2006MNRAS.371.1537M

Multi-object spectroscopy (MOS) instruments, such as the Two-degree Field (2dF) facility of the Anglo-Australian Observatory (AAO), have facilitated large-scale redshift surveys. Yet despite their acclaim, instrument design has been suspected of introducing subtle selection effects into surveys. Investigation into these selection effects has been overshadowed by instrument complexity. We identify the field configuration algorithm (FCA) used to select targets for observation as mainly responsible for such effects. An FCA can imprint artificial structure on observed target distributions, which may accrue over large angular scales, potentially to the detriment of statistical analyses applied to such surveys. We present here a new FCA developed for 2dF that is based on simulated annealing (SA), a generic method commonly used to solve constrained optimization problems. We generate synthetic fields and utilize mock 2dF volumes to contrast the behaviour of previous strategies with the SA FCA. The angular two-point correlation function and other sensitive techniques reveal that the new FCA achieves unprecedented sampling uniformity and target yield with improved target priority handling and observational flexibility over current FCAs. The SA FCA is generic enough to be used by current 2dF-like and potentially next-generation MOS instruments with little modification.