissioning phase matters if it's just 5-10% of the total, that's a rounding error in this kind of ballpark estimates.
- on the one hand, it misses some of the most critical aspects of these endeavors: it ignores from the get-go the effect of satellites in deployment/decommissioning/maintenance low orbits. This is fine if you want to address the steady-state ideal system, but it will never be like that in reality. It's like modeling traffic in a city with cars never stopping anywhere - it's useful for some basic understanding, but you'll never approach a realistic representation like that. In fact, it is reasonable to believe at least 5-10% of satellites will be in that low-orbit, low-drag situation where they are extremely visible and illuminated by barely less time than in the operational config, at any given time. Crudely extrapolating total/visible satellites for the steady case, this would mean between 1000-3000 satellites in total, of which 50-150 would be *very* visible in the sky at once, in the best case of homogeneous dispersion (obviously many more if they're still close together after deployment). Multiply that by 2-4 for the case of full-scale constellations.
Quote from: su27k on 03/07/2020 03:13 amQuote from: eeergo on 03/06/2020 10:50 am- on the one hand, it misses some of the most critical aspects of these endeavors: it ignores from the get-go the effect of satellites in deployment/decommissioning/maintenance low orbits. This is fine if you want to address the steady-state ideal system, but it will never be like that in reality. It's like modeling traffic in a city with cars never stopping anywhere - it's useful for some basic understanding, but you'll never approach a realistic representation like that. In fact, it is reasonable to believe at least 5-10% of satellites will be in that low-orbit, low-drag situation where they are extremely visible and illuminated by barely less time than in the operational config, at any given time. Crudely extrapolating total/visible satellites for the steady case, this would mean between 1000-3000 satellites in total, of which 50-150 would be *very* visible in the sky at once, in the best case of homogeneous dispersion (obviously many more if they're still close together after deployment). Multiply that by 2-4 for the case of full-scale constellations.I'm not sure why satellites in deployment/decommissioning phase matters if it's just 5-10% of the total, that's a rounding error in this kind of ballpark estimates.It's spelled out in my conversation with envy, but just as a summary:The impacts found by the study assume 1600-1100 *operational* satellites visible at once in the sky, but less than 300 would be brighter than mag 6. On the other hand, the numbers just calculated for satellites in low orbits and out of low-drag (lower visibility) would mean 100-150 high visibility satellites (mag 1-3, remember it's logarithmic). That's my point, even if the number of satellites undergoing non-low-drag operations is a correction to the total (and it's probably more towards 10% than 5%, as our calculations show), their effect wrt visible nuisance is very much as important.As NaN points out above me, it appears there's some level of mitigation that can be put in place based on DarkSat's results. That mitigation doesn't apply to non-low-drag sats, as abundantly shown by DarkSat too.
Quote from: eeergo on 03/07/2020 09:38 amQuote from: su27k on 03/07/2020 03:13 amQuote from: eeergo on 03/06/2020 10:50 am- on the one hand, it misses some of the most critical aspects of these endeavors: it ignores from the get-go the effect of satellites in deployment/decommissioning/maintenance low orbits. This is fine if you want to address the steady-state ideal system, but it will never be like that in reality. It's like modeling traffic in a city with cars never stopping anywhere - it's useful for some basic understanding, but you'll never approach a realistic representation like that. In fact, it is reasonable to believe at least 5-10% of satellites will be in that low-orbit, low-drag situation where they are extremely visible and illuminated by barely less time than in the operational config, at any given time. Crudely extrapolating total/visible satellites for the steady case, this would mean between 1000-3000 satellites in total, of which 50-150 would be *very* visible in the sky at once, in the best case of homogeneous dispersion (obviously many more if they're still close together after deployment). Multiply that by 2-4 for the case of full-scale constellations.I'm not sure why satellites in deployment/decommissioning phase matters if it's just 5-10% of the total, that's a rounding error in this kind of ballpark estimates.It's spelled out in my conversation with envy, but just as a summary:The impacts found by the study assume 1600-1100 *operational* satellites visible at once in the sky, but less than 300 would be brighter than mag 6. On the other hand, the numbers just calculated for satellites in low orbits and out of low-drag (lower visibility) would mean 100-150 high visibility satellites (mag 1-3, remember it's logarithmic). That's my point, even if the number of satellites undergoing non-low-drag operations is a correction to the total (and it's probably more towards 10% than 5%, as our calculations show), their effect wrt visible nuisance is very much as important.As NaN points out above me, it appears there's some level of mitigation that can be put in place based on DarkSat's results. That mitigation doesn't apply to non-low-drag sats, as abundantly shown by DarkSat too.They included the VLEO Starlink constellation and Sat Revolution constellation in the calculation, with an altitude of 340km and 350km respectively, which is very low, it's similar to the phasing orbit for the 550km Starlink. So the effect of having x number of satellites doing orbital raising/lowering can be approximated by increasing the size of these VLEO constellations by x. Total # of satellites between 340km and 350km is 32% of the total, adding 2400 more to it would just change the percentage to 38%. So unless the VLEO constellation has a huge disproportionate effect on the result, adding satellites in deployment/decommissioning phase would be similar to increasing the total # of satellites by 5-10%.
