3.4.1 Ecology and land use
3.4.2 Environmental change
3.4.3 Use of native plants
Livestock have been implicated as a major factor in environmental changes whereby 40% of the western Borana Plateau has endured bush encroachment and 19% has significant erosion. This perspective conforms to the mainstream view that pastoralists and their livestock can have widespread negative effects on natural resources, irrespective of climatic factors (see citations in Section 3.1: Introduction and Section 6.4.5: Equilibrial versus non-equilibrial population dynamics). Tendencies towards trends such as bush encroachment, however, have also been exacerbated by government policy.
Two major findings from the ecological mapping exercise are: (1) documentation of the high spatial diversity in forage resources conferred by the 104 ecological site types and (2) the high variability in access to the six ecological zones by residents of the 29 madda. To recap, sustainable rain-fed cultivation may be very viable on suitable landscapes in the subhumid and upper semi-arid zones as well as in bottomlands, Zonal criteria should be reviewed in the formulation of policy providing for cereal cultivation (see Section 7.3.2: Land-use policy and agronomic interventions). The subhumid and upper semi-arid zones have conditions that are conducive to woody encroachment and the arid zone requires the most flexibility in pastoral management due to patterns of past degradation as well as its variable availability of forage and water. This all implies that rather than embracing one comprehensive approach for range management and pastoral development, efforts need to be tailored, at minimum, to deal with region-specific issues at the madda level of resolution. This would require, however, a level of site-specific research and coordination that greatly exceeds the capabilities of local governmental and nongovernmental organisations. A reasonable compromise could promote collaborative, participatory approaches which involve the Borana leadership and repealing controversial land-use policies (see Chapter 8: Synthesis and conclusions).
Despite the diversity in ecological communities, the diversity of readily accessible landscape types with marked seasonal variation in access to strategic production resources (e.g. uplands vs bottomlands, or mountain ranges) appears low on the central Borana Plateau compared with other pastoral systems in East Africa for which use of different landscapes is well documented. This is largely because bottomlands, occur on <12% of the total area and because mountains are usually inaccessible to livestock because of poor water availability (see Section 2.4.1.7: Water resources). bottomlands, are also unevenly distributed. Coppock et al (1986a) and Ellis et al (1986) illustrated the critical role of forested riverine areas and mountain savannahs in promoting livestock species diversity and the perseverance of Turkana pastoral subsystems in arid north-western Kenya. Scoones (1991) noted the increasingly critical role of bottomlands, in African pastoral systems. They serve as sources of dry season forage and offer opportunities for the development of agropastoralism. As a result, bottomlands, are commonly a source of land-use conflicts where population pressure is acute (Scoones, 1991). On the central Borana Plateau bottomlands, are exploited mainly for collection of calf forages in dry seasons (Menwyelet Atsedu, 1990) while northern subhumid and upper semi-arid zones at higher elevations are used by cattle during extended dry seasons and drought (see Section 6.3.1.1: Livestock dispersal and herd composition). Use of reliable tula well systems during drought also leads to herds gathering in "fall-back' madda (see Section 2.4.1.7: Water resources Such trans-madda movements of cattle reduce the relevance of madda-specific estimates of carrying capacity during times of stress. Instead, resource managers need to consider population and resource-use dynamics at a level of resolution which encompasses the entire western half of the Borana Plateau (see Chapter 8: Synthesis and conclusions).
3.4.2.1 Review of ancillary work in Borana
3.4.2.2 Hypothesised cycle of grazing-induced vegetation dynamics
3.4.2.3 Short-term grazing effects
3.4.2.4 Population ecology of woody species
As perhaps the most ubiquitous consequence of heavy grazing pressure in the Borana plateau, bush encroachment is not unusual in semi-arid East Africa. Increases in woody vegetation have been commonly reported as a response to heavy grazing elsewhere (Norton-Griffiths, 1979; Cumming, 1982; Sabiiti and Wein, 1988; Belsky, 1989; a Tchie and Gakahu, 1989: Msafifi and Pieper. 1989).
Other investigators in the southern rangelands made similar conclusions regarding impacts of pastoralists on the environment. During the early 1970s when human and livestock numbers were just beginning to increase in the Did Hara region in response to pond development, consultants from AGROTEC/CRG/SEDES Associates (1974d: pp 85-87) noted that woody encroachment was a risk there. They also stated that woody encroachment had already advanced in a number of other semi-arid regions.
