NZSM Online

Get TurboNote+ desktop sticky notes

Interclue makes your browsing smarter, faster, more informative

SciTech Daily Review

Webcentre Ltd: Web solutions, Smart software, Quality graphics

Feature

Climate Extremes

Have recent shifts in regional weather patterns altered weather extremes like droughts, heat waves, and floods?

Georgina Daw, Dr Jim Salinger

The knowledge that our climate is changing, through global warming and natural climate shifts, presents us with a series of interesting questions.

For example, how will warmer temperatures affect evaporation, clouds and rain? Will wind patterns be affected? Which parts of the country will get drier or wetter, stormier or calmer? And will the extremes get better or worse?

To answer those questions, climate scientists first need to understand what has happened in the past. Has global warming already affected extreme temperatures and rainfall? Have recent shifts in regional weather patterns altered weather extremes like droughts, heat waves, and floods?

These extreme events can cause severe disruption to communities -- affecting life and limb, livelihood, and the environment. For example, the cost of the heavy rainfall in July 1998, which caused flooding in the Bay of Plenty and Waikato River basin, was estimated at $17.5 million.

Changes in extreme rainfall or extreme temperature can alter the risk of some hazards -- flooding, landslip, drought, fire -- and the viability of a human activity may be altered so that it is not profitable or sustainable any longer.

Change and Effects

Relatively few studies, either internationally or in New Zealand, have investigated trends or shifts in extreme climate. Yet given the disruption and economic losses which they can cause, it becomes very important, for future planning and mitigation of climate impacts, to establish whether New Zealand's climate is becoming more or less extreme. Recent research by scientists at NIWA (the National Institute of Water and Atmospheric Research) has provided answers to some of these questions.

The research's main intent is to find out if the changes in climate now being observed are having any effect on extreme weather events -- those disasters which hit human communities and economies so hard from time to time. We want to know whether changes in the mean climate are paralleled by changes in climate extremes.

Between 1951 and 1998, mean, maximum and minimum surface temperatures have warmed in the New Zealand region by 0.2oC, 0.1oC and 0.4oC, respectively.

The climate patterns in the Pacific region have shifted around 1950, and again in the mid-1970s, with resulting changes in average wind patterns, mean temperature and mean rainfall across New Zealand. It also appears possible that there was another abrupt change in climate pattern during 1998, but it will take several more years to determine whether this change is just a short-term aberration, or part of a longer-term shift.

From 1951-1975, the prevailing west or south-west winds weakened, with more winds from the east and north-east. Mean rainfall increased in the north of the North Island, and decreased in the south-east of the South Island. Average temperatures warmed in all regions of New Zealand.

In the period 1976-1998, a strengthened west to south-west circulation was dominant, with more settled climate over northern New Zealand. This climate pattern brought more rain and flooding to the west of the country in general, and in particular, to the West Coast of the South Island.

By contrast, the east of the country became significantly drier, with more droughts, as the winds from the west dropped their rain before they reached the east coast. However, in spite of more south-westerly flow (sourced from over the cold southern oceans), regional temperatures continued to increase.

Since 1998 there have been fewer westerly winds, and more easterly winds than normal. The warmest year on record globally was 1998, while both 1998 and 1999 were the warmest years ever observed for New Zealand.

Extreme Analysis

Recently we analysed extreme indices for statistically significant trends, over the period 1951-1998 (temperature) or 1951-1996 (rainfall). The extreme indices are based on percentiles, as opposed to a threshold amount (eg. days with rainfall more than 100 mm) because an arbitrary threshold would not be appropriate for all of the stations used, which come from climatically diverse regions around New Zealand.

Daily maximum and daily minimum temperature data from 20 stations around New Zealand were analysed. We calculated four annual indices of extreme temperature:

  • hot days (the maximum temperature 95th percentile i.e. the hottest 5% of days in a year)
  • cold nights (the minimum temperature 5th percentile i.e. the coldest 5% of nights in a year)
  • number of hot days (the annual number of days with maximum temperature above the long-term average 95th percentile value)
  • number of cold nights (the annual number of days with minimum temperature below the long term average 5th percentile value)

The hot days index reflects extremely high summer temperatures, and the cold night index reflects extremely low winter temperatures.

We also calculated three annual rainfall indices based on daily rainfall data from 22 New Zealand stations:

  • extreme rainfall intensity (daily rainfall 95th percentile i.e. the wettest 5% of daily rainfalls in a year)
  • number of extreme rainfalls (The annual number of daily rainfall exceeding the long-term average 95th percentile value)
  • dry spell length (maximum number of consecutive dry days in a year)

Climate Extremes Figure B (19KB)
Trends in New Zealand temperature 1950-1998
Note: Simplified illustration only, based on analysis of individual climate station records.

Hot Days and Cold Nights

Hot days show no statistically significant trends over the 1951-1998 period, except at Gisborne, where the hottest days have become hotter by about 1oC, and at Taupo, where the hottest days have become cooler by about 1oC.

