View all publications

2 The Impact of Climate Change on Flooding

2.1 Introduction

This chapter provides background information on the processes that lead to river flooding and how changes in climate might affect those processes. It provides a context for the technical methods in the following chapters. The chapter goes on to outline how predicted climate changes will affect flood flows, and concludes with a brief discussion of the implications of climate variability and change for some engineering design analyses.

2.2 Physical processes leading to flooding in New Zealand

The most common meteorological process leading to river flooding in New Zealand is heavy rainfall, which can greatly increase the water level in rivers and lakes and cause water to overflow into surrounding areas. When rainfall reaches the ground some of it runs into rivers by either surface or sub-surface pathways. The rest of the rainfall is stored in the soil and in surface depressions and lakes, or drains away to groundwater. Run-off from rainfall entering storage is also reduced by evaporation.

Flood run-off depends on many factors, including the amount, intensity and duration of rainfall, the topography, vegetation and soil characteristics of the catchment, the wetness of the catchment before the storm (referred to as the antecedent or initial conditions) and evaporation in the catchment. Rainfall-driven floods vary in both their duration and extent, and may result from:

  • brief localised events (eg, thunderstorms in urban areas or steep catchments)
  • storms lasting a day or two and causing flooding in limited areas
  • a repeated sequence of storms over a region, which saturate the soil and fill surface depressions and lakes, so that subsequent storms in the sequence produce more run-off because less water is able to be stored, which may lead to major widespread flooding.

Although the magnitude of a flood is often described using the peak water level or peak river flow rate during the flood, this is only a partial indicator of flood severity. The severity of inundation by floodwaters also depends on the volume of floodwater during the portion of the event when inundation takes place. For coastal river reaches, inundation is affected by sea level, including tides and storm surge.

In some parts of New Zealand, flooding can also be exacerbated by snow melt (eg, in Otago and Southland). Warm temperatures and rainfall on a deep snowpack can lead to rapid snow melt, which can occasionally be sufficient to cause a flood. Other types of flooding not considered in this manual include inundation by groundwater or high sea levels, or dam-break floods. Refer to the Coastal Hazards and Climate Change manual (Ministry for the Environment, 2008b) for details on coastal flood hazards and a discussion of sea-level rise.

The impact of floodwaters on communities also depends on non-meteorological factors, including how many people and what assets are at risk, and the effectiveness of flood protection and flood warning systems. The level of risk posed by the flood will help determine which of the methods described in later chapters should be used. For lower levels of risk, screening methods are more appropriate, while for higher levels of risk both screening and advanced methods should be used.

2.3 Decadal variability in rainfall and flooding

Flood hazards vary over time for reasons other than climate change. Natural shifts in climate on decadal time-scales will also impact on flooding. It is important to recognise these other sources of variability so that climate change impacts are kept in context. Flood hazards also change for reasons not directly associated with climate, such as land-use change (eg, urbanisation, deforestation). These impacts are outside the scope of this manual.

2.3.1 IPO influences on rainfall

Changes in New Zealand rainfall viewed over several decades show some association with changes in the phases of the Interdecadal Pacific Oscillation (IPO), a cycle of 15 to 30 years between warm and cool waters in the north and south Pacific. The IPO multi-decadal sea surface temperature pattern is similar to that of ENSO (the El Niño–Southern Oscillation), but with more variation in the extra-tropics, 4 especially in the north Pacific (Folland et al, 2002).

The IPO has been shown to be associated with long-term fluctuations in New Zealand’s climate (Salinger et al, 2001) and sea level (Goring and Bell, 1999). The increase in New Zealand temperatures around 1950 occurred shortly after the change from positive to negative phase IPO (see figure 1). The switch from negative to positive IPO in the late 1970s coincided with significant rainfall changes. Figure 2 maps annual rainfall changes between negative and positive IPO periods centred on 1978.

