CMBS4 NET forecasts follow-up

  Denis and John (2019-01-23)
Updated by Ben Racine for bath temperature T0 = 100 mK results (2019-02-25), and modified some plots.

Previous Postings


Back in 2017, we had performed an NET prediction for the standard CMB-S4 band definition. This was done using a newly developed python NET calculator (see details in the link above). To model the atmospheric emission, we had used a single atmospheric spectra derived from the 10-year MERRA2 median profiles for the relevant site. This 10-year MERRA2 median atmospheric profile had been fed to am to produce a 10-year median spectra for a few different sites including South Pole and the Chajnantor plateau.

Since we actually have the time series of the MERRA2 profiles at their full time-resolution (3hours) for these (SouthPole, and ChajnantorPlateau and more), we can actually do better than just take the 10yr MERRA2 median profile, we can compute all the relevant parameters as a time series every 3 hours: pwv, Tsky, Tx, tau, instrument and external loading (cmb+atmosphere), and of course NET. We can then display those results in time series or histograms and get more information than just from the median. This is all done using the exact same framework as that described in the 20170221 posting.


PWV plots

Simple plots showing the pwv tods, distribution and cumulative distribution for the SouthPole and Chajnantor plateau sites. Note that the tods are on different y-scales but the distributions are on the same scale. The dashed line is the value of the pwv from the 10year median MERRA2 profile.

Figure 1: Precipitable water vapour over the 1996-2018 period.

Tsky plots

From the atmospheric spectra we integrate the Tsk_rj over each band and plot the Tsky_rj tod and distribution. The tods and distributions for the 2 sites are on the same scale. For clarity, the distributions are broken down into 4 subplot with the same scale (0-30K_rj). The dashed lines is the Tsky_rj obtained from the 10yr median MERRA2 profile. We can make similar plots of the transmission coefficient (Tx) or the opacity (tau).

Figure 2: Sky temperature over the 1996-2018 period.

Internal and External photon loading plots

We then pass the atmospheric spectra into the NET calculator given the default instrument. One of the intermediate data product of the NET calculator is the photon loading both internal (from the instrument starting at the window down to the detector) and external (from cmb and sky). I plot the tods of the total loading (Qtot) in pW as well as their distributions per band. In the histograms of the total loading, the solid lines show the fixed instrument loading contribution and the dashed lines show the Qtot obtained from the 10yr median MERRA2 profile. Note the changing scales of the histograms for the different bands.

Figure 3: Total loading (Qtot) during the 1996-2018 period.

NET plots

Finally, the output of the NET calculator yields a value of NET in uKsqrt(s). We display below the tods and histograms. Note the scale of the highest band differs from that of the other bands.
We show the results for 2 different bath temperatures.
Figure 4: NET during the 1996-2018 period.

Note on Chile site.

Note that the coordinated used to compute the Chile atmospheric conditions are from the Chajnantor plateau, i.e. the ALMA site (lon=-67.76175°,lat=-23.0285°,alt=5060m), which is lower in altitude than the ACT site (lon=-67.788°, lat=-22.959°, alt=5190m).
See the 3D map here.