• Corpus ID: 207863253

Photometric Redshift Realism: A Technique for Reducing Catastrophic Outlier Redshift Estimates in Large-Scale Surveys

@article{Wyatt2019PhotometricRR,
  title={Photometric Redshift Realism: A Technique for Reducing Catastrophic Outlier Redshift Estimates in Large-Scale Surveys},
  author={M. Wyatt and Jack Singal},
  journal={arXiv: Astrophysics of Galaxies},
  year={2019}
}

• M. Wyatt, J. Singal
• Published 11 November 2019
• Physics
• arXiv: Astrophysics of Galaxies

We present results of using individual galaxies' effective redshift probability density information as a method of identifying potential catastrophic outliers in empirical photometric redshift estimation. In the course we develop a method of modification of the redshift distribution of training sets to improve both the baseline accuracy of high redshift (z>1.5) estimation as well as catastrophic outlier mitigation. We demonstrate these using two real test data sets and one simulated test data set spanning a wide redshift range (0 30%) of catastrophic outlier galaxies while simultaneously incorrectly flagging only a small percentage ( 10%) in the percentage of outlier galaxies greater than z=1.5 with only a small increase (<3%) in the percentage of outlier galaxies less than z=1.5 compared to the unmodified training set. In addition, we find that this modification can in some cases decrease the percentage of incorrectly identified non-outlier galaxies by almost 20%, while in other cases cause only a small (<1%) increase in this metric.