Reporting on a brief consulting survey, Pratt (1987a: pp 15,17) noted that given the favourable rainfall of the region, natural succession would lead to a dominance of woody plants in most communities regardless of grazing. He recognised the heightened soil erosion on slopes at higher elevations and speculated that it was caused by high stocking rates and the trekking of cattle to up-country markets. Pratt (1987a: p 18) also reported extensive bushland and thickets below 1500 m elevation. He hypothesised that the overall increase in woody plants appeared to be due to a gradual thickening of existing stands as a result of seedling recruitment rather than expansion of plants into new sites. Pratt (1987a: pp 17-18) concluded that although woody encroachment was marked, the rangeland was still in "reasonable" condition because: (1) patches of high-quality perennial grasses remained in heavily grazed sites that could respond well to adequate rainfall; and (2) similarity in grass cover between government ranches (managed at lower stocking rates) and adjacent communal rangeland. Pratt (1987a) felt that priority attention should be given to arrest trends toward rapid resource deterioration at higher elevations where population pressure was most acute.
Summarising another consulting survey, Hacker (1990: pp 5-24) also confirmed a high incidence of soil erosion and the presence of young woody plants in 68% of 801 survey sites. He noted that erosion appeared to be greatest on slopes comprised of red soil (i.e. yellow-red Haplic Xerosols) where grass cover was reduced by grazing, but that the amount of grass cover was not strongly associated with abundance of young woody plants. In contrast to the more favourable assessments three years earlier in the same region by Pratt (1987a), Hacker (1990) ranked 43% of his sample sites from poor to very poor grazing condition and felt that the government ranches had markedly higher grass cover than areas outside (note that the discrepancy between the range condition surveys of Pratt, 1987a and Hacker, 1990 may be explained by growth in the cattle population during 1987-90; see Section 7.2: A theory of local system dynamics). From a comparison of 15 sample points near Yabelo town Using aerial photos from 1967 and 1984, Hacker (1990: pp 27-37) estimated that woody canopy cover had increased only slightly (from 24.1 to 28.5%) and that cover had increased at only four and decreased at only one of the sites. He concluded that changes in woody cover in this particular region were thus more local than general in nature, which supports Pratt's (1987a) contentions.
In sum, the concensus appears to be that soil erosion and woody encroachment have occurred to varying degrees throughout the study area. While livestock pressure can probably be directly linked to soil erosion along trekking routes and high-elevation slopes, mechanistic ways that link cattle to woody encroachment are less clear. The tendency for woody plants to proliferate under grazing perturbation is greatest in the upper semi-arid and subhumid zones because higher soil moisture increases the likelihood of establishing woody vegetation (Tamene Yigezu, 1990).
In addition, woody encroachment could also be facilitated if grazing modifies the competitive relations among grasses and woody seedlings for soil moisture and nutrients, an interaction which may be most pronounced on fine-textured, rather than sandy, soils (Walker et al, 1981; Walker, 1985; Knoop and Walker, 1985). Most of the soils on the Borana Plateau are relatively fine-textured (Kamara and Haque, 1988).
Another factor that has probably facilitated woody encroachment may be the national policy banning burning of grazing and agricultural lands since the mid-1970s. Prior to this the Boran burned rangeland to improve forage quality and control woody plants and ticks (Coppock, 1990b). This policy constraint has been noted by several investigators Billé 1985; Pratt 1987a: p 18; Hacker, 1990: p 20). However, it is noteworthy that extensive woody encroachment predated the ban (AGROTEC/CRG/SEDES Associates, 1974d: pp 85-87; Billé et al, 1983). This suggests that traditional use of fire was only partially effective in controlling woody vegetation. The local ban on range fires was lifted by government administrators in 1990. This is discussed further in Section 7.3.1.4: Site reclamation.
