NGC 3079 热风手册（S09）NGC 3079 hot-wind handbook (S09)
项目一句话The project in one paragraph
NGC 3079 是侧向星暴星系（D = 19 Mpc，1′ = 5.53 kpc）。S09 用 10 个 XMM-Newton 指向、86 个 Einstein Probe FXT 指向与 Chandra 源普查研究其热外流：面亮度图与分区光谱测到 X 形双极风的"壁"到 ~55 kpc（55–66 kpc 为候选）、WSW 锥在 ~40 kpc 变为近等宽结构、90 kpc 内无"帽"、壁在加性背景读法下超压约 1.3–1.8 倍（若平软项是响应误差则约 2.0–2.1；晕本身只是上限，空白场检验 R004 支持加性背景：空白高纬 FXT 场同样有 5.2–5.7× 的平软项、γ 1.1–1.5 不低于 1、仅约 35–50% 的预期 CXB 沿 v1.30 剖面，"v1.30 渐晕太陡"不被支持）；壁的热能是现今可见风供能的 10.4–14.6 倍（加性读法，倾向；两种 FXT 模型合计 1/7–1/18）—— 解读为更早、更强爆发期的遗迹（CONDITIONAL）。可发表结果的状态截至 2026-10-16。
NGC 3079 is an edge-on starburst galaxy (D = 19 Mpc, 1′ = 5.53 kpc). Stage S09 studies its hot outflow with 10 XMM-Newton pointings, 86 Einstein Probe FXT pointings and a Chandra source census: maps and zoned spectra find X-shaped bipolar wind walls to ~55 kpc (55–66 kpc a candidate), a WSW cone turning near-constant-width at ~40 kpc, no cap within 90 kpc, walls over-pressured ~1.3–1.8× under the additive-background reading (~2.0–2.1× if the flat soft term were a response error; the halo itself is only an upper limit, and the blank-field test R004 supports the additive reading: blank high-latitude FXT fields carry the same 5.2–5.7× flat soft term, γ 1.1–1.5 not below 1, and only about 35–50% of the expected CXB follows the v1.30 profile, so "the v1.30 vignetting is too steep" is not supported), wall thermal energy 10.4–14.6× the visible present wind (additive case, favoured; 1/7–1/18 over both FXT models) — a relic of an earlier, stronger episode (CONDITIONAL). Publishable-result status as of 2026-10-16.
来源Source: cognition-os/insights/I021_s09_paper_plan_v2.md; .agent/jobs/2026-10-15-S09-FACTS.md; .agent/jobs/2026-10-16-fxt-blankfield-vignetting/REPORT.luna.md + REPORT.codexnim.md (R004); .agent/jobs/2026-10-16-redteam-rt14-wall-response/REPORT.codexnim.md (RT-14)
目标表（O1–O4）Objectives (O1–O4)
| 目标Objective | 状态与当前最佳数字（含来源）Status and best current number (with source) | 趋势Trend |
|---|---|---|
O1 论文 I：热外流 —— 壁、烟囱、无帽、超压遗迹（工作标题中的"66 kpc"可能更改，PI 决定）Paper I: the hot outflow — walls, a chimney, no cap, an over-pressured relic (the "66 kpc" in the working title may change; PI decision)L1–L10, L12–L14, L17; gates C1, C3–C5, C7, C9–C10 open; C2 done 2026-10-15 |
MEASURED 壁测到 ~55 kpc（联合 PN+FXT：39–55 kpc B = 2.48 ± 0.43 PW；含 0.3–0.5 PW 估计器系统学后 4.8–3.8σ）；55–66 kpc 为候选：2.09 ± 0.67，含估计器与掩罩系统学后 2.0–2.9σ；66 kpc 之外为上限 B < 0.75 PW（95%，仅统计）。超压 WSW 2.95 ± 0.84 → C2 后 2.87 —— CONDITIONAL 于 FXT 背景模型（RT-11/RT-12 → R004）：L12 为上限阶梯（0.87 / 0.93–1.07 / 1.04–1.81e-4）；空白场检验 R004 支持加性背景 —— 最佳拟合 q ~1.3–1.8、平衡晕 1.4e-4 在 2σ 内不被排除 —— 加性情形（R017/RT-18）：最佳晕（1.0e-4）下 13/13 个 WSW bin 都 q > 1（最小 1.12）；2σ 晕（1.8e-4）下仅外侧 4 个 bin（z ≥ 63.6 kpc，q 1.02–1.65）保持 q > 1，内侧 9 个 bin 为 0.62–0.92，平衡被允许；若平软项是响应误差则最佳拟合 q ~2.0–2.1、平衡被 2σ 排除（敏感性）；最不利叠加可降至 ~0.85。Walls measured to ~55 kpc (joint PN+FXT: B = 2.48 ± 0.43 PW at 39–55 kpc; 4.8–3.8σ with the 0.3–0.5 PW estimator systematic); 55–66 kpc a candidate: 2.09 ± 0.67, 2.0–2.9σ with estimator and mask systematics; beyond 66 kpc a limit B < 0.75 PW (95%, stat only). Over-pressure WSW 2.95 ± 0.84 → 2.87 after C2 — CONDITIONAL on the FXT background model (RT-11/RT-12 → R004): L12 is a limit ladder (0.87 / 0.93–1.07 / 1.04–1.81e-4); the blank-field test R004 supports the additive background — the best-fit q is ~1.3–1.8 and the balance halo (1.4e-4) is not excluded within 2σ — additive case (R017/RT-18): at the best halo (1.0e-4) q > 1 in all 13 WSW bins (minimum 1.12); at the 2σ halo (1.8e-4) only the 4 outer bins (z ≥ 63.6 kpc, q 1.02–1.65) keep q > 1, the inner 9 bins are 0.62–0.92, so balance is allowed; if the flat soft term were a response error the best-fit q is ~2.0–2.1 and balance is excluded at 2σ (a sensitivity case); the hostile stack can drop to ~0.85.来源Source: claims L7/L12/L13; I021; REDTEAM_QUEUE.md RT-01/RT-02/RT-05/RT-06/RT-08/RT-11/RT-12 |
↗ C2 完成后更稳↗ firmer after C2 |
O2 论文 II：径向 O/Fe（与 Liu et al. 2026 团队）Paper II: radial O/Fe (with the Liu et al. 2026 team)L11; gates C2 done, C5, C6 |
MEASURED (relative) 1T O/Fe：17–28 kpc 壁 8.26 → C2 PN+FXT 6.81（太阳仍被排除，dC(1) +5.4）；pooled 28–61 kpc 4.87 → 6.33。合并似然区间未校准（每 ON 区重复计一次 BG，e079_fit.py:220）。1T O/Fe: walls 17–28 kpc 8.26 → 6.81 with C2 PN+FXT (solar still rejected, dC(1) +5.4); pooled 28–61 kpc walls 4.87 → 6.33. Pooled-likelihood intervals are uncalibrated (BG counted once per ON region, e079_fit.py:220).来源Source: REPORT.hermes.md Table 2; session 2026-10-15.md (R1 astra) |
→ 待 C5/C6→ awaiting C5/C6 |
O3 物理解释（遗迹弱激波、能量来源、化学）Physical interpretation (relic weak shock, energy source, chemistry)L15, L16; P1/P2/P4/L1/L2/A1 (2026-10-15) |
CONDITIONAL 弱激波 Mach ~1.5–1.6（由 q 推定，非测量；需单相晕与光滑壁）。A1 审计 + RT-07：假设外延 R = 80–100 kpc时，100 Myr 情景每瓣 E0 = 3.40–4.94e57 erg、注入功率 1.08–1.57e42 erg/s（两瓣为 SN 功率的 2.7–3.9×）；对测得延伸 R = 55–66 kpc，同一模型给 1.4–2.5e57 erg、4.6–8.0e41 erg/s 每瓣，结束于 17–42 Myr 前。7–9e41 是对整个泡龄（151–176 Myr）的平均功率。全部能量 ∝ P_halo(47 kpc)/4.1e-14；L15 现在也受 C2 与 L12（上限）约束。Weak shock Mach ~1.5–1.6 (inferred from q, not measured; needs a single-phase halo and smooth walls). A1 audit + RT-07: assuming a reach R = 80–100 kpc, the 100 Myr episode gives E0 = 3.40–4.94e57 erg per lobe and injection power 1.08–1.57e42 erg/s (2.7–3.9× the SN power for two lobes); for the measured extent R = 55–66 kpc the same model gives 1.4–2.5e57 erg and 4.6–8.0e41 erg/s per lobe, ending 17–42 Myr ago. 7–9e41 is the power averaged over the full 151–176 Myr bubble age. All energies scale as P_halo(47 kpc)/4.1e-14; L15 is now also gated by C2 and by L12 (a limit).来源Source: I021 L15 (RT-07 triage); REDTEAM_QUEUE.md RT-07; A1 REPORT.sol.md; E098 record |
