Skip to content

fatihwf/AstroData-Testing-Pipeline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AstroData-Testing-Pipeline: A Multi-Stage Cosmological Data Analysis Pipeline

Python License AstroPhysics

An advanced computational framework for testing fundamental cosmological hypotheses using multi-messenger open data (CMB, Supernovae, Gravitational Waves, and LHC Open Data), alongside high-fidelity synthetic forecasts for next-generation observatories.

Project Overview

This project implements a robust statistical and computational pipeline to evaluate the limits of the Standard Cosmological Model (ΛCDM) against alternative theories such as Dynamic Dark Energy, Extra-Dimensional Gravity, and Conformal Cyclic Cosmology (CCC). Furthermore, it includes a cutting-edge Injection-Recovery framework to forecast the detection capabilities of future interferometers for beyond-GR anomalies.

Technical Stack

  • Physical Engines: CAMB (Einstein-Boltzmann solver), PyCBC (Gravitational wave informatics).
  • Data Processing: Healpy (HEALPix spherical mapping), Uproot & Awkward (ROOT file processing), Pandas, NumPy.
  • Statistical Inference: Emcee (MCMC), SciPy (Optimization), Grid-based Bayesian Inference, Savage-Dickey Density Ratios, Akaike/Bayesian Information Criteria (AIC/BIC).

Analysis Stages

Stage 1: Dynamic Dark Energy Model Selection

  • Scientific Goal: Testing for deviations from the cosmological constant (Λ) using a dynamic Equation of State parameter.
  • Dataset: Pantheon+ (1590 Type Ia Supernovae) & SDSS DR12 BAO.
  • Methodology: Bayesian inference via MCMC with analytical marginalization over absolute magnitude.

Stage 2: Multi-Messenger Gravity Leakage Test

  • Scientific Goal: Constraining the "leakage" of gravitational waves into extra dimensions using current observational limits.
  • Dataset: LIGO GW150914 Strain Data + SN + BAO.
  • Methodology: Waveform damping analysis and signal-to-noise ratio (SNR) consistency checks via joint MCMC.

Stage 3: LHC High-Energy MET Anomaly Detection

  • Scientific Goal: Searching for Missing Transverse Energy (MET) as a proxy for extra-dimensional particle decay.
  • Dataset: CERN ATLAS 13 TeV Monojet Open Data.
  • Methodology: High-dimensional phase-space analysis and background estimation using ROOT/Uproot.

Stage 4: Primordial Power Spectrum Dynamics

  • Scientific Goal: Modeling early universe leakage by adjusting the effective number of relativistic species and Dynamic Dark Energy.
  • Methodology: Acoustic peak simulation utilizing the CAMB Boltzmann solver.

Stage 5: Holographic Universe vs. Standard Inflation

  • Scientific Goal: Evaluating the 2D Holographic QFT Power Spectrum against 3D Inflation.
  • Dataset: Planck 2018 CMB (TT Spectrum).
  • Methodology: Custom power spectrum injection into CAMB and reduced χ²_ν optimization via Nelder-Mead.

Stage 6: Conformal Cyclic Cosmology (Hawking Points Search)

  • Scientific Goal: Searching for concentric rings of low variance in the CMB as remnants of a previous Aeon.
  • Dataset: Planck 2018 SMICA Full-Sky Map.
  • Methodology: Spherical image processing (HEALPix), radial autocorrelation, and rigorous Monte Carlo Look-Elsewhere Effect (LEE) / Bonferroni corrections.

Stage 7: Brane-Bulk Gravity Leakage Forecast (Injection-Recovery)

  • Scientific Goal: Forecasting the pipeline's sensitivity to detect macroscopic extra-dimensional gravitational leakage (2% phenomenological deviation) using next-generation observatories.
  • Dataset: 50,000-event High-Fidelity Synthetic Catalog (LISA/Cosmic Explorer era) utilizing Madau-Dickinson Star Formation Evolution.
  • Methodology: Ultra-high resolution 1D grid-based Bayesian inference, Savage-Dickey Bayes Factor calculation, and Lognormal instrumental noise modeling.

Results & Visualization

Stage Hypothesis Tested Finding Statistical Significance
Stage 1 Dynamic Dark Energy ΛCDM Preferred ΔBIC > 10
Stage 2 Extra-Dimensional Leakage No Leakage Detected η ≈ 0
Stage 3 LHC MET Anomaly Consistent with SM No 5σ excess
Stage 5 Holographic QFT Competitive Fit χ²_ν ≈ 1.13
Stage 6 CCC Hawking Points No Signal Detected 1.7σ (Noise consistent)
Stage 7 Brane-Bulk Leakage Forecast Signal Successfully Recovered > 22σ (BF > 10^110)

Overall Conclusion: Across Stages 1 through 6, using current multi-messenger observational data, the Standard Cosmological Model (ΛCDM) and the Standard Model of Particle Physics consistently emerged as the statistically preferred frameworks. Despite rigorous testing, no significant anomalies (≥ 5σ excesses) were detected in current datasets. However, Stage 7 demonstrates the profound capability of this computational architecture: the injection-recovery framework proves that the pipeline is mathematically primed to isolate and recover even microscopic deviations (e.g., 2% Brane-Bulk leakage) with overwhelming Bayesian evidence (Bayes Factor > 10^110) when next-generation multi-messenger data becomes available.


Sample Outputs

  1. MCMC Parameter Estimation (Dark Energy Analysis)

    Figure 1: Posterior distributions for dynamic dark energy model parameters.

  2. Holographic Universe vs. Standard Inflation

    Figure 2: Planck 2018 TT Power Spectrum fit against Afshordi QFT model and ΛCDM.

  3. Concentric Ring Variance Analysis (Conformal Cyclic Cosmology)

    Figure 3: Hawking Point search on Planck SMICA map with Monte Carlo LEE correction.

  4. Ultra-High Resolution Leakage Posterior (Injection-Recovery Framework)

    Figure 4: Grid-based Bayesian posterior successfully recovering the injected 2% gravitational leakage against a simulated Standard Model baseline.

Installation

git clone [https://github.com/fatihwf/AstroData-Testing-Pipeline.git](https://github.com/fatihwf/AstroData-Testing-Pipeline.git)
cd AstroData-Testing-Pipeline
pip install -r requirements.txt

License

Distributed under the MIT License. See LICENSE for more information.

📜 Citation

If you use this framework or data pipeline in your research or projects, please cite it as follows:

@misc{Goc2026Cosmo,
  author = {Fatih Gazi Göç},
  title = {AstroData-Testing-Pipeline: A Multi-Stage Cosmological Data Analysis Pipeline},
  year = {2026},
  publisher = {GitHub},
  journal = {GitHub Repository},
  howpublished = {\url{[https://github.com/fatihwf/AstroData-Testing-Pipeline](https://github.com/fatihwf/AstroData-Testing-Pipeline)}}
}

About

Bayesian MCMC cosmology pipeline: dynamic dark energy (w0-wa) vs ΛCDM using Pantheon+ SNe Ia & SDSS DR12 BAO. Full covariance matrix, Planck 2018 r_d prior, AIC/BIC model selection. Python | emcee | camb | astropy

Topics

Resources

License

Stars

3 stars

Watchers

0 watching

Forks

Releases

No releases published

Packages

 
 
 

Contributors