Atmospheric sulfur dioxide is produced mainly by anthropogenic activities like fossil fuel combustion and metal smelting as well as volcanic eruptions and degassing. Due to its generally low atmospheric background level it is an excellent marker for pollution events and volcanic activity. New UV remote sensing instruments like OMI on EOS-Aura and GOME-2 on MetOp-A make it possible to monitor SO2 emissions on a global scale and daily basis. They are sensitive enough to measure anthropogenic SO2 in the atmospheric boundary layer and SO2 from volcanic degassing. The GOME-2 instrument on the satellite MetOp provides operational near-real time measurements of the SO2 columns with a spatial resolution of 80x40 km² and a global coverage within about one day. Sulfur dioxide emissions are determined from solar backscatter measurements in the ultra-violet spectral range between 315 - 326 nm, applying the Differential Optical Absorption Spectroscopy (DOAS) method. This retrieval technique uses the high spectral resolution of the instruments to determine the total column density of SO2. The operational GOME-2 total column SO2 product is provided in the framework of EUMETSAT’s Satellite Application Facility on Ozone and Atmospheric Chemistry Monitoring (O3M-SAF). As the standard DOAS method is restricted to optically thin conditions, current research activities focus on the use of a direct fitting algorithm for SO2 for GOME-2 data, that can use shorter wavelength regions to achieve a higher sensitivity for low SO2 amounts typically observed in degassing situations. GOME-2 allows tracking of volcanic eruption plumes in near-real time (NRT) and monitoring of changes in volcanic degassing behavior. The ability to monitor changes in volcanic degassing behavior is of great importance for early warning of volcanic activity, as large increases in SO2 fluxes are often an indicator for new episodes of volcanic unrest. Relevant eruption parameters such as the location of the emission source, the time of the eruption and the emission height are derived by using backward trajectory ensemble matching technique. Further, GOME-2’s ability to measure SO2 at low altitudes makes it a valuable tool for monitoring air pollution in the boundary layer and determining changes and trends. In this contribution, we present exemplary results of SO2 retrieved from GOME-2 data, including analyses for air pollution and volcanic degassing detected by GOME-2. For recent volcanic eruptions results of the trajectory matching and of a particle dispersion model will be shown. We present the use of the operational GOME-2 SO2 for early warning of volcanic hazards within a new Volcano Fast Response System (Exupery). Further we show initial validation results for GOME-2 SO2 measurements and comparisons between volcanic SO2 retrieved using the DOAS method and direct fitting approach.