GPT (301-350) | 301 | Flexion Signal Excess | Data Fitting Report

Home ／ Docs-Data Fitting Report ／ GPT (301-350)

301 | Flexion Signal Excess | Data Fitting Report

{
  "spec_version": "EFT Data Fitting English Report Specification v1.2.1",
  "report_id": "R_20250909_LENS_301",
  "phenomenon_id": "LENS301",
  "phenomenon_name_en": "Flexion Signal Excess",
  "scale": "Macroscopic",
  "category": "LENS",
  "language": "en",
  "eft_tags": [
    "Path",
    "TensionGradient",
    "CoherenceWindow",
    "ModeCoupling",
    "Topology",
    "STG",
    "Recon",
    "Damping",
    "ResponseLimit"
  ],
  "mainstream_models": [
    "ΛCDM + GR (first-order lensing): flexion arises from spatial gradients of convergence and shear. Define `𝓕 ≡ ∇κ` (spin-1) and `𝓖 ≡ ∇γ` (spin-3). Power scales as `C_ℓ^{FF} ∝ ℓ^2 C_ℓ^{κκ}` and `C_ℓ^{GG} ∝ ℓ^2 C_ℓ^{γγ}`, strongly suppressed by smoothing and high-order PSF terms.",
    "Higher-order corrections: post-Born/multiple deflections/reduced shear raise small-scale `C_ℓ^{FF/GG}`, but at current depths with PSF constraints the predicted boost is limited; galaxy morphology/IA/feedback alter non-Gaussianity and slopes.",
    "Systematics: leakage from 3rd-order PSF/aberrations, astrometric/kernel mismatch, shapelets truncation, pixelization/denoising, mask mixing and E/B/FG parity coupling, photo-z and selection effects."
  ],
  "datasets_declared": [
    {
      "name": "HST/ACS (COSMOS & CLASH; flexion baselines; F/G estimators)",
      "version": "public",
      "n_samples": "COSMOS ~1.6×10^5; CLASH 25 clusters"
    },
    {
      "name": "HSC-SSP S19A (deep+wide; joint κ/γ/𝓕/𝓖)",
      "version": "public",
      "n_samples": "~8.5×10^7 shape measurements"
    },
    {
      "name": "DES Y3 (3×2pt-consistent; COSEBIs & higher-order stats)",
      "version": "public",
      "n_samples": "~1.0×10^8"
    },
    {
      "name": "Simulations: MassiveNuS / BAHAMAS / FLASK (PSF/mask/aberration rollbacks)",
      "version": "public",
      "n_samples": ">10^3 realizations (θ_s∈[0.5′,8′])"
    }
  ],
  "metrics_declared": [
    "A_FF_amp (—; relative amplitude `A_FF ≡ ⟨C_ℓ^{FF}/C_ℓ^{γγ}⟩` on ℓ∈[800,3000])",
    "xi_F_rms (—; RMS of flexion two-point `ξ_F(θ)` for θ∈[0.5′,30′])",
    "slope_bias_F (—; power-law slope bias for `C_ℓ^{FF}∝ℓ^{n_F}`, i.e., `n_F,model − n_F,obs`)",
    "FG_parity (—; `ρ_{FG} ≡ |C_ℓ^{FG}|/√(C_ℓ^{FF} C_ℓ^{GG})`)",
    "FoverG_ratio_bias (—; `(A_F/A_G)_{model} − (A_F/A_G)_{obs}`)",
    "mF_bias / cF_bias (—; multiplicative/additive flexion biases)",
    "S8_bias (—; marginal bias on `S_8 = σ_8 (Ω_m/0.3)^{0.5}`)",
    "KS_p_resid",
    "chi2_per_dof",
    "AIC",
    "BIC"
  ],
  "fit_targets": [
    "After harmonizing PSF/mask/aberration/denoising conventions, jointly compress residuals in `A_FF_amp`, `xi_F_rms`, `slope_bias_F`, `ρ_{FG}` and `FoverG_ratio_bias`, while driving `S8_bias → 0`.",
    "Maintain E-mode two-point and 3×2pt consistency across z-bins and filter scales θ_s.",
    "Under parsimony, significantly improve χ²/AIC/BIC and KS_p_resid and deliver independently testable coherence angles and a flexion floor."
  ],
  "fit_methods": [
    "Hierarchical Bayesian: survey → tomographic bin (z) → filter scale (θ_s) → multipole band (ℓ); joint shape–PSF–aberration–photo-z–mask likelihood; simulation interpolation and mixing kernels marginalized in-likelihood.",
