Dryland farming is a type of farming practiced in arid areas without irrigation by planting drought-resistant crops and maintaining a fine surface tilth or mulch that protects the natural moisture of the soil from evaporation. At a time of increased water stress in semi-arid regions, dryland farming will play a pivotal role for food security in the coming decades (Chatteron 1996; Squires and Tow 1991; Malcolm, Sale & Egan 1996; Drylandfarming.org 2013).
What is dryland farming?
Farming systems need to be compatible with both climate and environment if they are to be sustainable. They also need to be profitable for farmers.
London, Tunis and Adelaide all have about the same annual rainfall of 500 mm. Tunis and Adelaide are in the centre of dryland farming zones while London is typical of a humid temperate climate. The growing season in the temperate regions is limited by low temperatures and short days in winter. The summer moisture deficit in London rarely reaches 50 mm. In dryland farming zones such as Tunis and Adelaide the winters are mild and humid but the summers are hot and dry. The moisture deficit in summer is perhaps 1000 mm. Crop and pasture growth stops completely. Summer droughts occur in temperate farming regions from time to time but within dryland farming zones they are permanent feature of the farming year. A winter rainfall pattern is typical of Mediterranean dryland farming zones. These zones are found on the southern shore of the Mediterranean, West Asia, Southern Australia, Chile and California (Chatterton 1996; Drylandfarming.org 2013).
The map illustrates the difficulty of translating simple climatic parameters into farming zones. The pale green and orange coloured areas are clearly dryland farming zones. The Dark brown is clearly humid temperate but the dark green area classed as “high rainfall” Mediterranean which is an intermediate dryland farming zone. The summer droughts are shorter. Winter and spring rains are heavier and more reliable. A sustainable and profitable dryland farming system for these Mediterranean dryland regions was evolved in the 19th and 20th centuries by farmers in South Australia. It is based on self regenerating annual medicago (medic) and clovers and enables farmers to rotate crops of grain with pasture for sheep in a productive and low cost rotation. In its traditional Australian form the pasture is established and the seed reserves in the soil are safeguarded by ensuring that during seed bed preparation for the subsequent grain crop, deep ploughing is abandoned in favour of shallow cultivation. Shallow cultivation is energy and fuel efficient and saves time and labour. Shallow cultivation is essential as it protects the pasture seed and enables it to regenerate each spring without further seeding. Variations of this traditional rotation have been developed in North Africa that combine the medic with conventional fallowing. Nitrogen fertiliser is not used as the pasture produces sufficient soil nitrogen for grain crops. An application of superphosphate will ensure that the pasture remains abundant. Grazing is managed to ensure high levels of seed production. A rotation of grain and pasture becomes the basis of the farming system with a portion of the farmland in pasture and a portion in grain each year and variations depending on the climate and market signals. Dryland farming also refers to zones in the tropics where the growing season is limited by a seasonal drought. The rainfall in the tropics is during the summer while the winter is the dry season. Because temperatures and evapo-transpiration rates are much higher in summer tropical dryland farming requires about double the amount of annual rainfall for similar levels of production. Annual, self-regenerating, legume pastures are available for the dry tropics but they have not yet been developed into a practical farming system (Chatterton 1996).
The Green Revolution
Over the last half century the green revolution has transformed cereal production in the temperate rain fed regions and areas with irrigation. The Green Revolution is not “green” in the modern sense as it is a high input farming system that exploits some non-renewable resources. The three major inputs are water from ample rainfall or irrigation, large quantities of nitrogen fertiliser produced from petroleum and cereal cultivars that can convert the water and nitrogen into high grain yields. In dryland farming areas the water supply is unreliable. In the Mediterranean dryland farming zones the spring rainfall is critical for the development of good grain yields. The use of nitrogen fertiliser produces erratic results. The application of nitrogen encourages the cereal plant to tiller - that is to produce more side shoots. These shoots produce seed heads and then grain. If spring rains are poor the grain does not mature and because of the increased competition between tillers the yield may be lower than that from a crop that does not receive nitrogen fertiliser.