In the later (positive IPO) period, rainfall increased in the west and south of the South Island but decreased in the north and east of the North Island, relative to the earlier (negative IPO) period. This rainfall pattern is partly associated with the changing prevalence of El Niño versus La Niña events. In the positive IPO phase after 1977, more frequent El Niño events produced rainfall increases in the western South Island, whereas fewer La Niña events resulted in decreased rainfall in the Bay of Plenty and Northland. In other regions the observed effect of the IPO on rainfall is relatively unimportant (see green shading in figure 2). Decadal variability associated with the IPO can also have consequences for rainfall extremes (and therefore, in all likelihood, flooding).

Figure 1: Time-series of the Interdecadal Pacific Oscillation

Figure 1: Time-series of the Interdecadal Pacific Oscillation

Figure 1: Time-series of the Interdecadal Pacific Oscillation
This figure shows a time series of the Interdecadal Pacific Oscillation from the period from 1900 to 2006. There was a positive phase from 1925 to 1943 and again from 1978 to 1999, and a negative phase from 1944 to 1977.

Figure 2: Percentage change in average annual rainfall, 1978–1998 period compared to 1960–1977 period

Figure 2: Percentage change in average annual rainfall, 1978–1998 period compared to 1960–1977 period

Figure 2: Percentage change in average annual rainfall, 1978–1998 period compared to 1960–1977 period
This figure shows the percentage change in average annual rainfall for 1978 to 1998 compared to the period from 1960 to 1977 (as a percentage of the 1961-1990 average). The south and west of the South Island was more than eight percent wetter; the north of the South Island was up to eight percent wetter; the north and east of the North Island was up to eight percent drier; and in the remaining areas of New Zealand the differences were close to zero.

2.3.2 IPO influences on flooding

The decadal fluctuation in wind and rainfall over New Zealand also leads to variations in river flow and flooding. McKerchar and Henderson (2003) compared flood records for 1947–1977 (negative IPO) and 1978–1999 (positive IPO). They showed that a decrease in flood size occurred after 1978 in the Bay of Plenty region of the North Island. They also showed that increases in flood size and low-flow magnitude occurred in the South Island for most rivers with headwaters draining from the main divide of the Southern Alps and Southland. An example of this for the Rakaia River is shown in figure 3.

An important practical implication of these results is that in the north and east of the North Island, and in the south and west of the South Island, any analysis of extreme values relating to flood data should include an assessment of IPO influences. For these regions, short flood records lying mainly within the 1947–1977 period or mainly within the 1978–1999 period are likely to be biased and not representative of the long-term flood risk. Such flood records should be adjusted to compensate for the bias, using the observed decadal variability in flood magnitude at hydrologically similar sites with longer records.

Figure 3: IPO control of annual maximum floods for the Rakaia River: box plots comparing the distribution of annual maximum floods for the Rakaia River recorded at the Gorge/Fighting Hill sites for two periods, 1958–1977 and 1978–1999

Figure 3: IPO control of annual maximum floods for the Rakaia River: box plots comparing the distribution of annual maximum floods for the Rakaia River recorded at the Gorge/Fighting Hill sites for two periods, 1958–1977 and 1978–1999

Figure 3: IPO control of annual maximum floods for the Rakaia River: box plots comparing the distribution of annual maximum floods for the Rakaia River recorded at the Gorge/Fighting Hill sites for two periods, 1958–1977 and 1978–1999
Shows a box and whisker depiction of the annual maximum floods for the Rakaia river. The record has been split into two periods, the first for 1958-1977 - the negative phase of the IPO. The second period is for 1978-1999 - the positive phase of the IPO. The plots show the similar means for each of the two period (of around 2200 m3/s¬), but that the positive phase has a wider spread of flood sizes. The negative phase has 25 and 75 percentiles at 2000 and 2600 m3/s, where the positive phase has 1800 to 3000 m3/s. The extreme events for the negative phase lie around 1500 m3/s for the lowest, and 3000 m3/s for the highest. In contrast, the positive phase has extremes of 900 m3/s and 5800 m3/s.

Source: Adapted from McKerchar and Henderson, 2003.

Notes: The box plots compare the minima, the 25, 50 and 75 percentile values, and the maxima for the two periods. Twenty-five per cent of the floods from 1978 to 1999, during the positive phase of the IPO, were larger than anything observed from 1958 to 1977.

