Characterization of Starlink Direct-to-Cell Satellites In Brightness Mitigation Mode

Anthony Mallama; Jay Respler; Richard E. Cole; Scott Harrington

REVIEW 3 major objections 5 minor 5 references

Characterization of Starlink Direct-to-Cell Satellites In Brightness Mitigation Mode

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Starlink Direct-to-Cell satellites in brightness mitigation mode remain about 2.0 times brighter than Starlink internet spacecraft at a common distance, with a mean apparent magnitude of 5.16.

desk verdict Useful updated brightness data for DTC Starlinks, but the 'mitigation mode' means likely mix in Blue Zone/maneuver events and need an exclusion analysis before the headline is trusted. read the letter →

arxiv 2502.03651 v1 pith:IKKFMVNO submitted 2025-02-05 astro-ph.IM

Anthony Mallama , Richard E. Cole , Jay Respler , Scott Harrington This is my paper · ORCID

classification astro-ph.IM

keywords StarlinkDirect-to-CellsatellitebrightnessmitigationphotometrylowEarthorbitlightpollutionphasefunctionconstellation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Starlink's Direct-to-Cell (DTC) satellites, built to provide phone service from orbit, remain bright even after SpaceX adjusted their orientation into a brightness-mitigation mode. Using 551 electronic and visual observations from July through December 2024, the paper reports a mean apparent magnitude of 5.16 and a mean magnitude of 6.47 when all observations are adjusted to a common distance of 1,000 km. That makes the DTC spacecraft about 2.0 times brighter than Starlink internet satellites at the same distance, and although they have faded since early 2024 (when the mean apparent magnitude was 4.62), they still exceed both the magnitude-7 level that contaminates research astronomy and the magnitude-6 level of naked-eye visibility. The paper also presents a physical reflection model that accounts for the brightness and its dependence on solar phase angle. If these measurements are right, DTC satellites will continue to interfere with astronomical imaging and casual sky viewing unless further mitigation steps are taken.

What carries the argument

The central mechanism is the brightness-mitigation attitude itself, described by a physical model in which the spacecraft is roughly a flat Earth-facing surface (the DTC antenna and chassis base) plus two vertical chassis edges. In the mitigation orientation the solar panels are turned edge-on to the Sun, suppressing their reflection, and the model assumes the long axis of the spacecraft stays parallel to its velocity vector. The model treats all reflections as Lambertian, fits two parameters (overall normalization and the vertical-to-horizontal surface contribution), and adds a bright-Earth specular term that explains the upward turn in the phase function at phase angles above about 110 degrees. The resulting third-order polynomial phase functions reproduce the observed 1000-km magnitudes well enough for the model to predict DTC brightness across the sky for any Sun and observer geometry.

What would settle it

Observe a set of DTC passes simultaneously with a standard V-filter telescope and with the comparison-star visual method used here, then compare the two magnitude scales; if the systematic offset between them exceeds about 0.1 magnitude, the paper's mean 1000-km magnitude of 6.47 and the conclusion that DTCs remain above the magnitude-7 research limit would need to be recomputed.

Watch

Extended reading notes

Core claim

The paper's central claim is that Starlink Mini Direct-To-Cell satellites observed in brightness-mitigation mode during July through December 2024 have a mean apparent magnitude of 5.16 (standard deviation 1.30) and a mean 1000-km-adjusted magnitude of 6.47 (standard deviation 1.32). For comparison, Starlink internet Minis have corresponding values of 6.36 and 7.22, so at a common distance the DTC spacecraft are 2.0 times brighter. The DTC magnitude distribution is skewed toward brighter values, with an excess starting around magnitude 3.5 apparent and 6.0 at 1,000 km; the authors explain this skew by arguing that the lower-orbiting DTCs experience roughly 20 times more atmospheric drag and are therefore taken out of the mitigation attitude more often for station-keeping maneuvers. The mean values are fainter than those recorded before July 2024, confirming that SpaceX's attitude changes did dim the spacecraft, but the DTCs still land above the magnitude-7 and magnitude-6 thresholds used to gauge impacts on research and naked-eye viewing.

