View all publications

5 Developing the scenarios

Key points:

When developing future scenarios, include the following –

  • Identify the scenario categories to be explored. The principal category will be climate change, but it may also be appropriate to consider changes in population and land use, for example.

  • Use the guidance in Table 5.1 in a staged approach to assessments. This table provides information for developing initial scenarios for screening assessments, and for undertaking more in-depth studies if a screening assessment indicates these are warranted.

  • Consider the cost involved with different scenario approaches and the relative sensitivity of the natural resource to the effects that are to be examined. This will influence the scenario approach to be taken (refer back to section 4.2 for guidance).

  • Use the published information on climate change scenarios, the CLIMPACTS system where appropriate, and consult relevant experts, to identify the specific scenarios to be used.

  • Identify additional expertise that will be required to quantify other scenarios and to quantify effects.

  • Remember always that, whichever scenarios are chosen, they will be bound by important assumptions and thus will provide information only on plausible futures.

5.1 Introduction

This chapter provides guidance on developing and applying scenarios. A scenario is a plausible description of how the future may develop, based on a coherent and internally consistent set of assumptions about key drivers. Climate, social and economic scenarios can be formulated that span the likely range of future conditions. These can then be used together with expert knowledge and models of the sensitivity of natural or managed systems to climate (the information outlined in chapter 4) to deduce a range of possible climate impacts on selected council activities and services.

Scenario analysis is an appropriate tool for effects assessment because it is not feasible to make a definitive single quantitative prediction of exactly how much a particular climatic element (for example, heavy rainfall intensity) will change over the coming decades. This is because rates of climate change will depend on future global emissions of greenhouse gases, which in turn depend on global social, economic and environmental policies and development. Incomplete scientific knowledge about some of the processes governing the climate, and natural year-to-year variability, also contribute to uncertainty about the future. Thus, it is necessary to consider a range of possible futures when assessing climate impacts and developing adaptation strategies.

As already outlined in chapters 1 and 4, we recommend a staged approach to assessing climate effects. For a particular council function or service, this begins with a straightforward initial screening assessment using simple initial estimates of how climate factors relevant to this function may change. A more detailed effects study is justified only if this initial analysis indicates that material climate change impacts or opportunities are likely, at least for the upper end of the scale of potential future climate changes. The current chapter provides guidance both on simplified scenarios for initial screening assessments, and on which scenarios to use in more detailed studies when these are justified.

In developing scenarios, factors other than those related to climate variables need to be considered. Climate change will occur along with many other changes (social, economic, environmental) that are expected to occur in the coming decades, many of which also carry uncertainties. It must also be remembered that climate change is an underlying long-term trend on which future variability in climate will be superimposed. Therefore, deciding on the scenarios to use, is an important step. The scenarios chosen will determine the bands of uncertainty that will be quantified, which will feed through to decision-making processes in terms of what to do and what not to do about climate change, and over what timeframe.

There are three broad categories of scenario that should be considered:

  • social

  • economic

  • physical/environmental.

Each of these is described briefly in this chapter. Some examples are provided to illustrate how scenarios can be developed and applied, drawing in part from the climate change information provided in chapters 2 and 3. When developing all scenarios, it is good practice to consider the range of uncertainty, which may encompass the upper and lower end of projected climate change, high and low population projections, or different scenarios for economic development.

5.1.1 Making use of this chapter

This chapter identifies the key aspects that need to be considered in developing scenarios for a climate change impact assessment, and provides some guidance to help develop a scenario study (or impact or effects assessment). The key things to consider are as follows.

  1. Climate change will not occur in isolation from other changes. Thus scenarios need to be developed for more than just climate (refer to section 5.2 below, and the examples).
  2. A staged approach to developing and applying scenarios of climate change is recommended. Refer to Table 5.1 for examples, and sources of information/expertise for more in-depth scenario studies.
  3. Be sure to consider the range of uncertainty (see the example in Figure 5.2).