The periods 1947–1977 and 1978–1999 correspond to distinct phases of the IPO. The positive IPO phase that began in the late 1970s has ended, although there is as yet no indication of a return to strong negative values of the IPO index (figure 1). However, it is likely there will be more La Niña activity, and fewer El Niño events, over the next two decades compared to the 1978–1999 period. This would favour rainfall reductions in the southwest of New Zealand and increases in the northeast (ie, the reverse of the change shown in figure 2).

2.4 Climate change impacts on flooding

2.4.1 Climate change scenarios

Climate change is expected to affect flooding through a range of mechanisms, including rainfall, temperature, sea-level and river channel changes. Although it is impossible to know with certainty how climate change will affect flooding in New Zealand, we can have confidence in certain trends, such as the trend for increasing rainfall in the west. This confidence stems from a wide array of historical analyses combined with simulation modelling of the past and future. Climate change modelling studies mathematically depict the physics of the Earth’s climate system and are driven by emissions scenarios. Because the robustness of our climate change projections are rooted in the scenarios used to drive the climate change modelling analyses, it is useful to understand how these scenarios differ from one another.

There are many ways to develop climate change scenarios, depending on the data sets available and the requirements for impacts modelling. The climate change impacts shown in the Climate Change Effects manual are based on statistical downscaling of global climate model projections. These models are driven by a range of IPCC emissions scenarios (Nakicenovic and Swart, 2000). The IPCC has selected six of these emissions scenarios, which are known as illustrative marker scenarios and identified as B1, B2, A1T, A1B, A2 and A1FI, in order of increasing influence on global temperature increase over the 21st century (IPCC 2007).

The emissions scenarios span a reasonable range of plausible futures and depend on changes in population, economic growth, technology, energy availability and national and international policies. We cannot indicate whether any one emission scenario is more likely than another, so the guidance in this manual takes account of all six SRES illustrative marker scenarios (IPCC 2000) while focusing on a ‘middle-of-the-road’ scenario termed the A1B scenario. (See section 2.1 in the Climate Change Effects manual for more discussion on emissions scenarios.) The scenario projections in the Climate Change Effects manual were developed from a multi-model ensemble of 12 global climate models, using the A1B emissions scenario to represent the changes in greenhouse gases. The consequences of other emissions were accounted for by a simple rescaling of the A1B downscaled projections.

2.4.2 How changes in climate may affect rainfall

Extreme rainfall

Any consideration of the effect of climate change on flooding must start with the effects on rainfall. Projected changes in both mean rainfall and rainfall extremes are relevant. The intensity of extreme rainfalls is associated with temperature increases (Ministry for the Environment, 2008a), and so a consideration of future temperature change is also necessary. This is addressed in detail in chapter 3, but the expected impacts are summarised briefly here.

As a result of climate change, heavier and/or more frequent extreme rainfalls are expected over New Zealand, especially where the mean rainfall is predicted to increase. The percentage increase in extreme rainfall depths is expected to be approximately 8 per cent per degree Celsius of temperature increase.

Changes in seasonal rainfall

Climate change is expected to lead to increases in extreme rainfall, especially in places where mean rainfall is also expected to increase. Therefore, changes in seasonal and annual rainfall patterns, as well as changes in extreme rainfall, will be important factors for understanding future flooding.

Changes in annual rainfall are discussed in the Climate Change Effects manual, and the 100‑year trends are shown in the low part of figure 2.3 of that manual.

The 100-year trends in projected seasonal rainfall for a mid-range (A1B) scenario, averaged over the 12 climate models downscaled for New Zealand, are shown in figure 4. The four maps in figure 4, one for each season, show percentage changes in rainfall over 2080–2099 relative to the baseline (model) climatology of 1980–1999. This calculation is made for each model separately, and the results are then averaged. In the winter and spring seasons all models show similar trends.

Substantial increases in mean seasonal precipitation are projected for the west of the South Island, with decreases in mean seasonal precipitation for the east and north of the North Island. These precipitation changes are associated with an increased westerly wind flow across New Zealand during these seasons. In summer and autumn, trends vary across models, but, on average, reductions in seasonal precipitation are indicated in the west of the North Island particularly.

