search

SOFIAObservatory / Recipes

Cookbook recipes for SOFIA data analysis
https://www.sofia.usra.edu/science/proposing-and-observing/data-products/data-resources
8 stars 6 forks source link

HAWC+ low SNR resampling/smoothing #53

Closed s-goldman closed 2 years ago

s-goldman commented 2 years ago

suggested by Lars Bonne

s-goldman commented 2 years ago

Thoughts from BG: Polarization vectors can only be positive and thus errors are nongaussian. Estimating the uncertainty using a markov chain results in uncertainties that are too high, and SNRs that are underestimated. Stokes U and Q are gaussian but not P or Q. Perhaps there is a way to average the uncertainty in stokes U and Q and include that within the uncertainty estimation for P and Q.

ask simon, enrique, archana, etc.

s-goldman commented 2 years ago

Vaillancourt_2006.pdf Naghizadeh-Khouei_Clarke_1993.pdf

s-goldman commented 2 years ago

How to calculate the polarization bias is already included in the current HAWC+ data inspection notebook.