We present results from a survey of weak Mg ii absorbers in the VLT/UVES spectra of 81 QSOs obtained from the ESO archive .
In this survey , we identified 112 weak Mg ii systems within the redshift interval 0.4 < z < 2.4 with 86 % completeness down to a rest-frame equivalent width of W _ { r } ( 2796 ) = 0.02 Å , covering a cumulative redshift path length of \Delta Z \sim 77.3 .
From this sample , we estimate that the number of weak absorbers per unit redshift ( dN / dz ) increases from 1.06 ~ { } { \pm } ~ { } 0.04 at < z > = 1.9 to 1.76 ~ { } { \pm } ~ { } 0.08 at < z > = 1.2 and thereafter decreases to 1.51 ~ { } { \pm } ~ { } 0.09 at < z > = 0.9 and 1.06 ~ { } { \pm } ~ { } 0.10 at < z > = 0.6 .
Thus we find evidence for an evolution in the population of weak Mg ii absorbers , with their number density peaking at z = 1.2 .
We also determine the equivalent width distribution of weak systems at < z > = 0.9 and < z > = 1.9 .
At 0.4 < z < 1.4 , there is evidence for a turnover from a powerlaw of the form n ( W _ { r } ) ~ { } \propto~ { } W _ { r } ^ { -1.04 } at W _ { r } ( 2796 ) < 0.1 Å .
This turnover is more extreme at 1.4 < z < 2.4 , where the equivalent width distribution is close to an extrapolation of the exponential distribution function found for strong Mg ii absorbers .
Based on these results , we discuss the possibility that some fraction of weak Mg ii absorbers , particularly single cloud systems , are related to satellite clouds surrounding strong Mg ii systems .
These structures could also be analogs to Milky Way high velocity clouds .
In this context , the paucity of high redshift weak Mg ii absorbers is caused by a lack of isolated accreting clouds on to galaxies during that epoch .