AEEA mobilizes research, data, and actions to prevent mass extinction events. 

减少全球灾难性风险，保护地球生命系统

We develop MEST-TPC/MEST-AI to test a cross-scale “steep–turning–flat” structural pattern across galaxies, lensing, CMB, and selected micro-domain datasets—publishing controls, reproducible pipelines, and falsifiable predictions on candidate invariants and long-term Earth-risk questions.

我们提出对偶张量不变性与结构场拟合框架（MEST-TPC/MEST-AI），在星系、透镜、CMB 与微观数据中检验“陡峭—拐点—平缓”的跨尺度结构形态，并以可复现对照研究追踪两项候选不变量及其对太阳系结构与灭绝周期问题的可证伪预测。 

AEEA Research Team advances a falsifiable cross-scale program grounded in a proposed Dual Tensor Conservation/Invariance principle. Building on this idea, we develop a computable structural-field fitting framework (MEST-AI / MEST-TPC) using a small family of comparable structural kernels (constrained tanh / logistic / arctan forms). Under unified uncertainty treatment and explicit control models, we test structural signatures across public datasets including galaxy rotation curves, strong gravitational lensing, CMB cold/hot spots and void profiles, and selected micro-domain scattering and high-energy spectral data.

Our working hypothesis is that nature may share a common cross-scale morphology—steep–turning–flat—with turning-point metrics that can be stress-tested and falsified. We further test whether the Solar System resides near a turning-point regime of the Milky Way, whether Solar System dynamics exhibits measurable flattening residuals under controls, and whether such turning-point environments can amplify periodic perturbations relevant to long-term Earth risk. We treat “~250 Myr extinction periodicity” as a testable statistical claim, publish reproducible pipelines, and invite independent replication.

AEEA 研究团队正在推进一条可证伪的跨尺度路径：以“对偶张量守恒/不变性”为核心原理，发展结构场方程与可计算拟合框架（MEST-AI / MEST-TPC），并在星系旋转曲线、强引力透镜、CMB 冷斑/热斑与空洞剖面，以及部分微观散射与高能谱数据中进行系统检验。我们提出自然界可能存在跨尺度的共同结构形态——陡峭—拐点—平缓，并在多数据域中寻找其一致的“拐点指标”与候选不变量（two candidate invariants）。

进一步地，我们提出并检验：太阳系可能位于银河系结构拐点附近，太阳系内部或存在速度/残差平坦化迹象；地球所处位置可能使轨道系统更易出现周期性扰动。由此出发，我们把“约 2.5 亿年灭绝周期及其与太阳系—银河系时间尺度的关系”作为重点检验对象，用统计鲁棒性测试与对照模型区分相关与因果，并以可复现发布作为研究底线。 

We propose a falsifiable working hypothesis: dark-sector effects may be represented as an effective description of spacetime structure/background dynamics, modeled as a background-wave/medium-like term within a structural-field framework. In this view, ordinary matter and the dark-sector contribution jointly shape the effective mass–potential profile of galaxies, giving rise to a quantifiable turning-point scale.

We further hypothesize—and will test under explicit controls—that orbital systems near such turning-point regimes may exhibit enhanced sensitivity to periodic drivers, producing statistically identifiable perturbation signatures (e.g., secular drift of orbital elements, spectral enhancement, or systematic residual patterns). If supported, this offers a candidate mechanism pathway linking astrophysical environments to long-term Earth-risk questions; if not, we will publish clear constraints and failure boundaries. 

我们提出一个可证伪的工作假说：暗部门效应可被视为时空结构/背景动力学的一种有效表述，并在结构场框架下用类“背景波/介质项”进行建模。在这种表述中，暗部门与普通物质共同决定星系的有效质量—势能分布，从而形成可量化的结构转折尺度（turning point）。

我们进一步提出并将检验：在星系结构转折区附近，轨道系统可能对外部或内部周期性驱动更敏感，表现为统计上可识别的扰动增强（例如轨道要素的长期漂移、频谱增强或残差结构变化）。如果该效应成立，它为“天体环境变化—地球长期风险”的研究提供一种候选机制路径；若在严格对照与不确定性传播下不成立，我们同样将给出明确的约束与失败边界。

AEEA Research: Falsifiable Cross-Scale Structure Science

We build and stress-test a unified structural framework (MEST-AI / MEST-TPC) across galaxies, lensing, CMB, and micro-domain datasets—publishing controls, negative results, and reproducible pipelines.

Falsifiable by design: every claim ships with controls and failure boundaries.

■ Reproducible outputs: indexed data + scripts + one-click figures/tables.

■ Cross-scale focus: from astrophysical structures to micro-domain observables.