The difference is operational attitude. The VLEO satellites will have their solar panels in the operational orientation where they minimally contribute to the satellite's brightness. Satellites in transit to a higher orbit have their solar panel in low-drag configuration, where the solar panel is brighter by a much larger degree than the reduction in brightness from the changes implemented in DarkSat (though future changes could help reduce brightness in transit). Essentially, even if DarkSat was perfectly black, a DarkSat satellite in transit is about as bright as any current Starlink satellite in transit.
Executive SummarySimulations of the Rubin Observatory Legacy Survey of Space and Time (LSST) observing cadence and the full 42,000 SpaceX satellite constellation show that as many as 30% of all LSST images would contain at least one satellite trail.Nearly every LSST image taken during twilight would be affected by at least one satellite trail.Measurements of the brightness of the current LEO satellites indicate that trails would saturate in LSST images and cause residual artifacts in the reduced data, if no mitigations are made.If all affected images were rendered useless by saturated trails, LSST would have to add an additional 4 years to its planned 10 year survey to mitigate this loss.The Rubin Observatory team is working constructively with SpaceX engineers on remediation solutions.
LSST has also now published a statement on the impact to the survey.https://www.lsst.org/content/lsst-statement-regarding-increased-deployment-satellite-constellationsQuoteExecutive SummarySimulations of the Rubin Observatory Legacy Survey of Space and Time (LSST) observing cadence and the full 42,000 SpaceX satellite constellation show that as many as 30% of all LSST images would contain at least one satellite trail.Nearly every LSST image taken during twilight would be affected by at least one satellite trail.Measurements of the brightness of the current LEO satellites indicate that trails would saturate in LSST images and cause residual artifacts in the reduced data, if no mitigations are made.If all affected images were rendered useless by saturated trails, LSST would have to add an additional 4 years to its planned 10 year survey to mitigate this loss.The Rubin Observatory team is working constructively with SpaceX engineers on remediation solutions.
Rubin Observatory is an extreme case for the sensitivity of astronomical observations to satellite constellations...
If satellites were fainter by at least about a factor of ten... the net fraction of lost LSST pixels would be in the range of 0.3%-3%, which corresponds to several months of observing time.
They don't specify the baseline brightness from which this estimate is derived, but Darksat was observed in the last few weeks to be 3-5x dimmer than others at 550 km in operational configuration, and 20-100x dimmer than the orbit-raising configuration.Also, there was a PDF linked tot hat page, which I've attached.
Tony Tyson of Vera Rubin Obs. says they would like to see Starlink satellites become 10-20 times dimmer than now; will still have satellite trails but can eliminate nonlinear imaging artifacts that the bright satellites cause.
Tyson adds he’s “cautiously optimistic” Starlink satellites will be darkened sufficiently, but still a work in progress. Doesn’t yet appear to be much data on how effective the “DarkSat” launched in January is.
