Notes. Effective number density is defined as η = A⁻¹(∑ w)²/(∑ w²), computed using the uncalibrated shape measurement weight, w, and the survey area in square-arcmin, A. Superscript indices specify whether a number refers to sources that are removed from the sample (‘r’) or kept in the sample (‘k’). ${\eta }_{\text{s}}^{\text{r}}$ is the effective number density of sources (from the full MASK ≤ 1 sample) that are flagged for removal by the selection s. $f_{\text{tot}}^{\text{r}}$ is ${\eta }_{\text{s}}^{\text{r}}$ expressed as a fraction of the full MASK ≤ 1 sample $\left({\eta }_{\text{MASK}}^{\text{k}}\right).{\eta }_{s,\text{ run }}^{\text{r}}$ is the effective number density of sources from the currently un-flagged population that are newly flagged after applying selection s (assuming flags are applied in row-order). $f_{\text{run }}^{\text{r}}$ is ${\eta }_{s,\text{ run }}^{\text{r}}$ expressed as a fraction of the currently un-flagged population. ${\eta }_{s,\text{ run }}^{\text{k}}$ is the effective number density of sources remaining un-flagged after the flagging of selection s. $f_{\text{run }}^{\text{k}}$ is ${\eta }_{s,\text{ run }}^{\text{k}}$ expressed as a fraction of the full MASK ≤ 1 sample.