Synchro small scales - Githubissues

giuspugl commented 3 years ago

I have finalized a notebook for synchrotron emission. Notice that this summarizes lot's of inputs and findings from @zonca, @bthorne93 and @NicolettaK .
This PR tries to address issue #96 and finalize the wark started in PR#73

Synchrotron maps

We use the Haslam map at 408MHz for the intensity template
for Q and U we adopt maps from the K-band of WMAP .
rescaling Haslam map to 23 GHz assuming a constant spectral index =-3.1 and from K to uK
converting Wmap maps to from mK to uK
we reduce noise by smoothing to 2 deg

Power law fit

We fit in the range
$\ell \in[1-,36]$
a power law for TT spectrum of the logpol tens i map
we use the spectral index power law also for EE and BB power spectra to avoid unphysical crossings of EE over TT and BB over TT at higher multipoles

Define Modulation maps

as suggested by @bthorne93 , the injected small scales are at different multipoles for intensity and polarization, we consider 2 different modulation maps

Modulation for polarization :

smooth i map to 5 deg
we saturate all the pixels >5.5 to 1.5
reduce the dynamic range to range from .5 to 1.5 with MinMax rescaling

Modulation for intensity :

smooth i map to 5 deg
for the pixels >5 MinMax rescaling from 1.1 to 2
elsewhere MinMax rescaling from .1 to 1.1

Beta_s small scales

we subtract the offset term from the $\beta_s$ map
we use the fitted power law from the study of Nicoletta to inject small scales on to the Bs map
we rescale the large scale Bs map by a factor ~1.572 which comes from the SPASS latest results
we modulate the small scales on the Bs map by the same modulation map adopted for intensity