Skip to main content.

Part Three: Assessing flood risk

This part covers:

  • an overview of the risk assessment process
  • how to qualitatively rate risks from future flooding.

The risk assessment process

A sound risk assessment process is fundamental to ensuring climate change is appropriately factored into the planning and decision-making processes. The purpose of risk assessment, in the context of climate change, is to identify risks and hazards caused or exacerbated by climate change and to evaluate their effects and likelihood. Climate change risks and responses can then be prioritised with more confidence and compared equitably with other risks, resource availability and cost issues.

A broad guide to risk management is presented in the international standard ISO 31000:2009 Risk Management − Principles and Guidelines.6 This is the overarching risk management approach recommended by the Ministry for managing risks associated with climate change.

A high-level, decision-making framework for flood risk management is set out in New Zealand standard NZS 9401:2008 Managing Flood Risk – A Process Standard. This standard sets out a framework, based on accepted best practice, which users can work through as they seek to address their flood management issues. The standard was developed to give guidance on flood risk management, but it is not a detailed technical document.

There are six steps in the risk assessment process:

  1. establish the context
  2. identify hazards and describe the risks
  3. analyse the risks
  4. evaluate the risks
  5. assess appropriate responses based on the risks
  6. communicate, consult, monitor and evaluate.

An overview of this risk assessment process is described in the companion publication, Preparing for Climate Change. More detail on the risk assessment process is contained in sections 4.2.3 and 6.5 of the Climate Change Effects and Impacts Assessment manual (available at www.mfe.govt.nz/publications/climate/climate change-effect-impacts-assessments-may08).

This publication does not go into detail on each individual step because this has been covered elsewhere in the documents and guidance described above. However, some aspects of how to determine changes in flood risk resulting from climate change are discussed. In particular, we will show you how to assess flood risk in terms of social, cultural, economic and environmental consequences. This will enable you to determine levels of risk by combining estimates of the consequences and likelihood of an event occurring, so that flood risk can be prioritised alongside other risks.

Determining flood risk

The next section describes the parts of the risk assessment process that relate to describing, analysing and evaluating the risks of flooding (steps 2 to 4 above). It describes how to:

  • rate the level of consequences of a flood (from insignificant to catastrophic)
  • rate the likelihood of a specific flood event occurring (rare to almost certain)
  • assign a risk level, given both the consequences and likelihood (low to extreme)
  • analyse the results to compare how your risk profile might change with climate change.

This process then enables you to compare any differences between catchments and ensure the priorities for flood risk management are based on a fair and comparable assessment of risk. In other words, the options used to treat risks may vary across a region or district, but the risk assessment process should not.

Rate the consequences of a flood

Flooding can have a range of different social, cultural, economic and environmental consequences. For each of the impacts that are identified you need to think about how you would rate the severity of those consequences using a five-step range for the expected consequences from ‘insignificant’ to ‘catastrophic’. For instance, ‘insignificant’ could mean the results have very little or no cost and only some inconvenience, whereas ‘catastrophic’ could mean financial viability over the long term is compromised, a major disruption in the community or loss of life.

What level of rating you apply to particular consequences should reflect the risks for a specific type of catchment or location. For example, large-scale flooding of rural land may affect relatively few people but can have significant economic consequences at a regional level. Flooding of urban areas is likely to affect more people and could result in serious public health and safety consequences, large business disruptions and significant social upheaval.

An example of consequence ratings is provided in Table 8 below. In practice, you are likely to undertake significant consultation and collaboration with stakeholders when establishing your own consequence ratings for an area.

table 1: An example of consequence ratings

Consequence Rating

Social

Cultural (eg, heritage sites, historic structures, archaeological sites, sites of importance to Māori, such as wāhi tapu)

Economic

Environment
Public safety Community disruption (eg, displaced people, social disruption, cancelled events, school closures) Local economy and growth Lifelines*
Catastrophic Fatalities, or serious near misses, affecting more than 1000 people Significant disruption;
international significance or concern
International significance or concern Regional decline leading to widespread business failure, loss of employment and hardship; significant long-term impact on the national economy Systemic failure of lifeline assets, including lost transport connections, water supply and power failure, and failed wastewater systems Long-term, widespread impacts; slow recovery
Major Some injuries or serious near misses, affecting more than 100 people High-level disruption;
national significance or concern
National significance or concern Regional stagnation such that businesses are unable to thrive and employment does not keep pace with population growth Failure of some lifeline assets (eg, power lines or road access)  that require significant recovery investment and long-term temporary lifeline services Medium- to long-term widespread impacts
Moderate Minor injuries, or serious near misses, affecting more than 10 people Moderate disruption;
regional significance or concern
Regional significance or concern Significant but temporary reduction in economic performance relative to current forecasts Partial failure of some lifeline assets that requires temporary measures to provide lifeline services Reversible medium-term local impacts
Minor Serious near misses, affecting fewer than 10 people Minor disruption;
local community significance or concern
Local community significance or concern Individually significant but isolated areas of reduction in economic performance relative to current forecasts Some short-term disruption of lifeline assets raising public health concerns Reversible short-term impacts on local area
Insignificant Appearance of a threat but no actual harm Individual significance or concern Individual or small significance or concern Minor shortfall relative to current forecasts Minor disruption to lifeline assets Limited impacts on minimal area

Rate the likelihood of a specific event occurring

Likelihood can be expressed either numerically as a percentage chance of an event occurring or described qualitatively from “almost certain” to “rare”. For instance, “almost certain” could mean that something has happened before and is expected to happen again in the next 12 months. “Rare” could mean although something has not happened before in your experience, it is in the realms of possibility.

