Availability of Fodder Trees and Shrubs Integrated Into Agricultural Systems in Eastern Hararghe, Ethiopia

Tree/shrub based feed resources and feeding systems in Eastern Hararghe are not studied well. Therefore, a study aimed at assessing the availability of fodder trees and shrubs integrated into farming system, available feed resources, and feeding systems was made in two districts of eastern Hararghe zone, Ethiopia. A total of 268 respondents from the two districts, both from lowland and highland agro ecology, were interviewed. Sørensen’s Similarity Index was used to assess species composition in relation to agroecology. The result revealed that about 67.2% of the respondents have not integrated any fodder trees and shrubs into their farmlands whereas only 32.8% of the respondents integrated fodder trees and shrubs into their farmlands. A total of 20 fodder and non-fodder tree species were identified. Regarding tree species composition, only 46% of tree species were found in both districts, whereas about 54% of the tree species were dissimilar between the districts. The feeding system significantly varied with agroecology. Zero grazing system with stall-feeding technique is common in the highland agroecology. However, in the lowland agroecology free grazing on communal grazing area and feeding crop residue are common. Factors like inadequate extension service, lack of planting material and land scarcity has resulted in reduction of farmers’ interest to integrate fodder trees and shrubs into their farming system.


Introduction
In Ethiopia, about 85% of the total population directly depend on agriculture and live inrural areas (Krause et al., 2007).The livestock sub-sector plays an important role in the Ethiopian economy.Livestock contributes to the livelihoods of 60-70% of the population, and beside this, the sector serves as source of food, income and employment, manure , and also the subsector serves as productive assets that allow households to be self reliant (Zinash & Tegegne, 2000;Ayele et al., 2002).Despite having good number of livestock, per unit productivity in Ethiopia is quite low (Shenkute et al., 2012).The current low productivity status of livestock in Ethiopia is caused by lack of forage.As grazing land becomes scarce due to expansion of agricultural land attributed to population growth and related factors, it aggravates the poor nutrition of livestock.In order to improve the productive and reproductive capacity of smallholder ruminant animals, there is a need to look at ways of extending the availability and quality of feedstuffs produced on smallholder farms.One potential way for increasing the quality and availability of feeds for smallholder ruminant animals in the dry season may be through the use of fodder trees and shrub.Fodder trees and shrubs have always played a role in feeding livestock (Azim et al., 2011).In Ethiopia, trees and shrubs have long been considered as source of feed for livestock in different parts of the country.Fodder trees and shrubs are an effective insurance against seasonal feed shortages or risk of drought, and hence vulnerability of farmers' livelihood to climate changes.They provide green feed when grasses and other herbaceous materials are dry and they provide the only source of energy and protein during drought when all other feed is absent (Lefroy et al., 1992).Trees and shrubs are important components of animal feeding and suppliers of protein especially in harsh environmental conditions when grazing is not

Sampling Procedure and Data Collection Method
The intention to conduct the survey at these districts was to find out the availability of fodder trees and shrubs integrated in to the farming systems at two districts which have different agroecological characteristics i.e.Babile falls in the lowland agroecology while Kersa represents mid and highland agroecology.
A multi stage sampling technique was applied to select sample households.In the first stage, one district from high land and one district from lowland areas was selected purposively that can represent the familiarity of the districts for integration/managing of fodder trees in their farmland .In the second stage, one peasant association (PA) was selected from each district.In the third stage, households' were selected using simple random sampling techniques.The number of respondents was determined based on the sample size formula developed by Yamane (1967) as follows, where, n = is the sample size for the research use (134); N is the population size (total number of households which is 202 from each PA); and e is the level of precision (0.05).A total of 268 respondents with 134 household heads each from Kerssa and Babile districts, respectively, were randomly selected and interviewed independently.
Data were collected from sample respondents, personal observations of physical features, and informal discussions with Development Agents (DAs).Both structured and semi-structured questionnaire was prepared for individual households.Before conducting the survey, the questionnaire was pre-tested by interviewing some households from other sites having similar agroecologies to that of the study areas.The necessary adjustment to the questionnaires was then made before fully duplicating and distributing to the enumerators.Regular monitoring was made by the researchers while enumerators were interviewing the respondents and daily evaluation of the filled questionnaires was undertaken throughout the data collection processes.The major contents of the questionnaire focused on the issues: the availability of fodder trees and shrubs integrated into agricultural system, factors affecting integration of fodder trees, feed resources and feeding systems.
Tree inventory was carried out to record all woody species found in the traditional agroforestry practices.The sample households that were selected for household survey were employed as a sample plot for inventory.
Farmers were asked about name of the trees inlocal language (Oromifa) and the use of the plant species.Identification of species was done in the field.Nomenclature of the plant species followed the Flora of Ethiopia (Hedberg & Edwards, 1995).

Data Analysis
To analyze the quantitative and qualitative data, statistical package for social sciences (SPSS 16 version) was used.Descriptive statistical methods such as frequency, percentage, mean, and standard deviation were used.For categorical variables, a chi-square test was used to test the difference between the agroecologiesat the significance level (0.05).The trees species composition between the different agroecologies was computed using the Sørensen's Similarity Index: IS S = 2c/a + b + 2c, where, IS S is Sørensen's Similarity Index; a representing the number of species unique to sample a; b representing the number of species unique to sample b; and c representing the number of species in common.

