This paper presents a near real-time evaluation of earthquake hazard after the shallow ML 6.2 Hualien shock of depth 10.6 km occurred 2350 LT, 6 February 2018 at epicenter (24.10°N, 121.73°E). Since a large aftershock intends to bring up a new aftershock sequence, we develop a time-magnitude hazard function for the double aftershock sequences. Note that the double-sequence aftershock hazard model can be regarded as a generalization of the Reasenberg-Jones (RJ) model and hence is denoted by DSRJ. For a near real-time evaluation of the aftershock hazard in the golden window particularly for emergent rescue work, the DSRJ or RJ model is estimated based on early aftershock sequence and then the occurrence rate and number of the forthcoming aftershocks in 3 days after the Hualien main shock are forecasted. Results of a data analysis show that the DSRJ model is better than the RJ model on near real-time assessment of short-term aftershock hazard. This provides an evidence that the aftershocks occurred in a short time after the Hualien main shock may not be a single aftershock sequence.
- Bayesian information criterion
- Double-sequence aftershock hazard model
- Maximum likelihood estimate
- Reasenberg-Jones model