1. 已确立的（Trusted）
1. Established (Trusted)
- v18 是科学基线：APEC 环境修复后、skyarea “修正之修正”确认原版正确（T80）。
- v18 is the science baseline: after the APEC environment fix and the skyarea correction-of-correction confirming the original was right (T80).
- 径向温度数值本身：R2–4 = 0.715±0.025，R4–6 = 0.730±0.104，R6–8 = 0.761±0.069 keV（扇区平均，图见下）。
- The radial temperature values themselves: R2–4 = 0.715±0.025, R4–6 = 0.730±0.104, R6–8 = 0.761±0.069 keV (sector means; figure below).
- 背景系统差 < 0.02 keV（T58）：3B vs 4.background 的 kT 差均值 0.0071 keV。
- Background systematics < 0.02 keV (T58): mean kT difference 0.0071 keV between 3B and 4.background.
- SNR ≥ 25（dof ≥ 8）才能可靠测 kT（T61，来自 247 结果审计）。
- SNR ≥ 25 (dof ≥ 8) is required for a reliable kT (T61, from the 247-result audit).
- P0 检验与 D45 NO-GO 本身是可信的实现/审计结果（T138/T143/T144）。
- The P0 test and the D45 NO-GO themselves are trusted implementation/audit results (T138/T143/T144).
- 31 区 WVT 几何/提取审计可信（T86–T88），但不是 kT 科学结果。
- The 31-region WVT geometry/extraction audit is trusted (T86–T88), but it is not a kT science result.
v18 扇区平均径向温度。灰色点只用单一方向（provisional）。红色虚线是控制器重算的加权常数 0.728 keV（p = 0.7097）。 The v18 sector-mean radial temperature. Grey points use a single direction only (provisional). The red dashed line is the controller-recomputed weighted constant 0.728 keV (p = 0.7097).
figure/radial_thermo_profiles_v18/radial_kT_profile_v18_sector_mean.csv；I061（p 值）
熵与冷却时间（v15 beta 密度 + v18 kT）。K 从 67 升到 407 keV cm²；等压冷却时间 10 → 137 Gyr，远超宇宙年龄——外区气体不会冷却。 Entropy and cooling time (v15 beta density + v18 kT). K rises from 67 to 407 keV cm²; the isobaric cooling time 10 → 137 Gyr, far beyond the age of the Universe — the outer gas will not cool.
figure/radial_thermo_profiles_v18/entropy_cooling_time_profile_v18_v15beta.csv
2. 暂定的（Provisional）
2. Provisional
- “内冷外热”梯度：常数模型 p = 0.7097，斜率 1.37σ —— 数据不拒绝等温（T133，D41 降级）。
- The “inner-cooler/outer-warmer” gradient: constant model p = 0.7097, slope 1.37σ — the data do not reject isothermality (T133, downgraded by D41).
- R4–6 方向差（风侧 − 盘侧 ≈ +0.15 keV，1.5–1.8σ）是唯一持续候选（I061）。
- The R4–6 contrast (wind − disk ≈ +0.15 keV, 1.5–1.8σ) is the only persistent candidate (I061).
- E030 的 8 区统计能力结果：盘侧 12.9 arcmin² 达 10%；北侧 11.6 arcmin² 仅 32.2%。
- E030's 8-region statistical capability: disk-side 12.9 arcmin² reaches 10%; north 11.6 arcmin² only 32.2%.
3. 有疑问的（Questioned）
3. Questioned
- Chandra+XMM 联合管线 8/20 PASS，R6–8 单元非物理：X1/X2 = 0.109 keV 开区间，J1/J2 双分支（0.14–0.20 与 1.5–5.9 keV），C1 = 2.04 keV。
- The Chandra+XMM joint pipeline is 8/20 PASS with unphysical R6–8 cells: X1/X2 = 0.109 keV open interval, J1/J2 bimodal (0.14–0.20 and 1.5–5.9 keV), C1 = 2.04 keV.
