Volacanoe desgassing emissions

Hi, I am a researcher and constantly use the reanalysis and forecast CAMS products. I am currently comparing total column emissions of the last years from the volcano Nevado del Ruíz (Colombia) with the literature, and surprisingly, using reanalysis data the magnitude of the emissions had a scale of 0.6 mg/m2 while using forecast is 20 mg/m2 (as windy shows and aligned with Sentinel). I used reanalysis data because it is more robust; however, the publication has supported the forecast without referring to the reanalysis product. I read that the Tc SO2 reanalysis may be weak because of the model resolution, but do you think I should use forecast data instead?