Geoscience Reference
In-Depth Information
of the methods used to infer climatic change and introduces some of the key concepts
analysed in later chapters. Our geographical focus is primarily on the tropical and
temperate deserts and their margins, although, as we shall see, the cold desert of
Antarctica has played a major role in the long-term desiccation of Australia (and
possibly even central Asia). Antarctica has the distinction of being the coldest, driest
continent on earth, with Australia being the second driest.
1.2 What is a desert?
Before proceeding further, it is worth considering what we mean by the term 'desert',
starting with a very simple definition. For the purposes of this topic, we can define a
desert as a region where the precipitation is too little and too erratic and the evaporation
is too high to allow many plants and animals to survive, except in a few favoured
localities. Indeed, the Arabic word
sahra
denotes a flat wasteland devoid of water,
to be traversed as quickly as possible. There is also an economic definition of a
desert as a region where viable agriculture is not possible without irrigation - but this
depends entirely on the type of crop being grown and begs the question of what is
viable.
Amore quantitative definition of aridity may be achieved using the ratio of precipit-
ation (P) to evaporation. In practice, because long-term measurements of evaporation
are rare for most deserts, evaporation is usually expressed as potential evaporation
(E
pot
). Potential evaporation may be calculated using the Penman (
1948
)formula,
but here again there are too few reliable meteorological measurements to allow this
approach to be widely used. The Thornthwaite (
1948
) formula is simpler to use
and requires fewer climatic parameters. According to this formula, when P
=
E
pot
throughout the year, the index is 0. When P
=
0, the index is
−
100, and when P
greatly exceeds E
pot
, the index is
>
100. Climates with an index below
−
40 are arid,
−
20 are subhumid.
A somewhat arbitrary classification of aridity is that used by both the World Met-
eorological Organization and the United Nations Environment Programme, in which
drylands are defined as those regions where the ratio of mean annual precipitation,
P, to mean annual potential evaporation, E
pot
, was less than 0.65 for the 1951-1980
period (UNEP,
1992a
,
1992b
). A modified version of the 1948 Thornthwaite formula
was used to calculate P/E
pot
. Using this approach, drylands are classed into hyper-
arid, arid, semi-arid and dry subhumid, as shown in
Tabl e 1 . 1
. Here again, we need
to remember that mean precipitation is an almost meaningless concept in regions
where the rainfall is so variable from year to year. It is also worth stressing that low
precipitation is a
necessary
but not a
sufficient
cause of aridity. In certain cold areas
of the world, such as Patagonia and Greenland, the rates of evaporation may be low
enough to compensate for the low rates of precipitation, allowing a relatively dense
plant cover and even peat bogs to exist in spite of a very low annual precipitation.
20 to
−
40 are semi-arid, and 0 to
−