{"id":"0ce6a68e-35b9-43e2-bffa-373d5669009f","arxiv_id":"1908.02298","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"An extension of the 4-galaxy-normal density plot method finds higher-quality planes of satellites around the Milky Way and M31, including a previously uncharacterized 18-member plane in M31 comparable to the Great Plane of Andromeda.","lead":"This paper refines the method for finding flat planes of dwarf satellite galaxies around the Milky Way and Andromeda, and reports a new 18-member plane around M31 that is roughly perpendicular to the known Great Plane of Andromeda. The sharper planes give theorists a more precise set of structures to reproduce in cosmological simulations.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 15° aperture used to rank satellite membership is untested; changing it could alter the M31-2-18 plane and thus the central claim.","rationale":"The paper is a careful, well-documented extension of an established method, and the comparisons with VPOS, GPoA, and Ibata-Conn planes provide useful external grounding. However, the aperture sensitivity is the least secure condition in the logical chain: the C_ps ranking is the sole input to the plane catalog, yet the only robustness check reported is for density-map bin size. A secondary internal inconsistency reinforces the need for revision: Sec. 4.2.1 states that the c/a difference between M31-2-18 and the GPoA is about 10%, but Tables 1 and 2 list c/a = 0.155 +/- 0.032 for M31-2-18 versus 0.107 +/- 0.005 for the GPoA, roughly a 45% difference, and Delta_RMS = 21.0 +/- 4.2 versus 13.6 +/- 0.2, roughly 54% larger. Even if the aperture test is clean, this quantitative comparison must be corrected or reworded before the 'comparable quality' headline is accepted. These are fixable issues rather than demonstrated fatal flaws, so conditional acceptance is appropriate.","tokens_in":15803,"tokens_out":9493,"duration_ms":94642,"concrete_test":"Re-run the full pipeline with aperture angles 10°, 20°, and 25° (and optionally 12.5° and 17.5°), keeping all other steps identical. For each density peak, compare the ordered C_ps lists, the resulting best plane membership, and the ToI parameters against M31-2-18 and the other named planes. If the 18 members of M31-2-18 and its c/a and Delta_RMS values are recovered within the quoted error bars, the concern is resolved; if membership or quality shifts materially, the plane is an artifact of the 15° choice and the central claim fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central result, especially M31-2-18, is produced by a pipeline whose decisive first step is the C_ps ranking of satellites (Sec. 3.1 iv). That ranking counts 4-galaxy normals within a hand-chosen 15° aperture around each density peak, and the same ranking fixes the ordered sequence of planes in Sec. 3.2. The 'best' planes are then read off where c/a and Delta_RMS start rising sharply with N_sat, so any change in the ordered list can change which plane is singled out and its ToI parameters. The paper explicitly checks only bin-size stability, not aperture stability; no argument or convergence test establishes that 15° is the appropriate scale. If the aperture were varied, the ordered satellite lists could change, and the 18-member M31-2-18 plane might lose members or its quality might no longer be comparable to the GPoA. Since the headline claim, the new-plane catalog, and the mass-independence conclusion are all downstream of this ranking, the manuscript's central claim is conditional on an unverified hand-chosen parameter.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends the '4-galaxy-normal density plot' method of Pawlowski et al. (2013) to generate, for each dominant over-density of satellite positions around the Milky Way and M31, an ordered collection of planes with increasing member number N_sat. Plane quality is quantified via the Tensor of Inertia outputs (c/a, b/a, ΔRMS, normal direction, and distance to host). Applying this to the same 27 MW and 34 M31 confirmed satellites used by Pawlowski et al., the authors single out several high-quality planes: MW-1-11, MW-1-14, M31-1-14, M31-1-16, M31-1-23, and M31-2-18. The headline claims are that M31-2-18 (N_sat=18) has quality comparable to the Great Plane of Andromeda (GPoA, N_sat=19), that this second M31 plane is roughly