This is not a startling new discovery. Experiments conducted at the Roseworthy Agricultural College in the South Australian dryland farming zone in the 1880's showed that the application of nitrogen fertiliser produced such variable results that farmers were not advised to use it. The effectiveness of nitrogen depends on the spring rainfall. Where the annual rainfall is above 500 mm in the Mediterranean dryland farming zone the response to nitrogen fertiliser is more reliable while below 500 mm the risks are much greater. In some years there will be no yield response or even a reduction in yield. The application of nitrogen is not only a technical question. The response curve is the core of the decision making process but the farmer also needs tocalculate the cost and benefit from the price of the fertiliser and the return from the cereal crop. Nitrogen fertilisers have generally been cheap over the last hundred years but have increased dramatically in price with increases in energy prices. Grain prices have generally fallen in real terms. Over the last decade or so both grain prices and fertiliser prices have increased. The true price of the nitrogen fertiliser is not reflected in the price paid by farmers and entered into their profit and loss accounts. Nitrogen fertiliser is solid or liquid energy from fossil carbon sources. When it is applied to the soil it is converted by soil bacteria into various compounds of nitrogen that can be absorbed by the plants. During the conversion process oxides of nitrogen are released into the atmosphere. These are about 300 times more polluting than carbon dioxide in terms of the green house effect. Instead of taxing nitrogen fertiliser to reflect these environmental costs governments are more likely to subsidise it. The amount of nitrogen lost to the atmosphere varies according to climatic conditions but inadequate rainfall and rising temperatures as found in dryland farming zones during the spring are conducive to higher levels of loss. The third factor in the decision making process is risk. Risks are greater for small farmers with few resources of capital or credit compared to the larger farmers. The risk of zero or negative returns from nitrogen fertiliser are greater in low rainfall areas compared to higher rainfall zones. In Australia farms are larger in lower rainfall zones but that is not the case in many other dryland farming regions of the world. For example in North Africa due to the colonial legacy farms are often smaller in the more marginal zones. These farmers are unlikely to take on the additional risks associated with nitrogen fertilizer (Chatterton 1996)..
Alternative sources of nitrogen
For more than one hundred years the scientific community has been searching for alternative sources of soil 
. One of the first options, promoted in the first half of the 20th century, was the the long cultivated fallow. The land was cultivated in the spring and left bare during the summer. In the autumn the cereal crop was sown. At first this was advocated as a means of carrying soil moisture over from one season to the next but this theory has been comprehensively discredited. However the cultivated fallow did work. It increased yields through better weed control and the mobilisation of organic nitrogen.
Table 1. Long term rotational trials conducted at the Waite Agricultural Research Institute in Adelaide.
|
Dryland farming rotations
|
Means yield of wheat Kg/ha
|
|
Period 1926 to 1951
|
Period 1952 to 1983
|
|
Continuous wheat (1 year)
|
874
|
692
|
|
Wheat-cultivated fallow (2 years)
|
2300
|
1403
|
|
Wheat - grain legume (peas) (2years)
|
1668
|
1420
|
|
Wheat – self-regenerating legume pasture - legume pasture no fallow (3 years)
|
Not included
|
2033
|
|
Wheat – self-regenerating legume pasture - legume pasture - legume pasture/cultivated fallow (4 years) The Zaghouan 4 year rotation
|
Not included
|
2402
|
(own illustration Chatterton 2013).
The rotational trials from the Australian dryland farming zone have been illustrated in the table above because there are no trials anywhere else in the dryland farming zones which cover a similar 57 year period. It can be seen that the fallow performed well for the first 25 years. The yield of 2300 Kg/ha may not seem much to Europeans but the average wheat yield in Algeria is currently 1400 Kg/ha. During the next 20 years the fallow rotation collapsed. The fallow had mobilised all the organic nitrogen. There was no renewal and yields declined. Soil structure was damaged and erosion increased. Due to failing yields and increasing erosion a growing number of economically farmers also questioned the benefits of using the land for two years to grow a single cereal crop and nothing else. Farmers in Australia responded to the economic and environment crisis by adopting rotations that include annual legume pastures. These pastures restored the soil fertility, improved soil structure and provided a substantial, additional income from sheep. As can be seen from the above table, these legume pasture rotations were only recognised by the scientific community in the 1950's.
In North Africa and West Asia the destructive effect of the long fallow was well recognised half a century ago and many projects were launched by international and national agencies to “suppress or eliminate” the fallow. Most were based on a scientific, top down approach unlike Australia where the farmers had been the source of innovation. The favoured approach was a rotation that included grain legumes or fodder crops such as vetch.
Technically they were not a good solution. While vetches and grain legume crops fix large amounts of nitrogen from the air most of the protein is removed with the grain or hay and they do little to increase the fertility of the soil for the next cereal crop. Table I above shows the poor results from a wheat – peas rotation.They failed to have a widespread impact because the project managers took little account of practical farming problems. The scientific institutions conduct research in terms of the impact of physical inputs such as fertilisers, herbicides fungicides etc on grain yields. They take no account of the input of the farmers' labour as research centres have abundant resources of labour and machinery. Farmers on the other hand have to juggle priorities within their limited resources of labour and machinery. A change from fallow-cereals to grain legume – cereals requires at least a four fold increase in tractor time during the critical autumn sowing period (Chatterton 1996).
