Composite Seismogenic Sources (CSS)
The maximum earthquake magnitude of the CSS is estimated as the magnitude value in the moment magnitude scale, corresponding to the largest possible rupture that the fault can host based on its dimensions.
The relation between the rupture size and the moment magnitude value is derived from the fault scaling relations by Leonard (2010, 2014). Following the definition and geographic distribution of stable continents by Johnston (1994), also adopted by Leonard (2010, 2014), the faults are classified as belonging to the interplate or stable continent categories. The variability of all parameters is considered by including the following: 1) the rake, which determines whether the dip-slip or the strike-slip relation is used; 2) the width resulting from the combinations of the upper depth, lower depth, and dip angle; 3) the length (end-to-end for coherence with the scaling relations) considering the top and bottom traces, and 4) the standard deviation of the scaling relation. To explore this variability, we randomly sampled the range of all these parameters and developed a distribution of magnitude values from which we retained the 98th percentile.
Subduction Sources (SDS)
The maximum earthquake magnitude of the SDS is estimated as the magnitude value in the moment magnitude scale, corresponding to the largest possible rupture that the subduction interface can host based on its dimensions, constrained by the upper and lower seismogenic depths.
The relation between the rupture size and the moment magnitude value is derived from the fault scaling relations by Allen & Hayes (2017) in the case of oceanic subduction and by Leonard (2010, 2014) in the case of continental subduction. The variability of all parameters is considered by including the following: 1) the upper depth, 2) the lower depth, and 3) the standard deviation of the scaling relation. The interface geometry, however, is kept fixed. We retained the 98th percentile of the variability distribution.