The Effect of Acacia tortilis and Balanites aegyptica Trees on Fodder Quality: The Case of Mieso District, West Hararghe, Oromia National Regional State, Ethiopia

Acacia tortilis and Balanites aegyptica trees are a multipurpose tree species that produces diverse socioeconomic and ecological benefits. These trees are grown commonly on farm land in Mieso district but scientific information is not yet documented about their effect on fodder quality. Thus, this study was initiated to investigate the effect of these two tree species on leaf foliar macronutrients and proximate chemical concentrations for animal feed in Mieso District, Oromia, Ethiopia. Representative foliar samples from A. tortilis and B. aegyptica trees replicated four times were collected. From each sampled tree, fully matured and expanded green leaves in all compass directions from different crown positions were collected and evenly mixed to form 1 Kg of fresh leaves per sampled tree. The collected leaf samples were air dried. After air drying, the leaf samples were oven dried at 80°C till constant weight was achieved. Moreover, the leaf Dry Matter (DM), Crude Protein (CP), Crude Fibre (CF), Ether Extract (EE) and ash of the samples were determined. With the regard to leaf macronutrient, only leaf foliar nitrogen concentration was revealed significant variation between the two species whereas among the leaf proximate chemical composition %DM, %CP, %DCP and %CF were showed significant variation (<0.05) between the two species. Thus, A. tortilis and B. aegyptica trees have the potential to provide quality animal feed that contribute to the productivity of the livestock through the provision of their nutritive value of the fodder, and thus, retaining of this tree on farm land with proper management like pruning, pollarding, lopping and crown opening could able to contribute to the livelihood of the community through improving the productivity of livestock.


Introduction
Agroforestry is recognized as a land use option in which trees provide both products and environmental services. Akinbile et al. [1] described agroforestry as a deliberate integration of woody perennials with agricultural crops and/or animals on the same land management unit, with the aim of increasing income through the use of economic trees. It is alternative land management system which has immense potential to solve global challenges like deforestation, unsustainable cropping practices, loss of biodiversity, increased risk of climate change, rising hunger, poverty and malnutrition [2]. A scattered tree on crop land has a positive influence on maintaining soil fertility via addition of nutrient to soil through biological nitrogen fixation and efficient nutrient cycling [3][4][5]. As compared to treeless area, the microclimatic variables such as air temperature, soil temperature and soil moisture beneath tree canopy modified due to the influence of trees on radiation flux, air temperature and wind speed [6].
Agroforestry involves managing interactions between tree and non-tree components to produce diversified sustainable production system. The magnitude of positive and/or negative influence of trees on crop yield and /or livestock depends on management variables, canopy and root architecture, spatial and temporal arrangement, age and size of the tree and ecological type [5]. Therefore, understanding of the biophysical interactions of trees and livestock different ecological settings is vital to properly manage farm trees [7]. In this context, scattered A. tortilis and B. aegyptica trees on farm field are common in Mieso District, and thus, information on their interaction with animal via provision of quality feed need to be assessed so as to manage the system properly and enhance its productivity. In this regard, scientific information about the effect of A. tortilis and B. aegyptica trees on fodder qualities was not so far documented. Thus, this study was initiated to reveal the effects of A. tortilis and B. aegyptica trees leaves nutritional quality as animal feed.

Altitude and Climate
The district's altitude ranges between 900-1700m.a.s.l., and agro-ecologically classified as lowland with mean annual temperature of 24°C -28°C and mean annual rainfall of 400 to 900mm, with an average of about 790 mm 8 . The area receives a bimodal rainfall where the small rains are between March and April while the main rains are between July and September.

Agriculture
The district has two types of farming systems namely; pastoral and mixed (crop/livestock) farming systems [8]. The Sorghum (bicolor), Maize (Zea maize), Haricot bean (Pisum vulgares, Chat (Catha edulis) are grown in the area in which sorghum covers the majority of the cultivated lands. Also cattle, camel and goat are the dominant livestock in this district. Fattening of large and small ruminants are the main sources of households' livelihood. On the other hand, recurrent drought is common problem of the area that hinders the productivity of crop and livestock, and relief aid is a regular source of livelihood for many rural families.

Descriptions of the Studied Trees
Two Peasant associations were purposefully selected based on extensive distribution of A. tortilis and B. aegyptica trees on agricultural field. Subsequently, thirty individuals of A. tortilis and B. aegyptica trees having approximately similar size and from uniform site condition were marked and numbered from one up to thirty. Of them, four individuals' A. tortilis and B. aegyptica trees were selected randomly for this study.