A note on timeframes

Climate has effects over a range of timeframes. On an interannual basis, there are variations in climate that can be affected in some years by El Niño or La Niña events. On a decadal basis, there are fluctuations associated with, for example, the Interdecadal Pacific Oscillation (IPO: see chapter 3). Climate change, in the context of this Guidance Manual, relates to changes that are only now becoming apparent as underlying trends and that will be manifested increasingly over the next hundred years.

Local government planning can also occur over a range of timeframes. Infrastructure investments (such as flood protection) generally consider 50- to 100-year timeframes, which are consistent with climate change. Other planning activities (a good example might be biosecurity) are not presently closely linked to climate timeframes. Marrying decision-making timeframes with climate science timeframes needs to be an implicit part of an assessment of climate change effects.

Box 5.1: Screening assessment

In this Guidance Manual, we recommend an initial screening assessment for a particular council function, activity or service, to decide whether a more detailed climate change effects assessment and formal risk analysis is warranted. A screening assessment can be done for a particular function or service (Roadmap R1) or across all council activities (Roadmap R2).

The first step of a screening assessment is to identify whether a particular function or service is important to your council and whether it might be sensitive to climate change. This can be done by simply ticking the ‘yes’, ‘maybe’ or ‘no’ column for questions in the screening assessment table provided below, taking into account the context of evolving risk over time and the life of the project. More detail on the characteristics to be evaluated is given in section 7.2. If you answer ‘yes’ or ‘maybe’ to Question 1 or 2, as well as to any of Questions 3–6, you should then probably answer ‘yes’ to Question 7 and undertake a scenario-based initial screening analysis using climate scenario guidance from the screening assessment column of Table 5.1. This analysis may well be all that is needed. However, if the results of this analysis lead you to answer ‘yes’ or ‘maybe’ to Question 8, a more in-depth risk assessment is appropriate, using more detailed scenario guidance from the right-hand column of Table 5.1.

Characteristic

Question

Yes

Maybe

No

Current driver

1.   Is there an existing problem that may be exacerbated by climate change? (eg, recurrent inundation)

 

 

 

Future driver

2.   Is there a foreseeable problem that may be caused or exacerbated by climate change?

 

 

 

Complexity

3.   Is this a complex issue?
(eg, locating a new suburb as opposed to locating one house)

 

 

 

Location

4.   Is the location sensitive to climate change?
(eg, a flood plain as opposed to bedrock hillside)

 

 

 

Duration

5.   Is it a permanent long-term change?
(eg, locating a new suburb as opposed to permitting development of a campsite)

 

 

 

Extent

6.   Does it involve a lot of infrastructure and services provided? (eg, in an urban area as opposed to being remote rural)

 

 

 

7.   Is an initial screening analysis using a screening scenario from Table 5.1 justified?

 

 

 

8.   Does this scenario-based initial screening analysis indicate that material climate change impacts might occur? (A screening analysis must be performed in order to answer this question.)

 

 

 

9.   Should a full risk assessment be done for this issue?

 

 

 

5.2 Developing scenarios

5.2.1 Social scenarios

The most obvious social scenarios relate to demographic changes, of which changes in population size and distribution are probably the most commonly used. Future population changes are likely to have a significant influence on the demand and supply of local government functions and services, and consequently on the natural resources that are managed by local government.

5.2.2 Economic scenarios

Regional and district councils are increasingly focusing on economic development goals and how these relate to their functions and services. Future trends in activities such as agriculture, industrial development and tourism will all have consequences for local governments and the resources they manage. In predominantly rural regions, changes in land use could have significant effects on resource demand and supply. For example, a trend towards land-use intensification could lead to increased demand for water. Similarly, growth in industry and tourism will have significant effects in various regions.

5.2.3 Physical/environmental scenarios

The predominant physical or environmental scenarios that will need to be developed relate to possible future changes in climate. As described in Box 5.1, we recommend a staged approach to impact assessment, which involves a preliminary screening assessment followed by a more detailed analysis if justified by the screening process. This requires two levels of scenario development: simple initial scenarios, and more detailed scenarios for in-depth analysis.