The legume pastures come to North Africa and West Asia
During the 1970s and early 1980s most countries in North Africa and West Asia introduced legume pasture rotations based on the Australian model. These rotations were first developed in the state of South Australia and had their major impact between the 1930s and 1970s. Since then yields have increased substantially due to other technical innovations.
Table 2
|
|
South Australia
|
Algeria
|
|
1930s before the Dark Green Farming Revolution
|
1970s after the Dark Green Revolution
|
1970s and still waiting.
|
|
Wheat production – '000 tonnes
|
888
|
1327
|
1270
|
|
Wheat yield – Kg/ha
|
735
|
1139
|
624
|
|
Sheep numbers '000
|
8500
|
18961
|
8357
|
(own illustration Chatterton 2013)
While cereal yields increased, the really important impact was the very large increase in sheep numbers. The additional sheep were fed on cheap nutritious pasture grown in rotation with cereals or on non-arable land.
The introduction of the legume pastures (mostly based on medic - a common name for annual species of Medicago) to North Africa and West Asia was a great technical success. The pastures grew vigorously, the farmers had more and better sheep and cereal yields increased. This was hardly surprising as the medic cultivars had been collected by Australians from North Africa and West Asia over the previous several decades.
Legume pastures increase soil fertility but they also provide a break in the rotation which helps reduce cereal diseases and they improve the soil structure. Yield increases are greater than the simple nitrogen levels would indicate. In a series of trials in Iraq, for example, medic pastures produced a one tonne yield advantage over and above that expected by nitrogen levels alone. It is difficult to extrapolate from these results to other zones as different farming areas have different disease burdens and soil structures.
These annual legume pastures do not have a downside. Obviously if the spring rains fail the crop fails. The farmer does not lose money to the same degree as a farmer who used nitrogen. The legume nitrogen came free as a bye-product of a pasture that was profitable in its own right. When spring rains are poor, the crop fertilised with artificial nitrogen yields less than one with no nitrogen. This does not occur when the nitrogen come from legumes. The legume nitrogen is in an organic form. It breaks down slowly. There are no recorded instances where a cereal crop after medic yielded less.
The new medic farming system failed to become widely adopted in North Africa and West Asia except in Libya for farming reasons. The project managers failed to take account of the need for increased farming knowledge. Understanding grazing management and new form of shallow cultivation proved to be the weak links. In Libya, Australian farmers were employed as extension agents and these new techniques were rapidly adopted by the Libyan farmers.
While the lack of farming skills was the immediate cause of the failure to achieve wide scale adoption there were other more general barriers within the government and international bureaucracies themselves. Changing a complete farming system is not the same as introducing a single fertiliser or herbicide. It requires a much longer time scale yet most projects lasted only five years or in some cases as little as three.
Agricultural bureaucracies both administrative and research are divided into sections. The first major division is between animals and crops. The medic pasture-cereal rotation is truly integrated. The sheep and crops are dependent on each other. It is striking that the most successful introduction in Libya was managed not by the Ministry of Agriculture but by a general development authority.
Most senior officials and researchers within the dryland farming zones have post graduate qualification from outside their own country. They go to France, USA, Sweden, Russia – in fact any country in the world except those with relevant climatic similarities. These qualification add little to their knowledge of dryland farming and nothing of the medic pasture – cereal rotation. Pastures have gone out of fashion in the temperate northern zones. FAO abolished its pasture section some decades ago and if northern universities teach pastures at all their courses only include perennial grasses and legumes not the self-regenerating, annual legumes such as medic or stylo used in dryland farming zones.
The consequence is that ministries and research centres can provide little support. They may even resist the new rotations as they see them as a threat to their existing expertise. Commercial interests also see new farming systems as a threat to their markets for nitrogen fertiliser, sheep feed and deep ploughs (Chatterton 1996)..
New look nitrogen fertiliser
One approach to the problem of the poor response to nitrogen fertiliser in dryland farming areas has been to try and mimic the slow release from organic nitrogen from legume pastures. Less fertiliser is applied at sowing and then further applications are made during the growing season using Near Infra Red analysis of the crop to determine requirements.
This approach is expensive, time consuming and well beyond the resources of skill and investment for small farmers. It does nothing to tackle the problem of poor soil structure, soil erosion and environmental pollution from the production and use of nitrogen (Chatterton 1996)..
The dry tropics
The medic group of pasture plants are suited to winter rainfall zones but there are legumes with similar characteristics suited to the dry tropical regions. Unfortunately the practical farming systems based on these species are not nearly as well advanced as those for the Mediterranean dryland zones.