Leaf Sampling and Chemical Analysis
Representative foliar samples from A. tortilis and B. aegyptica trees replicated four times in were collected. From each tree, fully mature and expanded green leaves in all compass directions and from different crown positions were collected and evenly mixed to form 1 Kg of fresh leaves per sampled trees. The collected leaf samples were air dried, and after air drying, the leaf samples were oven dried at 80°C till constant weight was achieved. Sampled leaves were ground and analyzed for N, P and K contents following standard methods described by Anderson and Ingram [9].
Moreover, samples of leaf Dry Matter (DM), Crude Protein (CP), Crude Fibre (CF), Ether Extract (EE) and ash were determined according to AOAC [10]. The DM was determined by drying the samples at 80°C till constant weight was achieved. N was determined by micro Kjeldhal method. The %N was used for the estimation of CP. The EE in a sample was determined by extracting with diethyl ether at 60°C. For CF, a sample was reflexed first with 1.25% H 2 SO 4 and subsequently with 1.25% NaOH for 30 minutes each to dissolve acid and alkali soluble component present in it.
The residue containing CF was dried to a constant weight and the dried residue was ignited in a muffle furnace, then loss of weight on ignition was calculated to express it as CF content. Ash sample was ignited in muffle furnace at 550°C to burn all the OM and the leftover material was weighed as ash. Nitrogen free extract (NFE) was determined on dry matter basis as: Besides, the procedure used by El-Beheiry (2009) was employed for calculation of digestible crude protein (DCP) on DM basis as: (2)

Statistical Analysis
Statistical differences were tested using analysis of variance (ANOVA) following the general linear model (GLM) procedure of SAS Version 9.0 at 5% significant level. Tukey's honest significance difference (HSD) test was used for mean separation for those their analysis of variance showed statistically significant differences (p < 0.05).

Leaf Macronutrient Concentrations
The leaf foliar macronutrient concentrations of N, P and K of the two species are depicted in Figure 2 below. In the current study, the foliar N concentration of A. tortilis and B. aegyptica trees were found to be significantly (P<0.0001).
Bars indicated by letters are statistically significantly different at (p <0.05).