Figure 4 shows only the 12-model average, and there are large differences between models (see the range in tables 2.4 and 2.5 in the Climate Change Effects manual). Seasonal changes at selected grid points, co-located with cities and towns, are illustrated in figure 5. For many locations and seasons, either increases or decreases in precipitation are possible, sometimes in excess of ± 20 per cent. However, figure 5 also shows there is model agreement in some places. For example, in winter, models tend to agree on precipitation increases at sites exposed to the west (eg, Ruakura, Paraparaumu, Nelson and Queenstown) and decreases at sites exposed to the east (eg, Gisborne and Christchurch). The precipitation changes are very dependent on the atmospheric circulations developed within the models; the H symbol on various panels of figure 5 shows that where one particular model lies within the 12-model distribution can vary with season and location. This large uncertainty is something that has to be accepted in climate change projections at this time. Future work will attempt to quantify the uncertainty range (eg, as a frequency or probability distribution).

Figure 4: Projected percentage changes in seasonal mean precipitation for 2080–2099 relative to 1980–1999, averaged over 12 climate models for a mid-range (A1B) emissions scenario

Figure 4: Projected percentage changes in seasonal mean precipitation for 2080–2099 relative to 1980–1999, averaged over 12 climate models for a mid-range (A1B) emissions scenario

Figure 4: Projected percentage changes in seasonal mean precipitation for 2080–2099 relative to 1980–1999, averaged over 12 climate models for a midrange (A1B) emissions scenario
This figure shows four seasonal maps of projected changes in seasonal mean rainfall (in percentage) over New Zealand for 2090 relative to 1990 (twelve model average) for the A1B scenario.

For summer, the map shows increases in seasonal mean rainfall of up 10 percent in Hawkes Bay and parts of the East Cape. Smaller increases of up to 7.5 percent are projected in most parts of Marlborough and Canterbury, the Wairarapa and southern Bay of Plenty. Decreases of up to -5 percent are shown in the east of the North Island and Kapiti region, the northern West Coast and parts of Southland. Smaller decreases over Stewart Island, the Waikato Region and south Westland are projected.

For autumn, increases of up to 7.5 percent in seasonal mean rainfall are projected for Marlborough, Canterbury, Otago, Westland, Western Bay of Plenty, and from East Cape down through Hawkes Bay and the Wairarapa. Decreases of up to -5 percent are shown for Northland, Waikato, Wanganui, most of the central North Island and the region around the northern end of the Southern Alps.

For winter, marked increases in seasonal mean rainfall are projected, particularly in the west, with the West Coast of the South Island, Southland, parts of Taranaki, Manawatu and the Waikato increasing by over 10 percent. Decreases in the east of up to -7.5 percent in Northland and from East Cape down through Hawkes Bay and the Wairarapa are expected, as well as down the east coast of the South Island as far as Canterbury. Smaller decreases in the Bay of Plenty and parts of the Waikato are shown.

For spring, marked increases in seasonal mean rainfall of over 10 percent are projected for the West Coast of the South Island and Southland. Smaller increases of up to 5 percent are expected for Otago, the north-west of the South Island and most of Taranaki and the Manawatu. Marked decreases of more than -7.5 percent are shown in Northland, Auckland, Coromandel, East Cape, Hawkes Bay and parts of the Wairarapa. Smaller decreases in the central North Island and Marlborough/Canterbury regions are expected.

Figure 5: Projected percentage changes in seasonal mean rainfall at selected New Zealand locations, for a mid-range (A1B) emissions scenario

Projected Ruakura Precipitation: 2090

Projected Gisborne Precipitation: 2090

Projected Paraparaumu Precipitation: 2090

Projected Nelson Precipitation: 2090

Notes: The vertical coloured bars highlight the range over the 12 climate models considered, with stars indicating individual models. The symbol ‘H’ marks the location of one particular model (known as ukmo_hadcm3) within the distribution.

