Dust plays a pivotal role in the chemical enrichment of the interstellar medium. In the era of mid-/high-resolution spectra and multiband spectral energy distributions, testing extinctions against gas and dust-phase properties is becoming possible. In order to test relations between metals, dust, and depletions, and comparing those to the Local Group (LG) relations, we build a sample of 93 {gamma}-ray bursts and quasar absorbers (the largest sample so far) which have extinction and elemental column density measurements available. We find that extinctions and total column density of the volatile elements (Zn,S) are correlated [with a best fit of dust-to-metals (DTM) 4.05x10^-22^mag.cm^2^] and consistent with the LG DTM relation. The refractory elements (Fe,Si) follow a similar, but less significant, relation offset about 1dex from the LG relation. On the assumption that depletion on to dust grains is the cause, we compute the total (gas+dust-phase) column density and find a remarkable agreement with the LG DTM relation: a best fit of 4.91x10^-22^mag.cm^2^. We then use our results to compute the amount of 'intervening metal from unknown sources' in random sightlines out to redshifts of z=5. Those metals implicate the presence of dust and give rise to an average 'cosmic dust dimming' effect that we express as a function of redshift, CDD(z). The CDD is unimportant out to redshifts of about 3, but because it is cumulative it becomes significant at redshifts z=3-5. Our results in this paper are based on a minimum of assumptions and effectively relying on observations.

Cone search capability for table J/MNRAS/482/2731/table1 (GRBs data used to estimate dust-to-metals ratios and N(Fe,Si)dust)

Cone search capability for table J/MNRAS/482/2731/table2 (QSO DLAs and sub-DLAs data used to estimate dust-to-metals ratios and N(Fe,Si)dust)