Environmental research aimed at monitoring and predicting O2 depletion is still lacking or in need of improvement, in spite of many attempts to find a relation between atmospheric gas content and climate variability. The aim of the present project is to determine accurate historical sequences of the atmospheric O2 depletion by using the telluric lines present in stellar spectra. A better understanding of the role of oxygen in atmospheric thermal equilibrium may become possible if high-resolution spectroscopic observations are carried out for different airmasses, in different seasons, for different places, and if variations are monitored year by year. The astronomical spectroscopic technique involves mainly the investigation of the absorption features in high-resolution stellar spectra, but we are also considering whether accurate measures of the atmospheric O2 abundances can be obtained from medium and low resolution stellar spectra.