BoM - Heatwaves

Heatwaves are probably the most under-rated weather hazard in Australia, essentially because they are viewed as a 'passive' hazard in contrast to the more widely studied catastrophic hazards such as tropical cyclones and earthquakes. According to Coates (1996), heatwaves kill more people than any other natural hazard experienced in Australia. In a study on the consequences of heatwaves, Andrews (1994) reported that in the period between 1803 and 1992, at least 4287 people died as a direct result of heatwaves. This was almost twice the number of fatalities attributed to either tropical cyclones or floods over much the same time frame. In the United States, heatwaves are the second greatest cause of human mortality resulting from a natural hazard, killing more people than hurricanes, tornadoes, lightning and floods combined. Only the low winter temperatures have killed more people. Recent examples of heatwaves which have caused a large number of deaths include, Brisbane in January 2000 when 22 people died, and New York City in 1980, when 1600 deaths were attributed to a heatwave and economic losses totalled 15 billion dollars.

In Western Australia a total of 392 deaths have been caused by excessive heat between 1807 and 1994, compared to 1250 in New South Wales. This equates to a death rate per 100,000 of 0.68 in Western Australia, ranking second to South Australia (1.14) and above New South Wales (0.41). However there has been a decrease in the number of heat-related fatalities in Australia. From 1973 to 1992 the heat-related death rate per 100,000 has declined to just 0.06 in Western Australia.

In addition to the official statistics, there are many more heatwave-associated deaths caused mainly by heart disease and strokes, particularly in the elderly. On the other hand, not all heat-related deaths occur on days of extremely high temperatures suggesting that other factors, in addition to maximum temperature are significant in influencing mortality.

The impact of heatwaves extends further than just mortality rates. High temperatures are linked to:
· increased hospital admissions relating to heat stress, dehydration, or as a result of heat exacerbating existing conditions;
· increased rates of certain crimes particularly those related to aggressive behaviour such as homicide;
· increased number of work-related accidents and reduced work productivity; and
· decreased sports performance.

An extract from The West Australian (21 February, 1995), for example, illustrates the affect of a heat wave in Perth:
Perth's heatwave caused record power consumption and overstretched the ambulance service yesterday ... A spokesman said the demand was caused by elderly people collapsing and having heart problems in the stifling heat. There were also many more assaults and brawls than normal in the early hours of Sunday when Perth sweltered through the hottest February night for ten years. Ambulances were called out 70 times, compared with an average of 30 ... Royal Perth and Sir Charles Gairdner hospitals reported treating people for dehydration, severe sunburn and heat exhaustion. Many people who came in with other illnesses also complained that the heat was making their problems worse.

High temperatures can also cause significant economic losses through livestock/crop losses and damage to roads, railways, bridges, power reticulation infrastructure and electrical equipment (EMA, 1998). Heatwave conditions also lead directly to significant increases in demand for electricity to power domestic air conditioners, water consumption and retail sales of cold drinks. During a heatwave in February 2004, the power supply was unable to cope with demand resulting in power outages across Perth.

In assessing heatwaves as a natural hazard, consideration should also be given to the interaction of high temperatures and population vulnerability, which is a result of both the social and physiological systems. The distribution of heat-deaths is complex, the most vulnerable being the elderly, the sick and infants living in low socio-economic urban areas during early summer heatwaves. Physiological adjustments primarily include acclimatisation to high temperatures, which is an issue for the increasing number of tourists arriving from the northern hemisphere winter.

Historically there have been significant improvements to decrease the vulnerability of the population through the use of air-conditioners, better housing design, better clothing, a trend towards more people working indoors, education and temperature forecasts to seven days. However, the percentage of elderly people has increased, with those aged 65 and over comprising 11.2% of the population in Western Australia in 2001.

