2. Foundations for Hazard and risk assessment

Question 1

3 Components of risk assessment

Answer 1

Identification
Analysis
Evaluation

Question 2

Define risk identification

Answer 2

Recognise and describe the risks (may be associated with a single hazard or multi-hazards) and the scope of the assessment

Question 3

Define risk analysis

Answer 3

Understand the nature and sources of the risk and estimate the level of risk (a quantitative or qualitative calculation of H, E, V and risk)

Question 4

Define risk evaluation

Answer 4

Compare risk with risk criteria (acceptable, tolerable, unacceptable)

Question 5

What are the main components of hazard analysis?

Answer 5

Frequency/magnitude relationship, initiation process and travel or transmission of hazard

Question 6

What are the 6 scales of assessment for different hazards?

Answer 6

Global
Continental/large countries
National
Provincial
Municipal
Community

Question 7

What are statistical models? (empirical, black-box, empirical-statistical)
Method:
Data:
Assumption:
What can they be used to determine

Answer 7

Method: Derive a statistical relationship between observed hazard drivers and resulting hazard event. Hazard modelled using statistical relationship to transform inputs to outputs.
Data: observed hazard events with associated triggers (probabilities) and system properties (prep factors)
Assumption: repeatability and stationarity of the system (past is key to future). Only applicable in similar settings

Used for hazard magnitude frequency curve (probability). Cumulative probability curves, use distributions.

Question 8

What are physically based models? (deterministic or grey/white box models)
Method:

Answer 8

Method: hazard processes are explicitly represented by physics-based equations solved using numerical modelling (e.g. FE)
Data: measured physical parameters of the system
Assumptions: the model and input data describe the physical system

In reality env systems can never be fully defined, use a mixture of measured and empirical inputs calibrated against obs –> grey box

Question 9

Other types of hazard model (4)

Answer 9

Qualitative - info evaluated and analysed verbally based on judgement
Quantitative - info is recorded analysed and evaluated used numerical scales and techniques
Heuristic - experts use indicators/decision rules to assess hazard on semi-quantitative indices
Stochastic and probabilistic - input param values sampled from prob distribution, model run multiple times with different input values to represent full parameter space and account for the effects of uncertainty
Spatially distributed - implemented in Geographical Information Systems

Question 10

Sources of model uncertainty (6)

Answer 10

1) Aleatory
2) Epistemic (system dynamics)
3) Epistemic (forcing and response data)
4) Epistemic (disinformation)
5) Semantic/linguistic
6) Ontological

Question 11

2 methods of attributing value to human life

Answer 11

Human capital method:
Based on an individual’s lost future earning capacity in the event of accident or death. The life of a child has the highest value; zero value for people who are unable to work.

Willingness to pay approaches:
Measures risk aversion in terms of how much people would be prepared to pay to avoid a certain reduction in their chance of accident or premature death. Assessed by questionnaires.

Question 12

Aim of risk management?

Answer 12

Bring risk levels down to societally and economically acceptable, or at least ‘tolerable’ levels. Cost-benefit analyses used to prioritise risk reduction resources. Principle: make risks As Low As Reasonably Practicable (ALARP).

Question 13

In a risk assessment we first….. and then……

Answer 13

Identify the hazard scenario of interest - can be individual events or different hazards caused by the same event (cascades)

Then select the modelling approach depending on hazard type, purpose of assessment, scale of analysis and data and model availability.

Question 14

Define aleatory uncertainty

Answer 14

Inherent randomness of the system. Uncertainty with the stationary statistical characteristics (natural variation). May be structured but can be reduced to a stationary random distribution

Question 15

Define epistemic uncertainty (system dynamics)

Answer 15

Incomplete or insufficient knowledge of the system and lack of data. Uncertainty arising from lack of knowledge about how to represent the catchment system in terms of both model structure and parameters.

Question 16

Define epistemic uncertainty (forcing and data response)

Answer 16

Uncertainty arising from lack of knowledge about the forcing data or the response data with which model outputs can be evaluated. May be because of commensurability or interpolation issues when not enough info is provided by the observational techniques to adequately describe variables required in the modelling process.

Question 17

Define epistemic uncertainty (disinformation)

Answer 17

UC in either system representation or forcing data that are known to be inconsistent or wrong.

Question 18

Define semantic or linguistic uncertainty

Answer 18

Uncertainty about what statements or quantities in the relevant domain actually mean.

Question 19

Define ontological uncertainty

Answer 19

Uncertainty associated with different belief systems.

Question 20

Purposes and functions of risk assessments (4)

Answer 20

1) Policy - high lvl lobbying
2) Planning - hazard zoning and hotspot analysis
3) Management - high-level of hazard characterisation needed
4) Operational - basis for hazard warning and response

All require different scales (global - regional) and different modes of analysis (diff model types)

Question 21

Aim of risk management

Answer 21

To bring risk levels down to societally and economically acceptable, or at least ‘tolerable’ levels. C-B analysis used to prioritise risk reduction resources. Principle is to make risks ALARP.

Question 22

Typical risk modelling frameworks

Answer 22

QRA
Event Tree Analysis
Risk Matrices
Risk Indices

Question 23

What is quantitative risk analysis (QRA)

Answer 23

• Components of the risk equation are quantified for at least three hazard scenarios of different frequencies
• Vulnerability is on a scale of 0 to 1.
• The expected losses associated with each hazard scenario are plotted to create a risk curve: log frequency on y-axis, loss on x-axis
• The area under the curve gives the total losses for all scenarios
• Epistemic uncertainty may be accounted for with probabilistic analysis to give ranges of losses

Question 24

What is event tree analysis?

Answer 24

This method is useful where one hazard triggers another (cascading multi-hazards)

All possible chains of events are considered

The resulting probabilities are calculated for each possible outcome (the end of each branch)

Question 25

What are risk matrices (or Consequence-Frequency Matrices CFM)?

Answer 25

Not all risks can be numerically quantified due to uncertainties or lack of data

Expert judgement can be used to estimate hazard frequency, magnitude or intensity and the consequences (exposure and vulnerability)

Question 26

What are indicator based approaches?

Answer 26

Often used where the components of risk cannot all be numerically quantified due to lack of data; the analysis being carried out over extensive areas; and/or the need to represent qualitative information on vulnerabilities or capacities of at-risk elements.

The area of interest is sub-divided into units that can be assessed against indices (e.g. census enumeration districts and associated building types and poverty indicators).

Indices are standardised from 0 to 1 and weighted according to expert judgement before being combined to give hazard and consequence and the final risk index.