Mini trust line（可信度条）

Open-data first · Control models required · Versioned releases · Independent review welcome

1.2 Three Data-Domain Cards（三张数据域卡片）

建议版式：三卡横排（桌面）/竖排（手机），每卡含一句定位 + 三条“我们做什么” + 按钮。

Card 1 — Galaxy Dynamics（星系动力学）

Title: Galaxy Rotation Curves
One-liner: Testing whether a shared “steep–turning–flat” structural profile can explain rotation-curve transitions under controls.
We do

• Unified fitting protocol (multi-start, uncertainty, residual diagnostics)
  
  

• Function-family benchmarks (tanh / logistic / arctan + constrained forms)
  
  

• Model selection and controls (AIC/BIC/CV; empirical profiles as baselines)
  
  

Button: Open Galaxy Validation → /research/domains/galaxies

Card 2 — Gravitational Lensing（引力透镜）

Title: Strong Lensing Systems
One-liner: Quantitative lensing tests under explicit control models, with reproducible error evaluation.
We do

• Standardized data pipeline (units, uncertainties, versioning)
  
  

• Comparative lens models (control vs structural-field models)
  
  

• Residual maps + error metrics + disconfirmation criteria
  
  

Button: Open Lensing Validation → /research/domains/lensing

Card 3 — CMB & Large-Scale Structure（CMB/空洞/冷斑热斑）

Title: CMB Structures & Voids
One-liner: Testing structural turning-point signatures in CMB cold/hot spots and void profiles with controls and null tests.
We do

• Profile extraction + uncertainty propagation
  
  

• Null models and sensitivity analyses (selection/threshold/foreground controls)
  
  

• Turning-point metrics and reproducible pipelines
  
  

Button: Open CMB/Voids Validation → /research/domains/cmb

1.3 Methods Block（方法区块）

Section title

Our Engine: MEST-AI + MEST-TPC

Short paragraph

We operationalize structural hypotheses using MEST-AI (automation for fitting, benchmarking, and reporting) and MEST-TPC (Mass–Energy–Spacetime Turning-Point Tensor Computation), a standardized approach to quantify turning points, transition scales, and cross-domain invariants under strict controls.

4 个方法卡（小卡）

1. Unified Data Pipeline
   
   

   • data sources → cleaning → units → uncertainty model → versioning
     
     

2. Reproducible Fitting Protocol
   
   

   • multi-start optimization, confidence intervals, residual diagnostics
     
     

3. Model Selection & Controls
   
   

   • AIC/BIC/CV/Bayesian evidence; baseline empirical models required
     
     

4. Integrity & Disconfirmation
   
   

   • publish negative results; define failure boundaries; independent review
     
     

Methods CTA

• Methods & Reproducibility Hub → /research/methods
  
  

• Benchmark Library → /research/benchmarks
  
  

• Open-Source Tools → /research/tools
  
  

1.4 Latest Updates（最新动态）

建议 3 条动态，显示日期 + 标题 + 1 句摘要 + 链接。内容可由你们后续发布填充。

Update 1
[YYYY-MM-DD] Fit-Function Benchmark v1 Released
Summary: tanh/logistic/arctan families compared under unified initialization and residual diagnostics.
Link → /research/updates/fit-benchmark-v1

Update 2
[YYYY-MM-DD] Two-Constants Validation Report v1 (Draft) Open for Review
Summary: cross-object consistency tests with explicit controls and failure boundaries.
Link → /research/updates/two-constants-v1-draft

Update 3
[YYYY-MM-DD] CMB Turning-Point Pipeline (Alpha) Published
Summary: profile extraction + null tests + reproducible figure generation.
Link → /research/updates/cmb-tpc-alpha

1.5 CTA Block（行动号召区块）

Title

Help Us Stress-Test the Science

3 个 CTA 卡

A) Join as Collaborator
Build methods, controls, and theory-to-observable mappings.
Button → /get-involved/collaborate

B) Contribute Data or Replications
Share datasets, replication notebooks, or independent benchmarks.
Button → /get-involved/contribute-data

C) Join as Independent Reviewer
Audit assumptions, controls, and reproducibility packages.
Button → /get-involved/reviewer

Sponsor CTA（赞助入口）

Support the 2026 Research Themes → /support/sponsorship

2) “可证伪命题”清单+ 验证入口链接

每条命题都用 Claim → What would falsify it → Minimum tests → Link 的固定格式。
这些内容考虑放在首页下方折叠区（Accordion）或 “Validation” 页面索引。

Claim 1 — Dual Tensor Conservation / Invariance as a Formal Principle

Claim: A dual tensor conservation/invariance principle can be axiomatized and yields testable observable relations.
Falsifies if: no coherent axiomatization leads to falsifiable implications, or implications fail across independent datasets.
Minimum tests: definitions → propositions → derivations → 3–5 predictions mapped to public datasets.
Verify: /research/themes/dual-tensor-invariance

Claim 2 — Three structural fit-function families are sufficient and comparable

Claim: A small family of structural kernels (e.g., tanh/logistic/arctan with constraints) provides stable, identifiable fits under unified protocols.
Falsifies if: multi-start solutions diverge, parameters are non-identifiable, or residual structure systematically worsens vs controls.
Minimum tests: benchmark across datasets with AIC/BIC/CV; identifiability diagnostics; residual spectra.
Verify: /research/themes/fit-function-optimization

Claim 3 — Cross-domain “two constants” exist as candidate invariants

Claim: Two candidate invariants persist across object classes and data sources (with quantified uncertainties and failure boundaries).
Falsifies if: distributions are inconsistent across domains after controls, or scaling relations are not robust to pipeline/uncertainty changes.
Minimum tests: distribution stability; sensitivity analysis; holdout validation; controls against baseline models.
Verify: /research/themes/two-constants-validation

