CHARACTERIZATION OF RANGELAND RESOURCES AND DYNAMICS OF THE PASTORAL PRODUCTION SYSTEMS IN THE SOMALI REGION OF EASTERN ETHIOPIA

Abstract

additional index words: Biodiversity, degradation gradient, drought simulation feeding, rangeland condition, grazing behaviour, grazing/browsing capacity, key fodder species, soil characteristics, soil degradation, pastoral perceptions he study was conducted in the Somali region of eastern Ethiopia, with an arid to semirid climate. The study aimed at the characterization of the rangeland resources, assessing ne current condition of the rangeland, understanding pastoral perceptions on rangeland degradation and developing drought feeding strategies for livestock. hree experimental sites, representative of the three predominant vegetation types of eastern Ethiopia were selected. They were the arid Asbuli grassland (used as grazing area for large and small ruminants), the arid Aydora open savanna (bush-grassland: entirely used for grazing/browsing by all types of livestock), and the semi-arid Hurso closed savanna (bushland: selected for its importance as browsing for camels and goatsV degradation gradient was identified in each of the three vegetation types, where the botanical composition was surveyed and the rangeland condition assessed. The perceptions of the pastoralist on rangeland degradation were also quantified. The dry matter production of both the herbaceous and woody layer was determined, while the grazing and browsing capacity calculated. Key forage species were identified and the grazing pattern of various livestock species along the degradation gradient studied. The oil seed bank regeneration potential was assessed in a greenhouse experiment and the response of cattle, sheep and goats to a simulated drought, in terms of reduced fodder, was conducted under controlled conditions. he results of the study confirmed the existence of severe rangeland degradation that occurred since 1944 and which was aggravated after the 1974 drought. This contributed to an increase in the number of poor households. The average livestock holding per household declined from 809 Tropical Livestock Units (TLU) before 1974 to 483 TLU after 1974. Livestock holding shifted from a predominance of cattle to small ruminants, which are able to utilize the degraded rangeland more effectively. Camels are now the most important livestock species in terms of milk and meat production, mainly due to their ability to tolerate drought. The abundance of herbaceous plants, basal cover, dry matter production and grazing capacity was found to be higher in the benchmark sites compared to the other rangeland conditions. There w?as also a corresponding increase in percentage bare ground, soil compaction and soil erosion along the degradation gradients. Over-grazing and overiti lization through continuous grazing of the herbaceous layer were identified as the main causes of these differences. Rangeland condition was observed to significantly influence the grazing behaviour of livestock in terms of plant species selection, grazing intensity and intake per animal. When forage sources were adequately available, animals selected fever plant species. As forage resources declined the animals spend more time grazing and more species selected, including less palatable species. The number of bites, intake per bite and intake as a percentage of the animal's body mass also increased as the rangeland become more degraded. Acacia nubica and A. mellifera were identified as aggressive encroaching spccies in the Aydora open savanna and Hurso close savanna. The Aydora open savanna experienced extensive encroachment by woody plants with increasing plant densities across the degradation gradient. The Hurso closed savanna experienced an opposite trend where severe deforestation and a loss of valuable browse species occurred, mainly as a results of over cutting of the woody plants for firewood, charcoal making, construction and the clearing of the land for planted crops. The study on the soil seed bank of soil collected along the various degradation gradients showed a high abundance of plant seed present in the soil, confirming the potential of the area for rangeland restoration. As expected the body weight losses of all livestock species in the controlled feeding trial ere highly correlated with the reduction in daily dry matter feed. More than 50% of the animals showed pronounced emaciation and physical weakness and 25% of the cattle and goats collapsed and died within ten weeks after the trial started. This explained the large scale mortalities of livestock during prolonged droughts. Drought must be accepted as part of the pastoral life and there should be an adequate arly warning system regarding livestock feed availability and strategies of appropriate mitigation strategies. More realistic stocking rates is the obvious solution to the avoidance of stock losses during droughts, but in view- of the well established culture of the pastoralists it is highly doubtful if they will be willing to reduce their animal numbers. In conclusion, the experimental results indicated the existence of genetic variability among the various Somali livestock breeds regarding the tolerance to feed shortages and in rates of compensatory growth. This demonstrates the opportunity for improving the genetic composition of the Somali herds through selection