Estimation of major livestock feed resources and feed balance in Moyale district of Boran Zone, Southern Ethiopia

The study was conducted to assess the major livestock feed resources and estimating annual feed production and feed balance in Moyale district of Boran Zone, Southern Ethiopia. A simple random sampling technique was employed to select the household’s (HHs) and 96 representative households were selected. Data were collected using group discussions, structured questionnaire, secondary data and personal observations. As it was identified in the study district, natural pasture, crop residues and agro-industrial by-product were major feed resources for the livestock. An average of 129,461.0156 tons of feed dry matter (DM) per year was produced in the district from the major available feed resources, and the demand for maintenance requirement of the livestock population in the district was 190,054.416 (tons DM/year) and this showed that a deficit off 60,593.4004 (31.88%) tons of DM per year in the district. Drought, feed shortage, water scarcity, disease and parasite, market and theft and predator were assessed to be the major livestock production constraints. Generally, the results from this study confirmed that the total dry matter produced from different feed resources in to the study area was not enough to satisfy the dry matter requirement of livestock to support the livestock production in to the study area, which suggest that the main focus needs to be refining the existing feed resources through restoration of tainted grazing areas, introduction compliant feedstuff production, improving feed utilization practices and introduce and promote the crop residue feed improvement.


INTRODUCTION
Feed is the most important input in livestock production and its adequate supply throughout the year is an essential prerequisite for any substantial and sustained expansion in livestock production (Samuel et al., 2008;Legesse et al., 2010). According to Sefa (2017) animal feeds including; natural pasture, fodder crops, fodder trees, crop residues and non-conventional feeds are used in different parts of Ethiopia. Green fodder (grazing) is *Corresponding author. E-mail: Hassanhajj870@gmail.com. Tel: +251921563942/+251947217941. Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License the major type of feed (54.59%) followed by crop residues (31.60%), hay (6.81%) and industrial byproducts (1.53%), (CSA, 2017). Alemayehu et al. (2017) reported that feed in terms of both quantity and quality is bottleneck to livestock production in Ethiopia. This problem of feed shortage is more aggravated during the dry season (Zewdie, 2010). Even during years of good rainy season, forage is not sufficient to feed livestock in the highlands (Melese et al., 2014). Data from different parts of the country also indicated that available feed satisfy about 78.2% of demand (Sefa, 2017).
To obtain improvement in animal production and productivity, an assessment should be done on the types and sources of livestock feed (Endale, 2015). Further studies that aim to integrate feeds that have better nutritive values into the feeding system are required to evaluate feed intake, digestibility, level of inclusion (supplementary feeds), animal's responses, and antinutritional factors for more efficient utilization of these indigenous and well adapted feed resources for sustainable animal production (Deribe, 2015).
The pastoral population occupies a large area of Ethiopia mainly the arid and semi-arid lowlands that are characterized by high spatial and temporal variability in rainfall distribution and pattern (Kula et al., 2016). In these areas, livestock production is dependent mainly on natural vegetation composed of herbaceous and woody species. There is a marked seasonal fluctuation in the availability and quality of the natural vegetation. Availability of such information is paramount importance in designing the development strategies, research plans and intervention options for both livestock production and natural resource management. Therefore, this study was initiated with the objective to assess the major available feed resources, estimating annual feed production and feed balance in Moyale district of Boran Zone, Southern Ethiopia.

Sampling procedures and data collection
This study was carried out in Moyale district of Borana Zone, Southern, Ethiopia. The district is located at 775 km south of Addis Ababa and has an area of 14,810 km 2 . The altitude of the district ranges from 1150 to 1350 m above sea level. Out of the total 16 kebeles of the district, 5 kebeles were selected purposively, based on feed resources availability and livestock production potential. Simple random sampling technique was employed to select the households (HHs). The total number of households sampled for the study was calculated based on the formulas given by Cochran (1977) , where, n, desired sample size; Z, abscissas of the normal curve (The acceptable likelihood of error of 10%): 1.96. The value of Z at 90% confidence interval; P, estimated proportion that one is trying to estimate in the population, that is, 50% ; q, 1-p; d, desired absolute precision level at 90% confidence interval. The primary data was collected through interview and it was supported by secondary data from associations, developmental workers, experts, local administrations, pastoralist and rural development office of the study district, and non-governmental organizations.