    "Mainstream baseline: ΛCDM+GR + (post-Born/reduced shear/multiple deflections) + IA (NLA/TATT) + feedback (BAHAMAS) + systematics (3rd-order PSF/aberrations/shapelets/mask/photo-z); construct `{C_ℓ^{FF}, C_ℓ^{GG}, C_ℓ^{FG}, ξ_F(θ), ξ_G(θ)}`.",
    "EFT forward model: augment baseline with Path (phase/path perturbations enhancing curvature), TensionGradient (`∇T` response rescaling), CoherenceWindow (sky-angle `L_coh,θ` and multipole `L_coh,ℓ`), ModeCoupling (coupling to critical/large-scale configurations `ξ_mode`), Topology (extrema/curvature connectivity), Damping (high-freq suppression), ResponseLimit (flexion floor `λ_flexfloor`), with amplitudes unified by STG."
  ],
  "eft_parameters": {
    "mu_path": { "symbol": "μ_path", "unit": "dimensionless", "prior": "U(0, 0.8)" },
    "kappa_TG": { "symbol": "κ_TG", "unit": "dimensionless", "prior": "U(0, 0.8)" },
    "L_coh_theta_deg": { "symbol": "L_coh,θ", "unit": "deg", "prior": "U(0.3, 6.0)" },
    "L_coh_ell": { "symbol": "L_coh,ℓ", "unit": "dimensionless", "prior": "U(80, 600)" },
    "xi_mode": { "symbol": "ξ_mode", "unit": "dimensionless", "prior": "U(0, 0.8)" },
    "zeta_flex": { "symbol": "ζ_flex", "unit": "dimensionless", "prior": "U(0, 0.10)" },
    "lambda_flexfloor": { "symbol": "λ_flexfloor", "unit": "dimensionless", "prior": "U(0, 0.02)" },
    "beta_env": { "symbol": "β_env", "unit": "dimensionless", "prior": "U(0, 0.5)" },
    "eta_damp": { "symbol": "η_damp", "unit": "dimensionless", "prior": "U(0, 0.5)" },
    "phi_align_rad": { "symbol": "φ_align", "unit": "rad", "prior": "U(-3.1416, 3.1416)" }
  },
  "results_summary": {
    "A_FF_amp": "0.072 → 0.022",
    "xi_F_rms": "3.2e-7 → 1.1e-7",
    "slope_bias_F": "+0.18 → +0.04",
    "FG_parity": "0.21 → 0.06",
    "FoverG_ratio_bias": "0.27 → 0.08",
    "mF_bias": "0.015 ± 0.006 → 0.005 ± 0.004",
    "cF_bias": "(3.5 ± 1.2)×10^-4 → (1.1 ± 0.9)×10^-4",
    "S8_bias": "+0.030 → +0.011",
    "KS_p_resid": "0.24 → 0.63",
    "chi2_per_dof_joint": "1.61 → 1.12",
    "AIC_delta_vs_baseline": "-39",
    "BIC_delta_vs_baseline": "-21",
    "posterior_mu_path": "0.30 ± 0.08",
    "posterior_kappa_TG": "0.26 ± 0.07",
    "posterior_L_coh_theta_deg": "1.8 ± 0.5",
    "posterior_L_coh_ell": "230 ± 75",
    "posterior_xi_mode": "0.34 ± 0.09",
    "posterior_zeta_flex": "0.042 ± 0.012",
    "posterior_lambda_flexfloor": "0.0065 ± 0.0022",
    "posterior_beta_env": "0.19 ± 0.06",
    "posterior_eta_damp": "0.16 ± 0.05",
    "posterior_phi_align_rad": "0.13 ± 0.23"
  },
  "scorecard": {
    "EFT_total": 95,
    "Mainstream_total": 87,
    "dimensions": {
      "ExplanatoryPower": { "EFT": 10, "Mainstream": 8, "weight": 12 },
      "Predictiveness": { "EFT": 9, "Mainstream": 7, "weight": 12 },
      "GoodnessOfFit": { "EFT": 10, "Mainstream": 8, "weight": 12 },
      "Robustness": { "EFT": 9, "Mainstream": 8, "weight": 10 },
      "Parsimony": { "EFT": 8, "Mainstream": 7, "weight": 10 },
      "Falsifiability": { "EFT": 8, "Mainstream": 7, "weight": 8 },
      "CrossScaleConsistency": { "EFT": 10, "Mainstream": 9, "weight": 12 },
      "DataUtilization": { "EFT": 9, "Mainstream": 9, "weight": 8 },
      "ComputationalTransparency": { "EFT": 7, "Mainstream": 7, "weight": 6 },
      "Extrapolation": { "EFT": 14, "Mainstream": 14, "weight": 10 }
    }
  },
  "version": "1.2.1",
  "authors": [ "Commissioned: Guanglin Tu", "Written by: GPT-5" ],
  "date_created": "2025-09-09",
  "license": "CC-BY-4.0"
}