perpendicular to the GPoA and viewed nearly face-on from the Milky Way, and that stellar mass does not determine a satellite's membership in the best-quality planes.","tokens_in":15904,"tokens_out":3837,"duration_ms":40234,"significance":"If the method and results hold, the paper provides a systematic way to build and compare satellite plane candidates at fixed N_sat or fixed flattening, and it presents the first detailed analysis of a second substantial planar structure in M31. The Monte Carlo propagation of distance uncertainties into the plane parameters is a strength, as is the explicit comparison with previously published planes (classical, VPOS-3, VPOSall, Ibata-Conn-14, GPoA) using the same satellite sample. The mass-independence test (Fig. 5) is a useful falsifiable statement. However, the central claim of comparability between M31-2-18 and the GPoA is weakened by an internal numerical inconsistency and by the lack of a sensitivity test for the key aperture parameter that defines plane membership; these issues affect the manuscript's main conclusion as currently stated.","major_comments":[{"comment":"The 15-degree aperture angle used to compute the contribution C_ps of each satellite to a density peak is a free parameter, and the paper tests only bin-size stability (Sec. 3.1, final paragraph), not aperture stability. Because the ordered list of satellites by decreasing C_ps determines the membership of every plane in the collection, including M31-2-18, changing the aperture could alter which satellites are included and hence the c/a, ΔRMS, and normal directions of the singled-out planes. The authors should demonstrate that the recovered planes and their ToI parameters are stable when the aperture is varied (for example, 10 and 20 degrees), or provide a sample-independent criterion for choosing 15 degrees. Without this, the identification of M31-2-18 as a comparable-quality second plane is conditional on an untested hand-chosen parameter.","section":"Sec. 3.1(iv) and Sec. 3.2"},{"comment":"The statement that the differences between M31-2-18 and the GPoA are 'a 5% in N_sat, and a 10% in the c/a values' is inconsistent with the paper's own tables. From Table 1, c/a(GPoA)=0.107±0.005; from Table 2, c/a(M31-2-18)=0.155±0.032, which is a difference of roughly 45%, not 10%. In addition, ΔRMS is 13.6±0.2 kpc for the GPoA versus 21.0±4.2 kpc for M31-2-18 (a ~55% larger thickness), and D_M31 is 1.3±0.6 kpc versus 34.9±11.2 kpc. These differences are large enough that 'comparable quality' is not supported by the quoted numbers; the comparison should be reworded or the statistics revisited before the central claim can be accepted.","section":"Sec. 4.2.1 and Tables 1-2"},{"comment":"The paper reports a line-of-sight velocity dispersion of σ=90.20 km/s for M31-2-18 and immediately cites Fernando et al. (2017) to state that such a plane would be erased in a short timescale. This directly undercuts the physical significance of the newly identified plane, yet the paper does not explain how a spatially well-defined but dynamically short-lived structure should be interpreted. If 'quality' is intended purely as a spatial/morphological measure, that should be stated explicitly and distinguished from dynamical stability; if the plane is meant to be physically meaningful, the velocity-dispersion argument needs to be addressed rather than merely noted.","section":"Sec. 4.2.1"}],"minor_comments":[{"comment":"The citation 'CRAMR 1999' appears to be a typographical corruption of Cramér; the reference entry also contains formatting errors ('V ol.', 'Y .,' etc.) that should be corrected in a journal proof.","section":"Section 3.1 and References"},{"comment":"The caption reads 'M31's spin is marked with anX'; there is a missing space before 'X'. Additionally, the text states density maps are shown within l = [-90°, +90°], but the x-axis in Figure 1 extends beyond this interval; please clarify the coordinate range used.","section":"Figure 1 caption"},{"comment":"The abstract claims M31-2-18 shows 'a quality comparable to the GPoA', while the body notes that the GPoA has one more satellite and a lower c/a and is therefore strictly higher quality. The abstract should carry the same caveat as the body to avoid overstatement.","section":"Abstract and Section 4.2.1"},{"comment":"The statement that 'taking instead N_sat = 7 ± 2 to begin with does not alter our conclusions' is another parameter-sensitivity claim that is not demonstrated. Please provide the supporting test or remove the claim.","section":"Section 3.