Challenges and Hopes for the 2000s

The Rangelands of Arid and Semi-Arid Areas: A Review

Arid and semi-arid areas

Arid and semi-arid areas are defined as areas falling within the rainfall zones of 0-300 mm and 300-600 mm, respectively (FAO, 1987). Because of the short growing periods (1-74 and 75-119 growing days, respectively), these areas are not suitable for cultivation. Rainfall patterns are unpredictable and are subject to great fluctuations. One-year droughts are more frequent than multiyear droughts. The occurrence of drought is more frequent in the arid (lower rainfall) areas than in the semi-arid zones. For example, according to Ellis (1992), drought occurs every five years in the Turkana District of Kenya (200-500 mm annual rain), whereas it occurs every 8-12 years in the Massai region (300-700 mm rainfall). Severe multiyear droughts in Turkana decimated half the livestock and led to the temporary migration of 20% of the human population (Ellis, 1995). In contrast, single-year drought does not cause livestock mortality.

Historically, the rangeland dominating arid and semi-arid areas provided primary products (grasses, legumes and shrubs) which were converted into animal protein. The use of the resources for other purposes, such as fuel and building material, intensified with the increase in human population and with sedenterization. These rangelands maintained an ecological balance as a result of the natural defensive mechanisms typical of uncertain and highly erratic climates. Seasonal fluctuations influence the concentration and mix of herbivores, and multiyear droughts reduce the number of animals.

An assessment of the dynamics of drylands requires long-term data collection. However, based on the information available, it has been possible to conclude with a reasonable degree of certainty that plant biomass is greatly influenced by fluctuations in rainfall (Ellis, 1992, 1995; Scoones, 1995) (Figure 1).

Figure 1. Monthly and running mean normalized difference vegetation index

Source: Nysonyoka Territory, Turkana District, Kenya (Ellis, 1995, p. 39).

Recovery continues even after severe drought in spite of the short-term influences of seasonal and one-year droughts until the onset of another multiyear drought. Also, it is an accepted fact that the influence of this type of biomass variation is typical of the drylands environment and is much stronger than the influence of the changes induced by livestock. Such a pattern is common to most arid and semi-arid areas (e.g., Africa), except where government policies (and herders' incomes) allow for continuous use of the rangelands through the support of external factors (e.g., the trucking of water, the year-round availability of subsidized grain and stable livestock values compared with the changing value of local currencies). The latter is common in the Near East and North Africa countries.
Historic data (1960s to present) collected from 36 dryland countries (Annex 1) using AGROSTAT estimates (FAO, 1996) have been used in comparing the trend in livestock and human populations with the changes in permanent pastures (Figure 2). Area-wise, the permanent pasture remained unchanged over three decades, whereas the human population increased 2.6 times. In spite of some fluctuations, livestock numbers increased from 400 million head in 1961 to 600 million in 1995. The important conclusion reached from observation of the trends over three decades is that arid and semi-arid rangelands are currently subject to heavier grazing pressure than was the case 30 years ago.
The observations derived from FAO statistics are supported by extensive surveys carried out on livestock populations in the drylands and wetlands of Mali, Niger, Nigeria and Sudan (Wint and Bourn, 1994). Analysis of the results points to a strong relationship between human activity (density and cultivation) and the intensity of livestock-production activities. However, in spite of the progressive settlement of pastoralists and changes in herd composition, breed and animal size, the movement from extensive systems to intensive (zero-grazing) systems has always been slow whereas the supply of natural feed resources has remained static or increased very slowly (Figure 2).

Figure 2. Trends in human/livestock population and permanent pasture in 36 arid and semi-arid developing countries

NOTE: Permanent pasture: lands used permanently (five years or more) for herbaceous forage crops either cultivated, or growing wild (wild prairie or grazing lands) (FAOSTAT, 1993). Rangelands in arid and semi-arid areas include permanent pastures and land not suited for other uses. Livestock: cattle, goats, sheep and camels.

Source: FAO (1996).

Back to Top

Land degradation and erosion in arid and semi-arid areas

Because livestock is the major user of primary production in arid and semi-arid regions, degradation has always been attributed to this subsector (Sidahmed and Yazman, 1994). The United Nations Environment Programme (UNEP) singled out human impact and, specifically, livestock grazing as the cause of the irreversible degradation which has prevailed during the past two decades (Pearce, 1992). According to the World Resources Institute (WRI, 1992) "overgrazing is the most pervasive cause of soil degradation . . . . In Africa and Australia, overgrazing causes 49 and 80 percent, respectively, of soil degradation, mainly in semi-arid and arid regions." Although the share of responsibility on the part of other influences (the introduction of exotic species, fuel-wood harvesting, the suppression of the natural fire cycle, wildlife degradation and the conversion of rangelands to croplands or human settlements, etc.) has been emphasized in subsequent UNEP publications (WRI, 1994), overgrazing has always been considered the most important factor (Figure 3).