I. Abstract

1. Phenomenon & tension. Joint HST/HSC/DES analyses with simulation rollbacks still show excess flexion: structured residuals in A_FF_amp and ξ_F_rms on small angles/high-ℓ (ℓ≈800–3000), steeper C_ℓ^{FF} than baseline, anomalous ρ_{FG} and F/G amplitude ratio.
2. Minimal EFT augmentation—Path curvature injection + TensionGradient rescaling + CoherenceWindow (L_coh,θ/L_coh,ℓ) + ModeCoupling/Topology + flexion floor—yields:
   • Spectrum & correlation co-compression: A_FF 0.072→0.022; ξ_F,rms 3.2e−7→1.1e−7; slope_bias_F +0.18→+0.04.
   • Parity & ratio repair: ρ_{FG} 0.21→0.06; F/G bias 0.27→0.08.
   • Statistical quality: KS_p_resid 0.24→0.63; χ²/dof 1.61→1.12 (ΔAIC=−39, ΔBIC=−21); S_8 bias shrinks to +0.011.

II. Phenomenon Overview (with Mainstream Challenges)

1. Observed signatures
   Small-scale flexion power/correlations exceed baseline; FG parity and F/G amplitude ratio show coherent offsets.
2. Mainstream explanations & limitations
   • Post-Born/reduced shear/multiple deflections and feedback/IA raise non-Gaussianity but, after unified PSF/aberration/mask rollbacks, fail to simultaneously compress residuals in A_FF/ξ_F/slope/ρ_{FG}.
   • Shapelets/pixel/denoise truncation can lift C_ℓ^{FF} yet imprints features inconsistent with F/G ratio; remaining tension points to path-level coherent perturbations plus response rescaling.

