| |
 |
| Paper: |
Improving Photometric Redshifts Using Different Photometric Parameters |
| Volume: |
408, The Starburst−AGN Connection |
| Page: |
215 |
| Authors: |
Wang, T.; Gu, Q.-S.; Huang, J.-S. |
| Abstract: |
We compute accurate photometric redshifts for a sample of ~ 80,000 SDSS-2MASS galaxies with known spectroscopic redshifts, aiming to find the physical parameters which determine the accuracy of photometric redshifts. We find that the photometric redshift derived form the artificial neural network photometric redshift method (ANNz) recover the spectroscopic redshift distribution very well with rms of 0.017. Our main results include that: using magnitudes directly as input parameters produces more accurate photo-z’s than using the color index; the inclusion of 2MASS (J, H, K) bands does not improve photo-z’s significantly while the inclusion of the concentration index can improve the photo-z’s estimation up to ~ 10 percent. Moreover, if we divide the sample into early- and late- type galaxies or red and blue galaxies, and estimate their photo-z’s respectively, we can derive photo-z’s more accurately. Finally, our analysis show that the outliers in each case we considered are correlated well with galaxy types, that is, most outliers are late-type (blue) galaxies. |
|
|
|
|
|