The numerical likelihood of the probability of a flood occurring within the design life of an asset being considered can be determined using table 2 below. That figure can then be used to determine a qualitative rating to express the likelihood of that flood occurring. Table 3 provides an example of likelihood ranges that you could use, or you may prefer to assign your own categories for describing the likelihoods.

Likelihood should be assessed in terms of the design life of the asset or infrastructure that is at risk from flooding. For example some buildings might more realistically have a 100-year lifespan,7 even though they are only required to be designed for a 50-year lifespan. Therefore, the probability that a damaging flood will occur within that longer 100-year time horizon should be considered. The risk to a subdivision should be analysed over a longer period of time, because once land has been developed for residential use it is more than likely to remain occupied for very long periods of time, if not permanently. For temporary assets (eg, a culvert) or temporary land uses (eg, a camping ground), a shorter time horizon may be appropriate.

To illustrate this process in action consider a flood that might occur once in 100 years (a 100-year event) and an asset that has a design life of 100 years. Using the two tables below, the numerical likelihood from table 2 would be 63 per cent and it would be considered as ‘likely’ to occur. Keep in mind that a “1-in-100” year event means that there is a 1 per cent chance of the event occurring in a single year, not that the event only occurs once every 100 years.

table 2: Likelihood of a flood occurring within a given time horizon.8
Average recurrence interval of flood (years) Design life − time horizon (years)
2 5 10 20 50 100 200 500 1000
2 75% 97% 100% 100% 100% 100% 100% 100% 100%
5 36% 67% 89% 99% 100% 100% 100% 100% 100%
10 19% 41% 65% 88% 99% 100% 100% 100% 100%
50 4% 10% 18% 33% 64% 87% 98% 100% 100%
100 2% 5% 10% 18% 39% 63% 87% 99% 100%
200 1% 2% 5% 10% 22% 39% 63% 92% 100%
500 0% 1% 2% 4% 10% 18% 33% 63% 100%
table 3: flood risk likelihood ratings
Rating Percentage chance that a flood with a given average return interval will occur within the design life
Almost certain > 85%
Likely 60%−84%
Possible 36%−59%
Unlikely 16%−35 %
Rare < 15%

Assign a risk level, given both the consequences and the likelihood

Finally, a qualitative description of risk can be assigned using the risk assignment matrix in Table 11. It enables you to combine the likelihood and consequence ratings for a given return period event as determined in the previous exercises, and assigns a risk value from low to extreme.

table 4: A risk assignment matrix for setting the level of risk, based on likelihood and consequence.
Consequence Rating Insignificant Minor Moderate Major Catastrophic  
Likelihood rating Almost certain Moderate High Extreme Extreme Extreme
Likely Moderate High High Extreme Extreme
Possible Low Moderate High Extreme Extreme
Unlikely Low Low Moderate High Extreme
Rare Low Low Moderate High High

To illustrate this in practice, if we take a hypothetical case of a 100-year event for an asset with a 100-year design life where the consequence rating determined in the previous step using table 1 was ‘moderate’, and the likelihood rating was considered ‘63 per cent – likely’, then the corresponding risk rating from table 4 would be classed as high.

To take this example further, if climate change were to increase the likelihood of the 100-year event occurring so that it in the future it would be the equivalent to a 50-year event (due to increased rainfall intensity), then the likelihood rating from table 3 would increase from ‘likely’ to ‘almost certain’ because there would now be an 87 per cent likelihood of the flood occurring within the asset’s design life. Again, if we assume the consequence rating for this level of flood still has a rating of ‘moderate’, then the risk level assigned to the event in table 4 would change from ‘high’ to ‘extreme’.

Analyse the results of the risk assessment

The next step is to use all of this information to analyse flood risks across a ‘quadruple bottom line’ that considers social, cultural, economic and environmental consequences. The following examples illustrate how a risk analysis can be undertaken to compare how climate change might alter flood risk over time.

Example 1: Current climate – 100-year flood event

To illustrate how a risk analysis could be undertaken, and how the risk analysis is altered by the effects of climate change, we have provided an example of a hypothetical flood (Table 5). In the current climate this hypothetical flood is assumed to flood homes and affect the performance of some lifeline9services, but does not cause injury or long-term economic or environmental consequences. Table 5 provides an example of how, in this hypothetical flood in the existing climate, the likelihood information from table 2 and table 3 and the consequences ratings from table 81can be combined to determine risk.

