Tutorials

The notebooks below are rendered directly from the repository’s notebooks/ directory (symlinked into this folder so Sphinx can pick them up). They ship with stored outputs, so the site renders the plots the notebook author saw — myst-nb does not re-execute them during an ordinary docs build (nb_execution_mode = "off" in docs/conf.py); only CI’s nightly lane re-executes with NB_EXECUTION_MODE=cache to catch drift.

To run any notebook yourself, clone the repo and install the notebooks extra:

git clone https://github.com/CyrilPitrou/primat
cd primat
pip install -e ".[notebooks,recommended]"
jupyter lab notebooks/

Standard results

• PRIMAT — Standard Plots — the Schramm diagram: primordial abundances vs. η_b with 1σ nuclear-rate uncertainty bands and observational constraints (YP, D/H, ³He/⁴He, ⁷Li/H).

• BBN abundance evolution — time evolution of A_i Y_i(t) for every nuclide from 1 s to 10⁵ s, for both the small (12-reaction) and large (~429-reaction) networks.

• Network Comparison: small vs small_parthenope — small vs. small_parthenope head to head.

• Animated BBN abundance evolution — animated GIFs of the small-network abundance evolution A_i Y_i(t), sweeping ΔNeff and Ω_b h² in turn.

Nuclear rates

• BBN reaction rates vs the Hubble rate — plots the tabulated rate ⟨σv⟩(T9) of any reaction in the network, with the master-grid reinterpolation overlaid, alongside the n↔p weak rates, all compared against the Hubble rate H(T) to show freeze-out.

Parameter scans

• Posterior on Ω_b h² from primordial abundances — a posterior on Ω_b h² from YP and D/H: scans Ω_b h² ∈ [0.020, 0.024] and computes Gaussian likelihoods from each observable.

• Primordial abundances vs extra relativistic species ΔNeff — abundances vs. ΔNeff: scans ΔNeff ∈ [−2, +2] to show how extra relativistic species shift YP and D/H.

• Primordial abundances vs neutrino degeneracy parameter ξ — abundances vs. neutrino degeneracy ξ = μ_ν/T_ν: scans ξ ∈ [−0.05, +0.05].

Uncertainty analysis

• Full Monte Carlo uncertainty budget — the full MC uncertainty budget: draws nuclear rates, τ_n, and Ω_b h² simultaneously, with histograms and a corner plot of the joint distribution of every observable. See also Rate variation and Monte-Carlo uncertainty.

• Sensitivity of BBN observables to physical parameters — sensitivity tables: the logarithmic derivative ∂ ln(observable) / ∂ ln(parameter) for each of the 12 nuclear rates, τ_n, G_N, Ω_b h², and ΔNeff, as formatted tables and a heat-map.

Note

Common conventions across the parameter-scan notebooks: a fixed MC seed (MC_SEED = 0) at every grid point so finite-sample MC bias cancels across the grid; observational constraints shown as grey bands; the baryon density Ω_b h² = 0.02242 ± 0.00014 (Planck 2018 + BAO) shown as a red vertical band; and num_mc = 500+ for publication-quality uncertainty bands (the notebooks default to 50 for speed).