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a methodology-driven paper whose central claim rests on an untested 15-degree aperture and on a numerical comparison that is internally inconsistent. I believe these issues are fixable in revision: a stability test for the aperture and a corrected, properly caveated comparison would make the paper a solid contribution. I do not see grounds for rejection at this stage, but the required changes go beyond local editing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear X,\n\nThe paper delivers a useful toolkit and a new candidate, but you should be skeptical of the headline comparability claim.\n\nWhat is genuinely new: the extension of the 4-galaxy-normal density method produces an ordered catalog of planes by increasing Nsat, which is a real step beyond the single-plane output of Pawlowski et al. (2013). It is clearly written, uses public data, and propagates distance uncertainties in a transparent way. The first detailed look at the second M31 density peak is interesting; M31-2-18 is a plausible new structure. The mass-independence test is straightforward and the null result is credible.\n\nThe soft spots are in the claims, not the machinery. First, the text says the c/a difference between M31-2-18 and the GPoA is 10%, but the tables give 0.155 vs 0.107 — about 45% larger. The RMS thickness is also ~55% larger (21 vs 13.6 kpc), and the plane sits ~35 kpc from M31’s center, unlike the GPoA’s ~1 kpc. So 'comparable quality' is a stretch. The error bars overlap only if you take the worst-case for the GPoA and best-case for M31-2-18. Second, the 15° aperture used to rank Cps is a hand-chosen parameter and the paper only tests bin size, not aperture. Since the plane membership is built from that ranking, the M31-2-18 result is conditional on that choice. It could dissolve or change under a different aperture. That is load-bearing, and the authors should either test it or justify it.\n\nSome circularity is inherent — the planes are selected by contribution to planar overdensities — but the comparison with independent plane definitions (classical, Ibata-Conn-14, GPoA) provides some anchor.\n\nWho is this for? Specialists in Local Group satellite planes and halo substructure. It will be a useful reference for the method and for the M31-2-18 candidate, but treat the comparability claim as provisional.\n\nRecommendation: worth a serious referee. Send it to review, and in the report ask for an aperture sensitivity test and a corrected comparability statement. With those, it would be a solid addition.","headline":"Careful method paper with a new M31 plane candidate, but the claimed 'comparable' quality to the GPoA is overstated and the 15° aperture is untested.","tokens_in":16562,"tokens_out":3071,"would_cite":true,"duration_ms":30673,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Andromeda hosts a second great satellite plane, nearly perpendicular to the first, with quality rivaling the Great Plane of Andromeda.","keywords":["satellite planes","Milky Way dwarf galaxies","Andromeda dwarf galaxies","Great Plane of Andromeda","four-galaxy-normal density plot","tensor of inertia","Local Group structure","M31-2-18"],"falsifier":"Recompute the $C_{ps}$ ranking and the iterative plane collection from the same satellite positions using apertures of $10^\\circ$ and $20^\\circ$; if M31-2-18 loses its 18 members, shifts its normal by more than its quoted $\\Delta_{\\rm sph}=3.16^\\circ$, or its $c/a$ rises well above $0.15$, the claim that it rivals the GPoA fails. A kinematic check would measure proper motions of the 18 members and test whether their orbital poles are as collimated with the plane normal as the GPoA's are.","tokens_in":15472,"feed_emoji":"🌌","tokens_out":7896,"duration_ms":73326,"temperature":0.7,"pith_summary":"The paper claims that an extended version of the 4-galaxy-normal density plot method can turn one density peak into a whole nested collection of candidate satellite planes, ordered by how strongly each satellite supports the plane. Applied to the confirmed Milky Way and Andromeda satellite samples, it singles out