Figure 5: Projected percentage changes in seasonal mean rainfall at selected New Zealand locations, for a mid-range (A1B) emissions scenario
Shows four sets of plots, one for each of Ruakura, Gisborne, Paraparaumu and Nelson. The plots show the projected percentage change in rainfall, for each season, for the mid-range A1B scenario. The results for each of the twelve models investigated are plotted to give a representation of inter-model variability. The model results vary considerably with many of the plots showing both positive and negative changes in rainfall - depending on which model is chosen. The results from one model are highlighted, and show that this model does not always lie in the same place with respect to the other model results.

 

2.4.3 Storminess and climate change

Although there is a perception that ‘increased storminess’ is likely under climate change, the evidence of observed changes over New Zealand is far from conclusive. The concept of ‘storminess’ can refer to the number of storms or to the intensity, which in turn could be judged on the basis of either strong winds or heavy rainfall. Changes in storminess will also affect the level of the sea through changes in storm surge and waves. This may be important for rivers and stormwater drainage near the coast. Also, storms can approach New Zealand from the sub-tropics and from mid-latitudes (extra-tropics), and different trends are possible in the two regions.

Storms from the sub-tropics

Since the 2001 Third Assessment report of the IPCC, a number of articles have been published about observed increases in intense tropical cyclones. These results are still being reviewed, and the IPCC’s Fourth Assessment was cautious in its conclusions:

There is observational evidence for an increase in intense tropical cyclone activity in the North Atlantic since 1970 … There are also suggestions of increased intense tropical cyclone activity in some other regions where concerns over data quality are greater (Section 1.1, page 30, IPCC, 2007).

Tropical cyclones that develop in the south-west Pacific can affect New Zealand. From 1971 to 2004, tropical cyclones in this region averaged nine per year, with no observed trend in either frequency (Burgess, 2005) or intensity (Diamond, 2006). In any case, only about one cyclone per year moves south of 30 °S and comes close enough to New Zealand to have a direct impact, and no resulting change in New Zealand’s storminess from ex-tropical cyclones has yet been detected.

For the future, the IPCC (2007) states that “it is likely 5 that future tropical cyclones will become more intense” which implies larger peak wind speeds and increased heavy precipitation. Some model studies suggest a decrease in the total number of tropical cyclones, but the IPCC assigns little confidence to such a projection at this stage.

Such changes in tropical cyclones are potentially important for New Zealand because these weather systems can transform into intense sub-tropical lows as they move south and bring widespread heavy rainfall and damaging winds, waves and storm surge to New Zealand (eg, Cyclone Giselle, better known as the Wahine storm, in April 1968, and Cyclone Bola in March 1988).

Storms from mid-latitudes

Several recent studies have been made of trends in southern hemisphere extra-tropical cyclones. Over the period 1979 to 1999, there has been about a 50 per cent increase in the number of explosively deepening cyclones (the so-called ‘weather bombs’) per year (Lim and Simmonds, 2002). These rapidly deepening systems occur mainly to the south of 50°S but can form in the western Tasman Sea in the winter season. Although they form a small percentage (around 1 per cent or less, depending on location) of the total number of cyclones, they can be important for New Zealand. A recent example of an explosive deepening cyclone in the Tasman was the storm that affected Northland on 26/27 July 2008: this storm registered the lowest pressure on record (962 hectopascals) of any storm approaching New Zealand, but did not cause large-scale flooding because of its rapid movement.

Changes in the number of southern hemisphere cyclones have also been documented (Simmonds and Keay, 2000). Over the 40-year period, 1958–1997, there has been a general reduction in the mean cyclone density over most regions south of 40°S, with the greatest reductions near 60°S, but little change in the Tasman Sea. At the same time, systems have become more intense on average in the Australian Bight and the Tasman Sea, and weaker over the eastern Pacific. Just why the reduction in overall numbers should be occurring is not well understood, although one modelling study (Zhang and Wang, 1997) has suggested that under moister conditions (as would occur in a warmer atmosphere) cyclonic eddies transfer energy poleward more efficiently, and thus fewer cyclones would be ‘required’ to effect the same energy transport.