Delighted to announce that my Starlink paper has been accepted for publication in Astrophysical Journal Letters and will be on the arxiv on Wed; and you can download the preprint version now at https://planet4589.org/space/papers/starlink20.pdf
The Low Earth Orbit Satellite Population and Impacts of the SpaceX Starlink ConstellationJonathan C. McDowellCenter For Astrophysics — Harvard & Smithsonian 60 Garden St,Cambridge, MA 02138, USA(Revised Mar 14, 2020; Accepted Mar 16, 2020)Submitted to ApJLABSTRACTI discuss the current low Earth orbit artificial satellite population and show that the proposed ‘megaconstellation’ of circa 12,000 Starlink internet satellites would dominate the lower part of Earth orbit, below 600 km, with a latitude-dependent areal number density of between 0.005 and 0.01 objects per square degree at airmass < 2. Such large, low altitude satellites appear visually bright to ground observers, and the initial Starlinks are naked eye objects. I model the expected number of illuminated satellites as a function of latitude, time of year, and time of night and summarize the range of possible consequences for ground-based astronomy. In winter at lower latitudes typical of major observatories, the satellites will not be illuminated for six hours in the middle of the night. However, at low elevations near twilight at intermediate latitudes (45-55 deg, e.g. much of Europe) hundreds of satellites may be visible at once to naked-eye observers at dark sites.Keywords: artificial satellites — night sky brightness — astronomical site protection — ground-based astronomy
Jeez. First observations and magnitude measurement of SpaceX’s Darksat https://arxiv.org/pdf/2003.07251.pdf -> Special 'darkening treatment' makes brightness go from 6.69 mag in g band to 7.57 mag (at 976km), essentially from 'terribly bright' to...'terribly bright' 😟 #Starlink
First observations and magnitude measurement of SpaceX’sDarksatJ. Tregloan-Reed1, A. Otarola2, E. Ortiz3, V. Molina1, J. Anais1, R. González1, J. P. Colque1, and E. Unda-Sanzana11 Centro de Astronomía (CITEVA), Universidad de Antofagasta, Avenida U. de Antofagasta 02800, Antofagasta, Chile e-mail: [email protected]2 TMT International Observatory, 100 West Walnut Street, Pasadena, CA 91124, USA.3 Departamento de Física, Universidad de Antofagasta, Avenida Angamos 601, Antofagasta, ChileReceived Month dd,yyyy; accepted Month dd,yyyyABSTRACTAims. Measure the Sloan g’ magnitudes of the SpaceX STARLINK-1130 (Darksat) and 1113 LEO communication satellites and determine the effectiveness of the Darksat darkening treatment at 475.4 nm.Methods. Two observations of the SpaceX STARLINK Darksat LEO communication satellite were conducted on 2020/02/08 and 2020/03/06 using a Sloan r’ and g’ respectively. While a second satellite, STARLINK-1113 was observed on 2020/03/06 using a Sloan g’ filter. The initial observation on 2020/02/08 was a test observation when Darksat was still manoeuvring to its nominal orbit and orientation. Based on the successful test observation, the first main observation was conducted on 2020/03/06 along with an observation of a second STARLINK satellite.Results. The calibration, image processing and analysis of the Darksat Sloan g’ image gives an estimated Sloan g’ magnitude of 7.57 ± 0.04 at a range of 976.50 km. For STARLINK-1113 an estimated Sloan g’ magnitude of 6.69 ± 0.05 at a range of 941.62 km was found. When scaled to a range of 550 km, a reduction of (55 % ± 4.8 %) is seen in the reflected solar flux between Darksat and STARLINK-1113.Conclusions. The data and results presented in this work, show that the special darkening “treatment” used by SpaceX for Darksat has reduced the Sloan g’ magnitude by 0.88 ± 0.05 mag (55 % ± 4.8 %), when the range is equal to a nominal orbital height (550 km). This result will serve members of the astronomical community modelling the satellite mega-constellations, to ascertain their true impact on both the amateur and professional astronomical communities. Concurrent and further observations are planned to cover the full optical and NIR spectrum, from an ensemble of instruments, telescopes and observatories.
Quote from: @DScolJeez. First observations and magnitude measurement of SpaceX’s Darksat https://arxiv.org/pdf/2003.07251.pdf -> Special 'darkening treatment' makes brightness go from 6.69 mag in g band to 7.57 mag (at 976km), essentially from 'terribly bright' to...'terribly bright' 😟 #Starlink
After the ESO study was published, SpaceX got in touch to ask for a meeting to talk about impact on ESO’s own wide-field facility, the VISTA telescope. “The problem won’t be as bad as it is for the Vera Rubin, but we’re still talking about 5% to 6% of images lost or affected during the twilight hours,” Williams told Business Insider.“We are encouraged by the fact that SpaceX reached out to us with respect to our most affected telescope and their cooperative and constructive suggestions so far. We are looking forward to working with SpaceX, in cooperation with other astronomy groups and governments, to find a mutually agreeable solution,” said Williams. He did not give detail about what suggestions SpaceX have given thus far.
“This is a continuing experiment,” Tyson said of the DarkSat observations, noting that measurements of its brightness were taken just the night before. The data from the small Chilean telescope analyzed in the arXiv preprint came primarily from a single night of observations in early March after DarkSat reached its operational orbit.Tyson, though, emphasized the cooperation between SpaceX and the astronomy community to reduce the brightness of future Starlink satellites. “We’ve had a really delightful collaboration going now for a couple months with SpaceX engineers,” he said. “There are a lot of ideas on the table for darkening their satellites. This is just the first.”
Same but with video: 4 @SpaceX Starlink satellites including #Starlink-1030 DARKSAT on 22 March. DARKSAT clearly somewhat fainter. Video camera pointed on fixed alt/az. Frames of 4 video sequences stacked. Brightest stars are Capella and epsilon Aur.@planet4589
I have put the final version of my Starlink paper at https://planet4589.org/space/papers/starlink20.pdf - this is the version that will go on arxiv tomorrow, and reflects the published ApJ version except for a proofing error that crept in to the latter.
