SHERIFS; open-source code for computing earthquake rates in fault systems and constructing hazard models

Modeling the seismic potential of active faults and their associated epistemic uncertainties is a fundamental step of probabilistic seismic-hazard assessment (PSHA). Seismic hazard and earthquake rate in fault systems (SHERIFS) is an open-source python code that builds hazard models including earthquake ruptures involving several fault sections or fault-to-fault (FtF) ruptures. It contains user-friendly tools to calculate the annual rate of FtF ruptures in a fault system based on the slip-rate estimates and accounting for associated background seismicity. SHERIFS applies a forward incremental approach following three rules: (1) the FtF ruptures allowed in the fault system are defined as input by the user and explored randomly, (2) the magnitude-frequency distribution of the modeled seismicity in the fault system must follow an imposed shape, and (3) the slip-rate budget attributed to each fault section must be preserved in the calculation if the first two rules allow it. Indeed, in some cases, a fraction of the slip-rate budget must be considered as being spent in non-mainshock events such as creep or postseismic slip. Background seismicity rates are defined by the hazard modeler as the ratio of seismicity occurring on the modelled faults for different ranges of magnitude. Given a coherent set of input hypotheses, SHERIFS allows end users to build the seismic-hazard fault model thanks to an interactive user-friendly interface. It aims to help interactions between field data collectors and hazard modelers to explore and weight epistemic uncertainties affecting the input hypotheses. To do so, SHERIFS includes tools to compare modeled earthquake rates with the available local data (earthquake catalog and paleoseismological data). This comparison can be used to weigh different hypotheses explored in a logic tree and discard the hypotheses that are not in agreement with the data. SHERIFS's outputs are in a format that can be used directly as inputs for PSHA in the OpenQuake engine (Pagani et al., 2014).