Socio-economic Characteristics of the Respondents
Interviewed respondents comprised 75.4% male and 24.6% female with age ranging from 20 to 72 years and an average age of 36.05±11.8years.The average family size of the respondents was 6.01±2.45;and the average land holding size was 1.34±1.08ha (Table 1).Note. *** superscript indicates significant difference between studied parameters of respondents.

Availability of Fodder Trees and Shrubs Integrated Into Farmlands
The result of the respondent's response for integration fodder trees and shrubs into their farmlands were not varied with agroecology(X 2 = 1.63; df = 1; p = 0.21) (Table 2).The result of household survey revealed that, about 67.2% of the respondents have not integrated any fodder trees and shrubs into their farmlands whereas only 32.8% of the respondents integrated fodder trees and shrubs into their farmlands.
In general, a total of twenty tree and shrub species were identified in the study area out of which seven species were primarily promoted for construction material, nine for fodder and four for fruit (Table 2).Note.* = common species in Kerssa and Babile districts; X = species found at one district only.
Fodder or browse production from trees is one of the benefits of agroforestry.Fodder trees and shrubs become then important as a source of energy and protein to keep the animals' body healthy, improve growth rates and even increase milk and wool production (Azim et al., 2011).Hence, it is clear that farmer's involvement in establishment of fodder trees and shrubs in the development of new agroforestry technologies is important, right from the beginning.
However, in the study area most of the farmers (i.e., about 67.2% of the respondents) have not integrated any fodder trees and shrubs into their farmlands.Roothaert, (2000) noted that farmer's decisions to allow trees to grow on the farm are influenced by such importance or personal preferences.In the study area factors such as inadequate extension services, lack of planting material and land scarcity respectively (Figure 2) are the reasons that were identified by the respondents for not integrating fodder trees and shrubs into their farmlands.Although farmers seem to have a clear idea of the feeding value of fodder trees, and interested to grow them on-farm, sen's on in ted to 4).In the study area among the identified species as fodder trees only Carissa edulis was found in common in both districts.Species used as dry season fodder like Acacia sieberiana, Ricinus communis and Sesbania sesban were identified only in Kerssa district.While, Acacia brevispica, Gardena ternifolia, Grewia bicolour, Terminalia brownie and Ziziphusspina-christiwere found only in Babile district.Most of the identified fodder trees in the present study area were indigenous species, which are most suited to the local climatic and edaphic condition of the area.However, exotic species like Sesbania sesban also identified in the highland agroecology conditions.Sesbania sesban is an important multipurpose fodder shrub exotic to Ethiopia and is originally from east Africa (Mekoya, 2008).Some of the fodder species identified in the study area (Table 2) are similar to those described by many authors in Ethiopia (Shenkute et al., 2012;Beyene, 2009;Teferi, 2006;Abule, 2003).Shenkute et al. (2012) documented Acacia tortilis, Ficus gnaphalocarpa, Balanites aegyptica, Dichrostachys cinerea and Grewiabicolour as most widely utilized browse species in Mid Rift Valley of Ethiopia.Beyene (2009) in his study in south western Ethiopia indicated Rhusna talensis, Grewia ferruginea, Bauhinia farea, Deinbolliakilimandscharicaand Acacia seyal as the common browse in the Gembella National Regional State, Ethiopia.Acacia species andGrewia mollis were also reported as the most commonly utilized species in the Rift Valley of Ethiopia (Abule, 2003).Teferi (2006) also documented Ziziphus spina-christi, Acacia asak, Acacia lahai, Balanites aegyptiaca andTerminalia brownie as some of the most distributed browses species in northern Ethiopia.

Feed Resource and Feeding Systems
Grass and crop residues are the main sources of feed in both districts.However, the feeding system are significantly different with agroecology in the study area (X 2 = 98.8; df = 2; p = 000).In Kerssa district the main feeding system is stall-feeding.The major feed resources are sorghum and maize stover, straw, maize and sorghum leaves, thinned maize and sorghum seedlings and sterile plants, sweet potato leaves, haricot bean leaves and weeds grown in crop fields.Cattle, sheep and goats were observed tied with rope near crop fields or Khat fields.However, in Babile district livestock feeding is based on free grazing on communal grazing lands, road sides and crop residue (maize and sorghum stover, groundnut leaf and stem).The most common form of management in the area is mixture of grazing (communal land, crop land) and stall-feeding.Particularly in the khat based agroforestry system of Kerssa district, zero grazing system with stall-feeding technique is the most common practice.This agrees withthe idea that grazing land in eastern Hararghe is very little and usually fewer animals are found (Gebregziabher & Gebrehiwot, 2011).More intensive utilization of tree forages and shrubs is dependent on the type of feeding systems practiced.Table 3. Farmers' response on integration of fodder trees and shrubs and feeding systems in two districts in eastern Hararghe, Ethiopia But an important problem diagnosed in the study area specifically at Kerssa is lack of fodder during the dry season (Figure 3).as fodder by few farmers need further investigation in terms of chemical and tannin analysis, palatability and digestibility in order to enhance their utilization in the future.
** superscript indicates significant difference between studied parameters of respondents.

Table 1 .
The socio-economic profile of sampled household (n = 268) in Kersa and Babile districts, eastern Ethiopia

Table 2 .
Frequency of tree species identified in Kerssa and Babile districts, eastern Ethiopia