Robust Satellite Techniques (RST) for monitoring earthquake prone areas by satellite TIR observations: The case of 1999 Chi-Chi earthquake (Taiwan)

For more than 13. years a multi-temporal data-analysis method, named Robust Satellite Techniques (RST), has been being applied to satellite Thermal InfraRed (TIR) monitoring of seismically active regions. It gives a clear definition of a TIR anomaly within a validation/confutation scheme devoted to verify if detected anomalies can be associated or not to the time and location of the occurrence of major earthquakes. In this scheme, the confutation part (i.e. verifying if similar anomalies do not occur in the absence of a significant seismic activity) assumes a role even much important than the usual validation component devoted to verify the presence of anomalous signal transients before (or in association with) specific seismic events. Since 2001, RST approach has been being used to study tens of earthquakes with a wide range of magnitudes (from 4.0 to 7.9) occurred in different continents and in various geo-tectonic settings.In this paper such a long term experience is exploited in order to give a quantitative definition of a significant sequence of TIR anomalies (SSTA) in terms of the required space-time continuity constraints (persistence), identifying also the different typologies of known spurious sequences of TIR anomalies that have to be excluded from the following validation steps. On the same basis, taking also into account for the physical models proposed for justifying the existence of a correlation between TIR anomalies and earthquakes occurrence, specific validation rules (in line with the ones used by the Collaboratory for the Study of Earthquake Predictability - CSEP - Project) have been defined to drive the validation process. In this work, such an approach is applied for the first time to a long-term dataset of night-time GMS-5/VISSR (Geostationary Meteorological Satellite/Visible and Infrared Spin-Scan Radiometer) TIR measurements, comparing SSTAs and earthquakes with M >4 which occurred in a wide area around Taiwan, in the month of September of different years (from 1995 to 2002). In this dataset the Chi-Chi earthquake (MW=7.6) which occurred on September 20, 1999 represents the major, but not unique, event. The analysis shows that all identified SSTAs occur in the pre-fixed space-time window around (in terms of time and location) earthquakes with M >4. The false positive rate remains zero even if only earthquakes with M >4.5 are considered. In the case of the Chi-Chi earthquake, 3 SSTAs were identified (all within the established space-time correlation window), one of them appearing about 2weeks before and very close to the epicentre of the earthquake just along the associated tectonic lineaments. The wide considered space-time window, together with the high seismicity of the considered area, surely positively conditioned the achieved results, so that further analyses should be carried out by using longer datasets and different geographic areas. However, also considering the coincidence with other (possible) precursor phenomena, independently reported (particularly within the iSTEP project) at the time of the Chi Chi earthquake, achieved results seem already sufficient (at least) to qualify TIR anomalies (identified by RST) among the parameters to be considered in the framework of a multi-parametric approach to a time-Dependent Assessment of Seismic Hazard (t-DASH).