→ 讨论级→ discussion level |
O4 技术短文：EP-FXT 软弥漫辐射（可选）Technical note: EP-FXT soft diffuse emission (optional)L2, L2b, L3, E084-B; gate C10 |
OPTIONAL FXTDAS/CALDB v1.30 点源谱与 XMM 一致到 0–10%（0.4–1 keV）；MOS 卷帘梯度与 MOS1 CCD4 软超出已表征；RT-08/RT-11：FXT 图含一个未渐晕的平软项（0.49 cts/pix，0.5–1.3 keV，粒子模型 5×）；R004 空白场检验（luna + codexnim 一致）判定为加性背景（空白高纬场同样有 5.2–5.7× 的平软项、γ 1.1–1.5 不低于 1、仅约 35–50% 的预期 CXB 沿 v1.30 剖面；RT-12 曾倾向响应/渐晕误差，其论据是硬带渐晕部分缺失 17–51σ 与平软项 CXB 谱形 χ² 6.3/3 vs 粒子 255/3，但空白场不支持"v1.30 渐晕太陡"）—— 正是技术短文的素材。非首篇 EP-FXT 弥漫辐射文章（Tang et al. 2026, Abell 1795）。FXTDAS/CALDB v1.30 point-source spectra match XMM to 0–10% at 0.4–1 keV; the MOS roll-dependent gradient and the MOS1 CCD4 soft excess are characterised; RT-08/RT-11: the FXT map carries an unvignetted flat soft term (0.49 cts/pix, 0.5–1.3 keV, 5× the particle model); the R004 blank-field test (luna + codexnim agree) settles it as an additive background (blank high-latitude fields carry the same 5.2–5.7× flat soft term, γ 1.1–1.5 not below 1, and only about 35–50% of the expected CXB follows the v1.30 profile; RT-12 had favoured a response/vignetting error on the grounds that the hard-band vignetted part is missing at 17–51σ and the flat term is CXB-shaped, χ² 6.3/3 vs 255/3 for a particle spectrum, but the blank fields do not support "the v1.30 vignetting is too steep") — exactly the material for the technical note. Not the first EP-FXT diffuse paper (Tang et al. 2026, Abell 1795).来源Source: I021 (technical note); E084; S09 E085; REDTEAM_QUEUE.md RT-08/RT-11/RT-12; .agent/jobs/2026-10-16-fxt-blankfield-vignetting/REPORT.luna.md + REPORT.codexnim.md (R004) |
→ 可选→ optional |
组会一页纸（2026-10-16）Group-meeting one-pager (2026-10-16)
各组会记录（2026-10-15 / 16 / 17）→The full meeting log (2026-10-15 / 16 / 17) →
- 本周结论：FXT 平软项已由空白场检验（R004，luna + codexnim 两个执行者一致）判定为加性背景，"v1.30 渐晕太陡"不被支持。于是 L12 上限取加性读法 1.04–1.81e-4、平衡晕（1.4e-4）在 2σ 内不被排除；L13 最佳拟合 q ~1.3–1.8（加性情形 R017/RT-18：最佳晕 1.0e-4 下 13/13 个 WSW bin 都 q > 1，最小 1.12；2σ 晕 1.8e-4 下仅外侧 4 个 bin（z ≥ 63.6 kpc，q 1.02–1.65）保持 q > 1，内侧 9 个 bin 为 0.62–0.92，平衡被允许；响应误差敏感性 q ~2.0–2.1），状态 CONDITIONAL。This week's verdict: the FXT flat soft term is settled as an additive background by the blank-field test (R004, luna + codexnim agree), and "the v1.30 vignetting is too steep" is not supported. So L12 takes the additive upper 1.04–1.81e-4 and the balance halo (1.4e-4) is not excluded within 2σ; the L13 best-fit is q ~1.3–1.8 (additive case R017/RT-18: at the best halo 1.0e-4 q > 1 in all 13 WSW bins, minimum 1.12; at the 2σ halo 1.8e-4 only the 4 outer bins, z ≥ 63.6 kpc, q 1.02–1.65, keep q > 1, the inner 9 bins are 0.62–0.92, so balance is allowed; the response-error sensitivity q ~2.0–2.1), status CONDITIONAL.
- 能量缺口仍在（R017/RT-18）：现今可见风只供 WSW 壁热能的 1/10.4–1/14.6（加性读法，倾向；两种 FXT 模型合计 1/7–1/18）；解读为更早、更强爆发期的遗迹。The energy deficit survives (R017/RT-18): the visible present wind supplies only 1/10.4–1/14.6 of the WSW wall thermal energy (additive case, favoured; 1/7–1/18 over both FXT models); read as a relic of an earlier, stronger episode.
- 15′ 不是帽（RT-15）：12–17′ 凸起在 72 个扇区中排第 4、14–16′ 排第 7（~1.4–1.5σ），小轴两个扇区骑在全方位角平均上（ENE +0.9、WSW +2.1，平均 +1.5 对 +1.4）；15′ 环凸起 +0.56 PW 是 16 个半径中最大但给定 16 个半径不显著。L20 仍是唯一类帽候选。The 15′ bump is not a cap (RT-15): it ranks 4th of 72 sectors at 12–17′ and 7th at 14–16′ (~1.4–1.5σ), and the two minor-axis sectors straddle the all-PA mean (ENE +0.9, WSW +2.1, mean +1.5 vs +1.4); the 15′ ring bump +0.56 PW is the largest of 16 radii but not significant given 16 radii. L20 remains the only cap-like candidate.
- 指向框架已澄清（R005）：RA_PNT 在星系以北 2.29–2.32′（24 个 ObsID）只是指向记录；核到光轴的偏移为 0.2–0.76′（中值 0.5′，84 个成图 ObsID，roll 1–360°）。瞄准点偏移论证已从全站删除。The pointing frame is clarified (R005): RA_PNT 2.29–2.32′ north of the galaxy (24 ObsIDs) is only a pointing record; the nucleus-to-optical-axis offset is 0.2–0.76′ (median 0.5′, 84 map ObsIDs, rolls 1–360°). The aim-point-offset argument has been deleted site-wide.
- RT-17（R015）：FXT-B 探测器偶极不是 12–17′ 南北不对称的解释。预测 N−S 差 +1.12 ± 0.14 PW 对实测 +2.72 PW（比值 0.41）；按 roll 分组在 5.1σ 否证符号，相位旋转在 12.5σ / 18.3σ 被否证。该不对称在天球框架下相干但只有 1.8–1.9σ，仍是候选。RT-17 (R015): the FXT-B detector dipole is NOT the explanation for the 12–17′ north-south asymmetry. The predicted N−S difference is +1.12 ± 0.14 PW against an observed +2.72 PW (ratio 0.41); a split by roll rejects the sign at 5.1σ and the phase rotation is rejected at 12.5σ / 18.3σ. The asymmetry is sky-frame coherent but only at 1.8–1.9σ, so it stays a candidate.
- RT-16 + RT-19（R014 + R016）：a/a_CXB 缺口未解释（UNEXPLAINED）。在同一口径下（1.3–2.0 keV、10–24′、同一 ARF 滤镜、同一 CXB），NGC 3079 给出 −0.012 ± 0.023，空白场给出 +0.30 ± 0.04（A）/ +0.28 ± 0.04（B）。NGC 3079 的点源扣除值 0.24 的 0.34 缺口，但主导源是真的：一个 199σ 的 Chandra 点源（cid 134）承担环内源计数的 66%，且源模型处于其 χ² 最小值处 —— 所以该缺口未解释。不要引用 NGC 3079 的通用硬带 a/a_CXB。RT-04 排队。RT-16 + RT-19 (R014 + R016): the a/a_CXB gap is UNEXPLAINED. On the same footing (1.3–2.0 keV, 10–24′, the same ARF filter, the same CXB) NGC 3079 gives −0.012 ± 0.023 and the blank fields +0.30 ± 0.04 (A) / +0.28 ± 0.04 (B). The NGC 3079 point-source subtraction is worth 0.24 of the 0.34 gap, but the dominant source is real (a 199σ Chandra point source, cid 134, carrying 66% of the source counts in the ring) and the source model is at its χ² minimum — so the gap is UNEXPLAINED. Do not quote a generic hard-band a/a_CXB for NGC 3079. RT-04 is queued.