Billé and Assefa Eshete (1983b: p 34) noted the traditional importance of the hypothesised cycle of patch use by cattle and shifts in vegetation composition in maintaining a long-term dynamic stability and diversity of the Borana ecosystem. Factors that could disrupt the cycle, such as higher human and cattle populations that re-use sites before grass recovery has taken place (as at Goff), or higher populations of browsing stock (e.g. goats) that prevent establishment of woody vegetation, could negatively affect the sustainability of pastoral production. The same would hold if fire were too frequent (see Section 7.3.1.4: Site reclamation). Woody encroachment could be beneficial in the long run by forcing pastoralists to migrate from overgrazed sites thereby providing the degraded areas with an opportunity to rehabilitate themselves Billé and Assefa Eshete, 1983b).
In a rigorous sense, the elements that underpin the hypothesis of a cycle of grazing-induced dynamics of woody encroachment remain untested although some of its key elements are supported by other work in Borana and research from similar systems elsewhere:
Migration of pastoralists in response to condition of the local resource base. Observations that the Boran occupy areas until they are severely overgrazed is only supported by the anecdotal reports of Billé and Assefa Eshete (1983b) for Medecho madda. Pastoral response to advanced stages of woody encroachment can only be surmised from interviews. Herd owners have said that instead of attempting to control local woody encroachment they would prefer to move elsewhere (D. L. Coppock, ILCA, unpublished data). This is a logical response for households that have considerable labour constraints (Mulugeta Assefa, 1990), as benefits accrued to individuals from implementing difficult tasks such as bush clearing in communal grazing areas are probably only a fraction of the cost. Although the Boran are semi-settled, they can be highly mobile when the need arises. Encampments usually change location once every five to eight years and this is commonly related to the search for better grazing (Donaldson, 1986; Cossins and Upton, 1987). Herd owners have reported that today, however, it is increasingly difficult to relocate to improved grazing sites because of the general increase in the numbers of both people and cattle (see Section 7.1.3: Review of dynamics and past interventions). They also think they are becoming more sedentary than in previous generations and this is reportedly due to the attraction of roads, markets and permanent water development that encourage settlement (Menwyelet Atsedu, 1990).
The role of livestock grazing and corralling in relation to nutrient redistribution. Heavy grazing could serve to concentrate a significant quantity of nutrients from surrounding foraging areas because cattle are corralled nightly at encampments, but this has not been quantified. Tonnes of manure are neatly piled into small hills at each Borana encampment and this often represents many years of accumulation (Donaldson, 1986). Elsewhere in pastoral Africa it is well known that corralling confines animals for about 50% of the time and consequently serves to concentrate large quantities of nutrients (Jahnke, 1982: p 35; Powell, 1986; Coppock et al, 1988). In addition to nutrient transport, frequent defoliation of perennial grasses during growing seasons could serve to deplete nutrients in the top soils of foraging areas by stimulating translocation (Botkin et al, 1981).
The role of woody encroachment in facilitating recovery of the herbaceous layer. If nutrient depletion in top soils is a factor in the decline of herbaceous productivity under heavy grazing on the Borana Plateau, there is other evidence that woody encroachment could have some beneficial effects on recovery of herbaceous vegetation by improving soil fertility. First, since woody seedlings can grow deep roots quickly (Gates and Brown, 1988; Solomon Kebede, 1989; Tamene Yigezu, 1990) they could establish and tolerate nutrient-depleted top soils to a higher degree than shallower-rooted grasses. Second, woody plants have been shown to have a potentially significant role in nutrient turnover and the replenishment of soil fertility in semi-arid savannahs. Billé and Corra (1986) made some initial calculations of nutrient budgets for sites on the central Borana Plateau and estimated that leaf litter from at least 200 woody plants/ha could provide roughly 57% and 21 % of the annual turnover for nitrogen and phosphorus, respectively.
Hatton and Smart (1984) described the effects of a 24-year exclusion of wild herbivores from a Ugandan savannah Under natural conditions, elephants prevented the persistence of woody populations through their feeding and clearing activities. Once elephants were excluded, however, an increase in acacias occurred and these extracted nutrients from the soil profile and deposited litter on the soil surface. Compared to unprotected sites, top soils of protected sites showed up to a five-fold increase in exchangeable cations, a 50% increase in nitrogen and up to a 30% increase in organic matter. Similarly, Radwanski and Wickens (1967) in the Sudan and Radwanski (1969) in northern Nigeria found Acacia albida and Azadirachta indica, respectively, to improve the fertility of top soils. Bosch and Van Wyk (1970), Charley and West (1975) and Weltzin and Coughenour (1990) also found woody plants to improve soils and/or micro-climate for herbaceous plants. Dreyfus and Dommergues (1981) highlighted the potential role of nitrogen fixation by native woody legumes in improving soil fertility when colonising new sites.
