Risk Calculators

General

Hazard Getters

Hazard input management for Risk calculators.

exception openquake.engine.calculators.risk.hazard_getters.AssetSiteAssociationError[source]

Bases: exceptions.Exception

class openquake.engine.calculators.risk.hazard_getters.GroundMotionGetter(imt, taxonomy, hazard_outputs, assets)[source]

Bases: openquake.engine.calculators.risk.hazard_getters.HazardGetter

get_epsilons()[source]

Expand the underlying epsilons

class openquake.engine.calculators.risk.hazard_getters.Hazard(hazard_output, data, imt)[source]

Bases: object

Hazard objects have attributes .hazard_output, .data and .imt. Moreover you can extract the .hid (hazard_output.id) and the .weight associated to the underlying realization. The hazard .data is a numpy array of shape (N, R) where N is the number of assets and R the number of seismic events (ruptures) or the resolution of the hazard curve, depending on the calculator.

hid

Return the id of the given hazard output

weight

Return the realization weight of the hazard output

class openquake.engine.calculators.risk.hazard_getters.HazardCurveGetter(imt, taxonomy, hazard_outputs, assets)[source]

Bases: openquake.engine.calculators.risk.hazard_getters.HazardGetter

class openquake.engine.calculators.risk.hazard_getters.HazardGetter(imt, taxonomy, hazard_outputs, assets)[source]

Bases: object

A HazardGetter instance stores a chunk of assets and their associated hazard data. In the case of scenario and event based calculators it also stores the ruptures and the epsilons. The HazardGetter must be pickable such that it should be possible to use different strategies (e.g. distributed or not, using postgis or not).

attr assets:The assets for which we want to extract the hazard
Attr site_ids:The ids of the sites associated to the hazards
get_data()[source]

Shortcut returning the hazard data when there is a single realization

get_hazards()[source]

Return a list of Hazard instances for the given IMT.

hid

Return the id of the hazard output, when there is a single realization

class openquake.engine.calculators.risk.hazard_getters.RiskInitializer(taxonomy, calc)[source]

Bases: object

A facility providing the brigde between the hazard (sites and outputs) and the risk (assets and risk models). When .init_assocs is called, populates the asset_site table with the associations between the assets in the current exposure model and the sites in the previous hazard calculation.

Parameters:
  • hazard_outputs – outputs of the previous hazard calculation
  • taxonomy – the taxonomy of the assets we are interested in
  • calc – a risk calculator

Warning: instantiating a RiskInitializer may perform a potentially expensive geospatial query.

calc_nbytes(epsilon_sampling=None)[source]
Parameters:epsilon_sampling – flag saying if the epsilon_sampling feature is enabled
Returns:the number of bytes to be allocated (a guess)

If the hazard_outputs come from an event based or scenario computation, populate the .epsilons_shape dictionary.

init_assocs()[source]

Stores the associations asset <-> site into the database

init_epsilons(epsilon_sampling=None)[source]
Parameters:epsilon_sampling – flag saying if the epsilon_sampling feature is enabled

Populate the .epsilons_shape and the ._rupture_ids dictionaries. For the calculators event_based_risk and scenario_risk also stores the epsilons in the database for each asset_site association.

openquake.engine.calculators.risk.hazard_getters.expand(array, N)[source]

Given a non-empty array with n elements, expands it to a larger array with N elements.

>>> expand([1], 3)
array([1, 1, 1])
>>> expand([1, 2, 3], 10)
array([1, 2, 3, 1, 2, 3, 1, 2, 3, 1])
>>> expand(numpy.zeros((2, 10)), 5).shape
(5, 10)
openquake.engine.calculators.risk.hazard_getters.haz_out_to_ses_coll(ho)[source]
openquake.engine.calculators.risk.hazard_getters.make_epsilons(asset_count, num_samples, seed, correlation)[source]
Parameters:
  • asset_count (int) – the number of assets
  • num_ruptures (int) – the number of ruptures
  • seed (int) – a random seed
  • correlation (float) – the correlation coefficient

Classical PSHA Calculator

Core functionality for the classical PSHA risk calculator.

class openquake.engine.calculators.risk.classical_risk.core.ClassicalRiskCalculator(job)[source]

Bases: openquake.engine.calculators.risk.base.RiskCalculator

Classical PSHA risk calculator. Computes loss curves and loss maps for a given set of assets.

static core(workflow, getter, outputdict, params, monitor)

Celery task for the classical risk calculator.

