OpenQuake for Advanced Users

Latest (master) PDF version: https://docs.openquake.org/oq-engine/advanced/OpenQuakeforAdvancedUsers.pdf

Contents:

• Introduction
• Common mistakes: bad configuration parameters
  • The quadratic parameters
  • Maximum distance
  • pointsource_distance
  • The linear parameters: width_of_mfd_bin and intensity levels
  • concurrent_tasks parameter
• Tips for running large hazard calculations
  • Reducing a calculation
  • Classical PSHA for Europe (SHARE)
  • GMFs for California
  • Collapsing the GMPE logic tree
• Tips for running large risk calculations
  • Understanding the hazard
  • region_grid_spacing
  • Collapsing of branches
  • Using collect_rlzs=true in the risk calculation
  • Splitting the calculation into subregions
  • Trimming of the logic-trees or sampling of the branches
  • Disabling the computation of the epsilon matrix
• The asset loss table
  • The Probable Maximum Loss (PML) and the loss curves
  • Aggregating by multiple tags
• Advanced logic trees
  • extendModel
  • The logic tree demo
• The concept of effective realizations
• Reduction of the logic tree
• How to analyze the logic tree of a calculation without running the calculation
• Source Specific Logic Trees
  • Extracting the hazard curves
• Logic tree sampling strategies
• Rupture sampling: how does it work?
  • The case of multiple tectonic region types and realizations
  • The difference between full enumeration and sampling
• Extra tips specific to event based calculations
  • Sampling of the logic tree
  • Convergency of the GMFs for non-trivial logic trees
• The concept of “mean” ground motion field
  • Mean ground motion field by GMPE
  • Mean ground motion field by event
  • Single-rupture estimated median ground motion field
• Site-specific classical calculations
• Extracting data from calculations
  • Plotting
  • Extracting ruptures
• Reading outputs with pandas
  • Example: how many events per magnitude?
• Parametric GMPEs
  • Signature of a GMPE class
  • GMPETable
  • File-dependent GMPEs
  • MultiGMPE
  • GenericGmpeAvgSA
  • ModifiableGMPE
• MultiPointSources
• MultiFaultSources
• How the parallelization works in the engine
  • How to use openquake.baselib.parallel
• Special features of the engine
  • The custom_site_id
  • The minimum_distance parameter
  • GMPE logic trees with weighted IMTs
  • Equivalent Epicenter Distance Approximation
  • Ruptures in CSV format
  • max_sites_disagg
• The point source gridding approximation
  • Application: making the Canada model 26x faster
• How the hazard sites are determined
• Extended consequences
  • discrete_damage_distribution
  • The EventBasedDamage demo
  • The ScenarioDamage demo
• The conditional spectrum calculator
• Correlation of Ground Motion Fields
• Scenarios from ShakeMaps
  • Running the Calculation
  • Correlation
  • Performance Considerations
• New risk features
  • Taxonomy mapping
• Risk profiles
  • Single-line commands
  • Caveat: GMFs are split-dependent
• Developing with the engine
  • Prerequisites
  • The first thing to do
  • Understanding the engine
  • Running calculations programmatically
  • Case study: computing the impact of a source on a site
  • Working with GMPEs directly: the ContextMaker
  • Working with verification tables
• Architecture of the OpenQuake engine
  • Components of the OpenQuake Engine
  • Design principles
• Some useful oq commands
  • oq zip
  • Importing a remote calculation
  • plotting commands
  • prepare_site_model
  • Reducing the source model
  • Comparing hazard results
• Limitations of Floating-point Arithmetic