来源Source: .agent/jobs/2026-10-16-fxt-blankfield-vignetting/REPORT.luna.md + REPORT.codexnim.md (R004); .agent/jobs/2026-10-16-redteam-rt14-wall-response/REPORT.codexnim.md (RT-14); .agent/jobs/2026-10-16-redteam-rt15-wording-audit/REPORT.hermes.md (RT-15); .agent/jobs/2026-10-16-redteam-rt17-ns-sky-vs-det/REPORT.codexnim.md (RT-17); .agent/jobs/2026-10-16-redteam-rt16-a-cxb-gap/REPORT.codexnim.md (RT-16) and .agent/jobs/2026-10-17-redteam-rt19-source-model/REPORT.codexnim.md (RT-19); 07_REVISION_LOG.md R014/R015/R016; I022_paper1_wording_suggestions.md; claims L12/L13/L14/L20
未决问题表Open problems
| 问题Problem | 为何重要Why it matters | 最佳假设Best hypothesis | 下一步检验Next test | 负责Owner | 状态Status |
|---|---|---|---|---|---|
| 10–25′ 离轴的 FXT 响应（渐晕）：平软项已定为加性背景，但 a/a_CXB 不一致（顶行；R004 = luna + codexnim 一致）The FXT response (vignetting) at 10–25′ off-axis: the flat term is settled as an additive background, but a/a_CXB disagrees (top row; R004 = luna + codexnim agree) | 决定 L12/L13/L15/L16。R004 加性读法（倾向）：L12 上限 1.04–1.81e-4、平衡晕 1.4e-4 在 2σ 内不被排除、L13 最佳拟合 q ~1.3–1.8；响应误差敏感性：L12 上限 0.93–1.07e-4、平衡被 2σ 排除、L13 最佳拟合 q ~2.0–2.1；无额外项 0.87e-4decides L12/L13/L15/L16. Under the R004 additive reading (favoured): the L12 upper is 1.04–1.81e-4, the balance halo (1.4e-4) is not excluded within 2σ and the L13 best-fit q is ~1.3–1.8; under the response-error sensitivity: the L12 upper is 0.93–1.07e-4, balance is excluded at 2σ and the L13 best-fit q is ~2.0–2.1; with no extra term 0.87e-4 | R004（2026-10-16，21 个空白高纬场，877 ks，两执行者一致）：空白场直接拟合 a × 渐晕曝光 + b × 像素，0.5–1.3 keV 的逐场散差均值 γ = 1.41 ± 0.12（A）/ 1.12 ± 0.09（B）（luna）与 1.25 ± 0.33（A）/ 1.54 ± 0.25（B）（codexnim），两者都不低于 1；b/粒子 = 5.33 ± 0.12（A）/ 5.07 ± 0.13（B）—— 与 NGC 3079 的 5× 一致；只有约 35–50% 的预期 CXB 沿 v1.30 剖面。故平软项为加性背景，"v1.30 渐晕太陡"不被支持（RT-12 曾倾向响应误差，论据是硬带渐晕部分缺失 17–51σ、平软项 CXB 谱形 χ² 6.3/3 vs 粒子 255/3）。仍待定（RT-16 + RT-19，R014 + R016，未解释）：同一口径（1.3–2.0 keV、10–24′、同一 ARF 滤镜、同一 CXB）下 NGC 3079 的 a/a_CXB 为 −0.012 ± 0.023，空白场为 +0.30 ± 0.04（A）/ +0.28 ± 0.04（B）；点源扣除值 0.24 的缺口，但主导源是 199σ 的 Chandra 点源、源模型在 χ² 最小值处，所以缺口未解释；不要引用 NGC 3079 的通用硬带 a/a_CXBR004 (2026-10-16, 21 blank high-latitude fields, 877 ks, both executors agree): the blank fields fit a × vignetted exposure + b × pixels directly; the 0.5–1.3 keV field-scatter mean γ = 1.41 ± 0.12 (A) / 1.12 ± 0.09 (B) (luna) and 1.25 ± 0.33 (A) / 1.54 ± 0.25 (B) (codexnim), neither below 1; b/particle = 5.33 ± 0.12 (A) / 5.07 ± 0.13 (B) — matching the 5× of NGC 3079; only about 35–50% of the expected CXB follows the v1.30 profile. The flat soft term is therefore an additive background and "the v1.30 vignetting is too steep" is not supported (RT-12 had favoured a response error on the grounds that the hard-band vignetted part is missing at 17–51σ and the flat term is CXB-shaped, χ² 6.3/3 vs 255/3 for a particle spectrum). Still Pending (RT-16 + RT-19, R014 + R016, unexplained): on the same footing (1.3–2.0 keV, 10–24′, the same ARF filter, the same CXB) NGC 3079's a/a_CXB is −0.012 ± 0.023 against the blank fields' +0.30 ± 0.04 (A) / +0.28 ± 0.04 (B); the point-source subtraction is worth 0.24 of the gap, but the dominant source is a 199σ Chandra point source and the source model is at its χ² minimum, so the gap is unexplained; do not quote a generic hard-band a/a_CXB for NGC 3079 | 找出未解释的 a/a_CXB 缺口（NGC 3079 天区自身的 CXB 亏损、残差/权重口径）；FXT 团队的离轴渐晕验证；或带 CXB 先验的天空+平场+晕联合拟合find the unexplained a/a_CXB gap (a CXB deficit in the NGC 3079 sky itself, or a residual/weighting convention); an FXT-team off-axis vignetting validation; or a joint sky + flat + halo fit with a CXB prior | PI decision; R004 REPORT.luna.md + REPORT.codexnim.md; .agent/jobs/2026-10-16-redteam-rt16-a-cxb-gap/REPORT.codexnim.md (RT-16); .agent/jobs/2026-10-17-redteam-rt19-source-model/REPORT.codexnim.md (RT-19); REDTEAM_QUEUE.md RT-08/RT-11/RT-12/RT-16/RT-19 | R004 已出（21 个空白场，两执行者一致）：加性背景；RT-16 + RT-19 后 a/a_CXB 缺口 UNEXPLAINED（open）R004 reported (21 blank fields, both executors agree): additive background; after RT-16 + RT-19 the a/a_CXB gap is UNEXPLAINED (open) |
| PN 径向标定（X-only PN 图、逐指向径向项）PN radial calibration (an X-only PN map, a per-pointing radial term) | L2b：PN 无 MOS 式平面（< 0.17 PW/arcmin，2σ）但径向标定未测；公共模径向误差不被任何 N−S 检验约束；牵动 L7、L12、L21L2b: PN shows no MOS-like plane (< 0.17 PW/arcmin, 2σ) but its radial calibration is untested; a common-mode radial error is not bounded by any N−S test; carries L7, L12, L21 | RT-09：N−S 径向差 +0.23 ± 0.13、全减 N/S +0.26 ± 0.13 PW/arcmin；−0.245 残差把 PN-only 晕上限抬到 7.2–14.1e-5（平衡晕不被排除）RT-09: N−S radial difference +0.23 ± 0.13, all-minus-N/S +0.26 ± 0.13 PW/arcmin; a −0.245 residual lifts the PN-only halo upper to 7.2–14.1e-5 (the balance halo not excluded) | X-only PN 图 + 逐指向径向项拟合an X-only PN map with a per-pointing radial term | REDTEAM_QUEUE.md RT-09; I021 L2b | OPEN |
| 合并 on-off 似然把 BG 每 ON 区计一次Pooled on-off likelihood counts BG once per ON region | pooled kT/O/Fe 区间（E096 §d、C2 pooled）未校准pooled kT/O/Fe intervals (E096 §d, C2 pooled) are uncalibrated | 复合似然，偏差方向未知composite likelihood, bias direction unknown | 在引擎副本中改为每相机一个共享 BG 项并重拟 E096 pooled O/Fefix one shared BG term per camera in an engine copy and refit the E096 pooled O/Fe | e079_fit.py:220; session 2026-10-15 NEXT | OPEN |
| E098a 注入功率 vs 平均功率措辞E098a injection vs mean power wording | L15 的能量叙事；页 v10 面板 P 数字错误the L15 energy narrative; page v10 panel P numbers are wrong | A1 已给出正确措辞（未应用到记录）A1 supplies the correct wording (not yet applied to the record) | 修 E098 行 61 与 I021 L15fix E098 line 61 and I021 L15 | A1 · REPORT.sol.md | OPEN |
| WWLM 谱区受 MCG 9-17-9 影响WWLM spectral region exposed to MCG 9-17-9 | 28–39 kpc WSW 壁 kT/O/Fe（该区 18–30% 像素在伴星系 1.5–2′ 内）the 28–39 kpc WSW wall kT/O/Fe (18–30% of its pixels lie within 1.5–2′ of the companion) | P5：L7 与烟囱边缘不受影响；+1.4–1.7 PW 超出仅 ~2σ_nullP5: L7 and the chimney edge are unaffected; the +1.4–1.7 PW excess is only ~2σ_null | 用 2′ MCG 掩罩重拟 WWLMrefit WWLM with a 2′ MCG mask | P5 (REPORT.sonnet.md); C2 follow-up | OPEN |