Load-bearing premise

The load-bearing premise is that the brightness calibration is accurate: the electronic measurements are taken to lie within 0.1 magnitude of the standard V-band brightness scale (a claim passed along from a private communication), and the visual estimates made by comparing a satellite to nearby stars are taken to approximate that same scale; if either calibration is systematically off, the reported mean magnitudes and the 2.0-times brightness ratio would shift.

Editorial extensions

If this is right

Because the mean apparent magnitude is 5.16, DTC satellites will be visible to the unaided eye in reasonably dark skies, exceeding the nominal magnitude-6 limit.
Because the mean 1000-km-adjusted magnitude is 6.47, DTC satellites remain brighter than the magnitude-7 level that contaminates professional astronomical images.
The factor-of-2.0 brightness gap relative to Starlink internet spacecraft at a common distance isolates the contribution of the DTC antenna and lower orbital altitude to the overall satellite-brightness problem.
The fitted phase functions give a predictive tool for when a given DTC pass will be brightest, which could be used to schedule observations around the worst transits.
More frequent station-keeping maneuvers at lower altitude mean published orbital elements will be outdated more often, making it harder for astronomers to avoid DTC satellites.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Beyond the paper: if the 0.1-magnitude calibration claim for the electronic photometry is optimistic, the mean 1000-km magnitude of 6.47 could shift by a few tenths, which would move the DTCs closer to or farther from the magnitude-7 threshold; a dedicated V-band cross-calibration would settle this.
The maneuver-frequency explanation predicts a testable pattern: the blue-zone brightening events should become more common during solar maximum, when drag rises, and rarer at solar minimum.
The same three-surface model could be transferred to other planned direct-to-cell constellations in low Earth orbit to estimate their impact on astronomy before launch.
Observing the same DTC pass simultaneously with a calibrated filtered camera and with the visual comparison-star method would check the core assumption that visual magnitudes approximate the V-band; the result would directly rescale the 2.0-times ratio.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

The paper is a useful measurement update: after SpaceX changed DTC attitudes, the authors report mean apparent magnitude 5.16 and 1000-km adjusted magnitude 6.47 from 551 observations taken July through December 2024, roughly a magnitude fainter than their earlier 4.62/5.50. That is the result worth knowing, and it is directly relevant to observation planning and satellite policy.

The good: the sample is large for this kind of work, the paper cleanly separates early and late data, the MMT9 and visual methods are consistent with the authors' prior publications, and the data are in the SCORE database. The distance-adjusted comparison to internet Minis (2.0x brighter) is a clear, testable statement. The physical model is a reasonable exercise, and the residual plots show the authors noticed and modeled the bright-Earth reflection rather than ignoring it.

Now the soft spots, in decreasing order of importance. First, the central claim says the means characterize "brightness mitigation mode," but nothing in Section 3 excludes the Blue Zone and maneuver-brightened observations that Section 4.2 identifies as not in the low-brightness orientation. Section 5 even attributes the bright skew to removal from mitigation. If those events are in the mean, the quoted 6.47 is a mixture of attitudes, not the mitigation-mode value. The fix is simple: report how many Blue Zone or maneuver events are in the sample and recompute the means without them. This matters because a shift of half a magnitude would move the mean beyond the mag 7 research threshold. Second, calibration rests on a private communication for MMT9 and on visual estimates; a cross-check between the two methods would strengthen the numbers. Third, the physical model has two fitted parameters and is not independently validated, so I would treat it as illustrative rather than predictive.

Despite these issues, the basic empirical result—DTCs have faded since mid-2024 but remain above the IAU thresholds—is probably right, and the paper deserves a serious referee. My own verdict would be conditional: ask for the exclusion analysis before accepting the headline. This is not a manufactured flaw; it is in the paper's own text.

I would bring this to a reading group as a short, policy-relevant data point and would cite it if I were working on satellite brightness. Send it to peer review, with the sample-selection question as the key revision.

Referee Report

3 major / 5 minor

Summary. The paper characterizes the visible brightness of Starlink Mini Direct-to-Cell (DTC) satellites observed between July and December 2024, after SpaceX reportedly placed them in brightness mitigation attitudes. Using 551 electronic (MMT9) and visual magnitude estimates, the authors report a mean apparent magnitude of 5.16 and a mean distance-adjusted (1000 km) magnitude of 6.47, concluding that DTC satellites are 2.0 times brighter than Starlink internet satellites at a common distance. The paper also presents a physical model with Lambertian surfaces for the chassis and DTC antenna, including a bright-Earth reflection term, and discusses the impact of DTC brightness on astronomy and naked-eye visibility.