For the future, the IPCC (2007) states that “extra-tropical storm tracks are projected to move poleward, with consequent changes in wind, precipitation and temperature patterns”. There is no comment in the IPCC report on what this could mean for New Zealand.

2.4.4 How changes in climate may affect flooding

Extreme rainfall

The major impact of climate change on New Zealand river floods is expected to come through increases in extreme rainfall (in coastal river reaches sea-level rise will also affect inundation). As noted above, the relationship between rainfall intensity and flood magnitude depends on several factors and is not linear; so, for example, an 8 per cent increase in rainfall intensity does not necessarily lead to an 8 per cent increase in flood peak discharge, which does not necessarily lead to an 8 per cent increase in flood inundation. In many cases, the increases in flow and inundation will need to be estimated using our understanding of how the rainfall–run-off inundation processes are related to an increase in rainfall. This could be based on computer modelling of flow and inundation.

In the Westport flood study (see chapter 6), rainfall increases of 3 per cent, 5 per cent and 33 per cent for different temperature scenarios caused modelled peak river flow to increase by 4 per cent, 10 per cent and 37 per cent, respectively. As a consequence, flood inundation was estimated to increase from 4 per cent of Westport township being inundated under the current climate, to 13 per cent, 30 per cent and 80 per cent, respectively, for each of the three temperature scenarios used in that study: (i) mid-low for 2030, (ii) mid-high for 2030 (also used as mid-low for 2080), and (iii) mid-high for 2080.

Those results are specific to the Westport situation and depend on the assumptions made in that study. To work through the impacts of a change in extreme rainfall in your situation, methods such as those outlined in chapters 4 and 5 need to be applied to the specific flood risk.

Initial conditions

The changes in seasonal rainfall outlined above imply wetter initial conditions (eg, wetter soils, higher lake levels) in places and during seasons where seasonal rainfall is increasing faster than seasonal evaporation. This would be expected to increase flood magnitude. Conversely, if seasonal rainfall is projected to decrease, then initial conditions would be expected to be drier and floods smaller. Increases in temperature and wind are also likely to increase evapotranspiration. 6 This may be important for estimating flows for water resource considerations, but is less likely to be important for extreme flooding events. These changes can be taken into account either by adjusting the initial conditions of event-based simulation models (eg, via the curve number in TP108, or the run-off coefficient in the Rational Method), or by using continuous simulation models (eg, TopNet), which keep track of the initial conditions internally.

Snow versus rain

For locations that currently receive significant snowfall, the projected increases in temperature suggest a shift towards increasing rainfall instead of snowfall. This means that for rivers where the winter precipitation currently falls mainly as snow and is stored until the snow-melt season there is the possibility of larger winter floods. It is also likely the spring melting will occur earlier and faster. These impacts have not yet been quantified, but are in addition to the temperature-driven increases in extreme rainfall that result from a warmer atmosphere. They can be accounted for by simulation models such as a regional climate model linked to a catchment model. An example of this is given in figure 6.

The two maps in figure 6 show the change in winter rainfall (left) and snowfall (right) between 1980 and 2080, as simulated by NIWA’s regional climate model (RCM) under the medium-high A2 emissions scenario. They suggest that changes in total winter precipitation in the Southern Alps may be dominated by a significant shift from snowfall to rainfall. They also suggest that spill-over onto the eastern side of the main divide may be affected by this change, since rain is less easily transported by the dominant westerly winds. The net change in total precipitation (see figure 4) is consistent with an increase in precipitation on the west coast of the South Island and a decrease in the North Island due to changes in large-scale atmospheric circulation. The snow simulation in the RCM is not yet validated, and so the values shown are currently considered to be illustrative only.

Figure 6: Projected changes in winter rain (left) versus winter snow (right), simulated for a medium-high emissions scenario (A2) using NIWA’s regional climate model

Figure 6: Projected changes in winter rain (left) versus winter snow (right), simulated for a medium-high emissions scenario (A2) using NIWA’s regional climate model

Note: The snow simulation in the regional climate model is not yet validated, and as such the values shown are currently considered to be illustrative only.