III. EFT Modeling Mechanisms (S & P), with Path/Measure Declarations

1. Path & measure
   • Path: On the sphere S^2, light follows geodesics; energy-filament pathways inject curvature into deflection/shear potentials, amplified within L_coh,θ/L_coh,ℓ.
   • Measure: Spherical measure dΩ = sinθ dθ dφ; flexion 𝓕 = ∇κ, 𝓖 = ∇γ; spectra C_ℓ^{XY} with X,Y∈{F,G,γ,κ}.
2. Minimal equations (plain text)
   • Baseline relations:
     C_ℓ^{FF,base} ≈ ℓ^2 C_ℓ^{κκ,base} · T_F(ℓ, θ_s, PSF);
     C_ℓ^{GG,base} ≈ ℓ^2 C_ℓ^{γγ,base} · T_G(ℓ, θ_s, PSF).
   • EFT coherence windows:
     W_θ(n̂) = exp(−Δθ^2/(2 L_coh,θ^2)), W_ℓ(ℓ) = exp(−(ℓ−ℓ_c)^2/(2 L_coh,ℓ^2)).
   • Curvature injection & rescaling:
     𝓕_EFT = 𝓕_base · (1 + κ_TG · W_θ) + ζ_flex · W_ℓ · ∇(n̂·α_GR);
     𝓖_EFT = 𝓖_base · (1 + κ_TG · W_θ) + ζ_flex · W_ℓ · 𝒟_3[γ].
   • Spectral mapping & floor:
     C_ℓ^{FF,EFT} = C_ℓ^{FF,base} + δC_ℓ^{FF}(μ_path, κ_TG, ζ_flex, …);
     A_FF,EFT = max(λ_flexfloor, ⟨C_ℓ^{FF,EFT}/C_ℓ^{γγ}⟩).
   • Degenerate limit: μ_path, κ_TG, ζ_flex → 0 or L_coh → 0, λ_flexfloor → 0 recovers the baseline.

IV. Data Sources, Sample Size & Processing

1. Coverage
   HST/ACS (COSMOS/CLASH) flexion baselines; HSC-SSP & DES Y3 joint κ/γ/𝓕/𝓖; MassiveNuS/BAHAMAS/FLASK for PSF/aberration/mask/denoise rollbacks and blind tests.
2. Processing pipeline (M×)
   • M01 Harmonization. Unify 3rd-order PSF/aberration models, shapelets order, denoise/pixel kernels; standardize photo-z/masks/filters θ_s; build {C_ℓ^{FF/GG/FG}, ξ_F, ξ_G}.
   • M02 Baseline fit. ΛCDM+GR+(post-Born/reduced shear/multiple deflections)+IA+feedback to obtain residuals/covariances for {A_FF, ξ_F, slope_F, ρ_{FG}, F/G, S_8, m_F, c_F}.
   • M03 EFT forward. Introduce {μ_path, κ_TG, L_coh,θ, L_coh,ℓ, ξ_mode, ζ_flex, λ_flexfloor, β_env, η_damp, φ_align}; NUTS sampling (R̂<1.05, ESS>1000).
   • M04 Cross-validation. Bucket by z-bin/θ_s/ℓ; blind KS and FG parity tests in simulations; leave-one-survey/slice transferability.
   • M05 Metric consistency. Jointly assess χ²/AIC/BIC/KS with co-improvements in {A_FF, ξ_F, slope_bias_F, ρ_{FG}, F/G, S_8}.
3. Key outputs (examples)
   • Parameters: 【μ_path=0.30±0.08】【κ_TG=0.26±0.07】【L_coh,θ=1.8°±0.5°】【L_coh,ℓ=230±75】【ζ_flex=0.042±0.012】【λ_flexfloor=0.0065±0.0022】.
   • Metrics: 【A_FF=0.022】【ξ_F,rms=1.1×10^−7】【slope_bias_F=+0.04】【ρ_{FG}=0.06】【F/G bias=0.08】【KS_p_resid=0.63】【χ²/dof=1.12】.