However, there are smaller trends which, while not passing the statistically significant test, still show a useful pattern. In the north and north-east of the North Island, the hottest days have become even hotter, by between 0.5oC and 1oC. Elsewhere over New Zealand, however, the hottest days generally are cooler than before, by between 0.3oC and 0.6oC. It is notable that an abrupt change in the hot days index occurred early in the 1950s, with a marked increase in the hot day temperature in the north and west of the North Island, and also the West Coast of the South Island, at that time.

A similar geographical pattern is observed over New Zealand for trends in the number of hot days. Gisborne recorded a significant increase of around 10 hot days per year. Taupo, on the other hand, recorded a significant decrease of similar size. In the north and north-east of the North Island there have been smaller increases in the number of hot days, while there have been smaller decreases in most other parts of New Zealand.

It is notable that while over the past 50 years the national maximum temperature rose by 0.1oC, at most of the stations where records were studied, the hottest days were not as hot as before, and there were not as many hot days. Only in the north and north-east of the North Island are the increases in number and temperature of hot days slightly comparable to the change in maximum temperature recorded nationally.

This result reinforces the idea that extreme climate may respond differently to change than mean climate does. It also illustrates that trends in temperature extremes appear to be regionally diverse.

The cold nights index shows a significant increasing trend across most stations, with a warming of the coldest days of between 0.6oC and 1.8oC over the 1951-1998 period. The number of cold nights also decreased significantly over the second half of the century, with a decline in the number of cold nights of between 10 and 20 days per year.

These trends are consistent with those observed globally towards fewer extremes of low temperature. However, the New Zealand trends are larger than we might expect from the observed New Zealand minimum temperature increase (0.4oC). Around 1970, a shift occurred in northern and central areas of the North Island, with the coldest nights warming significantly after this time.

Wet and Dry

Extreme rainfall intensity shows very few statistically significant trends over the 1951-1996 period. Significant decreases are observed for Auckland, Taihape and Lincoln (Canterbury) of between 15% and 25%. Smaller decreases are recorded in Wellington and the Wairarapa, and also near Alexandra. Smaller increases in rainfall intensity are observed at Milford Sound, Queenstown and New Plymouth.

Since 1951, the number of extreme rainfalls significantly decreased over the east coast of both islands, and in Auckland, with a decline of up to 4 days per year. Elsewhere, the trend is for a small increase in the number of extreme rainfalls, apart from Queenstown which shows a significant increase of about 2.5 days per year. A change in climate pattern occurred around 1977 in the south-west of the South Island, with a marked increase in the frequency of extreme rainfall in this region after that time.

There appears to be a geographical "pattern" evident, in both the intensity and frequency of extreme rainfall. Reductions in intensity and frequency generally occur in areas sheltered from the west. Increases occur in those areas that are exposed to west or south-westerly winds. This type of "east/west" pattern in the extreme rainfalls reflects the effects the mountain ranges have on the prevailing west or south-westerly winds.

In general, the dry spell length has decreased across New Zealand over the period studied. At Hokitika, Milford Sound, Queenstown, Rotorua and Ruakura, the dry spell duration has shortened by between 2.5 and 4.5 days per year.

However, there has been a small trend towards longer dry spells in Gisborne, Wellington, Blenheim, Timaru and Dunedin.

Climate Extremes Figure A (12KB)
Trends in New Zealand rainfall 1950-1996
Note: Simplified illustration only, based on analysis of individual climate station records.

Trends

Very few significant trends over the period 1951-1998 are observed in the hot days value or frequency. In contrast, extreme cold temperatures have significantly warmed over most of the country, and also become less frequent. These cold night trends are consistent with those observed globally towards fewer extremes of low temperature. However, the New Zealand trends are larger than might be expected from the observed New Zealand minimum temperature increase. Both the hot day and cold night responses in tandem show that New Zealand temperatures over the last 50 years have generally become less extreme.

High rainfall and dry spells have generally become less extreme, also. Extreme rainfall displayed no broad trend towards an increase in intensity or frequency, as generally foreshadowed by models of climate warming.

Rather the opposite occurred, with significant decreases in the number of extreme rainfall events, and at a few locations, decreases in extreme intensities. However, the mean warming over the considered period (1951-1996) has been slight (0.2oC). A shortening of the dry spell duration was observed over much of New Zealand, but most strongly in western areas. Smaller increases in the dry spell length occurred in some eastern stations.

Of interest is the regional nature of the trends -- the implication that extremes respond to changes in mean climate depending on geographical region within New Zealand. The extreme temperature trends appear to be regionally distinct for the north and east of the North Island. The rainfall and dry spell extremes show a "east/west" pattern, with western areas generally showing an opposite response to eastern districts.

Both of these geographical characteristics indicate that extremes respond to climate patterns and winds around New Zealand. Abrupt changes in some of the extreme indices, around the time of major climate shifts in the Pacific region, agree with this.

Temperatures, rainfall and dry spells have generally become less extreme in the second half of the century. However, given the record warmth and rainfall volatility of the last two years, and the possibility that another climate shift occurred in 1998, it will be necessary to examine the latest data, to see whether the trends in climate extremes have indeed changed.

Georgina Daw and Dr Jim Salinger have been assisted by Isobelle Leleu of Meteo-France in this work.

Georgina Daw is based at NIWA Auckland.