Ground Temperatures

Temperatures at the surface of the ground are not quite the same as the screen temperatures (at 1.5 m) which are normally recorded. The difference depends on the season, the bareness of the surface, the wind, the amount of cloud and the time of day. At night, the ground surface is colder than air at screen height, and conversely when the net radiation is positive (Figure 3.12). As regards season, the mean difference between ground and screen minimum temperatures at Alice Springs is 4.5 K in July and 5.1 K in January, the ground being colder. Likewise, measurements at Cowra (NSW) showed monthly mean differences of 2.1 K in winter but 5.4 K in March, i.e. there is again a greater difference in the warmer months. The surface of bare ground in Pretoria (South Africa) at dawn in summer can be 12 K less than the screen minimum.

The effect of wind is to equalise screen and surface temperatures by linking them convectively. Clouds also reduce the temperature difference, especially precipitating clouds, because net radiation is reduced both at night and during the day (Section 2.8). For instance, the ground minimum in Sydney is typically 6 K cooler on a dry clear calm winter's night, but 2.9 K when daily rainfall is 1-5 mm, and only 1.6 K on wetter days (when there is more cloud).

As to the daily maximum at ground level, it exceeds the screen maximum by several degrees if there is little wind or cloud. The surfaces of sand or asphalt can be dangerously hot in strong sunshine; temperatures near 80°C have been recorded in deserts, which is 22 K higher than the world's record screen temperature. Chapter 5 explains such large differences as due to (i) a strong net radiation flux to the surface (i.e. low albedo and no cloud), (ii) little conduction of heat downward from the surface (i.e. a poorly conducting ground material, such as dry sand), (iii) no evaporative cooling from a moist surface (Chapter 4) and (iv) only slight convection between ground and screen, due to an absence of wind. Of these four factors, only the first and last reflect the weather, whilst the second and third depend on the soil and vegetation of the site. To reduce the effects of the site itself at a weather station, we standardise by measuring ground temperatures on a short-clipped, well-watered lawn, unobstructed by surrounding buildings or trees. This is known as the standard ground temperature.

The ground maximum can be lowered by increasing its albedo, to reflect solar radiation away. For instance, soil covered with white polythene in Tanzania was observed to be 13 K cooler than soil covered by black polythene. A layer of white chalk on black soil at Pune in India lowered the temperature by 14 K one clear afternoon in summer, whereas black soil on white sand warmed the surface by 10 K. Maori farmers in New Zealand reduce the albedo with charcoal to warm the soil, in order that plant germination be accelerated. Another way of achieving high surface temperatures is to select slopes facing the Sun (Chapter 5).

Ground-surface temperatures, averaged globally and over a long time, tend to be around 2 K higher than screen temperatures over land surfaces. For example, a difference of 4 K has been observed at 2,819 m in Ecuador. This positive difference shows that the ground heats the air, on the whole. Nevertheless, there are exceptions. For instance, mean screen temperatures are above ground-surface averages in Norway, where the oncoming air has been warmed by the Atlantic Gulf Stream.

Subterranean Temperatures

The temperature of the soil near the surface is important in agriculture, e.g. in affecting the germination of tomatoes (Note 3.F). Similarly, maize seeds are planted only when the ground has reached about 10°C.

Regular measurements are usually made at depths of 0.3 m, 1.2 m and, occasionally, at 0.1 m, 0.5 m and 3 m. The results show that the daily pulses of heat from the surface are attenuated with depth, so that there are hardly any daily fluctuations of temperature beyond half a metre or so, depending on the nature of the ground (Note 3.K). Annual fluctuations reach almost 20 times further (Figure 3.16).

The decrease of daily temperature range with depth makes it possible to live comfortably underground all the year round, where surface extremes are harsh. People at Coober Pedy (South Australia) live in dugouts excavated by the opal miners, where the steady temperature of 20°C is much more agreeable than the January mean maximum of 37°C, or than the frosty mornings in winter.

Diurnal and annual cycles of temperature not only decrease in amplitude with increasing depth within the soil, but also lag behind that


i i i i




/ 50mm \

/ S \

' \

/ / 0.2m «L^



S 10

S 10



/""NO. 1m / 1m \

* 10m \ ^


winter solstice summer solstice winter solstice

Figure 316 Typical variations of ground temperature at various depths (a) during a day, and (b) during a year.

at the surface (Figure 3.16). For the annual cycle, the lag is about six months at 10m, which means that the ground is colder at 10 m depth in summer than in winter. Likewise, the inside of a solid wall about 50 cm thick is coolest at around midday and warmest at midnight, promoting a useful stabilisation of indoor temperatures.

