Agricultural production and productivity predominantly decreased due to soil fertility loss. The causes for soil fertility problem are loss of organic matter, nutrient depletion, removal of top soil by erosion, and decline of physical soil properties (Zelleke et al., 2004). This mainly due to intensive agriculture with limited fertilization rate, affordability problem to purchase high priced inputs and lack of awareness towards integrated soil fertility management on smallholder farms. However, farmland trees have the potential of providing organic outputs used as an input in soil fertility improvement and enhancing agricultural production.

Leaf litter is an organic matter and nutrient return obtained from plants that improve soil fertility through decomposition by releasing of nutrients into the soil. Decomposition is a means in which dead organic biomass is converted into its constituents like carbon dioxide, mineral nutrients, and influenced by physical, chemical, and biological barriers. The breakdown of litter fall contributes the recycles of nutrients for plant use (Aerts et al., 1997). Knowledge concerned with the decomposition of organic biomass has great importance in the management of agro-ecosystem (Cardona and Sandeghian, 2005). Decomposition primarily influenced by the environmental conditions in which decay takes place, the quality of leaf litter in chemical composition and its ability to release nutrients, and the nature and abundance of decomposing agents that exist in the system (Polyakova and Billor, 2007). The decomposition of leaf litter is depending on the leaves' Physico-chemical properties such as lignin content, tannin, lignin to nitrogen ratio, physical leaf roughness, and physical barriers. The decomposition rates are determined from the mass loss placed on the litterbag. The rate of decomposition is different between dry and wet seasons (Nigatia et al., 2014). The decomposition process breaks down the leaf litter into inorganic nutrients as well as stable humus (Bragazza et al., 2008). The imbalance between NPK and the rate of litter decomposition contributes to significant organic matter accumulations in the soil systems (Clymo et al., 1998). The main fraction of NPK is delivered to the soil as dead organic matter through above and belowground biomass (Bragazza et al., 2008). The nutrients transfer and energy from living biological components to the soil is closely related to litter fall and is the beginning point for nutrient cycling. Litter fall act as a major interface of nutrient cycling between plants and soils (Berg, 2000; Gindaba et al., 2004). The decomposition of organic biomass produces organic matter as an important means for soil formation as well as the cycling of nutrients (Onyekwelu et al., 2006; Pandey et al., 2007). Considerable variations in leaf litter decomposition rates were observed among species (Koukoura et al., 2003). The regulation of ecosystem carbon storage and nutrient cycling maintained through the process of decomposition (Wardle, 2004; Santiago, 2007). Some studies also suggested that in the initial stage of litter decomposition, nitrogen is a good predictor of decomposition rate, whereas, in later stages, chemical compounds such as lignin can play a more important role (Liu et al., 2007). Therefore, this study was initiated to know the role of farmland trees’ leaf litter nutrient composition, decomposition and nutrient releasing potential to improve soil fertility. The objectives of this study focused on leaf litter of Cordia africana, Terminalia brownie, and Vitex doniana tree species in nutrient composition, rate of decomposition and important nutrient releasing potential in to the soil system.

Description of study area
The study was conducted at the main station of Jinka Agricultural Research Centre located in Debub Ari District, South Omo zone. The altitude ranged from 1383 meters above sea level (m.a.s.l.). The soil type is characterized as Cambisols with fine to very fine particles, pH ranges of 4.87 to 6.18, and strongly acidic to slightly acidic (Mesfin et al., 2017) under good range for crop production with appropriate management practices. The study site has a bi-modal rainfall pattern with a shorter rainy season from March to May and the longest rainy season from August to November. The total annual rainfall is 1,272.4 ± 250.7 mm. The annual mean minimum and maximum temperatures are 16.3 ± 0.9 and 27.7 ± 1.4°C.