Figure 2. Mean Values of Leaf Macronutrient as influenced by different tree species.
The analysis of variance for the foliar macronutrient concentration revealed that the macronutrients elements that form important constituent of plant tissues, foliar concentration (N, P and K) differed between A. tortilis and B. aegyptica tree species. Nitrogen, as measure of protein quality and as essential source of feed for microbial growth in the rumen, was significantly affected by A. tortilis and B. aegyptica tree species. The average values of nitrogen foliar concentration for A. tortilis and B. aegyptica trees were 37.30mg/g & 28.88mg/g respectively. Which means, the leaf foliar nitrogen concentration of A. tortilis was significantly higher than the average value for B. aegyptica trees.
It had been reported by Evans [11] that some tree species have nutrient rich leaves, which readily breakdown and release the nutrients to the organisms body or add to soil fertility. The mean foliar N content of the A. tortilis (37.30 mg g -1 ) and B. aegyptica (28.88mg/g) in the current study are above the required level for soil fertility improvement. Palm et al. [12] suggested that organic material with N contents above 25 mg g -1 would be considered good quality and suitable as replacement of mineral fertilizer for growth of fodder trees. Net immobilization of N can be expected if the species with N concentration below the critical levels (25 mg g -1 ) are used as organic fertilizer resources [13]. The average leaf N concentration for A. tortilis and B. aegyptica in this study were higher than the average leaf N (21.83 mg g -1 ) concentration of F. thonningii reported by Daniel et al [14] from Tigray and Enideg [15] from Gondar. Furthermore, it was also higher than the foliar N content of C. africana and C. macrostachyus (12.80 and 12.90 mg g -1 , respectively) found by Jiregna et al. [16]. However, the foliar N content of the present study for B. aegyptica was lower than those reported by Kindu et al. [13] for B. polystachya (36.66 mg g -1 ), C. palmensis (36.50mg g -1 ), H. abyssinica (30.07 mg g -1 ) and S. gigas (34.20 1mg g -1 ). The variation in N concentration among the different species could be ascribed to generic differences of the species, climate, geographical location/agroecology, and time of day, aspect of the tree, tree age, and position of leaves in the crown, age of foliage, internal nutrient balance and effects of diseases [11].
With regard to phosphorus, even though it was not significant, the foliar P concentration was slightly affected by A. tortilis and B. aegyptica tree species. The average values of phosphorus foliar concentration for A. tortilis and B. aegyptica trees were 2.99mg/g & 2.81mg/g respectively. Even though there was slight variation, the leaf foliar P concentration of A. tortilis was significantly higher than the average value for B. aegyptica tree. The foliar P content of the present study is within the recommended normal requirement range (1.2 -4.8 09 mg g -1 ) of ruminants for normal growth and reproduction [17]. Moreover, the P concentration also met the requirement of P deemed from a quality green biomass 2.5 mg g -1 as replacement of inorganic fertilizer suggested by Palm et al. (2001). The average leaf P concentration (2.99mg/g and 2.81mg/g) for A. tortilis and B. aegyptica respectively of in the present study were higher than the mean leaf P concentration (2.05 mg g -1 ) reported from Gondar by Enideg [15]. The foliar P concentration was also higher than that reported for species of C. africana and C. macrostachyus (1.90 and 2.40 mg g -1 , respectively) obtained by Jiregna et al. (2005). However, the foliar P concentration found in this study was lower than those reported for B. polystachya (4.71 mg g -1 ), D. torrida (3.76 mg g -1 ), H. abyssinica (3.71 mg g -1 ) and S. gigas (4.75 mg g -1 ) from Ethiopian highlands [13]. It was also found to be extremely low in the leaves of fodder trees studied, compared to the critical level (7.7 g P/kg DM) suggested by NRC [17].
Even though it was not significant, the foliar K concentration was slightly affected by A. tortilis and B. aegyptica tree species. The average values of foliar K concentration for A. tortilis and B. aegyptica trees were 45.25mg/g & 45.75mg/g respectively. Even though there was slight variation, the leaf foliar Potassium concentration of B. aegyptica was not significantly higher than the average value for A. tortilis tree. The mean foliar K concentration (39.95 mg g -1 ) in the present study was found to be higher than the mean foliar K concentration (9.10 mg g -1 ) reported by Enideg [15] from Gondar. It was also higher than the foliar K content of C. africana and C. Macrostachyus (11.5 and 8.70 mg g -1 , respectively) than those found by Jiregna et al. [16], and higher than those reported by Kindu et al. [13] for B. polystachya (21.55 mg g -1 ), C. palmensis (14.93 mg g -1 ), D. torrid (27.00 mg g -1 ), and H. abyssinica (21.22 mg g -1 ), whereas found to be lower than the mean foliar K concentration of S. gigas (55.50 mg g -1 ) reported by the same authors. With regard to Potassium (K) is an essential element to the growth and developments of an animal according to NRC [17]. Deficiencies in K in cattle can result in reduced intake, weight loss and stiff joints. Cattle stressed due to long transport distances may necessitate increased levels of K to replenish lost body reserve (Ibid).