planes that are thinner or more populous than the ones previously used as benchmarks: an 11-satellite Milky Way plane far flatter than the classical plane of the brightest 11 satellites, a 14-satellite version with the same thinness, and a newly identified second Andromeda plane, M31-2-18, with 18 members whose flattening approaches that of the 19-member Great Plane of Andromeda. The paper further reports that satellite stellar mass shows no correlation with plane membership, so mass-based pre-selection would miss the best planes.","feed_headline":"One more great satellite plane turns up in Andromeda","feed_subtitle":"An 18-member plane, seen nearly face-on, is as thin as Andromeda's known 19-satellite great plane.","key_machinery":"The load-bearing machinery is the extended 4-galaxy-normal density plot. The original method fits a plane to every four-satellite combination, repeats this for 100 random distance realizations per satellite, weights each resulting normal vector by $\\log((a+b)/c)$, and projects them onto a sphere so that an over-density marks the normal to a dominant satellite plane. The extension counts, for each density peak, the total weighted contribution $C_{ps}$ of all 4-galaxy-normals lying within a $15^\\circ$ aperture around the peak, ranks satellites by decreasing $C_{ps}$, and then iteratively fits tensor-of-inertia planes to the first 7, 8, 9, ... satellites in that order. Each over-density therefore becomes a catalog of planes of increasing $N_{\\rm sat}$ whose quality is compared through $N_{\\rm sat}$ and the flattening ratio $c/a$ (or rms thickness $\\Delta_{\\rm RMS}$), requiring $b/a \\sim 1$ to confirm a genuinely planar, rather than filamentary, configuration.","core_discovery":"The central discovery is that the two dominant 4-galaxy-normal over-densities in each galaxy, when expanded into ordered plane collections, reveal higher-quality planes than the previously reported benchmarks. In the Milky Way, a particular combination of 11 satellites (MW-1-11) and one of 14 (MW-1-14) are nearly twice as flat as the classical plane of the 11 most luminous satellites, with short-to-long axis ratios $c/a \\approx 0.095$ compared with $c/a \\approx 0.18$. In M31, the peak that contains the Great Plane of Andromeda yields improved 14- and 16-satellite planes (M31-1-14 and M31-1-16), while the second, previously unstudied peak yields M31-2-18: 18 satellites with $c/a \\approx 0.155 \\pm 0.032$, viewed nearly face-on from the Milky Way, roughly perpendicular to the GPoA, and of quality comparable to the GPoA's $N_{\\rm sat}=19$, $c/a = 0.107 \\pm 0.005$. The paper also finds that a satellite's contribution to the relevant density peaks is uncorrelated with its stellar mass, meaning the best planes are not assembled from the most massive satellites.","pith_inferences":["A direct test of the method's robustness would be to repeat the $C_{ps}$ ranking with apertures of, say, $10^\\circ$ and $20^\\circ$ around each density peak; if the 18-member set and normal direction of M31-2-18 persist, the plane is stable, and if not, the comparability claim weakens.","Two roughly orthogonal high-quality planes in one galaxy suggest a single coherent accretion stream cannot be the whole story; models may need two separate infall directions or a process that creates near-orthogonal families of dwarf orbits.","The same extended method could be applied to other nearby groups, such as Centaurus A, to ask whether multiple orthogonal satellite planes are common in galaxy groups or peculiar to M31.","With future proper motions for M31 dwarfs, one could test whether the M31-2-18 members corotate as a set; the paper's line-of-sight velocity dispersion of about 90 km/s would then be compared with the full 3D orbital poles."],"forward_implications":["The Milky Way's best plane is not the classical one: with $N_{\\rm sat}=11$ and $c/a\\approx 0.095$, MW-1-11 is much flatter, and adding three more satellites to make MW-1-14 keeps the same thinness, so any census restricted to the brightest satellites systematically misses the most planar structure.","M31 contains two comparably good planes, not one: the 19-member GPoA and the newly found 18-member M31-2-18, which are roughly perpendicular and share ten satellites.","Because M31-2-18 is viewed nearly face-on, its $c/a$ and $\\Delta_{\\rm RMS}$ carry larger distance-error bars than the GPoA's, yet it still reaches comparable quality, so the second plane is not an artifact of the binary's viewing geometry.","Stellar mass is not correlated with $C_{ps}$ for either galaxy, so plane membership is not determined by satellite mass; any successful formation model must populate planes without preferring massive dwarfs."],"supporting_citations":[{"why":"Supplies the 4-galaxy-normal density plot method, the MW/M31 satellite samples, and the VPOS-3 and GPoA benchmark planes that this paper extends.","marker":"Pawlowski et al. (2013)"},{"why":"Supplies the tensor-of-inertia plane-fitting technique and the $\\Delta_{\\rm sph}$ collimation measure used for plane quality.","marker":"Metz et al. (2007)"},{"why":"Supplies the 'Nearby dwarf galaxy database' positions, distance uncertainties, and stellar masses for the satellite samples.","marker":"McConnachie (2012)"},{"why":"Defines the PAndAS M31 satellite plane whose 14-satellite version is the baseline that M31-1-14 improves on.","marker":"Ibata et al. (2013)"},{"why":"Provides the edge-on viewing geometry of the GPoA and supports the interpretation of M31 Peak 1.","marker":"Conn et al. (2013)"},{"why":"Gives the perpendicular velocity-dispersion threshold above which satellite planes disperse in about 2 Gyr, used to comment on M31-2-18 stability.","marker":"Fernando et al. (2017)"}],"fun_headline_variants":["New Andromeda satellite plane rivals the Great Plane","Andromeda's second great plane of satellites found","18-satellite plane in Andromeda matches known great plane","New plane of 18 satellites in Andromeda rivals the great plane","Andromeda's hidden satellite plane nearly as thin as its famous one"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The ranking that decides which satellites belong to each plane depends on an untested choice of a $15^\\circ$ aperture around a density peak for summing each satellite's weighted contribution, and if a different aperture changed that ranking, the singled-out planes, especially M31-2-18, could change.","fun_headline_variants_meta":{"raw":{"variants":["New Andromeda satellite plane rivals the Great Plane","Andromeda's second great plane of satellites found","18-satellite plane in Andromeda matches known great plane","New plane of 18 satellites in Andromeda rivals the great plane","Andromeda's hidden satellite plane nearly as thin as its famous one"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000761,"raw_usage":{"total_tokens":3474,"prompt_tokens":1137,"completion_tokens":2337,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":753,"completion_tokens_details":{"reasoning_tokens":2250}},"tokens_in":753,"tokens_out":2337,"duration_ms":16445,"temperature":1.0,"reasoning_tokens":2250,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:48:15.732125+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the $C_{ps}$ ranking and the iterative plane collection from the same satellite positions using apertures of $10^\\circ$ and $20^\\circ$; if M31-2-18 loses its 18 members, shifts its normal by more than its quoted $\\Delta_{\\rm sph}=3.16^\\circ$, or its $c/a$ rises well above $0.15$, the claim that it rivals the GPoA fails. A kinematic check would measure proper motions of the 18 members and test whether their orbital poles are as collimated with the plane normal as the GPoA's are.","supporting_citations":[{"cited_title":"S., Kroupa P., Jerjen H., 2013, , 435, 1928","cited_arxiv_id":null,"evidence_quote":"Supplies the 4-galaxy-normal density plot method, the MW/M31 satellite samples, and the VPOS-3 and GPoA benchmark planes that this paper extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the tensor-of-inertia plane-fitting technique and the $\\Delta_{\\rm sph}$ collimation measure used for plane quality."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the PAndAS M31 satellite plane whose 14-satellite version is the baseline that M31-1-14 improves on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the edge-on viewing geometry of the GPoA and supports the interpretation of M31 Peak 1."},{"cited_title":"F., Ibata R","cited_arxiv_id":null,"evidence_quote":"Gives the perpendicular velocity-dispersion threshold above which satellite planes disperse in about 2 Gyr, used to comment on M31-2-18 stability."}],"review_version":1}