To illustrate how climate change might alter the current flood risk, we have reanalysed the flood risk for two climate change scenarios for 2100. These are illustrated in table 6 and table 7.

Example 2: 100-year flood event in 2010 – higher flow

In the first of the climate change examples (Table 6), the hypothetical 100-year climate change flood event has a greater magnitude than in the existing climate. Therefore, the consequences of the climate change flood are greater than the consequences of the current climate flood, although the likelihood of the event occurring is no greater.

In this hypothetical example, under the climate change scenario the consequences for public safety, community concern, economy and the environment remain unchanged, but the cultural risk has shifted from high to extreme, and the lifeline risk has also moved from high to extreme.

In the lifeline case, suppose a bridge is close to the threshold for failure. In the existing climate flood event, flooding may cause temporary closure of the road, but in the larger climate change event the flood may lead to failure of the bridge, causing the road to be closed for a prolonged period. In another example, a marae with nationally significant cultural value may be protected by a stopbank and only prone to local ponding in the existing climate flood event. However, in the larger climate change flood the stopbank could be predicted to overtop, flooding the marae and possibly leading to the destruction of a nationally significant taonga. The increase in risk reflects the increase in flood magnitude with climate change.

Example 3: Flood event in 2100 – flooding more frequent

In the third climate change example (table 7) there is an increased likelihood of a flood event occurring in 2100 that has consequences equivalent to the existing flood event. In other words, the 50-year flood event with climate change might have the same consequences as the 100-year flood event in the current climate. The likelihood of these consequences occurring has therefore increased, and hence the risk has increased.

Under the 50-year climate change flood event, public safety and lifelines risk has moved from high to extreme; cultural risk remains at high, because the change from ‘likely’ to ‘almost certain’ does not change the risk category when considering a consequence rating of minor. Community concern and economic and environmental risks remain unchanged. The increase in risks reflects the increase in the likelihood of a flood occurring with climate change.

table 5: Hypothetical risk assessment, current climate 100 - year flood event (Example 1)
 

Social

Cultural

Economic

Environment
Public safety Community disruption Local economy and growth Lifelines
Design lifetime horizon 100 years 10 years 100 years 10 years 100 years 10 years
Consequences
rating
Moderate –
minor injuries, or serious near misses affecting more than 10 people
Major – high-level disruption;
national significance or concern
Minor – local community significance or concern Minor – individually significant but isolated areas of reduction in economic performance relative to current forecasts Moderate – partial failure of some lifeline assets that requires temporary measures to provide lifeline services Minor – reversible short-term impact on local area
Likelihood
rating
Likely Rare Likely Rare Likely Rare
Risk High High High Moderate High Low
table 6: Hypothetical risk assessment, 100-year flood event in 2100 – higher flow
(Example 2)
 

Social

Cultural

Economic

Environment
Public safety Community disruption Local economy and growth Lifelines
Design lifetime horizon 100 years 10 years 100 years 10 years 100 years 10 years
Consequences
rating
Moderate – minor injuries, or serious near misses affecting more than 10 people Major – high-level disruption;
national significance or concern
Major – national significance or concern Minor – individually significant but isolated areas of reduction in economic performance relative to current forecasts Major – failure of some lifeline assets that require significant recovery investment and long-term, temporary lifeline services Minor – reversible short-term impact on local area
Likelihood
rating
Likely Rare Likely  Rare Likely Rare
Risk High High Extreme Moderate Extreme Low
table 7: Hypothetical risk assessment, flood event in 2100– flooding more frequent
(Example 3)

 

Social

Cultural

Economic

Environment

Public safety Community disruption Local economy and growth Lifelines
Design lifetime horizon 100 years 10 years 100 years 10 years 100 years 10 years
Consequences
rating
Moderate – minor injuries or serious near misses affecting more than 10 people Major – high-level disruption;
national significance or concern
Minor – local community significance or concern Minor – individually significant but isolated areas of reduction in economic performance relative to current forecasts Moderate – partial failure of some lifeline assets that requires temporary measures to provide lifeline services Minor – reversible short-term impact on local area
Likelihood
rating
Almost certain Unlikely Almost certain Unlikely Almost certain Unlikely
Risk Extreme High High Moderate Extreme Low

|


6 www.iso.org/iso/catalogue_detail.htm?csnumber=43170 (previously known as NZS 4360:2004).

7 The Building Act 2004 and Building Code currently require residential buildings and community care facilities to be built at a higher elevation than the flood level of a 1-in-50-year event (ie, they set the minimum height). The Building Code does not yet require a flood protection standard for commercial buildings.

8 The probability Pe that a certain-size flood occurring during any period will exceed the 100-year flood threshold can be calculated using Pe = 1 – [1-(1/T)]n where T is the return period of a given storm threshold (eg, 100-yr, 50-yr, 25-yr, and so forth), and n is the number of years.

9 Lifeline services include telecommunications, power, gas, water and roading,