Leaf Proximate Chemical Compositions
Results of the analysis of variance for proximate chemical compositions of leaf dry matter (DM), crude protein (CP), digestible crude protein (DCP), ash, ether extract (EE), crude fibre (CF) and nitrogen free extract (NFE) of A. tortilis and B. aegyptica trees are presented in Figure 3.  respectively. In this study, the leaf dry matter content of A. tortilis was statistically significantly lower than the average value for leaf dry matter content of B. aegyptica trees. Reported values of Dry matter for those current experimental tree species and other tree species are differed. For instance, lower values (21.63 and 40.61 %) of F. thonningii than the present study were reported by Ogunbosoye and Babayemi [18] and Tegbe et al. [19] respectively. Likewise Ojokuku et al. [20] reported lower value of DM (19.8%) for F. capensis while Nkafamiya et al. [21] found lower values (19.01%, 14.12%) for F. asperifolia and F. sycomorus, respectively and Daniel et al. [14] for Ficus polita (30.40%). On the contrary, Azim et al. [22] found higher value (50.5%) for F. religiosa.
Crude Protein (% CP) was found to be significantly varied between the species studied. Crude protein is an important requirement to support optimum microbial growth of rumen. The leaf proximate chemical composition %CP of A. tortilis and B. aegyptica trees showed significant differences (P<0.0001) (Figure 3). The average values of %CP for A. tortilis and B. aegyptica trees were 23.3% & 18.05% respectively. The average value of crude protein for A. tortilis statistically significantly higher than the average value of % crude protein of B. aegyptica tree. The mean CP of the present study by far exceeds the minimum protein requirements of ruminants 7%, below which the feed intake might be depressed as it is not sufficient to meet the needs of the rumen bacteria [17]. Comparing the CP of the current study (23.3% and 18.05%) for A. tortilis and B. aegyptica trees respectively were higher than other research result reported for F. thonninigii [23]. Other results on Ficus species also revealed substantial difference. While higher (20.27%) and comparable (17.24%) values of crude protein were reported for species of F. asperifolia and F. sycomorus respectively [20,21]. However, Azim et al. [23] and Daniel and Agena [24] found lower (3.63%, 11.7% and 15.70%, respectively) CP value for F. capensis, F. religiosa and F. polita. The substantial variation between the different browse tree species could be genetic whereas the variation within species could be ascribed to climate or season, geographical location or agroecology, time of day, aspect of the tree, tree age, and position of leaves in the crown, age of foliage, internal nutrient balance, and effects of diseases, soil type and age.
Digestible crude protein (% DCP): The leaf digestible crude protein (% DCP) of A. tortilis and B. aegyptica trees showed significant difference (P<0.0001) with the average values of 18.14% & 13.25% respectively (Figure 3). In this case, the average value of percent digestible crude protein for A. tortilis was statistically significantly higher than the average value of percent digestible crude protein of B. aegyptica tree.
Ash (%): The mean values of leaf ash content of A. tortilis and B. aegyptica trees did not show significant variation (P<0.063). The average values of leaf ash content of for A. tortilis and B. aegyptica trees were 10.18% & 11.47% respectively ( Figure 3). Although the average values of leaf foliar concentration of ash was not significant, the average value of A. tortilis was slightly lower than the average value of B. aegyptica trees.
Ether Extract (% EE): No significant variation was shown in the mean values of leaf Ether Extract (% EE) content of A. tortilis and B. aegyptica trees (P<0.063). The average values of leaf ash content of for A. tortilis and B. aegyptica trees were 5.01% & 7.25% respectively (Figure 3). Although the average values of leaf foliar concentration of Ether Extract was not significant, the average value of A. tortilis was slightly lower than the average value of B. aegyptica trees.
Crude Fibre%: The leaves Crude Fibre contents (% CF) of A. tortilis and B. aegyptica trees were revealed statistically significant (P<0.0002) differences between the species. The average values of leaf crude fibre for A. tortilis and B. aegyptica trees were 10.12% & 20.89% respectively. In this study, the leaf crude fibre content of A. tortilis was statistically significantly lower than the average leaf crude fibre content of B. aegyptica trees.
Nitrogen Free Extract (%NFE): The mean values of leaf nitrogen free extract (% NFE) content of A. tortilis and B. aegyptica trees were not showed significant variation (P<0.063) between species. The average mean values of leaf nitrogen free extract content of A. tortilis and B. aegyptica trees were 42.25% & 44.35% respectively. Although the leaf foliar concentration of nitrogen free extract was not statistically significant the two species, the average value for A. tortilis was slightly lower than the average value for B. aegyptica trees. The variation in DM composition of the different browse might be due to the generic differences, climate or season, geographical location or agroecology, time of day, aspect of the tree, tree age, and position of leaves in the crown, age of foliage, internal nutrient balance, effects of diseases, soil type and soil age.

Conclusion and Recommendation
Among the leaf foliar macronutrients considered in this study, the leaf foliar nitrogen concentration of A. tortilis was statistical significantly higher than B. aegyptica. With the regard to proximate chemical composition, dry matte, crude protein, digestible crude protein and crude fiber were also revealed statistical significant variation between the two species. Tree phenology is different in different season thus its nutritional quality may be different for different season and age. Therefore, further study will be important to know the effect of this tree on microclimatic factors under its canopy in different seasons. Because of its role in ameliorating microclimate and improving soil fertility under its canopy, retaining of A. tortilis and B. aegyptica trees on crop land is important for the farmers with appropriate component management like lopping and crown opening. financial support which helped us to undertake this research activity. We would also like to thank staff of Mieso Office of Agriculture and Natural Resource Management, the enumerators and farmers who participated during the field data collection. Furthermore, our gratitude goes to all staff members of Oda Bultum University for their facilitation of any logistic required for this work, and all our intimate friends who encouraged us during this work.