Table 4.3 summarises the sources of information currently available, for assessing the effects of changes in climate. Generally, making use of the sources requires specific climate values (or statistics based on these) as input. These statistics, or methods for deriving them, are not always immediately apparent from chapter 2. Therefore, some further advice on obtaining appropriate climate parameters is provided in Table 5.1.

Scenarios for initial screening assessment. The second column of Table 5.1 outlines how to obtain region-specific values of climate parameters for use in these initial scenario studies, based largely on numbers available from this Guidance Manual. The emphasis is on mid-range climate projections. If use of these mid-range values in a screening assessment indicates that material climate change impacts or opportunities are plausible, then a more detailed analysis is recommended (Figure R1). If the mid-range scenario does not reveal any significant impacts, it is good practice to also examine the effects resulting from a scenario near the upper bound of possible future climate changes. This initial screening analysis is essentially a climate sensitivity study. It may also be useful to examine historical data, perhaps carrying out statistical analysis (as used in the North Shore City example in section 5.5 below), or use data from past events (eg, floods, droughts, warmer years) as analogues for the future.

Scenarios for more detailed studies. If the initial screening assessment suggests that material climate impacts are plausible, then more detailed scenarios are often needed for the subsequent detailed study. These may rely on a more complex physical or statistical modelling approach, draw on detailed analyses of current climate statistics in a location, and cover the high and low ends of the downscaled (to New Zealand) IPCC SRES scenario bands. These detailed scenarios should be considered across timeframes that are relevant to the particular function or natural resource being considered. Guidance on developing these more detailed scenarios is provided in the third column of Table 5.1.

Table 5.1: Values for, or sources of, climate parameters suggested for use in scenario analysis.

Climate factor For screening assessment scenarios For detailed study scenarios

Mean temperature

Mid-range 2040 and 2090 projections (upper panels Figure 2.3; central values from Tables 2.2 and 2.3)

Low, mid and high scenarios from ranges given in Tables 2.2 and 2.3, or approach a science provider for regional numbers

Frost occurrence

For 2090, two top panels of Figure 2.8. For 2040, use mid-range CLIMPACTS1 or move current seasonal frequency distribution of daily minimum temperature to the right by seasonal mean change2

Use CLIMPACTS to develop low, medium and high scenarios for frost changes, and/or approach a science provider for regional numbers

Extreme high temperatures

For 2090 use lower two panels of Figure 2.8

Use CLIMPACTS to develop low, medium and high scenarios for maximum temperatures and/or approach a science provider for location-specific weather generator-based results

Growing degree-days (GDDs)

Use CLIMPACTS for a mid-range scenario

Use CLIMPACTS to develop low, medium and high scenario changes for GDDs; approach a science provider for location-specific projections

Winter chilling

 

Approach a science provider for weather generator-based location-specific projections

Mean rainfall (annual, seasonal)

Mid-range 2040 and 2090 projections (lower panels Figure 2.3; central values from Tables 2.4 and 2.5)

Low, mid and high scenarios from ranges given in Tables 2.4 and 2.5

Heavy rainfall3

Use factors from Table 5.2 with 5, 10, 50, 100 year ARI values from HIRDS4 or from local data analyses

Obtain assistance from a science provider with site-specific applications of the gamma function analysis outlined in Appendix 3, or obtain updated guidance based on modelling results published after this Guidance Manual

Flood

Use factors from Table 5.2 with the rainfalls used to drive the design floods

Approach specialist hydrologists for targeted advice

Water deficit (for irrigation)

 

Use weather generator in CLIMPACTS for locations of interest, for low, middle and high greenhouse gas scenarios

Snow

Assume snowline rises by 140 m for each 1°C increase in annual average temperature

Requires research and development of linked spatial weather generator/snow budget modelling software for future projections

Strong winds

Increase 99th percentile wind speed by 10% for 2090

Changes in the frequency of strong winds and ARI of damaging winds are still very uncertain. Consult with a science provider if screening indicates possible problems

Sea level, coastal hazard

Refer to the Coastal Hazards and Climate Change manual (Ministry for the Environment 2008)

Refer to the Coastal Hazards and Climate Change manual (Ministry for the Environment 2008)

Notes for Table 5.1: These are suggestions for scenario analyses, and not firm scientific predictions. Entries in this table – especially for strong winds and heavy rainfall – are likely to be revised as science and modelling develop further. Many of the entries in the ‘initial screening study’ column focus on 2040 and 2090. For other planning horizons within the coming century, climate factors for screening studies can be estimated by interpolating between present, 2040 and 2090 values.

