A robust satellite data analysis technique (RAT) has been recently proposed as a suitable tool for satellite TIR surveys in seismically active regions and already successfully tested in different cases of earthquakes (both high and medium–low magnitudes).In this paper, the efficiency and the potentialities of the RAT technique have been tested even when it is applied to a wide area with extremely variable topography, land coverage and climatic characteristics (the whole Indian subcontinent). Bhuj–Gujarat's earthquake (occurred on 26th January 2001, MS ∼ 7.9) has been considered as a test case in the validation phase, while a relatively unperturbed period (no earthquakes with MS ≥ 5, in the same region and in the same period) has been analyzed for confutation purposes. To this aim, 6 years of Meteosat-5 TIR observations have been processed for the characterization of the TIR signal behaviour at each specific observation time and location.The anomalous TIR values, detected by RAT, have been evaluated in terms of time–space persistence in order to establish the existence of actually significant anomalous transients. The results indicate that the studied area was affected by significant positive thermal anomalies which were identified, at different intensity levels, not far from the Gujarat coast (since 15th January, but with a clearer evidence on 22nd January) and near the epicentral area (mainly on 21st January). On 25th January (1 day before Gujarat's earthquake) significant TIR anomalies appear on the Northern Indian subcontinent, showing a remarkable coincidence with the principal tectonic lineaments of the region (thrust Himalayan boundary).On the other hand, the results of the confutation analysis indicate that no meaningful TIR anomalies appear in the absence of seismic events with MS ≥ 5.