# LBNL-2016: Neutrino Forecasting

## Forecasting Needs for Neutrinos

Discussion leader: Joel Meyers

There are two compelling targets for neutrinos that are right at the level of sensitivity for CMB-S4. It is important that our forecasts realistically address our ability to meet these targets. The purpose of this discussion is address what aspects of forecasting could impact our sensitivity to neutrinos mass and N_{eff}.

- Short introduction on major theoretical targets for M
_{ν}(0.058 eV) and ΔN_{eff}(0.027) (< 5 minutes)

- Current status and necessary improvements of forecasts (10 minutes)

- Discussion of l
_{min}, τ, etc. (5-10 minutes)

- Role of delensing for N
_{eff}(5 minutes)

- External data sets, BAO, etc. (5 minutes)

## Session Summary

Baseline neutrino forecasts focus on two theoretically-motivated thresholds:

- Sum of neutrino masses: 58 meV

- N
_{eff}: 0.027

We should not lose sight of the value of Y_{p}, especially due to the complementarity with BBN.

# Neutrino Mass

Optical depth has a large impact on our ability to constrain neutrino mass

- We should move beyond parametrization in terms of l
_{min}

- Include forecasts with higher noise data at low l (perhaps from S3, CLASS, balloons, etc.)

BAO breaks degeneracies shrinking neutrino mass error bars by about a factor of three compared to CMB alone

- How robust are the BAO forecasts?

- Should we include forecasts for a range of BAO experiments? (e.g. BOSS BAO versus DESI BAO)

- Should we forecast constraints including galaxy power spectrum, red-shift space distortions, etc.?

Foregrounds do not seem to significantly impact constraints on neutrino mass (or N_{eff}) in current forecasts

- Are there more important foregrounds or systematics that we are missing?

# N_{eff}

How do forecasts on N_{eff} change when allowing Y_{p} to vary? To what degree is the degeneracy in the damping tail broken?

How much more information on N_{eff} and Y_{p} exists at yet higher l?

- Probably outside of scope of S4 due to noise, foregrounds, non-Gaussian lenses, etc.

Delensing sharpens peaks and reduces lensing covariance, thus leading to better N_{eff} constraints.

- Forecasts should include realistic T and E delensing to account for these effects