  1. CLIMPACTS is an integrated assessment model developed by the International Global Change Institute (IGCI, University of Waikato) and a consortium of CRIs (see Glossary). An ‘open-framework’ version of the model, called ‘SimCLIMtm’, is now available which allows users to develop their own model for any area and spatial resolution. To find out more about CLIMPACTS or SimCLIMtm, contact CLIMsystems at http://www.climsystems.com/../../index.php (3 April 2008).
  2. This requires site-specific historical temperature data.
  3. As explained in section 2.2.4, there is still considerable uncertainty about the likely size of future changes in heavy rainfall events. The heavy rainfall guidance provided here should continue to be viewed as interim.
  4. HIRDS is the High Intensity Rainfall Design System available on CD from NIWA, or (HIRDS Version 3) via the NIWA website.

Table 5.2 shows recommended percentage adjustments per 1°C of warming to apply to extreme rainfalls when you are developing screening assessment scenarios. This is a new table and supersedes the corresponding table in the previous edition of this Manual. Note that preliminary analysis of NIWA regional climate model results indicates that increases substantially higher than the upper limit of 8% given in this table are possible in limited areas.

As indicated in Table 5.1, current extreme rainfall rates for selected locations, durations and average recurrence intervals (ARIs) can be obtained from analysis of historical rainfall datasets from particular sites, or from the HIRDS CD. For temperature, use the projected changes in annual mean temperature from the rightmost columns of Tables 2.2 and 2.3, or from Figure 2.3. At least two screening calculations should be undertaken – for low and high temperature change scenarios. A worked example of the application of this information is provided in Appendix 4. In carrying out such site-specific analyses, one should also bear in mind the uncertainties in return period estimates for the present climate. In many places, rainfall records cover a past period of only a few decades, so that design rainfall estimates for 50- or 100-year ARIs contain statistical assumptions and data-based uncertainties.

Table 5.2: Factors for use in deriving extreme rainfall information for screening assessments.

ARI (years)

Duration

2 5 10 20 30 50 100

< 10 minutes

8.0

8.0

8.0

8.0

8.0

8.0

8.0

10 minutes

8.0

8.0

8.0

8.0

8.0

8.0

8.0

30 minutes

7.2

7.4

7.6

7.8

8.0

8.0

8.0

1 hour

6.7

7.1

7.4

7.7

8.0

8.0

8.0

2 hours

6.2

6.7

7.2

7.6

8.0

8.0

8.0

3 hours

5.9

6.5

7.0

7.5

8.0

8.0

8.0

6 hours

5.3

6.1

6.8

7.4

8.0

8.0

8.0

12 hours

4.8

5.8

6.5

7.3

8.0

8.0

8.0

24 hours

4.3

5.4

6.3

7.2

8.0

8.0

8.0

48 hours

3.8

5.0

6.1

7.1

7.8

8.0

8.0

72 hours

3.5

4.8

5.9

7.0

7.7

8.0

8.0

Note: This table recommends percentage adjustments to apply to extreme rainfall per 1°C of warming, for a range of average recurrence intervals (ARIs.). The percentage changes are mid-range estimates per 1°C and should be used only in a screening assessment. The entries in this table for a duration of 24 hours are based on results from a regional climate model driven for the A2 SRES emissions scenario. The entries for 10-minute duration are based on the theoretical increase in the amount of water held in the atmosphere for a 1°C increase in temperature (8%). Entries for other durations are based on logarithmic (in time) interpolation between the 10-minute and 24-hour rates. Refer to the discussion in section 2.2.4.

