Maps2Cell Project List

From CMB-S4 wiki
Revision as of 16:18, 30 July 2020 by Fcyrracine (talk | contribs) (→‎High priority (unordered):)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search

We welcome volunteers interested in any and all of these areas. If this is you, please email Marilena and Christian to discuss how you can contribute.

High priority (unordered):

  • Scan strategies and fksy: Different options and implications for Neff. Explore the tradeoffs between scan strategy and the final sigma(Neff) constraint, considering scanning efficiency, Galactic cuts, performance degradation at low elevation and so on. Also discuss the implications for transient analyses. Present the optimal strategy. [Preliminary studies exist in postings here and here

Interested parties: Neil Goeckner-Wald, Reijo Keskitalo, Federico Bianchini, (your name here)

  • Beam errors and how well they need to/can be determined. Consider realistic beam errors (perhaps inspired by past experiments). Quantify the relationship between these errors and final Neff constraint. Discuss how these errors can be constrained (eg point sources), and calculate likely constraining power. [Preliminary study in posting here

Interested parties: Abigail Crites, Dan Grin, Neil Goeckner-Wald (can contribute thoughts and plots from SPT3G beams), Francis-Yan Cyr-Racine (your name here)

  • Beam systematics: The beam uncertainty in CMB-S4 is likely to be dominated by systematic errors given the raw sensitivity and large number of bright radio sources in the survey. Estimate the level of systematic errors such as frequency dependence of the beam, and investigate mitigation strategies.

Interested parties: Francis-Yan Cyr-Racine and his graduate student (TBD) (your name here)

  • Check on kSZ/tSZ/CIB/galactic dust/etc in Neff with component separation vs template fitting vs ell-cut (bias, errors) -- Foregrounds (both galactic and extragalactic) can bias Neff and increase the errors. A few schemes have been used to handle this in the past -- component separation, templates, throwing away some ells. Look at the relative performance of these schemes as far as bias and sigma. Are there multiple competitive schemes?

Interested parties: Srini Raghunathan, Yuuki Omori, Sasha Rahlin, Neil Goeckner-Wald (interested, but haven't thought much about this) (your name here)

Medium Priority (unordered):

  • Study how foregrounds and instrumental systematics will propagate through a delensing analysis to Neff. Presuming there’s a significant effect, look at mitigation methods, and how the hi-ell mitigation strategies may follow or be different from the strategy for low-ell delensing and r.

Interested parties: (your name here)

  • Study the requirements on gain (absolute, l-dependence, elevation-dependent,...) and leakage terms (T->T, T->P, P->P) for Neff. This can also account for polarized beams or errors in the estimated transfer function.

Interested parties: (your name here)

  • Study the impact of beam asymmetry and detector time constants on the Neff constraints. Also look at whether scan strategy/cross-linking can mitigate any effects.

Interested parties: (your name here)

Other ideas (unordered):

  • How well can CMB-S4 constrain the tSZ power spectrum and bispectrum? What can we learn from such constraints?

Interested parties: (your name here)

  • Can we save CPU-hours by simulating TOD-spectra for subsets of the total data volume? By what factor can one decimate the data volume before it becomes an issue -- what impacts does such a strategy have on Neff constraints and recovering the hi-ell power spectrum?

Interested parties: (your name here)

  • How should CMB-S4 combine the lensing + CMB power spectra constraints to get a combined likelihood for the total data set?

Interested parties: (your name here)

  • How to estimate the CMB-S4 bandpower covariance matrix? How many sims are needed for each scheme? What data products? What does uncertainty on the covariance due to the Neff constraints?

Interested parties: (your name here)