| C5：kT 的前景/参考区系统学C5: foreground/reference systematics for kT | L10 的 kT(z)（→ L13 压力、L14 预算、L15 Mach）误差仅统计；参考区选择把外壁 kT 移动 0.08–0.10 keV；39 kpc 处 kT 突变未确立the L10 kT(z) errors (→ L13 pressure, L14 budgets, L15 Mach) are statistical only; the reference choice moves the outer-wall kT by 0.08–0.10 keV; the 39 kpc kT step is not established | RT-13：C5 未过（前景冻结/自由、多个参考区、0.2 vs 0.3 keV T0 转换、外壁 2T/CX 未测）RT-13: C5 open (foreground frozen/free, several reference regions, the 0.2 vs 0.3 keV T0 conversion, outer-wall 2T/CX untested) | 前景冻结 vs 自由 + 多参考区重拟 kT(z)；外壁 2T/CX 模型refit kT(z) with the foreground frozen vs free and several reference regions; outer-wall 2T/CX models | I021 C5; REDTEAM_QUEUE.md RT-13 | OPEN |
| NGC 3079 的 a/a_CXB 与空白场不一致（R004 遗留；RT-16 + RT-19 后仍未解释）NGC 3079's a/a_CXB disagrees with the blank fields (an R004 leftover; still unexplained after RT-16 + RT-19) | R004 已把平软项定为加性背景，但 NGC 3079 的 a/a_CXB 与空白场符号相反；这决定加性背景是否足以解释 NGC 3079 的全部 FXT 平软项，牵动 L12/L13/L15/L16R004 settles the flat soft term as an additive background, but NGC 3079's a/a_CXB has the opposite sign to the blank fields'; this decides whether an additive background explains all of NGC 3079's FXT flat term and carries L12/L13/L15/L16 | RT-16 + RT-19（R014 + R016）：缺口未解释。同一口径（1.3–2.0 keV、10–24′、同一 ARF 滤镜、同一 CXB）下 NGC 3079 为 −0.012 ± 0.023（合并，n = 14，χ² 64.6/12），空白场为 +0.30 ± 0.04（A）/ +0.28 ± 0.04（B），缺口 0.34（A）/ 0.30（B）。NGC 3079 的点源扣除值 0.24（合并），但主导源是真的：XMM 源 0（r 14.1′，PA 338.5°）承担环内源计数的 66%，是一个 199σ 的 Chandra 点源（cid 134，2952 ± 55 counts，POINT，r90 21.6″），源模型处于其 χ² 最小值处（x1.0），把它整块掩掉只给出 −0.085 ± 0.025（只掩 r < 60″ 核为 −0.000 ± 0.023），都到不了空白场的 +0.27/+0.30；只有宣称这个 199σ 源是假的才达到 +0.264 ± 0.025。因此缺口不是源模型的错误，也不是已测试过的仪器/响应/延展发射机制 —— 未解释。不要引用 NGC 3079 的通用硬带 a/a_CXB。RT-16 + RT-19 (R014 + R016): the gap is unexplained. On the same footing (1.3–2.0 keV, 10–24′, the same ARF filter, the same CXB) NGC 3079 gives −0.012 ± 0.023 (combined, n = 14, χ² 64.6/12) against the blank fields' +0.30 ± 0.04 (A) / +0.28 ± 0.04 (B), a gap of 0.34 (A) / 0.30 (B). The NGC 3079 point-source subtraction is worth 0.24 (combined), but the dominant source is real: XMM source 0 (r 14.1′, PA 338.5°) carries 66% of the source counts in the ring and is a 199σ Chandra point source (cid 134, 2952 ± 55 counts, POINT, r90 21.6″), and the source model sits at its χ² minimum (x1.0); masking it entirely gives only −0.085 ± 0.025 (masking just the r < 60″ core gives −0.000 ± 0.023), neither of which reaches the blank fields' +0.27/+0.30 — only declaring this 199σ source spurious reaches +0.264 ± 0.025. So the gap is not a source-model error and not any instrumental, response or extended-emission mechanism that was tested — it is UNEXPLAINED. Do not quote a generic hard-band a/a_CXB for NGC 3079. | 找出未解释的缺口：NGC 3079 天区自身在 10–25′ 的真实 CXB 亏损，还是残差/权重口径；FXT 团队的离轴渐晕验证find the unexplained gap: a genuine CXB deficit in the NGC 3079 sky itself at 10–25′, or a residual/weighting convention; an FXT-team off-axis vignetting validation | .agent/jobs/2026-10-16-redteam-rt16-a-cxb-gap/REPORT.codexnim.md (RT-16); .agent/jobs/2026-10-17-redteam-rt19-source-model/REPORT.codexnim.md (RT-19); 07_REVISION_LOG.md R014/R016 | RT-16 + RT-19 已出（R014 + R016）：缺口 UNEXPLAINED（open）RT-16 + RT-19 reported (R014 + R016): the gap is UNEXPLAINED (open) |
| FXT-B 的探测器坐标方位偶极 vs 12–17′ 的南北不对称The FXT-B detector-frame azimuth dipole vs the 12–17′ north-south asymmetry | 影响主轴晕剖面（L12）：NNW 边（PA 345）+2.61 PW 对 SSE 边（PA 165）−0.6 PW；15′ 处的凸起来自盘边而非帽（RT-15：12–17′ 排第 4/72、14–16′ 排第 7/72，~1.4–1.5σ）affects the major-axis halo profile (L12): the NNW side (PA 345) +2.61 PW against the SSE side (PA 165) −0.6 PW; the 15′ bump comes from the disk side, not a cap (RT-15: rank 4/72 at 12–17′, 7/72 at 14–16′, ~1.4–1.5σ) | 属宽的南北不对称（PA 330–30° +2–3 PW，PA 150–210° 约 0；E074 已知的北向偶极）。RT-17（2026-10-17，R015）：探测器偶极不是解释。正确的 DET→天球几何下，探测器偶极对 12–17′ 的预测 N−S 差为 +1.12 ± 0.14 PW（168 个 obs+模块、427.1 ks 曝光加权），而实测为 +2.72 PW（北 +2.33、南 −0.39），比值 0.41。按 roll 分组直接否证探测器固定的不对称：模块 b 的 roll 对比（低 roll 减高 roll）为 −4.145e-05 ± 1.15e-05 cts s⁻¹ arcmin⁻²（−3.6σ），而偶极预测 +1.680e-05，观测减预测 = −5.1σ，符号被否证。天球偶极相位在 PA_PNT 0–45° 与 225–270° 两个 bin 间只移动 −16.4°（a）/ +14.6°（b），而探测器固定的不对称必须移动 +106.3°：相位旋转在 12.5σ / 18.3σ 被否证。该不对称在天球框架下是相干的（R_sky = 0.918（a）/ 0.971（b）对 roll 零假设 R_roll = 0.803），但只有 1.8–1.9σ，因此它本身仍是候选part of a broad north-south asymmetry (PA 330–30° +2–3 PW, PA 150–210° about 0; the known E074 north dipole). RT-17 (2026-10-17, R015): the detector dipole is NOT the explanation. With the correct DET→sky geometry the detector dipole predicts an N−S difference at 12–17′ of +1.12 ± 0.14 PW (168 obs+module, 427.1 ks exposure-weighted), against an observed +2.72 PW (north +2.33, south −0.39), a ratio of 0.41. A split by roll rejects a detector-fixed asymmetry outright: the module-b roll contrast (low-roll minus high-roll) is −4.145e-05 ± 1.15e-05 cts s⁻¹ arcmin⁻² (−3.6σ) where the dipole predicts +1.680e-05, so observed − predicted = −5.1σ, rejected on sign. The sky-frame dipole phase moves by only −16.4° (a) / +14.6° (b) between the PA_PNT 0–45° and 225–270° bins, where a detector-fixed asymmetry must move by +106.3°: the phase rotation is rejected at 12.5σ / 18.3σ. The asymmetry is sky-frame coherent (R_sky = 0.918 (a) / 0.971 (b) against the roll null R_roll = 0.803), but only at 1.8–1.9σ, so it stays a candidate | 是什么造成这个 1.8–1.9σ 的天球相干不对称（盘侧发射、源扣除残差、还是统计涨落）；确认 FXT-A 是否也有不可相干分量what produces this 1.8–1.9σ sky-frame coherent asymmetry (disk-side emission, source-subtraction residuals, or a statistical fluctuation); confirm whether FXT-A also carries an incoherent component | .agent/jobs/2026-10-16-redteam-rt17-ns-sky-vs-det/REPORT.codexnim.md; 07_REVISION_LOG.md R015; sessions/2026-10-16.md FOUND; artifacts/s09_fxt15_excess | RT-17 已出（R015）：偶极被否证；不对称本身 1.8–1.9σ，暂定RT-17 reported (R015): the dipole is rejected; the asymmetry itself is 1.8–1.9σ, provisional |
