Abstract
We present details of the Automated Radio Telescope Imaging Pipeline (ARTIP)
and results of a sensitive blind search for HI and OH absorbers at $z<0.4$ and
$z<0.7$, respectively. ARTIP is written in Python 3.6, extensively uses the
Common Astronomy Software Application (CASA) tools and tasks, and is designed
to enable the geographically-distributed MeerKAT Absorption Line Survey (MALS)
team to collaboratively process large volumes of radio interferometric data. We
apply it to the first MALS dataset obtained using the 64-dish MeerKAT radio
telescope and 32K channel mode of the correlator. With merely 40 minutes on
target, we present the most sensitive spectrum of PKS1830-211 ever obtained and
characterize the known HI ($z=0.19$) and OH ($z=0.89$) absorbers. We further
demonstrate ARTIP's capabilities to handle realistic observing scenarios by
applying it to a sample of 72 bright radio sources observed with the upgraded
Giant Metrewave Radio Telescope (uGMRT) to blindly search for HI and OH
absorbers. We estimate the numbers of HI and OH absorbers per unit redshift to
be $n_{21}(z\sim0.18)<$0.14 and $n_{\rm OH}(z\sim0.40)<$0.12, respectively, and
constrain the cold gas covering factor of galaxies at large impact parameters
(50 kpc $<\rho<$ 150 kpc) to be less than 0.022. Due to the small redshift
path, $\Delta z\sim$13 for HI with column density$>5.4\times10^{19}$ cm$^{-2}$,
the survey has probed only the outskirts of star-forming galaxies at $\rho>30$
kpc. MALS with the expected $\Delta z\sim10^{3-4}$ will overcome this
limitation and provide stringent constraints on the cold gas fraction of
galaxies in diverse environments over $0<z<1.5$.