Parameters:
  • workflow – A openquake.risklib.riskinput.RiskModel instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (e.g. a LossCurve)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance

For each calculation unit we compute loss curves, loss maps and loss fractions. Then if the number of units are bigger than 1, we compute mean and quantile artifacts.

getter_class

alias of HazardCurveGetter

output_builders = [<class 'openquake.engine.calculators.risk.writers.LossCurveMapBuilder'>, <class 'openquake.engine.calculators.risk.writers.ConditionalLossFractionBuilder'>]
validators = [<class 'openquake.engine.calculators.risk.validation.HazardIMT'>, <class 'openquake.engine.calculators.risk.validation.EmptyExposure'>, <class 'openquake.engine.calculators.risk.validation.OrphanTaxonomies'>, <class 'openquake.engine.calculators.risk.validation.ExposureLossTypes'>, <class 'openquake.engine.calculators.risk.validation.NoRiskModels'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasInsuranceBounds'>]
openquake.engine.calculators.risk.classical_risk.core.classical(workflow, getter, outputdict, params, monitor)[source]

Celery task for the classical risk calculator.

Parameters:
  • workflow – A openquake.risklib.riskinput.RiskModel instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (e.g. a LossCurve)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance

For each calculation unit we compute loss curves, loss maps and loss fractions. Then if the number of units are bigger than 1, we compute mean and quantile artifacts.

openquake.engine.calculators.risk.classical_risk.core.save_individual_outputs(outputdict, outs, params)[source]

Save loss curves, loss maps and loss fractions associated with a calculation unit

Parameters:
  • outputdict – a openquake.engine.calculators.risk.writers.OutputDict instance holding the reference to the output container objects
  • outs – a openquake.risklib.workflows.Classical.Output holding the output data for a calculation unit
  • params – a openquake.engine.calculators.risk.base.CalcParams holding the parameters for this calculation
openquake.engine.calculators.risk.classical_risk.core.save_statistical_output(outputdict, stats, params)[source]

Save statistical outputs (mean and quantile loss curves, mean and quantile loss maps, mean and quantile loss fractions) for the calculation.

Parameters:
  • outputdict – a openquake.engine.calculators.risk.writers.OutputDict instance holding the reference to the output container objects
  • outs – a openquake.risklib.workflows.Classical.StatisticalOutput holding the statistical output data
  • params – a openquake.engine.calculators.risk.base.CalcParams holding the parameters for this calculation

Classical PSHA BCR Calculator

Core functionality for the classical PSHA risk calculator.

class openquake.engine.calculators.risk.classical_bcr.core.ClassicalBCRRiskCalculator(job)[source]

Bases: openquake.engine.calculators.risk.classical_risk.core.ClassicalRiskCalculator

Classical BCR risk calculator. Computes BCR distributions for a given set of assets.

Attr dict vulnerability_functions_retrofitted:
 A dictionary mapping each taxonomy to a vulnerability functions for the retrofitted losses computation
static core(workflow, getter, outputdict, params, monitor)

Celery task for the BCR risk calculator based on the classical calculator.

Instantiates risklib calculators, computes BCR and stores the results to db in a single transaction.

Parameters:
  • job_id (int) – ID of the currently running job
  • workflow – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (in this case only BCRDistribution)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance
getter_class

alias of HazardCurveGetter

output_builders = [<class 'openquake.engine.calculators.risk.writers.BCRMapBuilder'>]
validators = [<class 'openquake.engine.calculators.risk.validation.HazardIMT'>, <class 'openquake.engine.calculators.risk.validation.EmptyExposure'>, <class 'openquake.engine.calculators.risk.validation.OrphanTaxonomies'>, <class 'openquake.engine.calculators.risk.validation.ExposureLossTypes'>, <class 'openquake.engine.calculators.risk.validation.NoRiskModels'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasInsuranceBounds'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasRetrofittedCosts'>]
openquake.engine.calculators.risk.classical_bcr.core.classical_bcr(workflow, getter, outputdict, params, monitor)[source]

Celery task for the BCR risk calculator based on the classical calculator.

Instantiates risklib calculators, computes BCR and stores the results to db in a single transaction.

Parameters:
  • job_id (int) – ID of the currently running job
  • workflow – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (in this case only BCRDistribution)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance

Event-Based PSHA Calculator

Core functionality for the classical PSHA risk calculator.

class openquake.engine.calculators.risk.event_based_risk.core.DisaggregationOutputs(assets_disagg, magnitude_distance, coordinate, fractions)[source]

Bases: object

class openquake.engine.calculators.risk.event_based_risk.core.EventBasedRiskCalculator(job)[source]

Bases: openquake.engine.calculators.risk.base.RiskCalculator

Probabilistic Event Based PSHA risk calculator. Computes loss curves, loss maps, aggregate losses and insured losses for a given set of assets.

agg_result(*args)
static core(workflow, getter, outputdict, params, monitor)

Celery task for the event based risk calculator.

