Overview

This repository contains the complete history of numerical calculations for the paper "Direct correlation of line intensity mapping and CMB lensing from evolution along the lightcone" [2507.17752]. Sorry for the mess. Specific references to key calculations are outlined below.

---

If you have any questions about the code, I'm happy to chat. Let me know at delon@stanford.edu

Angular Spectra

The angular spectrum of line-of-sight correlation functions is tabulate by 000.002 and 000.003 Once this is tabulate relevant spectra are computed:

• CMB lensing spectrum discussion of this computation can be found in 001.003 but generally since computing the CMB lensing spectrum is so quick, it is recomputed each time it is needed.
• Unfiltered LIM spectrum [CII][CO][Lya][HI]
• Unfiltered LIM x CMB lensing cross spectrum 009.015
• High-pass (Foreground) filtered LIM x CMB lensing cross spectrum 009.016
  • Same computation with the Limber approximation in 009.017
• High-pass (Foreground) filtered LIM spectrum 010.023-03-25 for all lines and experiments.
• High-pass filtered LIM noise spectrum 008.022
  • Mathematica notebook computing the analytical form of the filtered LIM noise spectra summands in terms of sine- and cosine-integrals can be found at 008.023.

The noise spectra for CMB lensing, stored in data/N0_*, is assumed to be only $N^{(0)}$ which we tabulate using LensQuEst for Planck and Simons Observatory.

Detectability

The signal-to-noise calculation in 011.007 assembles all the computed angular spectra to compute the detectability, specifically $d(\textbf{{\sf SNR}}^2)/d \ln \ell$, of a direct correlation.

Miscellaneous

File names of figures in the paper, accessible in the TeX Source on arXiv, reference which file generated that figure. For example Fig. 5 of the paper is named 013.000.IHi_kappa.pdf meaning it was generated by the file indexed as 013.000. Also of possible interest is LIMxCMBL/experiments.py which contains our instrumental noise models for LIM experiments considered in the paper.

Numerics for the toy model can be found in 015.* and code which generates components of the summary visualization (Fig. 1) can be found in 016.*