| 红队计划（free-redteam）Red-team programme (free-redteam) | L2b/L7/L8/L9/L10/L11/L12/L13/L14/L15/L20/L21 的独立复核；RT-01–RT-03、RT-05–RT-07、RT-08..RT-13（2026-10-15/16 完成）已入账；RT-04 排队independent re-checks of L2b/L7/L8/L9/L10/L11/L12/L13/L14/L15/L20/L21; RT-01–RT-03, RT-05–RT-07 and RT-08..RT-13 (done 2026-10-15/16) folded in; RT-04 queued | RT-01/RT-06：55–66 kpc 候选（2.0–2.9σ）；RT-03：3-D 烟囱 INFERRED；RT-02：q = 2.95 ± 0.84；RT-05（2026-10-15）：L12 → 上限、L13 → CONDITIONAL；RT-07（2026-10-15）：L14 改写、L15 → CONDITIONAL；RT-08（2026-10-16）：FXT 平软项 0.49 cts/pix，按响应误差 L12 上限 0.93–1.07e-4；RT-09（2026-10-16）：PN 无平面但径向未测，PN-only 不再单独排除平衡晕；RT-10（2026-10-16）：L8 无帽成立，作为 L15 约束条件性；L20 保持候选（−1.0 PW），L21 非帽证据；RT-11（2026-10-16）：按加性背景 L12 上限 1.04–1.81e-4，平衡晕不被排除；RT-12（2026-10-16）：曾倾向响应/渐晕误差（硬带渐晕部分缺失 17–51σ、平软项 CXB 谱形）—— 该模型下 L12 上限 0.93–1.07e-4、最佳 7.95–8.67e-5、平衡晕被排除，L13 q ~2.0–2.1；已被 R004 空白场检验推翻（见顶行）：平软项为加性背景；RT-13（2026-10-16）：L10 → CONDITIONAL（误差仅统计、C5 未过、一致区集内 17–28 → 41–61 kpc 仅 +0.03 ± 0.03 keV，39 kpc kT 突变未确立）RT-01/RT-06: 55–66 kpc candidate (2.0–2.9σ); RT-03: the 3-D chimney is INFERRED; RT-02: q = 2.95 ± 0.84; RT-05 (2026-10-15): L12 → a limit, L13 → CONDITIONAL; RT-07 (2026-10-15): L14 reworded, L15 → CONDITIONAL; RT-08 (2026-10-16): the FXT flat soft term is 0.49 cts/pix; as a response error the L12 upper is 0.93–1.07e-4; RT-09 (2026-10-16): PN shows no plane but its radial calibration is untested; PN-only no longer excludes the balance halo alone; RT-10 (2026-10-16): the L8 no-cap stands, conditional as an L15 constraint; L20 stays a candidate (−1.0 PW), L21 is not a cap piece; RT-11 (2026-10-16): as an additive background the L12 upper is 1.04–1.81e-4, the balance halo not excluded; RT-12 (2026-10-16): a response/vignetting error had been favoured (the hard-band vignetted part is missing at 17–51σ; the flat term is CXB-shaped) — in that model the L12 upper is 0.93–1.07e-4 with best 7.95–8.67e-5, the balance halo is excluded, and the L13 q is ~2.0–2.1; overturned by the R004 blank-field test (see the top row): the flat term is an additive background; RT-13 (2026-10-16): L10 → CONDITIONAL (errors stat only, C5 open, and in a consistent zone set 17–28 → 41–61 kpc is only +0.03 ± 0.03 keV; the 39 kpc kT step is not established) | RT-04（L11 pooled 似然，排队）——待定RT-04 (L11 pooled likelihood, queued) — pending | REDTEAM_QUEUE.md RT-01–RT-13 | RT-04 待定RT-04 pending |
| FXT 公共模径向标定（RT-08，已完成 2026-10-16）FXT common-mode radial calibration (RT-08, done 2026-10-16) | 一个仍被 2σ 允许的核为中心径向残差可把 L12 上限抬到平衡密度，并使 L13 超压失效；牵动 L7 FXT、L8、L12、L20a nucleus-centred radial residual still allowed at 2σ can lift the L12 upper to the balance density and kill the L13 over-pressure; carries L7 FXT, L8, L12, L20 | RT-08 发现 FXT 图有一个未渐晕的平软项（0.49 cts/pix，0.5–1.3 keV，粒子模型 5×，各波段皆有，非 Si K 线）；估计器无法吸收；按响应误差传播 → L12 上限 0.93–1.07e-4RT-08 found an unvignetted flat soft term in the FXT map (0.49 cts/pix, 0.5–1.3 keV, 5× the particle model, in every band, not the Si K line); the estimator cannot absorb it; propagated as a response error → L12 upper 0.93–1.07e-4 | R004 空白场检验（luna + codexnim）判定平软项为加性背景；RT-16 + RT-19 后 a/a_CXB 缺口仍未解释（见顶行），仍需 FXT 团队的离轴渐晕验证the R004 blank-field test (luna + codexnim) settles the flat soft term as an additive background; after RT-16 + RT-19 the a/a_CXB gap is still unexplained (see the top row), so an FXT-team off-axis validation is still needed | REDTEAM_QUEUE.md RT-08/RT-12/RT-16/RT-19; I021 C3 systematic; R004 REPORT.luna.md + REPORT.codexnim.md; .agent/jobs/2026-10-16-redteam-rt16-a-cxb-gap/REPORT.codexnim.md; .agent/jobs/2026-10-17-redteam-rt19-source-model/REPORT.codexnim.md | 已完成 2026-10-16；R004 判定加性背景；a/a_CXB 缺口 UNEXPLAINED（open）done 2026-10-16; R004 settles an additive background; the a/a_CXB gap is UNEXPLAINED (open) |
| "PN 干净"（RT-09，已完成 2026-10-16）"PN is clean" (RT-09, done 2026-10-16) | PN-only 是 L7、L12（PN/FXT 斜率）与 L21 的独立检验；若 PN 也有大尺度系统，独立检验失效PN-only is the independent check for L7, L12 (the PN/FXT slope) and L21; if PN also carries a large-scale systematic, the independent check fails | RT-09：PN 无 MOS 式平面（0.073 ± 0.043，2σ < 0.17 PW/arcmin）但径向标定未测；N−S 径向差 +0.23 ± 0.13、全减 N/S +0.26 ± 0.13；一个 −0.245 PW/arcmin 的径向残差把 PN-only 晕上限抬到 7.2–14.1e-5 —— PN 单独不再排除平衡晕RT-09: PN shows no MOS-like plane (0.073 ± 0.043, 2σ < 0.17 PW/arcmin) but its radial calibration is untested; N−S radial difference +0.23 ± 0.13, all-minus-N/S +0.26 ± 0.13; a −0.245 PW/arcmin radial residual lifts the PN-only halo upper to 7.2–14.1e-5 — PN alone no longer excludes the balance halo | X-only PN 图 + 逐指向径向项拟合（见顶行 PN 径向标定）an X-only PN map with a per-pointing radial term (see the top PN radial-calibration row) | REDTEAM_QUEUE.md RT-09; E085 | 已完成 2026-10-16；径向项 OPENdone 2026-10-16; radial term OPEN |
| 晕锚（C7）：FXT 公共模渐晕未测Halo anchor (C7): common-mode FXT vignetting untested | L13–L17 全部随 P_halo(47 kpc) 缩放L13–L17 all scale with P_halo(47 kpc) | 两台 FXT 模块一致；PN-only XMM 曾被认为独立排除平衡晕 —— RT-09 后不再成立（一个 −0.245 PW/arcmin 的径向残差即可把 PN-only 上限抬到 7.2–14.1e-5）both FXT modules agree; PN-only XMM was thought to exclude the balance halo independently — after RT-09 this no longer holds (a −0.245 PW/arcmin radial residual lifts the PN-only upper to 7.2–14.1e-5) | FXT 空白场/团场检验（只读）an FXT blank/cluster-field test (read-only) | I021 C7; E084 | OPEN |
| L21 轴候选（PN 55–66 kpc）L21 axis candidate (PN 55–66 kpc) | PN C = +2.8 ± 1.0，FXT +0.6 ± 0.8，仪器间 ~2σ 不一致PN C = +2.8 ± 1.0, FXT +0.6 ± 0.8, a ~2σ instrument disagreement | 候选，未确认candidate, unconfirmed | PN N/S 组、FXT A/B、一个谱检验（C12）PN N/S groups, FXT A/B, a spectral test (C12) | E086; I021 C12 | CANDIDATE |
骨架图Skeleton
来源Source: I021_s09_paper_plan_v2.md (papers, checks C1–C12)
未来 1–4 周计划（按 I021 与 session 2026-10-15）Plan for the next 1–4 weeks (per I021 and session 2026-10-15)
- C2 已完成（2026-10-15）。把 PN+FXT 的 kT 与 O/Fe 更新到页 v10 面板 D/P，并采用 A1 的能量措辞。C2 is done (2026-10-15). Update page v10 panels D/P with the PN+FXT kT and O/Fe, and adopt the A1 energy wording.