On the whole, the annual average temperature increases with depth as the result of radioactive decay in the interior of the Earth. The geothermal gradient is higher in regions with volcanoes, such as New Zealand, and is largely unaffected by variations of temperature at the surface. Nevertheless, small irregularities in the subterranean temperatures can be attributed to surface-temperature anomalies a long time ago. For instance, a kink in the temperature profile below 1,000 metres underground in New South Wales is attributed to the last Ice Age 18,000 years ago, when glaciers flowed in the Snowy Mountains. Likewise, temperature irregularities in boreholes a few hundred metres deep in eastern Canada show relative warmth about 1,000 years ago (the Little Climatic Optimum), coldness around AD 1600 (the Little Ice Age) and a warming by up to 3 K during the last century. This is discussed further in Chapter 15.


Frost is defined in various ways. Ground frost is said to occur when the ground minimum is below 0°C. There is black frost (i.e. there is no deposit of ice) if the air is so dry that the frozen ground surface fails to reach the dewpoint (a measure of atmospheric humidity, see Chapter 6). A frosty night means that the screen temperature falls to +2°C or less, which usually implies a ground frost, defined above. A heavy frost occurs when the screen minimum is 0°C or below, and a killing frost (i.e. one that kills plants) may be defined in terms of a screen minimum below -2°C. But it is better to quote the actual temperature and the level at which it is measured than to use such labels.

Hoar frost is a silvery-white deposit of tiny crystals formed by the direct deposition of water vapour onto surfaces colder than 0°C. Glaze is an ice coating caused by the freezing of

'supercooled' droplets of rain (Chapter 9) when they impact onto surfaces colder than 0°C.


Occasions of frost are due either to the advection of polar air, or to intense radiation cooling at night, or, most commonly, to both together. An advection frost affects a wide area, such as the whole of southern Brazil, where cold air can drive far north in the lee of the Andes (Chapter 13). Such frosts may cause enormous damage to Brazil's coffee crop. On the other hand, radiation frosts are more local, more common and more intense. They result from reduced sky radiation on clear, calm nights in winter. They are less likely in the warm conditions of low latitudes or within a few kilometres of the sea. Further inland, the drier air allows greater nocturnal cooling (Section 2.7), so, for instance, there are about thirty-three frosty nights annually at Alice Springs, in the centre of Australia, despite a latitude of only 23°S.


There is no frost north of 23°S in Brazil. Also, it is hardly known north of 20°S near the east coast of Australia, or 28°S on the west; the asymmetry is due partly to a warm ocean current down the west coast (Chapter 11) and partly to cold southerly winds over eastern Australia and warmer northerly winds over the western area in winter (Chapter 13).

Frosts are less frequent at lower levels. Observations in the Craigieburn Range (in New Zealand at 43°S) show a reduction of the average frost-free period by about thirteen days for each 100 m extra elevation. The chance of frost is increased in some low-lying localities by cold air draining from higher (i.e. colder) land into frost hollows (Note 3.L). These occur in parts of the Snowy Mountains of south-east Australia, so that daily minimum temperatures in a valley bottom may be less than what is needed for tree growth (Note 3.F). As a result, there is a reversed tree-line, with trees growing only above a certain level.

There has been a gradual reduction in the frequency of severe frosts during the past forty years, in many countries. This is partly due to more urban heating (Section 3.7), but is also evident in rural areas such as outback Queensland. Frosts are fairly common there, on account of the reduced cloudiness when winters are unusually dry.


Farmers are especially concerned about frosts. A notable frost in Brazil in July of 1975 reduced the country's coffee harvest by 65 per cent, seriously upsetting export trade.

The agricultural growing season is often defined in cool areas as the period between the last frost of spring and the first of autumn; crops are suitable there only if they reach maturity within that time (Notes 3.F and 3.I). The date for the start of such a growing season at Walgett (NSW) is within twelve days of 8 September, in half the years of a long record, and the end within twelve of 22 May, so the average frostfree period between is 255 days. It is about five weeks less at Tamworth, which is 270 m higher. Naturally, the periods may be quite different in any particular year. Sometimes the effective growing season is referred to instead, the period between (i) the date of all but the last 10% of killing frosts in spring, and (ii) the date of all but the earliest 10 per cent in autumn, in a long series of years.

The best protection against frost is proper site selection (Note 3.L). The risk may also be reduced by either training plants above the ground, covering the soil with black plastic (to absorb more heat during the day), or sheltering the crop within a glasshouse. Frost damage may be reduced by sprinkling water (Chapter 4) or large fans (Chapter 7).

Was this article helpful?

0 0
Solar Power

Solar Power

Start Saving On Your Electricity Bills Using The Power of the Sun And Other Natural Resources!

Get My Free Ebook

Post a comment