Significance. If the headline numbers are correct, the result is directly relevant to the ongoing debate about satellite constellation impacts: DTC satellites would remain brighter than the magnitude-7 research contamination limit and the magnitude-6 naked-eye visibility limit even in mitigation mode. The paper's strengths include using a large public dataset (551 observations from SCORE), clearly stating its modeling assumptions, and providing a quantitative comparison to previous work on early DTC and internet satellites. The physical model, although simple, is a useful framework for predicting satellite brightness and could be tested against future independent observations. However, as detailed below, the central mean values rest on assumptions about sample selection and calibration that need to be substantiated before the conclusions can be fully accepted.

major comments (3)

[Section 3 and Section 4.2, Figure 6] The paper's headline means (5.16 and 6.47) are stated to characterize DTC satellites 'in brightness mitigation mode,' but the manuscript does not demonstrate that the July-December 2024 sample consists exclusively of mitigation-mode observations. Section 4.2 explicitly identifies a set of 'Blue Zone' observations in which 'the solar panel is not in the low-brightness orientation' and which are 'much brighter than the majority,' and Section 5 attributes the bright skew of the same DTC distribution to spacecraft being 'removed from brightness mitigation mode more often than internet satellites' during station-keeping. Nowhere does the text state that these Blue Zone or maneuver-brightened observations were excluded from the Section 3 sample used to compute the quoted means. If they are retained, the quoted mean is a mixture of mitigation and non-mitigation attitudes, not the mean in mitigation mode. Given that the 1000-km mean of 6.47 is close to the magnitude-7 contamination limit, even a modest shift fainter would change the paper's central conclusion. The authors should report the number of Blue Zone and maneuver-related events in the sample and recompute the headline statistics with them excluded.
[Section 2] The photometric calibration rests on two unverified or weakly verified assumptions: MMT9 magnitudes are claimed to be within 0.1 magnitude of the V-band 'according to information in a private communication from S. Karpov as discussed by Mallama (2021),' and visual magnitudes 'approximate the V-band' by comparison with nearby stars. No cross-calibration between the two methods is presented, and no estimate of systematic error is given. The reported SDM of 0.06 reflects only formal scatter, not systematic calibration uncertainty. A systematic error of, say, 0.2 magnitude in either method would shift both the mean magnitudes and the derived 2.0x brightness ratio, so the calibration needs to be documented more rigorously (e.g., by comparing MMT9 and visual measurements of the same satellites at the same times or by referencing published V-band calibrations of the MMT9 system).
[Section 4.2, Table 1] The physical model is fitted to the same data used to claim model success, so the agreement between model and observations is partly guaranteed by construction. Section 4.2 states that the overall brightness normalization and the vertical/horizontal surface ratio are fitted to the data, and the bright-Earth contribution and the phase-function polynomial coefficients (Table 1) are also derived from the data. The paper does not provide an independent validation of the model (e.g., a withheld subset of observations or a prediction for a different geometry or satellite type). The residuals in Figures 6-8 also lack quantitative error bars or a goodness-of-fit measure. The model is a useful interpretive tool, but the statement in Section 6 that the model 'fits the observations' needs to be qualified as a fit with two or more free parameters, not an independent confirmation of the physical assumptions.

minor comments (5)

[Section 3, Figures 1 and 2] The histograms in Figures 1 and 2 do not indicate the sample sizes or the number of observations in the 'early' versus 'late' and 'MMT9' versus 'visual' subsets. This information is important for assessing the robustness of the quoted means and for interpreting the apparent skewness.
[Section 3, Table 1] The phase-function polynomial coefficients are given without uncertainties or the number of points used in the fit. Adding these would allow readers to judge the stability of the fit and the significance of the high-order terms.
[Section 4.1] The assumption that 'the long axis of the spacecraft, through the DTC antenna, remains oriented parallel to the orbit velocity vector' is presented without justification, and Section 4.2 later notes that SpaceX has stated the orientation is variable and not published. This assumption directly affects the modeled surface orientations and should be flagged as a source of model uncertainty rather than a fixed axiom.
[Section 5] The hypothesized explanation for the DTC skew (more frequent removal from mitigation during station-keeping) is supported only by a comparison of RMS residuals for a single DTC and a single internet satellite in Figure 9. The authors should either present statistics for more satellites or temper the claim that the greater variation 'suggests' more frequent station-keeping.
[General] The paper would benefit from a statement of the epoch and orbital parameters of the DTC satellites in the sample, as well as the distribution of phase angles and distances, so that readers can assess the representativeness of the sample with respect to the full DTC constellation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central magnitude statistics are direct photometric measurements, and the fitted physical model is not used to derive them.