Parameters:
  • job_id – the id of the current openquake.engine.db.models.OqJob
  • workflow – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (e.g. a LossCurve)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance
Returns:

A dictionary {loss_type: event_loss_table}
getter_class

alias of GroundMotionGetter

output_builders = [<class 'openquake.engine.calculators.risk.writers.EventLossCurveMapBuilder'>, <class 'openquake.engine.calculators.risk.writers.LossFractionBuilder'>]
post_process()[source]

Compute aggregate loss curves and event loss tables

pre_execute()[source]

Base pre_execute + build Event Loss Asset outputs if needed

validators = [<class 'openquake.engine.calculators.risk.validation.HazardIMT'>, <class 'openquake.engine.calculators.risk.validation.EmptyExposure'>, <class 'openquake.engine.calculators.risk.validation.OrphanTaxonomies'>, <class 'openquake.engine.calculators.risk.validation.ExposureLossTypes'>, <class 'openquake.engine.calculators.risk.validation.NoRiskModels'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasInsuranceBounds'>]
openquake.engine.calculators.risk.event_based_risk.core.disaggregate(outputs, ruptures, params)[source]

Compute disaggregation outputs given the individual outputs and params

Parameters:
  • outputs – an instance of openquake.risklib.workflows.ProbabilisticEventBased.Output
  • params – an instance of base.CalcParams
  • rupture_ids (list) – a list of openquake.engine.db.models.SESRupture IDs
Returns:

an instance of DisaggregationOutputs
openquake.engine.calculators.risk.event_based_risk.core.event_based(workflow, getter, outputdict, params, monitor)[source]

Celery task for the event based risk calculator.

Parameters:
  • job_id – the id of the current openquake.engine.db.models.OqJob
  • workflow – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (e.g. a LossCurve)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance
Returns:

A dictionary {loss_type: event_loss_table}
openquake.engine.calculators.risk.event_based_risk.core.save_individual_outputs(outputdict, outputs, disagg_outputs, params)[source]

Save loss curves, loss maps and loss fractions associated with a calculation unit

Parameters:
  • outputdict – a openquake.engine.calculators.risk.writers.OutputDict instance holding the reference to the output container objects
  • outputs – a openquake.risklib.workflows.ProbabilisticEventBased.Output holding the output data for a calculation unit
  • disagg_outputs – a DisaggregationOutputs holding the disaggreation output data for a calculation unit
  • params – a openquake.engine.calculators.risk.base.CalcParams holding the parameters for this calculation
openquake.engine.calculators.risk.event_based_risk.core.save_statistical_output(outputdict, stats, params)[source]

Save statistical outputs (mean and quantile loss curves, mean and quantile loss maps) for the calculation.

Parameters:
  • outputdict – a openquake.engine.calculators.risk.writers.OutputDict instance holding the reference to the output container objects
  • stats –
    class:

    openquake.risklib.workflows.ProbabilisticEventBased.StatisticalOutput holding the statistical output data

  • params – a openquake.engine.calculators.risk.base.CalcParams holding the parameters for this calculation

Event-Based PSHA BCR Calculator

Core functionality for the Event Based BCR Risk calculator.

class openquake.engine.calculators.risk.event_based_bcr.core.EventBasedBCRRiskCalculator(job)[source]

Bases: openquake.engine.calculators.risk.event_based_risk.core.EventBasedRiskCalculator

Event based BCR risk calculator. Computes BCR distributions for a given set of assets.

agg_result(acc, event_loss_tables)[source]

No need to update event loss tables in the BCR calculator

static core(workflow, getter, outputdict, params, monitor)

Celery task for the BCR risk calculator based on the event based calculator.

Instantiates risklib calculators, computes bcr and stores results to db in a single transaction.

Parameters:
  • job_id (int) – ID of the currently running job
  • workflow – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (in this case only BCRDistribution)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance
output_builders = [<class 'openquake.engine.calculators.risk.writers.BCRMapBuilder'>]
post_process()[source]

No need to compute the aggregate loss curve in the BCR calculator.

validators = [<class 'openquake.engine.calculators.risk.validation.HazardIMT'>, <class 'openquake.engine.calculators.risk.validation.EmptyExposure'>, <class 'openquake.engine.calculators.risk.validation.OrphanTaxonomies'>, <class 'openquake.engine.calculators.risk.validation.ExposureLossTypes'>, <class 'openquake.engine.calculators.risk.validation.NoRiskModels'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasInsuranceBounds'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasRetrofittedCosts'>]
openquake.engine.calculators.risk.event_based_bcr.core.event_based_bcr(workflow, getter, outputdict, params, monitor)[source]

Celery task for the BCR risk calculator based on the event based calculator.