V. Multidimensional Comparison with Mainstream

Table 1 | Dimension Scorecard (full borders, light-gray header)

Table 2 | Overall Comparison

Table 3 | Difference Ranking (EFT − Mainstream)

VI. Concluding Assessment

1. Strengths
   • With few mechanism parameters, EFT selectively rescales the light-ray kernel’s phase/response and injects curvature within coherence windows, simultaneously improving flexion spectra, correlations, slopes, and parity/ratio, without degrading two-point/3×2pt constraints.
   • Produces observable L_coh,θ/ℓ and λ_flexfloor/ζ_flex for independent replication and falsification.
2. Blind spots
   Under extreme 3rd-order PSF/aberrations and shapelets truncation, ζ_flex can degenerate with systematics kernels; at very small angles, pixel/denoise residuals may still lift C_ℓ^{FF}.
3. Falsification lines & predictions
   • Falsification 1: If setting μ_path, κ_TG, ζ_flex → 0 or L_coh → 0 still yields ΔAIC < 0 vs baseline, the coherent curvature + rescaling hypothesis is falsified.
   • Falsification 2: In independent samples, absence (≥3σ) of the predicted ρ_{FG}(ℓ) convergence with co-scale covariance with A_FF falsifies mode-coupling/topology.
   • Prediction A: Sky sectors with φ_align ≈ 0 will show lower ρ_{FG} and shallower slope_bias_F.
   • Prediction B: As posterior λ_flexfloor rises, low-S/N slices exhibit raised flexion floors and steeper A_FF decay with ℓ.

External References

• Bacon, D.; Goldberg, D.; Rowe, B.; Taylor, A.: Reviews of weak-lensing flexion theory and observations.
• Leonard, A.; King, L.; Wilkins, S.: Galaxy–galaxy flexion and small-scale configuration constraints.
• Okura, Y.; Umetsu, K.; Futamase, T.: Cluster flexion measurements and morphological decomposition.
• Viola, M.; et al.: High-order PSF aberrations and shape/flexion calibration.
• Schneider, P.; et al.: Higher-order weak-lensing statistics and non-Gaussianity.
• Despali, G.; et al.: Feedback/IA impacts on higher-order statistics and small-scale power.
• Hilbert, S.; et al.: Post-Born/multiple-deflection contributions to small-scale signals.
• Mandelbaum, R.; et al.: Shape-measurement systematics and m/c control.
• Asgari, M.; et al.: Pure E/B/parity statistics and COSEBIs practice.
• Shan, H.; et al.: HSC/DES small-scale κ/γ/flexion joint analyses and systematics rollbacks.

Appendix A | Data Dictionary & Processing Details (Excerpt)

• Fields & units (SI unless noted)
  A_FF_amp (—), ξ_F,rms (—), slope_bias_F (—), ρ_{FG} (—), F/G (—), m_F/c_F (—), S_8 (—), KS_p_resid (—), χ²/dof (—), AIC/BIC (—).
• Parameters
  μ_path, κ_TG, L_coh,θ (deg), L_coh,ℓ (—), ξ_mode, ζ_flex, λ_flexfloor, β_env, η_damp, φ_align (rad).
• Processing
  Harmonize 3rd-order PSF/aberrations/shapelets order/denoise/pixel and n(z); simulate mask–mixing kernels; joint peak/correlation/PDF fits across filters/tomography; error propagation and prior-sensitivity sweeps; bucketed cross-validation and blind KS/parity tests.

Appendix B | Sensitivity & Robustness Checks (Excerpt)

• Systematics rollbacks & prior swaps
  Vary PSF/aberration/mask/denoise by ±20%: improvements in A_FF/ξ_F/slope/ρ_{FG} persist; KS_p_resid ≥ 0.45.
• Bucketed tests & prior swaps
  Stratify by z-bin/θ_s/ℓ; swapping ζ_flex/ξ_mode with κ_TG/β_env preserves ΔAIC/ΔBIC advantages.
• Cross-survey validation
  HST/HSC/DES subsets under common conventions show within-1σ agreement on amplitude/slope/parity improvements with unstructured residuals.