full rationale

The paper's headline results are purely observational: mean apparent magnitude 5.16, mean 1000-km adjusted magnitude 6.47, and a 2.0x brightness ratio versus internet Starlinks are computed directly from MMT9 and visual magnitude measurements (Sections 2 and 3). These claims do not depend on the physical model, so there is no reduction of the central result to its own inputs. The physical model of Section 4 is explicitly described as fitting the observations: 'The overall predicted brightness of the spacecraft is normalized to match the overall observed brightness. The average relative contribution of the vertical and horizontal surfaces is also fitted to the data, so there are a total of two fitted parameters in the model.' This is honest curve-fitting, and the concluding statement 'A physical model for satellite brightness fits the observations' makes no predictive claim that would be circular; the later assertion that the model 'can be used to generate predictions' is extrapolation, but no specific predicted quantity is presented as an independent validation, so it is not a circular derivation. Self-citations to prior work (Mallama et al. 2024, Cole 2021) are used for context and model construction, but the load-bearing photometric claims depend on the current observations, not on those citations. The paper does contain a notable internal-validity concern, flagged here per the reviewing rule: Section 4.2 identifies 'Blue Zone' observations where 'the solar panel is not in the low-brightness orientation' and which are 'much brighter than the majority,' and Section 5 explains the bright skew by spacecraft being 'removed from brightness mitigation mode more often than internet satellites'; the paper never explicitly states that these non-mitigation observations are excluded from the Section 3 means. That is a possible sample-selection or construct-validity problem for the 'mitigation mode' label, but it is not a circularity of the kind where an output is equivalent to an input by definition or where a fitted parameter is renamed as a prediction. Therefore the circularity score is 0.

Assumptions & free parameters 4 free parameters · 7 assumptions · 0 invented entities

The central brightness statistics are direct measurements and do not depend on the fitted model. The model itself carries two explicitly fitted parameters and a phase-function polynomial, and the interpretation as 'mitigation mode' depends on an operator statement. The axioms listed are the background assumptions needed to turn raw photometry into V-band magnitudes and to interpret the brightness changes.

free parameters (4)

Physical model overall brightness normalization = not quantified in paper
Section 4.2: 'The overall predicted brightness of the spacecraft is normalized to match the overall observed brightness', so this is a fitted scale factor, not a first-principles prediction.
Relative contribution of vertical and horizontal surfaces = not quantified in paper
Section 4.2: 'The average relative contribution of the vertical and horizontal surfaces is also fitted to the data, so there are a total of two fitted parameters in the model.'
DTC phase function polynomial coefficients (order 0 to 3) = 3.365, 0.06376, -1.716E-4, -1.139E-6
Table 1 lists coefficients fit to the 1000-km magnitudes versus phase angle; these describe the observed phase function rather than predict it.
Bright Earth reflection scale = not stated
Section 4.2 describes adding a bright Earth term after the simple model under-predicts high phase angles; the scale of this term is presumably fitted to residuals, though not explicitly listed.

assumptions (7)

