A Pilot Study for the CSST Slitless Spectroscopic Quasar Survey Based on Mock Data

Abstract

The wide survey of the Chinese Space Station Telescope (CSST) will observe a large field of 17,500 $\text{deg}^2$. The GU, GV, and GI grism observations of CSST will cover a wavelength range from 2550 to 10000Å at a resolution of $R\sim 200$ and a depth of about 22 AB magnitude for the continuum. In this paper, we present a pipeline to identify quasars and measure their physical properties with the CSST mock data. We simulate the raw images and extract the one-dimensional grism spectra for quasars, galaxies, and stars with the r-band magnitudes of $18<\text{m}_{\text{r}}<22$ using the CSST Cycle 6 simulation code. Using a convolution neural network, we separate quasars from stars and galaxies. We measure the redshifts by identifying the strong emission lines of quasars. We also fit the 1D slitless spectra with QSOFITMORE to estimate the black hole masses and Eddington ratios. Our results show that the CSST slitless spectroscopy can effectively separate quasars with redshifts $z=0-5$ from other types of objects with an accuracy of 99\%. Among those successfully classified quasars, 90\% of them could have precise redshift measurements with $σ_{\mathrm{NMAD}}=0.002$. The scatters of black hole masses and Eddington ratios from the spectral fittings are 0.13 and 0.15 dex, respectively. The metallicity diagnosis line ratios have a scatter of 0.1-0.2 dex. Our results show that the CSST slitless spectroscopy survey has the potential to discover about 0.9 million new quasars and provide important contributions to AGN science and cosmology.