Grid-Based Probabilistic Earthquake Forecast for Iran

In this paper, a model of earthquake forecast is presented to assess the long-term probabilities of future earthquakes with moderate magnitudes for a region including Iran (latitude 25-40° and longitude 44-62°). The model estimates a coupled rate of magnitude, space and time for future seismicity using a spatial-temporal Poisson process. The smoothed spatial distribution of seismicity is measured by an adaptive kernel using the locations of past M ³ 4.5 earthquakes listed in the ISC catalog in the period of 1980 to 2014. The retrospective area skill score test has been carried out to check the significant of the results, using a spatially uniform reference model. At 95% confidence level, the model was not rejected by the test. Moreover, the results show a meaningful correlation between anomalies of the forecasted map and the epicenters of target events occurred from 2015 to 2016. Based on the results, it is concluded that the areas characterized by high forecasted rates of seismicity could be considered as the highly hazardous ones, most likely to seismic activation in the Iranian plateau.