Defining Heatwaves

There is no universal definition of a heatwave although in a general sense it can be defined as a prolonged period of excessive heat. The difficulty in defining a heat wave in Australia has been in establishing an appropriate heat index with an acceptable event threshold and duration, and relating it to the climatology of the area under investigation. Various heat or thermal comfort indices have been developed to evaluate heat-related stress combining air temperature and humidity, and in some cases, wind and direct sunlight. Two of the most widely used indices are the apparent temperature work of Steadman (1984) and the Relative Strain Index, RSI, derived by Belding and Hatch (1955) and discussed in the Goldfields-Eucla climatic survey (Bureau of Meteorology, 2000).

High temperatures in the Perth area typically correspond to low humidity values because the prevailing east to northeasterly winds originate from the dry inland parts of the state. A study of temperatures at Perth Airport at 3 pm indicates that for temperatures exceeding 35ºC, the relative humidity exceeds thirty per cent on about ten per cent of occasions. Also, for the same air temperatures, the apparent temperature exceeds the air temperature on only 12% of occasions. While more humid conditions can exist when the air temperature is closer to 30ºC and can provide some degree of discomfort, such days are not generally associated with heatwave conditions. As a result, for Perth the air temperature alone can provide a reasonable measure of heat stress.

Heatwave Risks

Heat stress makes us feel uncomfortable not so much because we feel hot, but rather because we sense how difficult it has become to lose body heat at the rate necessary to keep our inner body temperature close to 37ºC. The body responds to this stress progressively through three stages:

Heatwaves in the Perth area

Perth's summer patterns often follow a typical sequence. A ridge of high pressure south of the state combines with a deepening trough off the west coast to direct east to northeasterly winds over the Perth region. This pattern causes rising temperatures over successive days. The trough then moves inland allowing early seabreezes along the coast resulting in a cool change. A new ridge then develops to the south producing southeasterly then easterly winds and the sequence begins again. Prolonged spells of hot days occur when this pattern is slow moving, the high being maintained south of the state and the west coast trough remaining off the coast. On such occasions, the east to northeasterly winds prevent the early arrival of the seabreeze and cause temperatures well above the average. Figure 1 shows a typical heatwave weather chart from March 2003. Tropical cyclones off the west or northwest coast can also help to maintain the trough offshore resulting in high temperatures in southern Western Australia. These days are usually associated with hazardous fire weather conditions (see Bushfires).

Figure 1. Mean Sea Level Pressure analysis of a typical heatwave pattern in Perth. A large and slow-moving high south of Australia combining with a tropical cyclone off the Pilbara coast maintains hot northeast winds in Perth. Click on image to enlarge.

The occurrence of sequences of maximum temperatures of 35ºC or greater and 40ºC or greater at Perth Airport over fifty-nine summers (1944-45 to 2003-04) is shown in Table 1. There have been 185 occasions where maximum temperatures of 35ºC or greater were recorded on at least three successive days, about three times per summer on average; and 15 occasions when at least three successive days have occurred on which maxima reached 40ºC or more, about once every four years on average.

Table 1. The occurrence of sequences of heatwaves at Perth Airport over 59 summers (1944-45 to 2003-04)

Duration of Sequence (days)	Maximum Temperature greater than 35ºC	Maximum Temperature greater than 40ºC
3	91	11
4	39	2
5	21	1
6	18	1
7	4
8	3
9	2
10	3
11	1
12	0
13	1
Total	185	15

Table 2 is a listing of notable heatwaves showing data from Perth and Perth Airport. One of the greatest heatwaves occurred from 20 January to 6 February 1956, when for seventeen consecutive days the maximum temperature at Perth Airport remained above 32ºC and averaged 38.8ºC. For five consecutive days maxima remained above 40ºC. The hottest summer on record was in 1978 when the temperature averaged 34ºC over sixty-four days from January to March.

Table 2. Notable heatwaves in Perth.