Applications of climate change scenarios for screening and more detailed assessments are shown in Figures 5.1 and 5.2. In these examples, changes in the area of land suitable for kiwifruit have been calculated. An initial screening assessment, using a mid-range scenario, indicated that the Bay of Plenty climate could become unsuitable for kiwifruit by the end of this century (Figure 5.1). A more in-depth study was then carried out (Figure 5.2), to evaluate incremental changes over the next 100 years and to identify the range of uncertainty associated with these changes.76 This latter result gives more detailed information about when climate change could become critical for the kiwifruit industry in the Bay of Plenty.

Figure 5.1: An example of a screening assessment of kiwifruit suitability, using a mid-range climate scenario for the year 2095.

 See figure at its full size (including text description).

Figure 5.2: An example of a detailed scenario analysis showing changes in the area suitable for kiwifruit in the Bay of Plenty, using a range of scenarios over a 100-year timeframe.

 See figure at its full size (including text description).

The following examples describe the developed scenarios for some climate change studies that have been undertaken recently.

5.3 Example 1: Southland water resources

In a State of the Environment report for water, Environment Southland (2000) identified three main drivers of change that would have impacts on Southland’s freshwater environment in future. These were: environmental drivers (principally climate); population changes; economic development. Trends were identified in all three of these drivers as follows:

  • Environmental: The greatest change has been an increase in the average minimum temperature in Southland over the last 40 years. The daily temperature range is decreasing at a greater rate than elsewhere in New Zealand.

  • Population: Both urban and rural areas in Southland have been experiencing population declines since the late 1970s.

  • Economic: Agriculture is a major contributor to the Southland economy and accounts for 82% of the total land area in the region that is not conservation land. Agricultural activities are the largest contributor of nutrients, microbiological and other contaminants, to freshwater resources. Changes in land use can have a major effect on resultant environmental pressures. Over the last decade, there has been a rapid expansion of dairy farming and associated industry infrastructure. Other economic activities that could lead to increased pressure on freshwater resources in the future include tourism.

Thus, if Environment Southland were intending to conduct a study on the possible effects of climate change on Southland’s freshwater environment, it would need to consider changes in the above key drivers over the next 30–100 years. This would require some consideration of alternative scenarios for each driver (see Table 5.3 for examples).

In this example, some linkage has been made between the different climate change scenarios and different population and economic scenarios. The climate change information presented here is drawn from the information provided in chapter 2. It is important to reiterate that these are presented as plausible futures only and also that there will be sub-regional variation in climate change parameters.

Table 5.3: Scenarios for key drivers affecting Southland freshwater resources.

  Environment Population Economic

Scenario 1

Low scenario of climate change:

  • Slight temperature changes, in the order of 0 to 0.5ºC in most seasons

  • Slight increase in summer rainfall, decreases of –20% to –10% in other seasons

Downward trend in population stabilises, with low growth over the next 50–100 years

Moderate land-use changes, with slightly warmer and drier average conditions

Scenario 2

High scenario of climate change:

  • Temperature increases in the order of 3ºC, with greater increases in winter than in summer

  • Precipitation increases greater than 20% in all seasons, with likely increased proportion that falls as heavy rain

Downward trend in population stabilises, with more rapid growth over the next 50–100 years owing to more favourable climate (particularly for the agricultural sector, such as dairy farming)

Greater intensification of land use with warmer, wetter conditions

Note: Climate change scenarios for the 2080s only (from an earlier assessment than the current report) are used here, and are provided here in summary form.

5.4 Example 2: Water resource changes in three river catchments

The Ministry of Agriculture and Forestry commissioned Lincoln Environmental and NIWA “to quantify the potential change in agricultural water usage and availability due to climate change, and assess the implication of these changes on the potential pressures on water sources and water allocation issues” (Lincoln Environmental 2001).

Changes in three river catchments were studied: Rangitata in South Canterbury, Motueka in Nelson, and Tukituki in Hawke’s Bay.