Figure 6: Projected changes in winter rain (left) versus winter snow (right), simulated for a medium-high emissions scenario (A2) using NIWA’s regional climate model
The figure shows the projected changes in winter rain (left) versus winter snow (right), simulated for a medium-high emissions scenario (A2) using NIWA’s regional climate model. In the winter rainfall figure, the plot depicts a decrease in rainfall of around 1-2 mm/day north of Taupo and significant increase of over 4 mm./day for the Southern Alps. The snowfall plot shows a decrease in snowfall over the Southern Alps of over 4 mm/day.

Erosion and sediment transport

Changes in climate can also affect flood magnitude indirectly. For example, increases in rainfall intensity can lead to increases in erosion, and when the eroded material is delivered to the river system it may lead to changes in river channel shape and position, and consequent changes in the likelihood of inundation. For example, extra sediment may be deposited in the bed of a river, raising the level of the bed (known as aggradation), and thus reducing the flood-carrying capacity of the channel. As a result, for a given river flow rate less water can be conveyed by the channel and more water will overflow and cause inundation. The opposite situation may also occur, where an increase in floodwaters in a channel results in greater water velocities and hence increases the forces required to transport sediment. This can lead to increased erosion and down-cutting (degradation) in the channel. Within a river system patterns of aggradation and degradation vary in both time and space, making changes in channel position and level – and thus changes to flood-plain inundation – very hard to predict. Numerical morphodynamic modelling is likely to be required to quantify changes to sediment transport and channel morphology.

Sea-level rise

Another important impact of climate change on flood severity is through changes in sea level. Projected sea-level rise as a result of climate change is discussed thoroughly in the Coastal Hazards and Climate Change manual (Ministry for the Environment, 2008b), which suggests using a risk-based approach to assess sensitivity to different amounts of future sea-level rise. As part of the risk assessment process, the manual recommends considering the potential consequences of higher sea levels. The Ministry recommends planning for the following projection of future sea-level rise:

  • for planning and decision timeframes out to 2090–2099, a base value sea-level rise of 0.5 metres relative to the 1980–1999 average be used along with an assessment of potential consequences from a range of possible higher sea-level rise values. At the very least, all assessments should consider the consequences of a mean sea-level rise of at least 0.8 metres relative to the 1980–1999 average
  • for planning and decision timeframes beyond the end of this century, an additional allowance of 10 millimetres per year be used.

Refer to tables 2.2 and 2.3 on page 21 of the Coastal Hazards and Climate Change manual for more detail on this issue.

These potential sea-level changes mean the base water level in coastal river reaches may be significantly higher than the current level. Calculations of flood inundation should take this higher base level into account in coastal river reaches. (This matter is discussed further in chapter 5.)

2.5 Climate change or IPO: which is more important?

Climate variability due to the IPO (see section 2.3) will continue to be an important factor in determining flood risk. The impacts of climate change over the next 40 to 50 years will be similar in magnitude to the impacts of the IPO, and so both need to be considered when assessing changes in flood magnitude. Climate change trends will become more important, and will change the background state as time goes on, so the mean climatic conditions around which the variation is happening will continue to move further away from the historical mean conditions that have effectively been the basis for flood designs in the past.

2.6 Summary

  • The most common cause of river flooding is extreme rainfall.
  • Climate change is likely to lead to increases in extreme rainfall, especially in places where mean rainfall is expected to increase.
  • Changes in storminess as a result of climate change are harder to predict, but it is likely that tropical cyclones will be more intense, and such weather systems can transform into intense sub-tropical lows that bring heavy rainfall to New Zealand.
  • As the snowline rises, places that currently receive snow are likely to see a shift towards increasing rainfall instead of snowfall.
  • Sea-level rise will increase base levels for coastal river reaches.
  • The climate is naturally variable. On a decadal timescale New Zealand rainfall and flooding are affected by the Interdecadal Pacific Oscillation (or IPO) in some places. IPO effects may need to be considered when calculating flood risk.

4 The mid-latitudes (approximately 30°S–50°S).

5 In IPCC terminology, “likely” means a 66 per cent chance or greater of occurring.

6 The combined process of evaporation from the Earth’s surface and transpiration from vegetation.