Figure 3. Rangeland degradation since 1945

NOTE: Percentages indicate the contribution of overgrazing to total degraded area.

Source: WRI (1994).

However, recent results from the long-term monitoring studies of the International Livestock Centre for Africa (ILCA) in East and West Africa (Ellis, 1992; Hiernaux, 1993) have challenged this assumption and provided evidence that climate and not livestock is the main determinant of changes in arid and semi-arid environments and that rangelands are resilient and capable of recovery. For example, plant biomass in the Turkana region was greatly influenced by rainfall fluctuations (Figure 1) and has progressively recovered following the lengthy drought of 1979-80. ILCA studies concluded that "the strong seasonality of rangeland production in the Sahel limits the risk of overgrazing damaging the environment to short periods and consequently confined areas." Moreover, related studies in Mauritania concluded that Sahelian vegetation appears very resilient to natural and grazing stresses because of the strong dynamism of the seed production, dispersion and germination cycle (Carrière, 1989).

The results and the impact of studies carried out by ILCA and others were discussed in two consecutive workshops supported by the World Bank, the Commonwealth Secretariat, the Overseas Development Institute and others (Behnke and Scoones, 1992; Scoones, 1995) and confirmed the conventional wisdom and traditional practices of pastorals and nomads throughout history in these areas. Traditionally, pastorals adopted an opportunistic strategy of mobility and raised mixed species of stock with different preferences for standing vegetation, so as to optimize the use of the limited vegetation available (Sidahmed, 1993). The core of recent interpretations leading to so-called "rethinking range ecology" is appropriately based on the fact that degradation (or what was considered "desertification") is mainly influenced by rainfall patterns.

But the management of the growing vegetation is equally important. According to Behnke (1993), the resilience of dryland vegetation is an outcome of the marked fluctuations between wet and dry seasons. Whereas growing vegetation is vulnerable to damage by grazing and trampling, dry-season vegetation is far below demand and has its own defensive mechanisms, such as the protection of the living parts behind thorns or in the seeds. This was confirmed during more than ten years of range monitoring by ILCA in Mali. The results (Hiernaux, 1993) indicated the negative influence of repeated grazing during the growing season on the replenishment of the soil seed-bank. On the other hand, minor changes were found to be caused by the short-term effects of trampling, grazing and burning during the dry season. The studies identified close similarities between the changes induced by drought and those resulting from grazing pressure on the Sahelian rangelands, emphasizing the fact that it is rather the policies (that limit livestock mobility and cause seasonal dependence by providing feed inputs) that lead to degradation.

The conclusion reached by ILCA's long-term monitoring studies support the recent findings of other investigators in Sweden (Lund University, the Swedish Agency for Research Cooperation with Developing Countries), the UK (University College London), the US (Florida State University) and the World Bank (Nelson, 1988). The findings challenge the long-term claim by UNEP that grazing areas of arid and semi-arid regions, particularly those in the desert margins of the Sahara (Pearce, 1992), are subject to irreversible degradation. These reviews and others have raised concerns about the quality and the relevance of the research conducted by UNEP on desertification in the Sahel. These concerns relate to the absence of evidence of good data or reliable scientific standards.

For example, the satellite pictures examined by Ulf Hellden at Lund University in Sweden, as well as polar-orbiting meteorological satellite pictures analysed by Tucker and colleagues at the NASA Laboratory for Terrestrial Physics, indicate that the desert in the Sahara moves back and forth between its northern and southern boundary. According to the studies, these oscillations or north-south/south-north shifts are rainfall related. Furthermore, Nelson (1988) reported an astonishing recovery in what were considered irreversibly degraded areas in Ethiopia after rainfalls following the successive droughts of the 1980s.

Several ecologists have attributed degradation around water points to livestock concentration. However, this is a one-sided judgment because it excludes the benefits of the manure deposited in the circle just around the water point, as the animals come and go. Andrew Warren, a geographer at University College London, and Hellden at the University of Lund, Sweden, did not find any evidence of desertification around water points in Sudan. A review of the surroundings of 77 water points in central Sudan and 20 on the desert fringe of Senegal was unable to establish a relationship between degradation and animal concentration around the wells. This and other examples by Warren were brought to the attention of the UN in 1991. According to Warren "even the view that cattle watering points act as centres or poles of desertification is now questioned."