For this study, scenarios were developed for two of the three categories identified above. These were the environmental (climate and river flow changes) and economic (land-use changes) categories. However, as is explained below, the land-use changes themselves were generated principally from projected climate changes.

Climate and river flow changes

The main steps in developing these scenarios were:

  • gathering historical climate and river flow data for the period 1971–1995 for selected sites in each catchment

  • generating two climate change scenarios for 2050, and site changes (monthly) for precipitation, maximum temperature, minimum temperature, dew point temperature and wind run. Two different General Circulation Models were used for these scenarios, with the same greenhouse gas emissions scenario used for both

  • using a weather generator to synthesise 30 years of daily climate data for 2050 for key sites, based on the monthly changes in the parameters listed above

  • generating river flow scenarios, using the NIWA ‘Topnet’ model.

There were some important assumptions made with these scenarios:

  • The climate scenarios provided mean changes only for 2050, with no allowance for changes in the interannual variability of climate, eg, as a result of El Niño-Southern Oscillation (ENSO) events and IPO shifts.

  • There were two key assumptions with the use of weather generators, the first being the manner in which weather elements were simulated and the second being that monthly mean values only were changed, with no change in other properties (for example, no change in the typical variability and relative intensity of rainfall and temperature extremes).

  • There were two key assumptions with the river flow model: “no hydrologically significant changes in vegetation” and “no new diversion or abstraction of water, nor any new extraction of groundwater that sustains river flow”.

Land-use changes

Land-use changes were determined for each of the three catchments by calculating changes in mean monthly degree-days, and by consulting local experts.

In determining these changes, economics were held constant. That is, it was assumed that present economic trends for different crops and farming systems would hold for 2050. In general, the pattern presented was one of more intensive land use.

To complete this study, the scenarios of climate, river flow and land-use change were brought together to quantify possible changes in water demand and supply, using an irrigation scheme simulation model.

5.5 Example 3: Stormwater and wastewater effects in North Shore City

North Shore City Council commissioned a major study (as part of Project CARE) on its wastewater system (North Shore City Council et al 2003). As part of this study it was decided, for a relatively small incremental cost, to examine the possible effects of climate change on future wet weather overflows.

The approach taken by the consultants for North Shore City is summarised in Figure 5.3, and in the accompanying text. In brief, existing system performance was translated into expected future performance based on changing rainfall (extreme events) using a statistically established relationship between existing rainfall patterns and existing system performance.

A number of key points, relating principally to the steps taken in developing climate change scenarios, are presented below. These are taken the Executive summary of the North Shore City report:

  • The existing condition of the receiving environment was determined by hydrologic, hydraulic and water quality simulation of the system based on 17-year historical rainfall record.

  • A study was carried out to determine the rainfall changes due to global warming and phase changes in the Interdecadal Pacific Oscillation (IPO) for the planning scenarios. Global mean temperature was expected to increase by 1°C by year 2050 and this may be accompanied by more rainfall. The study suggested that there would be heavier, longer-duration extremes in IPO negative phase and heavier, shorter-duration extremes in IPO positive phase resulting in more rainfall.

  • A statistical approach was considered to evaluate the system performance due to IPO phase change and global warming, as the long-term hydrologic and hydraulic simulation using future rainfall time series is considered to be time-consuming, costly and may only provide a similar confidence level.

  • Expected future storm characteristics (year 2050) were estimated from the historical storm characteristics, and historical and predicted future rainfall IFD tables. Future storm characteristics were estimated separately for both positive and negative IPO phases, expected to be experienced by 2050.

  • Although climate change is well accepted by professionals worldwide, the analysis adopted in this study is based on a number of simplified assumptions with inherent uncertainties associated with modelling the effects of global warming. The results, therefore, should be used to assess trends more than provide absolute values, and their interpretation should be carried out by suitably qualified and experienced professionals.

Figure 5.3: Approach used to generate and apply future scenarios for the North Shore City Council study on future wet weather overflows.

 See figure at its full size (including text description).


76 Kenny et al 2000.