Methods of data analysis
The quantitative data collected from individual respondents were entered into the Microsoft Excel template and analyzed. Qualitative data were examined and summarized for each topic. Indexes were developed to provide the aggregated ranking of major feed resources in the study district [(5 × number of responses for 1 st rank + 4 × number of responses for 2 nd rank + 3 × number of responses for 3 rd rank + 2 × number of responses for 4 th +1× number of responses for 5 th )] divided by (4 × total responses for 1 st rank + 3 × total responses for 2 nd rank + 2 × total responses for 3 rd rank + 1 × total responses for 4 th rank) the higher the rank for a given reason, the greater its importance. All the collected data were arranged, organized and analyzed by using simple descriptive statistics such as mean, frequency and percentage by using SPSS version 20.

Estimation of feed balance
Feed demand in district was estimated by considering all ruminant and non-ruminant with exception of poultry livestock species reported by the district offices of agriculture and converting to tropical livestock units (TLU) using FAO (1987) methodology. The daily DM requirement for maintenance of one TLU is estimated to be 2.5% of the body weight (ILCA, 1991). Livestock to TLU conversion factors were used (Jahnke, 1982). The feed balance for district was estimated by subtracting the demand for maintenance requirement of the livestock population in the district (tons DM/ year) from the available feed DM (tons/ year) which was acquired from pastoral development office. The amount of feed DM attained annually from diverse land use types was calculated by multiplying the hectare of land under each land use types by its conversion factors (FAO, 1984). The quantities of available crop residues produced by farmers were estimated by converting crop yield to straw yield (FAO, 1987) and (De Leeuw and Tothill, 1990) and considering 10% wastage (Adugna and Said, 1994). The potential fodder yield of shrubs and trees are estimated by using an equation of Petmak (1983). Accordingly, leaf DM yield of fodder trees were predicted using the allometric equation of log W = 2.24log DT-1.50, where W = leaf yield in kilograms of dry weight and DT is trunk diameter (cm) at 130 cm height. Available concentrate feeds also considered.

Household characteristics
The household characteristics of the respondents are presented in Table 1. From the interviewed households 1.0, 93.8 and 5.2% were single, married and divorced, respectively. The present finding is little far apart with that reported by Bizelew et al. (2016) (Table 3).

Educational status and main source of income
Educational level and main source of income of respondents is presented in  (Mulugeta et al., 2015). The proportion of respondents capable of reading and writing was 19.8%; few of the respondents have attended elementary school (6.2%) and secondary school (9.4%) education. A similar result was reported for pastoral production system in Gambella regional state, southwestern Ethiopia (19.4, 26.1 and 7.8%), reading and writing, elementary school and secondary school, respectively, by Bizelew et al. (2016). The major sources of income for all respondents were livestock product sale, and both livestock and their product sale (79.2%) and (16.7%), respectively (Table  4). This finding were similar with a review of Kula et al. (2016) who stated that livestock rearing is the main means of livelihoods and source of income. Furthermore, the findings show that most of the respondents in the study area is pastoral and partially dependent on agronomic activities, the production of crop was too small and that is not exceed for house consumption.

Land holding
In Moyale district the mean total land holding per household was 1.4ha, out of which, the size of homestead, crop land and grazing land were 0.1, 0.9 and 0.4 ha, respectively (Table 3). There was no significant (P>0.05) difference in land holding among the studied kebeles. The total land holding of the study was comparable with total land holding of 1.34 ha and 1.3 ha respectively, reported by Bizelew et al. (2016) and Jimma et al. (2016) in Gambella regional state, southwestern Ethiopia and SNNPRS of Ethiopia, respectively. However, the present result was lower than the values of 1.8 ha reported by Gebreegziabher et al. (2016) in Humbo district of Wolaita zone, Southern Ethiopia and 3.23 ha reported by Kenenisa and Meles (2016) in Adami Tullu Jiddo Kombolcha district, Oromia regional state, Ethiopia. The present study indicated that the crop land was higher than other land holding types due to the grazing land and homestead area were occupied by people for settlement, there was illegal inhibition of land, locally called kaloo as informal reported by the respondents.
Out of the overall respondents, the majority (60.4%) reported that there was decrease in land holding size in the district while, 39.6% of the respondents reported no change. The reasons for the changes of land holdings were settlements (55.2%), tribal conflicts (2.1%), rainfall variability (2.1%) and illegal inhabitation (1.0%), ( Table  3). The results of the present study also showed that most of the respondents could access communal grazing lands.