Resilience is not only a feature in the grazing lands of the arid and semi-arid tropics, but also of continental and temperate dryland areas. For example, an IFAD environmental assessment mission (November 1992) to the Hainan Tibetan region of Qinghai Province could not find evidence to allow any firm conclusions regarding the claim that overgrazing is the cause of degradation of robust and resilient summer and winter grazing areas. According to the mission, species diversity has been kept intact, and regeneration would be possible, in spite of the increasing yak and sheep populations. As in the case of long droughts in the tropics, periodic very cold winters cause heavy losses in the animal population, leading to reduced grazing pressure and regeneration of the vegetative cover.

The situation in the Near East and North Africa region

Most of the land area in the region (62%) is classified as rangelands (FAO, 1991), and half of these rangelands are desert and semi-desert, with a limited contribution to controlled or reliable livestock production. The region has witnessed rapid socio-economic change in the last three decades. The areas most affected by this change are those rangelands which have been encroached by cultivation (Sidahmed, 1991). However, neither the reduction in the rural labour force as a result of out-migration to urban areas, nor the drought years of the 1970s and 1980s have brought about a reduction in the number of animals in the region, as might have been expected. On the contrary, both the livestock and human population increased sharply in the region between 1961 and 1995; from 105 million to 192 million and from 96 million to 245 million, respectively (Figure 4). On the other hand, the increase in the land used for permanent pasture was very small and was limited to Sudan, Saudi Arabia and Libya. Furthermore, human population numbers are expected to double between now and the year 2020 (Sidahmed, 1991), and a similar trend is expected for livestock.

Figure 4. Trends in human and livestock population and permanent pastures in 19 arid and semi-arid Arabic-speaking countries

NOTE: Permanent pasture: lands used permanently (five years or more) for herbaceous forage crops either cultivated, or growing wild (wild prairie or grazing lands) (FAOSTAT, 1993). Rangelands in arid and semiarid areas include permanent pastures and land not suited for other uses. Livestock: cattle, goats, sheep and camels.

Source: FAO (1996).

In contrast to the majority of the drylands of Africa, extensive livestock production in the Near East and North Africa region has shifted to systems which are heavily dependent on imported and subsidized feed grain (Sidahmed, 1992). The subsidies mostly favour large herders and absentee livestock owners in the high and middle-income countries such as Saudi Arabia, where only 20% of the basic feed resources are provided by range forages and the balance by barley grain. However, supplemental feeding is also very attractive to poor farmers, who find financial incentives from converting food grain into animal protein. This is particularly the case in countries where animal value is stable and highly rewarding compared with local currencies. For example, the price ratio of sheep to grain in Algeria exceeded 20 to 1 (Bayer and Waters-Bayer, 1995). In semi-arid central Tunisia, smallholders in normal years buy 40% of the feed they require and in the dry years up to 70%. However, the availability of supplemental feed does not take animals off the rangelands. Because the natural feed resources are cheap and since grazing is known to improve meat quality, animals are kept continuously on the range, which disturbs the natural balance and intensifies the degradation process.

There is still much work to be done to assess the effect of these changes on the sustainability of rangelands and their value for future generations. Limited observations by IFAD missions to Jordan indicate that the consumption of green feed doubled from the 1960s to the 1990s, leading to extensive depletion of the seed supply (IFAD, 1993). In that country, the range supplied 85% of the livestock feed in the 1950-60s, but total consumption was much lower than the 50% "proper utilization factor". Although the range contribution to animal feed in Jordan was reduced to only 40% in the 1990s, a sharp increase in animal numbers following a sharp rise in imported and subsidized feed grain (Figure 5) has led to the consumption of very large proportions of the standing vegetation (75-90%). According to this study, the area of rangeland subject to intense livestock grazing has been raised to the maximum limit by the establishment of wells and the trucking of water.

Furthermore, based on the modelling exercise used by Peter Harris of the IFAD mission in interpreting the available data, the overall forage-energy-utilization level increased from 35% in the 1960s to 67% at present. In simplified terms, this means that at least 290 kg of forage needed to be grown in the 1960s (when there was a more lenient, more sustainable utilization level and high carryover) in exchange for every 100 kg consumed to keep the rangeland in good condition, whereas, under the present regime (overgrazed ranges with little carryover), only 150 kg are grown for every 100 kg consumed. This might explain the cause of the 50% decline in range productivity in Jordan over the last 30 years. Unfortunately, this kind of information is rare and, when available, is based on distorted or uncertain data.

Figure 5. Trends in livestock numbers, livestock feed imports and forage production in Jordan

NOTE: The number of livestock is a total of sheep, goat, cattle, equine and camel numbers. Feed imports are expressed in terms of metabolizable energy. Forage production is mainly from range, but includes crop residues and local bran and grain.

Valid CSS! Valid XHTML 1.0 Transitional