Claim 4 — A universal steep–turning–flat structural pattern appears across systems

Claim: Many systems exhibit a shared “steep–turning–flat” structural morphology that can be quantified via turning-point metrics.
Falsifies if: turning-point signatures vanish under consistent definitions or are explained by selection/threshold effects.
Minimum tests: turning-point detection + null models + sensitivity to definitions and noise.
Verify: /research/methods/turning-point-metrics

Claim 5 — Galaxy outer “smoothing” reflects the structural pattern (not arbitrary fitting)

Claim: Outer flattening/smoothing in galaxy rotation curves is consistent with the structural turning-point model and outperforms controls in residual structure.
Falsifies if: control models match equally well with fewer assumptions; residual structure shows no improvement; parameters unstable.
Minimum tests: matched controls, AIC/BIC, residual diagnostics, multi-start stability.
Verify: /research/domains/galaxies-rotation-curves

Claim 6 — The Solar System lies near a Galactic turning point (testable dynamical signature)

Claim: The Solar System’s location in the Milky Way corresponds to a turning-point regime in the Galaxy’s structural profile, producing measurable dynamical constraints.
Falsifies if: no turning-point metric can be defined robustly for MW profiles, or constraints contradict independent MW measurements.
Minimum tests: MW profile reconstruction under multiple datasets; turning-point detection; uncertainty propagation; controls.
Verify: /research/domains/solar-system/galactic-turning-point

Claim 7 — Within the Solar System, there are hints of velocity-profile “flattening”

Claim: After unified definitions and controls, outer Solar System dynamics exhibits a systematic residual (Δv or equivalent) suggestive of flattening beyond baseline models.
Falsifies if: apparent flattening disappears when using consistent definitions/epochs, or is fully explained by selection effects and known perturbations.
Minimum tests: unify epoch + define r consistently + compute residuals vs baseline; selection-bias controls; publish null results.
Verify: /research/domains/solar-system/velocity-flattening-tests

Claim 8 — Earth may be near a Solar System structural turning point

Claim: Earth’s orbital/planetary environment is near a turning-point regime of a defined Solar System structural profile, implying specific constraints (not post hoc).
Falsifies if: no robust Solar System structural profile exists, or predicted constraints fail.
Minimum tests: define the profile; turning-point metric; pre-registered predictions; controls.
Verify: /research/domains/solar-system/earth-turning-point

Claim 9 — Extinction periodicity (~250 Myr) is statistically robust and linked to dynamical timescales

Claim: A ~250 Myr extinction periodicity (if present) is statistically robust under multiple tests and shows a non-trivial relation to Solar System–Galaxy dynamical timescales.
Falsifies if: periodicity fails robustness tests (multiple comparisons, dating uncertainties, alternative datasets) or coupling disappears under controls.
Minimum tests: rigorous periodicity tests; null models; dating-uncertainty propagation; explicit controls decoupling from dynamics.
Verify: /research/domains/extinction/periodicity-tests

Claim 10 — The 250 Myr extinction cycle vs ~240 Myr Galactic orbital timescale is a testable alignment

Claim: The proximity of ~250 Myr and ~240 Myr timescales reflects a testable alignment rather than coincidence.
Falsifies if: alignment is within expected coincidence rates under plausible uncertainty distributions, or if coupling mechanisms fail.
Minimum tests: uncertainty-aware alignment statistics; Monte Carlo coincidence rates; mechanism constraints.
Verify: /research/domains/extinction/timescale-alignment

三项“我们做什么”标签（3 bullets）

• 风险识别：监测与建模潜在全球性风险
  
  

• 科学研究：开放、可复现的研究与数据工作流
  
  

• 行动与协作：推动跨机构合作与公众参与
  
  

CTA（两个按钮）

• Primary：Donate / 支持我们
  
  

• Secondary：Read the Plan / 查看年度行动计划
  
  

Section A：我们关注的风险（3–6 卡片）

• 小行星/彗星撞击风险
  
  

• 超级火山与全球气候突变
  
  

• 生态系统崩溃与连锁灭绝
  
  

• 太阳/地磁与关键基础设施风险
  
  

• 新兴技术与系统性风险（以“风险治理”视角呈现）
  
  

每个卡片：一句话解释 + “Learn more”。

Section B：旗舰项目（Programs）

做成 3 个主项目卡（每个卡有“目标—方法—产出”）：

1. Early Warning & Modeling（预警与建模）
   
   

2. Open Research Library（开放研究库）
   
   

3. Public Education & Preparedness（公众科普与应对能力）
   
   

Section C：可信度与透明度（Trust）

用“数字化证据”呈现（没有数据也可先用占位，后续补齐）：

• 发布的研究/报告数量
  
  

• 合作机构/顾问人数
  
  

• 数据集/工具数量
  
  

• 年度资金去向摘要
  
  

加一个入口：Financials & Governance / 财务与治理

隐私政策、Cookie说明

Join AEEA to safeguard the future of life on Earth.