- ~13′ 面亮度过亮（Q-C）— 已判定为假象（Provisional，I065/T149）：v18 表保存的 flux 本身已是每 arcmin² 的面亮度（B020，Trusted）；旧的 flux/area 曲线把面积除了两次，抬高了外侧小面积 bin。按正确面亮度，Angle2 11–14′/8–11′ = 1.37；把外区 kT 冻结到 0.75 keV 后降到 1.07（Δstat +1.30）→ 剩余凸起与弱信号区源/背景简并一致。独立 QPB 扣除图像（0.5–2 keV）外/内 = 1.041±0.068，图像不显示超出。背景星系团 WARP J1240.4−1147（z = 0.193，12.3′，位于 R11-13 Angle2 内）在 v18 会话中加 APEC 成分拟合 norm 为 0。开放问题：外环谱的拟合方法（更宽 kT 范围 + profile 误差，或绑定径向模型，或用偏置指向）；15–30′ 对比（R′）尚无法从保存产品检验。
- The ~13′ surface-brightness excess (Q-C) — resolved as an artefact (Provisional, I065/T149): the v18 table's saved flux is already a per-arcmin² surface brightness (B020, Trusted); the old flux/area curve divided by area twice, inflating the small outer bins. With the correct SB, Angle2 11–14′/8–11′ = 1.37; freezing the outer kT to 0.75 keV lowers it to 1.07 (Δstat +1.30) → the residual bump is consistent with source/background degeneracy at low S/N. Independent QPB-subtracted images (0.5–2 keV) give outer/inner = 1.041±0.068 — the images show no excess. The background cluster WARP J1240.4−1147 (z = 0.193, 12.3′, inside R11-13 Angle2) fits to a norm of 0 as an added APEC in the v18 session. Open: the outer-ring spectral method (wider kT range + profile errors, or a tied radial model, or the offset pointings); the 15–30′ comparison (R′) is not yet testable from saved products.
- 多套拟合策略无共同验证（Q-A 本身）：注入–恢复试点（R2-4 Angle1，50 实现 × 2 策略）偏差 +0.0046/−0.0027 keV，条件覆盖率 0.76/0.80 — 无策略被认证（T153）。
- The fitting strategies lack common validation (Q-A itself): the injection–recovery pilot (R2-4 Angle1, 50 realizations × 2 strategies) gives biases +0.0046/−0.0027 keV and conditional coverages 0.76/0.80 — no strategy is certified (T153).
Q-C 的证据（修正后）：实线 = v18 表的裸面亮度（每 arcmin²，B020 正确估计量）；虚线灰 = 旧的 flux/area 双重归一曲线（Deprecated，B020）。橙色带是 0900170101 视场边缘（~13–15′）。Angle2 在 11–14′ 仍比 8–11′ 高 1.37 倍，但冻结 kT 后降到 1.07，且图像不显示超出 → 假象（I065）。 The Q-C evidence (corrected): solid = the v18 table's bare surface brightness (per arcmin², the correct estimator per B020); dashed grey = the old flux/area double-normalised curve (Deprecated, B020). The orange band is the FoV edge of 0900170101 (~13–15′). Angle2 at 11–14′ is still 1.37× the 8–11′ level, but freezing kT lowers it to 1.07 and the images show no excess → artefact (I065).
v18_profile_analysis/v18_complete_parameters.csv（flux_0.5-2.0keV 列 = 每 arcmin²；B020）；I065；E034
4. 已废弃的（Deprecated）
4. Deprecated
- 全部 100 个 pre-APEC-fix 自适应 bin kT（D17）。
- All 100 pre-APEC-fix adaptive-bin kT values (D17).
- HR→kT 带比值方法（T23）与 0/90/180/270 象限图（T67）。
- The HR→kT band-ratio method (T23) and the 0/90/180/270 quadrant map (T67).
- 把 Fe-L 色图像素当温度（T121）。
- Treating Fe-L colour pixels as temperatures (T121).
- 一切用 flux/area_arcmin2 构建的面亮度/EM/密度轮廓（B020：双重归一，Deprecated）；旧 all-method 面亮度表（面积换算错 484 倍，Deprecated）。
- Any SB/EM/density profile built from flux/area_arcmin2 (B020: double normalisation, Deprecated); the old all-method SB tables (area conversion wrong by 484×, Deprecated).