Available feed resources in Moyale district
In the present study, three major different feed resources were identified and categorized into (1) natural pasture (grasses, browse and herbaceous legumes), (2) concentrates (wheat bran) and (3) crop residues. However, the contribution of agro-industrial by products such as noug seed cake, linseed cake, molasses, brewery by products, non-conventional feed resources like food refusal, vegetable refusal, sugarcane residue and improved forage were uncommon and rarely used. These findings were similar to the reports of Yadessa et al. (2016) in Meta-Robi district, West Shewa zone, of Oromiya regional state, Ethiopia.

Seasonal variations in the availability of feed resources
The respondents were asked to identify and rank the major feed resources in both dry and wet seasons (Table  4). Accordingly, fodder trees, hay, crop residues, concentrates and stubble grazing were ranked 1 st , 2 nd , 3 rd , 4 th , and 5 th , respectively, during the dry season. During the wet season, natural pasture (1 st ), fodder trees (2 nd ), hay (3 rd ), crop residues (4 th ) and stable grazing and concentrates (5 th ) were ranked by the respondents. The major feed resources identified in the present study were in agreement with the reports of Emana et al. (2017) in Abol and Lare district of Gambella region, Ethiopia; Dawit et al. (2013) in Adami Tullu Jiddo Kombolcha district, Oromia regional state, Ethiopia and Kassahun et al. (2015) from Horro and Guduru districts of Oromia regional state. In consistent with results of the present study, Alemayehu (2004) reported fodder trees and shrubs are important feeds in Ethiopia especially in the arid and semi-arid areas; and Sefa (2017) reported crop residues are the main feeds during the dry season. Moreover, Andualem (2016) stated that, most browse species have the advantage of maintaining their greenness throughout the dry season when grasses dry up and deteriorate in quality and quantity.

Dry matter production from different land types
According to Pastoral Development Office (2016)

Grazing land dry matter production
Rendering to data adapted from pastoral development office of the study area (2016), the total hector of communal and individual grazing lands were 24,559 and 8,284.52 ha, respectively. Besides the total amount of DM available in natural pasture was determined by multiplying the average value of grazing land holding with conversion factor of 2t DM/ha/year for communal grazing land and 3tDM/ha/year for individual grazing land (FAO, 1984). Amount of DM obtained from communal grazing land and individual grazing land was factored into total communal grazing areas of the district. Accordingly, 49,119.16 and 24,853.56 tDM/ha/year were produced from communal grazing land and individual grazing land, respectively.

Forest land dry matter production
Conferring to data obtained from pastoral development office of study area (2016), the total hector of forest land 1,502 ha. Moreover, the total amount of DM available in forest land was determined by multiplying the average value of forest land holding with conversion factor of 0.7 t DM/ha/year (FAO, 1984). Amount of DM obtained from forest land was factored into total forest land areas of the district. Accordingly, 1051.4 t DM/ha/year were produced from forest land.

Fallow land dry matter production
According to data obtained from pastoral development office of study area (2016), the total hector of fallow land was 1,340.88 ha. Moreover, the total amount of DM available in fallow land was determined by multiplying the average value of fallow land holding with conversion factor of 1.8 tDM/ha/year (FAO, 1984). Amount of DM obtained from forest land is factored into total forest land areas of the district. Accordingly, 2413.584 t DM/ha/year were produced from fallow land.

Crop residues dry matter production
The crop and natural resource office of Moyale district (2016) report had demonstrated that 19,726 ha of land are covered by cropping land. The agro pastoral communities in study area currently have been produced crop residues from maize, teff and haricot bean. The total area of different crop types grown is 18,109, 1,154.8 and 461.92 ha, for maize, haricot bean and teff, respectively. The crop residues (38,296.64 tons) are the second dominant feed resource in Moyale district next to open grazing land with supply of 49119.16 tons of DM from crop residues to feed livestock. Crop residues dry matter productions are shown in Figure 2.

Crop aftermath dry matter production
Rendering to crop and natural resource office of ] divided by (4 × total responses for 1 st rank + 3 × total responses for 2 nd rank + 2 × total responses for 3 rd rank + 1 × total responses for 4 th rank) the higher the rank for a given reason, the greater its importance. Moyale district (2016) report had demonstrated that 19,726 ha of land covered by the cropping land. The quantities of available DM in crop aftermath grazing were determined by multiplying the available land by the conversion factors of 0.5 for grazing aftermath (FAO, 1987). Accordingly, 9,863 tons DM/ha/year was produced from crop aftermath.