Date	No. of days	*Perth Average Temperature (ºC)**	*Perth Highest Temperature (ºC)**	Perth Airport Averarage Temperature (ºC)	Perth Airport Highest Temperature (ºC)	Comments
16-27 January 1920	12	36.6	41.6
3-11 February 1933	9	40.2	44.6
20 January - 5 February 1956	17	37.1	43.7	38.8	43.7	Eight days over 40ºC at Airport
29 December 1961 - 13 January 1962	16	36.0	40.5	37.1	41.6
29 December 1964 - 5 January 1965	8	39.1	41.2	39.9	42.1	Six consecutive days over 40ºC at Airport
29 January - 5 February 1975	8	36.9	40.4	38.5	42.4
11 - 23 February 1975	13	37.1	40.9	36.4	40.9
3 January - 7 March 1978	64	33.2	44.7	34.0	44.2	Perth's hottest summer ever
5-17 February 1985	23	36.5	43.7	37.5	43.4
6-22 February 1988	17	35.8	39.0	36.6	40.8
27-31 January 1991	5	39.1	45.8	40.5	46.0
19-23 February 1991	5	39.1	46.2	40.5	46.7	Perth's highest temperature on record
1-16 February 1996	16	36.5	42.4	37.5	43.2
24 December 1999- 7 January 2000	15	35.6	39.3	36.2	40.4
21 February - 6 March 2001	14	34.7	39.7	35.3	40.1

* The station of Perth has undergone many changes throughout the observational record. The Perth station was moved from West Perth to East Perth in 1967, and then to Mt Lawley in 1993.

Sequences of three successive days with maxima of 35ºC or greater have commenced in all months between November and March, inclusive. Longer sequences of days with maxima of 35ºC or greater have been largely confined to the period from late December to mid-March.

The primary environmental influences on day-time temperature variations across the greater Perth area are: the ocean, the Darling Scarp, the Swan River estuary, and the nature and extent of urban development. The greatest single factor is proximity to the coast, because of the impact of the seabreeze on temperatures. Perth's reliable afternoon seabreeze, commonly referred to as the 'Fremantle Doctor', provides cooling relief from the east to northeast winds on hot summer days. The breeze initially arrives on the coast then progressively extends inland often decreasing the temperature on hot days below 30ºC within a few hours from onset. The degree of cooling varies according to the initial temperature, the time of onset and the strength of the seabreeze. Seabreezes typically take several hours to extend from the coast to the foothills, although there is significant variability in its movement inland. During very hot periods the seabreeze is delayed until at least mid-afternoon, and may not reach the coast at all on some days. On those days coastal suburbs may experience higher temperatures than elsewhere.

Figure 2 shows temperatures from a number of Perth sites during the heatwave from 1-16 February 1996. In general the temperatures follow the same behaviour day to day, however on any one day temperatures vary between sites. The highest temperatures tend to occur at Pearce and Gosnells. On most days, the coastal site of Swanbourne has the lowest temperature reflecting the influence of the earlier seabreeze. The difference in temperature between Pearce and Swanbourne can be up to nine degrees on some days. On other days (e.g. 2nd, 7-9th, 11th and 13th February), when the seabreeze is very late or does not reach the coast at all, Bickley, located in the hills, has the lowest recorded temperature reflecting the influence of altitude on temperature.

Figure 2. Maximum temperatures at Perth sites during the February 1996 heatwave. Click on image to enlarge.

Trends and future projections

In contrast to the decline in the number of heat-related fatalities, the temperature record suggests an increase in the number of hot days in the Perth area. Summer maximum temperatures are estimated to have increased at the rate of over one degree per 100 years using data between 1950-98. Collins et. al. (2000) determined a slight increase in the frequency of heatwave events (at least three successive days of temperatures 35ºC or greater) at Perth Airport of 0.4 events per decade from 1957 to 1996.

Concern has been expressed at the possibility of temperature rises as a result of human-induced changes to the climate system. Computer modelling by the CSIRO of the effect of an enhanced greenhouse effect on temperatures in the summer months suggests an increase in the number of days of 35ºC or greater at Perth and it is likely that the number of successive days of high temperatures would also increase.

For more details see Appendix B: City summaries on the Climate Change in Australia website at:

Heatwaves

The heatwave threat

Defining Heatwaves

Heatwave Risks

Heatwaves in the Perth area

Trends and future projections