5. 研究进展（2026-10-15 更新）
5. Research progress (updated 2026-10-15)
| job / 事项Job / item | 状态与主要结果（读取于 2026-10-15，多数未经评审）Status and main result (read 2026-10-15; most not yet reviewed) | ||
|---|---|---|---|
| fit-reliability-audit（Q-A，sol 完成） | fit-reliability-audit (Q-A, sol done) | 154 个会话免重拟合恢复；策略清单 12,209 条逐行引证；试点 50 实现 × 2 策略：偏差 +0.0046/−0.0027 keV，条件覆盖率 0.76/0.80；发现联合拟合简并（共享 kT + 各仪器自由 norm → Chandra norm→0）。无策略被认证（T153）。 | 154 sessions restored without refitting; a 12,209-line-quoted strategy inventory; pilot 50 realizations × 2 strategies: biases +0.0046/−0.0027 keV, conditional coverages 0.76/0.80; a concrete joint-fit degeneracy found (shared kT + free per-instrument norms → Chandra norm→0). No strategy certified (T153). |
| sb-profile-13arcmin + cluster-13arcmin-test（Q-C，完成） | sb-profile-13arcmin + cluster-13arcmin-test (Q-C, done) | 13′ 超出 = 归一错误 + 弱信号简并假象（I065，见上）；WARP 星系团拟合 norm 为 0，非成因。 | The 13′ excess = a normalisation + weak-signal degeneracy artefact (I065, above); the WARP cluster fits to a norm of 0 and is not the cause. |
| chandra-xmm-joint-status（Q-B，只读审计完成） | chandra-xmm-joint-status (Q-B, read-only audit done) | 8/20 PASS、12 PENDING 核实；停止原因 = R2 MOS2 面积探针运行时 HEASoft 未初始化；无被接受的单元。旧严格交集管线搁置（D46）。 | 8/20 PASS and 12 PENDING confirmed; stop cause = HEASoft not initialised during the R2 MOS2 area probe; no accepted cell. Old strict-intersection pipeline parked (D46). |
| gti-lightcurve-audit（完成） | gti-lightcurve-audit (done) | 保留 2.filter（I066）；MOS1 静态过亮（见新开问题）。 | Keep 2.filter (I066); the MOS1 quiescent excess (see new open problems). |
| source-mask-design（完成，2 阶段） | source-mask-design (done, 2 stages) | 45′ 内 958 个 X 射线源；mask 规则 v2（x=1：点源中位 30″，最大 360″；WARP 242″；1eRASS 星系团 644″）；遮蔽无法把残余压到 <5%，必须建模残余（E035）。 | 958 X-ray sources within 45′; mask rule v2 (x=1: point sources median 30″, max 360″; WARP 242″; the 1eRASS cluster 644″); masking cannot push residuals below 5%, so the residual must be modelled (E035). |
| 2d-intensity-map（Q-D，进行中） | 2d-intensity-map (Q-D, running) | 清单与首批马赛克构建中（仅用现有产品；无新事件过滤）。 | Inventory and first mosaic being built from existing products only; no new event filtering. |
| chandra-xmm-perinst-profile（Q-B 新路线，进行中） | chandra-xmm-perinst-profile (Q-B new route, running) | 逐仪器支持路线（D46）：每仪器各自覆盖 + 面亮度跨仪器联结（交叉标定 0.7–1.3 自由），kT 0.2–3 keV，profile 网格 ≥40 点；preflight 中，尚无新温度。 | Per-instrument-support route (D46): each instrument its own coverage + SB linked across instruments (free cross-calibration 0.7–1.3), kT 0.2–3 keV, profile grids ≥40 points; in preflight, no new temperatures yet. |
| virgo-rereduction-s0（进行中） | virgo-rereduction-s0 (running) | 引擎 v1.7.8 基线复现（0900170101 ODF 356/356 校验和通过）；按 D47 每阶段只改一个设计元素。 | Engine v1.7.8 baseline reproduction (0900170101 ODF 356/356 checksums pass); one design element changed per stage per D47. |
.agent/jobs/2026-10-14-*/REPORT.*.md、2026-10-15-chandra-xmm-perinst-profile/{BRIEF,REPORT.sol}.md、cognition-os/sessions/2026-10-14.md（读取于 2026-10-15；除已入 cognition-os 的 I065/I066/B020/E034/E035/T148–T153/D46/D47 外均未经评审）