domain assumption MMT9 photometry is within 0.1 magnitude of the V-band.
Section 2: based on a private communication from S. Karpov, not a published calibration verified in this paper.
domain assumption Visual magnitudes approximate the V-band when compared to nearby reference stars.
Section 2: method from Mallama 2022; no inter-comparison of visual and MMT9 results is given in this paper.
domain assumption The July to December 2024 observations represent the brightness mitigation attitude, while pre-June data did not.
Section 1: relies on SpaceX informing the authors that DTCs had not yet been placed in mitigation attitudes during the earlier data; this is operator testimony, not independently verified in the paper.
ad hoc to paper The long axis of the DTC spacecraft remains parallel to the orbit velocity vector.
Section 4.1: authors assume this even though they note SpaceX says orientation is variable and not published; if wrong, the model geometry is wrong.
domain assumption Solar panels are mostly edge-on to the Sun in mitigation mode, so their reflection is suppressed.
Section 4.1: based on SpaceX published statement and the Blue Zone interpretation; not directly measured in this paper.
domain assumption Atmospheric drag ratio between 350 km and 550 km is about 20.
Section 5: derived from MSISE-90 and the authors' choice of moderately high solar activity; used only for the maneuvering-frequency hypothesis.
domain assumption Surfaces can be modeled as Lambertian reflectors, with the Earth-shine specular term added separately.
Section 4.2: Cole 2021 model assumption; residuals show it is approximate.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Characterization of Starlink Direct-to-Cell Satellites In Brightness Mitigation Mode." pith.science (2026). https://pith.science/paper/IKKFMVNO

@misc{pith2026250203651,
  author       = {Pith},
  title        = {Pith review of: Characterization of Starlink Direct-to-Cell Satellites In Brightness Mitigation Mode},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IKKFMVNO}},
  note         = {Machine review of arXiv:2502.03651}
}

read the original abstract

The mean apparent magnitude of Starlink Mini Direct-To-Cell (DTC) satellites observed in brightness mitigation mode is 5.16, while the mean of magnitudes adjusted to a uniform distance of 1,000 km is 6.47. The DTCs have faded since early in 2024 because SpaceX subsequently adjusted the spacecraft attitudes to dim them. A physical model for satellite brightness that fits the observations is described.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

5 extracted references · 4 canonical work pages

[3]

https://arxiv.org/abs/2111.09735

Starlink satellite brightness -- characterized from 100,000 visible light magnitudes. https://arxiv.org/abs/2111.09735. Mallama, A. and Young, M

arXiv
[5]

Brightness Characterization for Starlink Direct-to-Cell Satellites

Brightness characterization for Starlink Direct-To-Cell satellites. https://arxiv.org/abs/2407.03092. 8

work page Pith review arXiv
[2021]

A sky brightness model for the Starlink 'Visorsat' spacecraft, arXiv, https://arxiv.org/abs/2107.06026 IAU Centre for the Protection of the Dark, Quiet Sky from Satellite Constellation Interference and 40 co-authors

work page Pith review arXiv
[2022]

The Method of Visual Satellite Photometry

The method of visual satellite photometry. https://arxiv.org/abs/2208.07834, Mallama, A., Cole, R.E., Harrington, S. and Respler, J

work page Pith review arXiv
[2024]

Call to Protect the Dark and Quiet Sky from Harmful Interference by Satellite Constellations

Call to protect the dark and quiet sky from harmful interference by satellite constellations. https://arxiv.org/abs/2412.08244. Karpov, S., Katkova, E., Beskin, G., Biryukov, A., Bondar, S., Davydov, E., Perkov, A. and Sasyuk, V

work page Pith review arXiv

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.

[1] [3]

https://arxiv.org/abs/2111.09735

Starlink satellite brightness -- characterized from 100,000 visible light magnitudes. https://arxiv.org/abs/2111.09735. Mallama, A. and Young, M

arXiv

[2] [5]

Brightness Characterization for Starlink Direct-to-Cell Satellites

Brightness characterization for Starlink Direct-To-Cell satellites. https://arxiv.org/abs/2407.03092. 8

work page Pith review arXiv

[3] [2021]

A sky brightness model for the Starlink 'Visorsat' spacecraft, arXiv, https://arxiv.org/abs/2107.06026 IAU Centre for the Protection of the Dark, Quiet Sky from Satellite Constellation Interference and 40 co-authors

work page Pith review arXiv

[4] [2022]

The Method of Visual Satellite Photometry

The method of visual satellite photometry. https://arxiv.org/abs/2208.07834, Mallama, A., Cole, R.E., Harrington, S. and Respler, J

work page Pith review arXiv

[5] [2024]

Call to Protect the Dark and Quiet Sky from Harmful Interference by Satellite Constellations

Call to protect the dark and quiet sky from harmful interference by satellite constellations. https://arxiv.org/abs/2412.08244. Karpov, S., Katkova, E., Beskin, G., Biryukov, A., Bondar, S., Davydov, E., Perkov, A. and Sasyuk, V

work page Pith review arXiv