GS-TEC: the Gaia spectrophotometry transient events classifier

We present an algorithm for classifying the nearby transient objects detected by the Gaia satellite. The algorithm will use the low-resolution spectra from the blue and red spectrophotometers on board the satellite. Taking a Bayesian approach, we model the spectra using the newly constructed reference spectral library and literature-driven priors. We find that for magnitudes brighter than 19 in Gaia G magnitude, around 75 per cent of the transients will be robustly classified. The efficiency of the algorithm for Type Ia supernovae (SNe I) is higher than 80 per cent for magnitudes G ≤ 18, dropping to approximately 60 per cent at magnitude G = 19. For SNe II, the efficiency varies from 75 to 60 per cent for G ≤ 18, falling to 50 per cent at G = 19. The purity of our classifier is around 95 per cent for SNe I for all magnitudes. For SNe II, it is over 90 per cent for objects with G ≤ 19. GS-TEC also estimates the redshifts with errors of σ_z ≤ 0.01 and epochs with uncertainties σ_t ≃ 13 and 32 d for SNe I and SNe II, respectively. GS-TEC has been designed to be used on partially calibrated Gaia data. However, the concept could be extended to other kinds of low-resolution spectra classification for ongoing surveys.