Hi Hyeuk,
1. If you mean the probability distribution used for sampling the number of times a particular rupture on a particular source will occur in a stochastic event set spanning the investigation time, it is not visible to the user—it is derived from the magnitude frequency distribution (MFD) of the source. The MFD is discretized into a set of equally spaced magnitude bins as specified by width_of_mfd_bin. For each magnitude bin, the corresponding annual occurrence rate is calculated. All ruptures of a particular magnitude on a particular source are assumed to be equally probable, and thus the corresponding annual occurrence rate is split equally amongst ruptures of the same magnitude. Depending upon the recurrence model, the corresponding probability distribution for the number of occurrences of each rupture generated by a particular source is obtained. Let me know if I misunderstood your question.
2. The aggregate loss curve for total economic losses (i.e., adding losses to structural and nonstructural components, contents and business interruption) is not computed by OpenQuake. The aggregate losses for each component are provided for each rupture in the stochastic event set, so this loss curve should be easy to obtain using a spreadsheet. Obtaining the loss map for total economic losses is not as trivial, as it will require post-processing of the asset loss tables which are currently available only in the HDF5 datastore.
3. For plotting the aggregate loss curve, calculating the average annual loss, and/or obtaining the losses corresponding to specified return periods, please refer to the
Probable Maximum Losses notebook in the Risk Module of the Risk Modeller's Toolkit.
Regards,
Anirudh