- 修 E098 行 61 措辞与 I021 L15（按 A1）；改正页 v10 面板 P 数字。Fix the E098 line-61 wording and I021 L15 (per A1); correct the page v10 panel P numbers.
- 用 2′ MCG 9-17-9 掩罩重拟 WWLM（C2 后续）。Refit WWLM with a 2′ MCG 9-17-9 mask (C2 follow-up).
- 在引擎副本中修复合并 on-off 似然（每相机一个共享 BG 项），并重拟 E096 pooled O/Fe 作比较；在此之前 L11 pooled 区间保持 PROVISIONAL/未校准。Fix the pooled on-off likelihood in an engine copy (one shared BG term per camera) and refit the E096 pooled O/Fe once for comparison; until then L11 pooled intervals stay PROVISIONAL/uncalibrated.
- C1 + C3（图集与端点系统学），然后 C5、C7；C10 文献刷新后起草论文 I。C1 + C3 (map set and extent systematics), then C5, C7; draft Paper I after the C10 literature refresh.
- C6 为论文 II（与 Liu et al. 团队，D2 之后）。C6 for Paper II (with the Liu et al. team, after D2).
本次更新What's new
| 日期Date | 变更Change |
|---|---|
| 2026-10-16 | 阶段 7（R004 空白场翻案 + RT-14 + RT-15）：FXT 平软项判定为加性背景 —— 空白场检验（R004，luna + codexnim 两个执行者一致，21 个空白高纬场 877 ks）给出 b/粒子 = 5.33 ± 0.12（A）/ 5.07 ± 0.13（B）（与 NGC 3079 的 5× 一致）、γ（逐场散差均值）1.41 ± 0.12（A）/ 1.12 ± 0.09（B）（luna）与 1.25 ± 0.33（A）/ 1.54 ± 0.25（B）（codexnim），两者都不低于 1、只有约 35–50% 的预期 CXB 沿 v1.30 剖面，因此"v1.30 渐晕太陡"不被支持，RT-12 的响应误差倾向被推翻。L12 翻为加性读法：2σ 上限 1.04–1.81e-4、平衡晕 1.4e-4 在 2σ 内不被排除（响应误差敏感性 0.93–1.07e-4、最佳 7.95–8.67e-5）。L13 翻为 q ~1.3–1.8（响应误差情形 q ~2.0–2.1），状态 CONDITIONAL。RT-14：能量缺口仍在，可见风供 WSW 壁热能 1/7–1/14（响应误差情形 1/7–1/18）。RT-15：15′ 凸起排第 4/72（12–17′）与 7/72（14–16′，~1.4–1.5σ），小轴两扇区骑在平均上，环凸起 +0.56 PW 为 16 半径中最大但不显著；L11 太阳排除 ~2.3σ（pooled，C2 后）。指向框架（R005）：RA_PNT 2.29–2.32′ N（24 ObsID）与核到光轴偏移 0.2–0.76′（中值 0.5′，84 个成图 ObsID，roll 1–360°）分开陈述；瞄准点偏移论证从全站删除。新发现：两个 FXT 模块相隔 90° 安装，FXT-B 有相干探测器坐标方位偶极（局部计数 6.4%）→ NGC 3079 指向 +9–12% 北向超出（12–17′ 不对称的候选贡献；2026-10-17 RT-17 已否证，见 RT-17 行）。首页新增组会一页纸（id="meeting-card"）与 <main>；未决问题表顶行改为 R004 结论，新增"NGC 3079 的 a/a_CXB 与空白场不一致"与"FXT-B 探测器坐标方位偶极"两行。RT-04 排队。Stage 7 (the R004 blank-field reversal + RT-14 + RT-15): the FXT flat soft term is settled as an additive background — the blank-field test (R004, luna + codexnim agree, 21 blank high-latitude fields, 877 ks) gives b/particle = 5.33 ± 0.12 (A) / 5.07 ± 0.13 (B) (matching NGC 3079's 5×), a field-scatter mean γ of 1.41 ± 0.12 (A) / 1.12 ± 0.09 (B) (luna) and 1.25 ± 0.33 (A) / 1.54 ± 0.25 (B) (codexnim), neither below 1, and only about 35–50% of the expected CXB follows the v1.30 profile, so "the v1.30 vignetting is too steep" is not supported and RT-12's response-error preference is overturned. L12 flips to the additive reading: a 2σ upper of 1.04–1.81e-4 with the balance halo (1.4e-4) not excluded within 2σ (the response-error sensitivity is 0.93–1.07e-4, best 7.95–8.67e-5). L13 flips to q ~1.3–1.8 (the response-error case q ~2.0–2.1), status CONDITIONAL. RT-14: the energy deficit survives — the visible wind supplies 1/7–1/14 of the WSW wall thermal energy (1/7–1/18 in the response-error case). RT-15: the 15′ bump ranks 4/72 at 12–17′ and 7/72 at 14–16′ (~1.4–1.5σ), the two minor-axis sectors straddle the mean, and the ring bump +0.56 PW is the largest of 16 radii but not significant; L11's solar rejection is ~2.3σ (pooled, after C2). The pointing frame (R005): RA_PNT 2.29–2.32′ N (24 ObsIDs) and the nucleus-to-optical-axis offset 0.2–0.76′ (median 0.5′, 84 map ObsIDs, rolls 1–360°) are now stated separately; the aim-point-offset argument is deleted site-wide. New finding: the two FXT modules are mounted 90° apart and FXT-B carries a coherent detector-frame azimuth dipole (6.4% of the local rate) → a +9–12% north excess for the NGC 3079 pointings (a candidate contributor to the 12–17′ asymmetry; rejected on 2026-10-17 by RT-17, see the RT-17 row). The home page gains a group-meeting one-pager (id="meeting-card") and <main>; the open-problems top row is rewritten to the R004 verdict and two rows are added ("NGC 3079's a/a_CXB disagrees with the blank fields" and "the FXT-B detector-frame azimuth dipole"). RT-04 queued. |
| 2026-10-15 | 手册初建（H-N3079-001）：§3 结果、§4 讨论、L1–L21 台账、§1/§2 占位。吸收 2026-10-15 的 C2（无 MOS2 外区谱）、P1 氧预算、P2 晕静力学、P4 NGC 3073、P5 MCG 9-17-9、A1 能量审计、R1 红队、L1 AGN vs 星暴、L2 混合层与 FUV 的结果。First handbook build (H-N3079-001): §3 Results, §4 Discussion, the L1–L21 ledger, §1/§2 placeholders. It folds in the 2026-10-15 results: C2 (outer spectra without MOS2), P1 oxygen budget, P2 halo hydrostatics, P4 NGC 3073, P5 MCG 9-17-9, the A1 energy audit, the R1 red team, L1 AGN vs starburst, L2 mixing layers vs FUV. |
| 2026-10-15 | 阶段 2（评审更正，并入正文）：红队 RT-01/RT-02/RT-03/RT-06 结果全文吸收 —— L7 改为"测到 ~55 kpc；55–66 kpc 候选（2.0–2.9σ 随系统学）；66 kpc 外上限 B < 0.75 PW"；L9 边缘形状 MEASURED、3-D 空心烟囱 INFERRED（转折 26–43 kpc）；L13 误差扩为 2.95 ± 0.84（含 28% 晕误差），持平依赖 z > 14 kpc 截断，ENE 2.6–3.6，最不利叠加仍 1.41 > 1。§4.8 三层表重排："就绪 = MEASURED 且无未过的门"，本版无合格行。图修复：fig_wall_profile 晕参考线（10× 偏高，改为 r = z 剖面并在图注说明）、fig_l7_zones 标注移位、fig_kt_ofe 标签/图例/网格边 ≥ 标记。首页首屏（1280×800）改为目标表+未决问题表优先。RT-04/RT-05/RT-07 待定。Stage 2 (review corrections folded into the text): the red-team RT-01/RT-02/RT-03/RT-06 results are absorbed throughout — L7 becomes "measured to ~55 kpc; 55–66 kpc a candidate (2.0–2.9σ depending on systematics); a limit B < 0.75 PW beyond 66 kpc"; L9 edge shape MEASURED, 3-D hollow chimney INFERRED (break 26–43 kpc); L13 error widened to 2.95 ± 0.84 (28% halo error included), flatness depends on the z > 14 kpc cut, ENE 2.6–3.6, the most hostile stack still 1.41 > 1. The §4.8 three-tier table is rebuilt on "ready = MEASURED and no open gate"; no row qualifies in this version. Figure fixes: fig_wall_profile halo reference line (10× too high; replaced by the r = z profile with a caption note), fig_l7_zones annotations moved, fig_kt_ofe labels/legend/grid-edge ≥ markers. The home first screen (1280×800) now leads with the objectives and open-problems tables. RT-04/RT-05/RT-07 pending. |
