A comparative analysis of Galactic extinction at low Galactic latitudes
We use near-infrared (J-K)-colours of bright 2MASS galaxies, measured within a 7"-radius aperture, to calibrate the Schlegel et al. (1998) DIRBE/IRAS Galactic extinction map at low Galactic latitudes ($|b| < 10^{\rm o}$). Using 3460 galaxies covering a large range in extinction (up to $A_K$ = 1.15 or E(B-V) ~ 3.19), we derive a correction factor $f = 0.83 \pm 0.01$ by fitting a linear regression to the colour-extinction relation, confirming that the Schlegel et al. maps overestimate the extinction. We argue that the use of only a small range in extinction (e.g., $A_K$ < 0.4) increases the uncertainty in the correction factor and may overestimate it. Our data confirms the Fitzpatrick (1999) extinction law for the J- and K-band. We also tested four all-sky extinction maps based on Planck satellite data. All maps require a correction factor as well. In three cases the application of the respective extinction correction to the galaxy colours results in a reduced scatter in the colour-extinction relation, indicating a more reliable extinction correction. Finally, the large galaxy sample allows an analysis of the calibration of the extinction maps as a function of Galactic longitude and latitude. For all but one extinction map we find a marked offset between the Galactic Centre and Anticentre region, but not with the dipole of the Cosmic Microwave Background. Based on our analysis, we recommend the use of the GNILC extinction map by Planck Collaboration (2016b) with a correction factor $f = 0.86 \pm 0.01$.