It is important to note, however, that whether interactions among woody and herbaceous plants are positive, neutral or negative depends on soil texture (Walker, 1985), soil fertility (P. N. de Leeuw, ILCA, personal communication), harshness of climate (Weltzin and Coughenour, 1990), density and maturity of the woody layer, and potential for species-specific competition based on compatibility of the seeds, juveniles and mature forms of the plants (Harper, 1977). Of particular importance for species-specific interactions are root morphology and distribution (Walker, 1985) and the degree to which woody crowns intercept light from the understorey. Whether plants are adaptable to drought and/or fire would also complicate interactions.
Examples of the diverse interactions among plants on the Borana Plateau include the apparently positive or neutral effects of A. tortilis Billé and Corra, nd) and A. drepanolobium (Tamene Yigezu, 1990) on the herbaceous layer as well as negative effects of A. horrida and A. seyal (Solomon Kebede, 1989). A major hypothesis to explaining these diverse interactions is variation in the root and crown morphologies of the woody plants. The most marked contrasts include those species with lateral roots, short boles and low spreading crowns (i.e. A. horrida) having the most negative effects and those with tap roots, tall boles and elevated open crowns (i.e. A. drepanolobium) having only minor effects.
Results of Menwyelet Atsedu (1990) on and off kalo (traditional reserves) sites were most notable in terms of the lack of significant effects that seven years of protection from continuous grazing had on soil chemistry, woody populations, herbaceous cover and species composition, or grass production. These findings are also striking because off-kalo sites occurred in the immediate vicinity of encampments where grazing pressure is expected to be high. Work also was conducted during the high-density phase of the cattle population in the interdrought cycle (see Section 7.2: A theory of local system dynamics). This suggests that these particular plant communities are resilient enough to cope with heavy use (as noted by Pratt, 1987a: pp 17-18). This could, in part, be to their being established on deep soils (Menwyelet Atsedu, 1990) and may not be widely generalisable. That forte production markedly increased off-kalo, however, suggests that the plant community did respond at least in a minor way to heavy grazing. Forbs have not been observed to be important dietary components for cattle in East African pastoral systems, although they may be more important small ruminants (Belete Dessalegn, 1985; Coppock et al 1987a). Forbs are low in fibre which is the major dietary constituent for energy for cattle (Kay et al, 1980). The increase in fortes may thus have no effect on cattle in the Borana system as long as grass production is not compromised. Food habits and nutritional studies would be required for an adequate assessment of the situation regarding small ruminants.
There are roughly 16 species of woody plant in the Borana Plateau that are thought to be encroachers (see Table B8, Annex B). Trials conducted by Tamene Yigezu (1990) on two species are thus limited in scope, but offer some potentially important insights. That heat treatment in the laboratory stimulated germination of A. brevispica seeds significantly more than other treatments serves as a caution for prescribed burning as a management tool for bush control in the upper semi-arid and subhumid zones. Field research involving heat treatment of A. brevispica should be designed to detect effects on seedling emergence. Heat scarification of seeds may well be an important survival strategy for woody plants in zones where plant communities have evolved under frequent fire disturbance.
Although limited in regional extent, livestock food habits observed on the Borana Plateau are typical of those recorded for the same livestock species elsewhere in East Africa in terms of relative emphasis on grazing or browsing (Migongo-Bake and Hansen, 1987; Coppock et al, 1986a). Our work documents the potentially high degree of dietary overlap between grazing cattle and sheep on a forage-class basis, while browsing goats and camels appear to have more distinctive diets. Although goats and camels share some preferences for certain woody species (Woodward, 1988), differences based on feeding height in the canopy are to be expected.
There were no studies of forage competition per se among livestock on the Borana Plateau that should increase in one livestock species leading to production constraints in another as a result of reduced intake of preferred forages. It is speculated, however, that the potential for competition exists between cattle and sheep in some situations. Cattle comprise over 90% of the livestock biomass on the central plateau and cattle productivity may be significantly compromised during average rainfall years when cattle density exceeds 25 head/km2, creating a situation for possible forage competition (see Section 7.2: A theory of local system dynamics).