Market opportunities
For the dryland farming zone of North Africa and West Asia one would think that market opportunities are limitless as this region is the major food deficit area of the world. Grain markets have been opened to international competition in most countries in the region and while prices have risen on average during the last decade they are extremely variable. In the past grain prices rose when crops were poor and fell when they were good. Now they can move in either direction according to the whims of world traders and the position of grain farmers has become more precarious. The ability of small farmers in these dryland farming zones to compete with the large grain exporting countries is doubtful.
Sheep provide greater opportunities than grain. Sheep meat is preferred in the West Asian and North African region. Prices are the highest in the world and even greater in terms of purchasing power. While this makes sheep meat a luxury for most of the urban population it makes sheep production extremely profitable for local farmers provided they can feed their flocks on cheap, nutritious pasture not expensive imported grain. While live sheep and sheep meat are imported from Australia and New Zealand, the local fresh product receives a premium price. There is no premium for local grain.There needs to be a technical-economic balance in development policies. The price of sheep in North Africa and West Asia has been high for many decades but farmers are not profiting because of a lack of access to cheap pasture technology. On other occasions technology has been pushed at farmers in spite of the poor prospects for profit. Sheep production and cheap legume pastures are a dream combination of technology and economics.
For the tropical dryland zones the market opportunities are not as good. The cereal market can collapse completely during a drought due to the importation of grain for famine relief. The livestock market has declined for Sudan and Somalia due to more controls on animal health imposed by Saudi Arabia and other importing countries of the Gulf. Other local urban markets in East and West Africa have been depressed by the importation of dumped European beef (Chatterton 1996)..
The future for dryland farming zones
It is time to change the focus of agricultural development in dryland farming zones towards an exclusive concentration on profits for the farmer. Profit has appeared in the list of project objectives for decades but national production targets have often received a much higher priority. In policy terms this means that Ministries of Agriculture must resist the temptation to exhort farmers to grow more wheat or any other product for reasons of national planning. Researchers must look at the cost of production as well as yield increases.
Mediterranean dryland farming
In the dryland farming zones of North Africa and West Asia more profitable farming is comparatively simple. Given the strong domestic demand, sheep production has considerable potential provided the costs are low. Annual legume pastures can provide good low cost feed. Non arable land presents a huge untapped resource for sheep production. The increase in sheep numbers from 8 million to 18 million (Table 2) in South Australia did not come just from medic pasture in rotation with cereals but also by the transformation of poor grazing on non arable land through the use of medic and sub clovers combined with phosphate fertiliser. The same potential exists in North Africa and West Asia.
On the arable land the Zaghouan rotation is unashamedly aimed at better profits for farmers. The rotation is an amalgam of existing fallow-cereal rotations and medic pastures. It is based on research from ICARDA and practical farming expertise. The rotation reduces the area under cereals from half the farm to a quarter. Yields are at least double. Farmers profits from cereals are increased substantially. The remaining three quarters of the land is used for medic pastures. Sheep numbers are increased four times. The cost of feeding is reduced and the weight of the sheep increased.
Tropical dryland farming
The farming situation in these zones is bleak. They are the poorest countries in the world. There is no high priced domestic market for livestock. Export markets in West Asia are difficult to access due to increased requirements for healthy animals. Urban markets on the west and east coasts of Africa are not affluent and have been depressed due to the dumping of subsidised beef from Europe. A truly integrated system of legume pasture, grazing livestock and crops is the most suitable option for farmers in the zone. It has low inputs (mostly phosphate fertiliser) and low risks. Whereas the North African and West Asian region had the Australian working model to use as a starting point for further modification and adaption (the Zaghouan rotation from Tunisia for example) there is no working model for this zone. The legume pastures with the necessary characteristics exist but they have not yet been merged into a working system.
The improvement of dryland farming in the tropical zone will depend on price signals as well as a sustainable technical package. Farmers will receive better prices if livestock markets in urban centres can be developed and if the UN Agencies expand the use of cash cards for famine relief rather than distributing free grain (Chatterton 1996).
References
Chatterton L. and Chatterton B. (1996). Sustainable dryland farming, Cambridge University Press
Malcolm, B.; Sale, P."; Egan, A. (1996). Agriculture in Australia - An Introduction. Australia: Oxford University Press.
Squires, V. and Tow, P., Dryland Farming: A Systems Approach - An Analysis of Dryland Agriculture in Australia (Sydney: Sydney University Press, 1991)
Further reading
Dryland Farming Organisation. www.drylandfarming.org
Great Green Wall for the Sahara and the Sahel Initiative www.greatgreenwallinitiative.org