Contribution of concentrate feed (wheat bran)
The quantity (DM basis) of concentrates (wheat bran) available for each household was obtained by interviewing the farmers during questionnaires survey. Accordingly, 1.3216 tons of DM was obtained from wheat bran in the study area; the contribution of wheat bran was very little as compared to other feed resource.

Trees and shrubs dry matter production
The potential fodder yield of shrubs and trees were estimated by measuring stem diameter using measuring tape and applying the equation of Petmak (1983). In quantifying tree feed resources from common property resources (   level similar approaches, as communal grazing area mentioned earlier, was used. Empirical evidence from WBISPP (2004) suggests that only about 75% of all available DM is accessible by livestock for use and therefore this study area was used the same accessibility factor to quantify total DM utilized by livestock from grazing and browsing areas. Accordingly the total DM production of shrubs and trees in the study was 3,862.35 tons obtained from communal and forest land.

Total tropical livestock unit (TLU) and their dry matter requirement
Based on the reported data of pastoral development office of livestock (2016) Figure 3.

Feed balance
The open grazing land, private grazing land, fallow land  ] divided by (4 × total responses for 1 st rank + 3 × total responses for 2 nd rank + 2 × total responses for 3 rd rank + 1 × total responses for 4 th rank) the higher the rank for a given reason, the greater its importance. and forest land, concentrate feed (wheat bran), indigenous browse, shrubs and crop residues were used to calculate feed supply for livestock in the study area. The total of 81,300.054 tons of DM per year was produced from different land use types with exception of crop land; concentrate feed (wheat bran) and aftermath grazing which produce 38,296.64, 1.3216 and 9,863 tons of DM per year, respectively.
As it had been calculated the total DM produced in the study area from different feed resources was 129,461.0156 tons and the demand for maintenance requirement of the livestock population in the district was 190,054.416 (tons DM/ year). The feed balance for the district was estimated by subtracting the demand for maintenance requirement of the livestock population in the district (tons DM/ year) from the available feed DM (tons/ year), and this showed that a deficit off 60,593.4004 (31.88%) tons of DM per year in the district. In general, the feed balance data showed that the DM produced in the study area per year was imbalanced with the minimum maintenance requirements of dominant livestock species. Similarly, in previous studies, challenges in Ethiopia showed that the dry season is characterized by inadequacy of grazing resources, because of which animals are not able to meet even their maintenance requirements and lose of substantial amount of their weight (Aster et al., 2012). This further recalls that there is need to introduce the feed improvement interventions in the study area in order to save the livestock.

Major constraints to livestock production
Livestock production in the study area has been primarily hampered by drought (mean rank 0.28), feed shortage (0.22), water scarcity (0.17), disease and parasite (0.16), market and theft (0.10) and predator (0.07) (Table 5). Furthermore, drought was one of the main constraints which lead the farmers to travel a long distance in search of feed. In agreement to the results of the present study, Hidosa and Tesfaye (2018) reported that climate change is one of the non-technical livestock feed production constraints, which has been affected livestock production through induce decline in pasture availability. Furthermore, Bizelew et al. (2016), Gebreegziabher et al. (2016), Kenenisa and Melese (2016) and Amistu et al. (2017) reported that feed shortage is one of the major constrains to livestock production in Ethiopia, which is in support of the results of the present study.

CONCLUSION AND RECOMMENDATIONS
The natural pasture, crop residues and agro-industrial byproduct are major feed resources in the study area. Drought is one of the main constraints which lead the farmers to travel along distance for feeding livestock. Moreover, as it had been calculated the total DM produced in the study area from different feed resources was 129,461.0156 tons and the demand for maintenance requirement of the livestock population in the district was 190,054.416 (tons DM/ year). The annual feed DM production in the district could not satisfy 31.88% of the DM requirement of livestock kept in the area. The study described that the contribution of the open grazing area is declining from time to time and livestock may not fulfill the DM requirements. Therefore, this calls for interventions that improve the productivity of declining grazing areas such as rehabilitations of retreated grazing area through the introduction and promoting area closures, over sown with locally adaptable legume forages and fertilization with livestock dungs and droppings. For more efficient utilization of the available feed resources such as crop residues feed treatments should be introduced and practiced.