| 2026-10-15 | 阶段 3（RT-05/RT-07 分诊并入）：L12 由测量降为上限（n_H < 0.9–1.4e-4，2σ；FXT 1.63σ、PN −0.33σ，非探测；公共模径向残差可抬至平衡密度 1.43e-4）；L13 改为 CONDITIONAL 于 FXT 径向标定（最佳拟合晕处 2.95 ± 0.84；0.2 keV 一致发射率 → ~2.55；最不利叠加在公共模 2σ 下 ~0.85）；L14 措辞改写（2.7e40 是 HK2020 可见成分；"可见风供能 1/7–1/14"；弃 ×10–40 与"无法建造"；壁质量可为扫积晕气体）；L15 改为 CONDITIONAL（单相晕、光滑壁、外延：R 80–100 kpc 给 3.40–4.94e57 erg、结束 51–76 Myr 前；R 55–66 kpc 给 1.4–2.5e57 erg、4.6–8.0e41 erg/s、结束 17–42 Myr 前；另受 C2 与 L12 约束）。§4.8 三层表重排（L12/L13/L15 移入"仅讨论级"）；未决问题表加 RT-08（FXT 公共模径向标定，在跑）与 RT-09（PN 干净性，在跑）行，RT-05/RT-07 标记完成，RT-10 排队。图修复：fig_kt_ofe 图例移入绘图区（不再压标题）、标签写全"wall/int."；fig_wall_profile 左板加多刻度标签、晕画为上限带。RT-04/RT-08/RT-09/RT-10 待定。Stage 3 (the RT-05/RT-07 triages folded in): L12 downgraded from a measurement to a limit (n_H < 0.9–1.4e-4, 2σ; FXT 1.63σ, PN −0.33σ, not a detection; a common-mode radial residual can lift it to the balance density 1.43e-4); L13 becomes CONDITIONAL on the FXT radial calibration (2.95 ± 0.84 at the best-fit halo; a consistent 0.2 keV emissivity → ~2.55; the hostile stack ~0.85 under the common-mode 2σ); L14 reworded (2.7e40 is HK2020's visible component; "the visible wind supplies 1/7–1/14 of the energy"; ×10–40 and "cannot build" dropped; the wall mass can be swept-up halo gas); L15 becomes CONDITIONAL (single-phase halo, smooth walls, reach: R 80–100 kpc gives 3.40–4.94e57 erg ending 51–76 Myr ago; R 55–66 kpc gives 1.4–2.5e57 erg and 4.6–8.0e41 erg/s ending 17–42 Myr ago; also gated by C2 and L12). The §4.8 three-tier table rebuilt (L12/L13/L15 moved to "discussion only"); the open-problems table gains RT-08 (FXT common-mode radial calibration, running) and RT-09 (PN cleanliness, running) rows, RT-05/RT-07 marked done, RT-10 queued. Figure fixes: fig_kt_ofe legend moved inside the axes (no longer over the title), labels spelled "wall/int."; fig_wall_profile left panel multi-tick labels, halo drawn as a limit band. RT-04/RT-08/RT-09/RT-10 pending. |
| 2026-10-16 | 阶段 4（RT-08..RT-11 并入）：L12 改为上限阶梯 —— n_H(47) < 0.9–1.8e-4（2σ），取决于 FXT 背景模型：0.87e-4（无额外项）/ 0.93–1.07e-4（平软项按响应误差，RT-08）/ 1.04–1.81e-4（按加性背景，RT-11；最佳拟合 0.93–1.38e-4）；平衡晕在加性模型下不被排除，哪个模型正确待定（PI/FXT 标定）。L13 改为 CONDITIONAL 于 FXT 背景模型：无额外项时 2.95 ± 0.84；加性项下最佳拟合 q ~1.3–1.8、平衡在 2σ 内（Claude 算术）。L2b 改写：PN 无 MOS 式平面（< 0.17 PW/arcmin，2σ）但径向标定未测；PN-only 不再单独排除平衡晕。L8（RT-10）：无环平均薄壳帽成立，但作为 L15 约束是条件性的（帽亮 ~ ∝ n_H²，仅当 n_H(47) ≥ ~4.5e-5 排除 R 55–70 kpc 壳；标准壳密度下 90–100 kpc 远壳即使 5e-5 也不排除）；L20 保持候选（−1.0 PW）；L21 非帽证据、被 PN 模板系统学削弱 ~0.4 PW。fig_wall_profile 2σ 上限带扩到 ×3.3（1.81e-4/5.5e-5），图注说明每条边属于哪个模型。未决问题表：新顶行"FXT 平软项：加性背景还是响应误差？"与"PN 径向标定"行；RT-08..RT-11 标记完成（2026-10-15/16）。RT-04 待定。Stage 4 (RT-08..RT-11 folded in): L12 becomes a limit ladder — n_H(47) < 0.9–1.8e-4 (2σ), depending on the FXT background model: 0.87e-4 (no extra term) / 0.93–1.07e-4 (flat soft term as a response error, RT-08) / 1.04–1.81e-4 (as an additive background, RT-11; best fit 0.93–1.38e-4); the balance halo is not excluded under the additive model, and which model is right is open (PI/FXT calibration). L13 becomes CONDITIONAL on the FXT background model: 2.95 ± 0.84 with no extra term; under the additive term the best-fit q is ~1.3–1.8 and balance lies inside 2σ (Claude arithmetic). L2b rewritten: PN shows no MOS-like plane (< 0.17 PW/arcmin, 2σ) but its radial calibration is untested; PN-only no longer excludes the balance halo alone. L8 (RT-10): the no rim-averaged thin-shell cap stands, but as an L15 constraint it is conditional (cap brightness ~ ∝ n_H²; excludes an R 55–70 kpc shell only if n_H(47) ≥ ~4.5e-5; with the standard shell density a 90–100 kpc far shell is not excluded even at 5e-5); L20 stays a candidate (−1.0 PW); L21 is not a cap piece and is weakened ~0.4 PW by PN template systematics. fig_wall_profile 2σ band widened to ×3.3 (1.81e-4/5.5e-5), the caption says which model each edge belongs to. Open problems: a new top row "the FXT flat soft term: additive background or response error?" and a "PN radial calibration" row; RT-08..RT-11 marked done (2026-10-15/16). RT-04 pending. |