Given such hypothesised intra-species pressure for cattle, inter-species pressure from cattle to sheep may also be likely during the high-density phases of cattle population. During other times disease may be the most pervasive constraint for sheep production. In contrast, disease may be the most chronic limiting factor for goats and camels given their low population densities and abundant browse resources (see Sections 5.3.7.1: Sheep and goats and 5.3.7.2: Camels). These perspectives are important given the apparent opportunities for increasing livestock species diversity as a development strategy among the Boran (Section 7.3.3.2: Camels, donkeys and small ruminants). While marketing opportunities may best justify promoting them (Section 4.4.3: Livestock supply to markets), sheep are the least complementary species to cattle in terms of food habits on a forage-class basis. By virtue of their food habits, the scope to increase goat and camel populations appears greater. Neither, however, are able to control bush encroachment according to Borana informants (see Section 7.3.1.4: Site reclamation).
Results from Woodward (1988) and Woodward and Coppock (1989) indicate an abundance and a variety of browse species although not all are suitable forage based on morphological features and/or variable concentrations and types of polyphenolic compounds. Woodward (1988) noted that species higher in polyphenolics tended to be more abundant in this upland study site. While this suggests that past browsing pressure may have led to the persistence of less palatable species, site-specific effects may also be a factor. Coley et al (1985) theorised that plants on nutrient-poor soils have "more to lose" from excessive herbivory, and may consequently invest more resources into concentrating defensive secondary compounds. Conversely, plants growing on more fertile sites may tolerate more herbivory because nutrients lost to herbivores may be more readily replaced. These plants may thus invest fewer scarce resources in secondary compounds. This suggests a hypothesis that browse species growing in bottomlands, may be more palatable than those growing in uplands (see Section 2.4.1.3: Soils).
Importantly, the mere presence of tannins does not indicate that forage has inferior feeding value; nutritional variation occurs with respect to types and concentrations of polyphenolics and the level of intake. Forages therefore need to be tested on a case-by-case basis using feeding trials. It is envisioned that small quantities of various types of tanniferous acacia leaves and dry dehiscent fruits could be useful as protein supplements for calves on grass hay diets (see Section 7.3.1.3: Forage improvements). This underscores the claim that, given the generally poor results from numerous establishment trials with exotic forages (Hodgson, 1990), more benefits could accrue from identifying promising indigenous grasses and trees and attempting to promote these more widely in the Borana system. Some valuable grasses are Pennisetum, Chrysopogon, Cenchrus, Chloris, Cynodon and Themeda spp to others (also see Section 2.4.1.5: Native vegetation). Valuable woody plants are A. brevispica, Euclea shimperi, A. tortilis, Dichrostachys cinera, Rhus natalensis, Pappea capensis, A. etbaica, Grewia spp, Ormocarpum mimosoides, Balanites spp, Cadaba spp and Caparis tomentosa among others. Those regarded as encroachers and with little apparent value for animals or households include A. drepanolobium, Albizia amare, A. horrida and A. mellifera. Strategies to make economic use of encroaching species are reviewed in Section 7.3.1.4: Site reclamation.
Not surprisingly, the Boran have an excellent knowledge of native vegetation, a common fact among rural people in Africa that has spurred interest in ethnobotany (Morgan, 1981; Stiles and Kassam, 1984; Marx and Wiegand, 1987; Mathias-Mundy and McCorkle, 1989). In this regard, plants used as traditional medicines for people or animals may have a role in developing more sustainable health practices in situations where imported drugs are expensive or unavailable (Tafesse Mesfin, 1990). Research to confirm stated properties of various plant compounds is thus required. Where range management goals focus on reduction or promotion of certain plant species, information should be solicited regarding household use. Forage value of plants should not be the sole criteria for management objectives. Government policy also impinges on the use of plants for cultural items and/or handicrafts. Blanket bans against taking out certain wooden handicrafts from the Borana Plateau to market in Addis Ababa are based on the premise that the tree species involved are endangered (Hodgson, 1990). While caution is commendable, policy makers need updated information on the abundance and trends of key tree species to guide policy. Pastoral households should not be deprived of needed ancillary income from handicraft sales if the species in question are abundant and increasing.
