Outlier fraction as a function of the |(zMUSE−pz) / (1 + zMUSE)| threshold adopted. The blue solid line shows the trend for BPZ photo-zs and the dashed blue line that using the best EAZY photo-zs as derived in Sect. 4. The dotted green line shows the fraction of galaxies that are outliers both relative to EAZY and BPZ but where EAZY and BPZ agree. The fraction of objects that have redshifts inconsistent with the PDF of the photo-z are shown by the dashed orange horizontal line. The outlier fractions measured by Ilbert et al. (2009) are shown by the green squares, with the one at a fraction of 0.007 corresponding to the i+< 22.5 sample in COSMOS and the higher value at 0.2, for a faint subsample 24 <i+< 25. The outlier fraction measured by Hildebrandt et al. (2010) is shown by the pink lozenges with the lozenge arbitrarily placed at the median outlier fraction and the error spanning the range of outlier fractions found in their study. The measurement at a threshold of 0.15 has been shifted to 0.14 to avoid overlap with other data. The outlier fractions found in Dahlen et al. (2013) are shown by the purple triangles. The solid triangle shows the median outlier fraction and the full range found, while the open triangles show the results for their 24 <F160W< 26 (at 0.16) and 26 <F160W< 28 (at 0.28) subsamples. The outlier fractions reported for GOODS-S in Bezanson et al. (2016) is shown as the grey enlongated rectangle with error bars covering the range from comparison to grism redshifts (lowest) to spectroscopic redshifts (highest).
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.