Instantiates risklib calculators, computes bcr and stores results to db in a single transaction.

Parameters:
  • job_id (int) – ID of the currently running job
  • workflow – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (in this case only BCRDistribution)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance

Scenario Calculator

Core functionality for the scenario risk calculator.

class openquake.engine.calculators.risk.scenario_risk.core.ScenarioRiskCalculator(job)[source]

Bases: openquake.engine.calculators.risk.base.RiskCalculator

Scenario Risk Calculator. Computes a Loss Map, for a given set of assets.

agg_result(*args)
static core(workflow, getter, outputdict, params, monitor)

Celery task for the scenario risk calculator.

Parameters:
  • workflow (list) – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (in this case only LossMap)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance
getter_class

alias of GroundMotionGetter

output_builders = [<class 'openquake.engine.calculators.risk.writers.LossMapBuilder'>]
post_process()[source]
validators = [<class 'openquake.engine.calculators.risk.validation.HazardIMT'>, <class 'openquake.engine.calculators.risk.validation.EmptyExposure'>, <class 'openquake.engine.calculators.risk.validation.OrphanTaxonomies'>, <class 'openquake.engine.calculators.risk.validation.ExposureLossTypes'>, <class 'openquake.engine.calculators.risk.validation.NoRiskModels'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasInsuranceBounds'>, <class 'openquake.engine.calculators.risk.validation.ExposureHasTimeEvent'>]
openquake.engine.calculators.risk.scenario_risk.core.scenario(workflow, getter, outputdict, params, monitor)[source]

Celery task for the scenario risk calculator.

Parameters:
  • workflow (list) – A openquake.risklib.workflows.Workflow instance
  • getter – A HazardGetter instance
  • outputdict – An instance of writers.OutputDict containing output container instances (in this case only LossMap)
  • params – An instance of base.CalcParams used to compute derived outputs
  • monitor – A monitor instance
openquake.engine.calculators.risk.scenario_risk.core.workflow_argdict(loss_type, assets, gmfs, epsilons)[source]
Parameters:
  • loss_type – loss type string
  • assets – list of N assets
  • gmfs – list of N ground motion arrays
  • epsilons – list of N epsilon arrays
Returns:

a dictionary with the loss_type and three lists with N - M elements, where M is the number of assets without value (usually 0).

Scenario Damage Calculator

Core functionality for the scenario_damage risk calculator.

class openquake.engine.calculators.risk.scenario_damage.core.ScenarioDamageRiskCalculator(job)[source]

Bases: openquake.engine.calculators.risk.base.RiskCalculator

Scenario Damage Risk Calculator. Computes four kinds of damage distributions: per asset, per taxonomy, total and collapse map.

Attr dict fragility_functions:
 A dictionary of dictionary mapping taxonomy -> (limit state -> fragility function) where a fragility function is an instance of openquake.risklib.scientific.FragilityFunctionContinuous or openquake.risklib.scientific.FragilityFunctionDiscrete.
agg_result(*args)
static core(workflow, getter, outputdict, params, monitor)

The core calculation celery task function

get_risk_model()[source]

Load fragility model and store damage states

getter_class

alias of GroundMotionGetter

output_builders = []
post_process()[source]

Save the damage distributions by taxonomy and total on the db.

validators = [<class 'openquake.engine.calculators.risk.validation.HazardIMT'>, <class 'openquake.engine.calculators.risk.validation.EmptyExposure'>, <class 'openquake.engine.calculators.risk.validation.OrphanTaxonomies'>, <class 'openquake.engine.calculators.risk.validation.NoRiskModels'>]
openquake.engine.calculators.risk.scenario_damage.core.scenario_damage(workflow, getter, outputdict, params, monitor)[source]

Celery task for the scenario damage risk calculator.

param workflow:A openquake.risklib.workflows.Workflow instance
param getter:A HazardGetter instance
param outputdict:
 An instance of writers.OutputDict containing output container instances (in this case only LossMap)
param params:An instance of base.CalcParams used to compute derived outputs
param monitor:A monitor instance
Returns:A matrix of fractions and a taxonomy string