| 2026-10-16 | 阶段 6（PI 范围变更 12:04Z：新增 §2 数据处理、论文 I 措辞建议、FXT 15′ 检验）：§2 数据处理写入 intro_data.html（§1 仍为占位）—— 2.1 XMM（12 个归档指向、剔除 0110930201 与 0802710401、ODF 重处理与相机级耀发 GTI、PN gtibug B003 与"clean"文件未做耀发筛选 B012、MOS1 CCD4 软超出 L2、MOS 卷帘梯度 L2b 与"r ≳ 5′ 只用 PN"规则）、2.2 FXT（指向数两个来源都列：目录 59 个独立 ObsID / v1.30 再处理 86 个指向、地图用 84 个曝光每模块；FXTDAS/CALDB v1.30 与其"updated the vignetting calibration"；瞄准点 2.29–2.32′ N、roll 248–313°；465 个源扣除；平软项 0.49 cts/pix 与正在跑的空白场渐晕检验）、2.3 Chandra（只用源普查，19 个伪条目 L4）、2.4 图与单位（PW、4″/8″ 网格、PAGESET、CLMASK、E080 补丁、E078 blob、场散因子 k）、2.5 on-off 谱引擎、2.6 三条已知局限。新增页面 wording.html（措辞建议，标注"建议（未经评审）"；I022 标题选项 A/B/C、逐条摘要措辞、避免词、故事顺序、15′ 讨论文字），各页导航加入口。FXT 15′ 检验（PI 问题）：14–16′ 的凸起来自 NNW 盘边（PA 345 +2.61 PW，72 扇区中排第 4，~1.4σ），属宽南北不对称；小轴（帽应在处）与全方位角平均持平；瞄准点偏移预测相反符号；不是帽证据，L20 仍是唯一类帽候选（新图 fig_fxt15）。未决问题表新增两行：FXT 空白场渐晕检验（运行中）与 FXT 12–17′ 南北不对称（OPEN）。RT-04 待定。Stage 6 (the PI scope change of 12:04Z: a real §2 data reduction, Paper I wording suggestions, the FXT 15′ check): §2 Data reduction written into intro_data.html (§1 stays a placeholder) — 2.1 XMM (12 archived pointings, 0110930201 and 0802710401 excluded, ODF reprocessing with camera-level flare GTIs, the PN gtibug B003 and the un-flare-filtered "clean" files B012, the MOS1 CCD4 soft excess L2, the MOS roll gradient L2b and the "PN only at r ≳ 5′" rule), 2.2 FXT (both pointing counts quoted: 59 independent ObsIDs in the inventory against 86 pointings in the v1.30 reprocessing, with the maps using 84 exposures per module; FXTDAS/CALDB v1.30 and its "updated the vignetting calibration"; the aim point 2.29–2.32′ N, rolls 248–313°; 465 sources subtracted; the 0.49 cts/pix flat soft term and the running blank-field vignetting test), 2.3 Chandra (source census only, 19 spurious entries L4), 2.4 maps and units (PW, the 4″/8″ grids, PAGESET, CLMASK, the E080 patches, the E078 blobs, the field-scatter factor k), 2.5 the on-off spectral engine, 2.6 three known limits. New page wording.html (wording suggestions, labelled "Suggestion (not reviewed)"; I022's title options A/B/C, the claim-by-claim abstract wording, the words to avoid, the story order and the 15′ discussion text), with a nav entry on every page. The FXT 15′ check (the PI question): the 14–16′ bump comes from the NNW disk side (PA 345 +2.61 PW, rank 4 of 72 sectors, ~1.4σ) and belongs to a broad north-south asymmetry; the minor axis (where a cap would lie) is level with the all-azimuth mean; the aim-point offset predicts the opposite sign; not cap evidence, L20 remains the only cap-like candidate (new figure fig_fxt15). The open-problems table gains two rows: the FXT blank-field vignetting test (running) and the FXT 12–17′ north-south asymmetry (OPEN). RT-04 pending. |
| 2026-10-16 | 阶段 5（RT-12/RT-13 并入）：RT-12：FXT 平软项倾向为响应/渐晕误差 —— 1.3–4 keV 渐晕部分 ~0 而 v1.30 响应下标准 CXB 预期在 17–51σ（两模块）；平软项谱形为 CXB（χ² 6.3/3）而非粒子（255/3）；CALDB 渐晕文件混合模块、无公开离轴验证；未解：只携带预期 CXB 计数的 20–30%。L12 阶梯出现倾向情形：响应误差（倾向）下 2σ 上限 0.93–1.07e-4、最佳 7.95–8.67e-5、平衡晕在该模型下被 2σ 排除；加性敏感性 1.04–1.81e-4、平衡不被排除；无额外项 0.87e-4。L13：倾向模型下最佳拟合 q ~2.0–2.1（2.95 × 5.77/7.95–8.67，Claude 算术）、仍 > 1；加性敏感性下 q ~1.3–1.8、平衡在 2σ 内；仍 CONDITIONAL。首页段落改为"倾向 FXT 模型下超压约 2–3 倍（若平软项是加性背景则约 1.3–1.8）"。L10 → CONDITIONAL（RT-13）：kT 0.25–0.29（8–28 kpc）/ 0.18–0.26 keV（41–61 kpc）；一致 C2 PN+FXT 区集内 17–28 → 41–61 kpc 仅 +0.03 ± 0.03 keV（1.1σ，仅统计）；C5 未过（参考区移动外壁 0.08–0.10 keV）；39 kpc 处 kT 突变未确立 —— L13 依赖的 kT step 随之动摇。§4.8：L10 移入"仅讨论级"；未决问题表顶行改为"10–25′ 离轴的 FXT 响应（渐晕）"（倾向解释 + 离轴验证为决定性输入），新增"C5：kT 的前景/参考区系统学"行；RT-12/RT-13 标记完成。RT-04 待定。Stage 5 (RT-12/RT-13 folded in): RT-12: the FXT flat soft term is favoured to be a response/vignetting error — in 1.3–4 keV the vignetted part is ~0 although a standard CXB through the v1.30 response predicts it at 17–51σ (both modules); the flat term has a CXB spectrum (χ² 6.3/3), not a particle one (255/3); the CALDB vignetting file is mixed-module with no published off-axis validation; open: it carries only 20–30% of the expected CXB counts. The L12 ladder gains a favoured case: under the response error (favoured) the 2σ upper is 0.93–1.07e-4 with best 7.95–8.67e-5 and the balance halo is excluded at 2σ in that model; the additive sensitivity is 1.04–1.81e-4 with balance not excluded; no term 0.87e-4. L13: in the favoured model the best-fit q is ~2.0–2.1 (2.95 × 5.77/7.95–8.67, Claude arithmetic), still > 1; under the additive sensitivity q is ~1.3–1.8 with balance inside 2σ; still CONDITIONAL. The home paragraph now reads "over-pressured ~2–3× in the favoured FXT model (~1.3–1.8 if the flat soft term were an additive background)". L10 → CONDITIONAL (RT-13): kT 0.25–0.29 (8–28 kpc) / 0.18–0.26 keV (41–61 kpc); in a consistent C2 PN+FXT zone set 17–28 → 41–61 kpc is only +0.03 ± 0.03 keV (1.1σ, stat only); C5 open (the reference choice moves the outer walls by 0.08–0.10 keV); the 39 kpc kT step is not established — the kT step L13 leans on is shaken. §4.8: L10 moves to "discussion only"; the open-problems top row becomes "the FXT response (vignetting) at 10–25′ off-axis" (the favoured explanation, with the off-axis validation as the decisive input), a new "C5: foreground/reference systematics for kT" row is added; RT-12/RT-13 marked done. RT-04 pending. |
数据截至Data as of: 2026-10-17T03:2xZ （阶段 8b：RT-17（R015）判定 FXT-B 探测器偶极不是 12–17′ 南北不对称的解释；RT-16（R014）+ RT-19（R016）判定 a/a_CXB 缺口未解释；合并 γ 改为逐场散差均值。最新记录 = RT-16/RT-17/RT-19 的 REPORT.codexnim.md 与 07_REVISION_LOG.md R014/R015/R016；仍待定 = 未解释的 a/a_CXB 缺口、1.8–1.9σ 的南北不对称、FXT 团队离轴验证；RT-04 排队）(stage 8b: RT-17 (R015) rules out the FXT-B detector dipole as the explanation for the 12–17′ north-south asymmetry; RT-16 (R014) + RT-19 (R016) leave the a/a_CXB gap unexplained; the pooled γ is replaced by the field-scatter means. Newest records = the RT-16/RT-17/RT-19 REPORT.codexnim.md and 07_REVISION_LOG.md R014/R015/R016; still Pending = the unexplained a/a_CXB gap, the 1.8–1.9σ north-south asymmetry and an FXT-team off-axis validation; RT-04 queued)