Galaxy Spectral Types and Their Applications in the 2dFGRS

The 2dF Galaxy Redshift Survey (2dFGRS) has already measured over 200,000 redshifts of nearby (median redshift z~0.1) galaxies. This is the single largest set of galaxy spectra ever collected. It allows us to subdivide the survey into subsets according to the galaxy intrinsic properties. We outline here a method (based on Principal Component Analysis) for spectrally classifying these galaxies in a robust and independent manner. In so doing we develop a continuous measure of spectral type, eta, which reflects the actual distribution of spectra in the observed galaxy population. We demonstrate the usefulness of this classification by estimating luminosity functions and clustering per spectral type.