Experimental design and Treatments
Leaf litter
The research work was conducted in Jinka agricultural research center on the main station so it was easy to monitor and attend the experiment. For this study, three dominant farmland tree species were selected and used as treatments. The treatments were labeled and sampled for leaf litter of Cordia african, Vitex doniana, and Terminalia brownie, and an equal amount was distributed in each block for each species. Randomized Complete block design (RCBD) was employed with four replications. Fresh and senesced leave samples were collected from mature trees after shaking gently or picked up from the ground after falling. To estimate the decomposition rate it is important to measure litter mass loss, an estimate obtained using litterbags (Bragazza et al., 2008). The litterbag technique consists in confining dried litter in mesh bags that are placed on the ground and collected to measure the remaining litter mass and associated litter nutrients in a fixed time interval (Bragazza et al., 2008). The litterbag size of 20cm X 15cm was made from 2mm diameter polyester mesh to allow the entering of micro-faunas. The number of litterbags was 12 (three species X four replications) and placed on each block. The C: N content needs to be examined before experimentation (from the collected soil sample and consequent analysis before burring of the leaf samples).

Data collection
The buried litterbags were collected at specified time intervals to determine the decomposability and nutrient release the potential of each species. The initial and after incorporation leaf litter samples were measured. It was grounded and packed with equal weight to analyze OC, total nitrogen, available phosphorus, available potassium Sulfur, Zink, and C: N ratio, this helped to estimate the released nutrients from leaf litter at a given time. This activity was conducted during the wet season the rate of decomposition is different during different seasons (Nigatia et al., 2014).

Data analysis
One-way Analysis of variance (ANOVA) was computed among the tree species and patterns of litter decomposition. Mean separation will be tested for determining the treatment differences. LSD was tested to determine between time and patterns of nutrient release. SAS software (version 9.1) and Microsoft excel speared sheet was employed for data analysis and organization.

Laboratory analysis for soil samples before (composite) and after experimentation was done in Jinka agricultural research center soil lab. The analysis of leaf samples for each tree species was done in Deber Birhan Agricultural Research Center (DBARC) soil and water laboratory. The loss in dry mass of litter samples was calculated as the difference between the initial dry mass and remaining mass at each sampling time. The rate of decomposition was calculated using the percentage of mass loss divided by respective days of sample collection (Hasanuzzaman and Hossain, 2014). The decay constant for leaf litter was calculated using the negative exponential decay model (Olson, 1963) as follows:
K = ln(X / Xo)/t

Where Xo is the initial weight and X is the remaining weight at time exponential function. Half-life calculated using equation 2 (Daldoum et al., 2010).

t1/2= ln (2)
             k

Decomposition rates between the compositions and soil moisture between the treatments was compared by analysis of variance

Decomposition rate and nutrients in leaf biomass (litter) of studied tree species
The current research study was conducted in Jinka agricultural research center on the main station so it was easy to monitor and attend the experimental site. The result revealed that decomposition and amount of nutrients constituted in the leaf litter vary from species to species (Table 1 and 2). This could be due to the quality of leaf litter, the amount & type of nutrients constituted in it, and the environmental condition that affects the process of decomposition. For instance, cordia africana decomposes faster than that of Terminalia brownie and Vitex doniana, which was expressed by decomposition rates of 0.00871, 0.00839, and 0.00598, respectively (Table 3).

Change in soil fertility status after incorporated and decomposed leaf litter
The leaf litter of studied tree species result revealed that required a specified time to decompose and resulted in a change of soil physicochemical properties. The soil nutrients level of pH, OC, C: N ratio, av. P and K showed variation after incorporating leaf litter than the soil result sampled before. The leaf litter incorporated in the soil showed certain nutrients change in their level this due to the easily decomposing and releasing tendency of nutrients into the soil profile (Table 4).

Decomposition rate and nutrients in leaf biomass (litter) of studied tree species
Retained tree species found on farmland play a significant role in terms of soil improvement through litter fall and recycling of nutrients in the soil system. Each studied species has its own contribution to soil fertility improvement through addition of decomposable leaf litter fall and the potential of releasing available nutrients into the soil solution. For instance, leaf litter of cordia africana provides a high level of nitrogen (Table 1) than the other species whereas the carbon composition in Terminalia brownie and Vitex doniana leaf litter was higher. This result is in line with the findings (Mahari, 2014) who reported there was variation in leaf litter nutrient composition and decomposition rate of cordia africana and Croton macrostachyus. Leaf litter of Cordia africana decomposes faster than the other tree species due to high Nitrogen and lower carbon-to-nitrogen ratio composition. The present study results are in agreement with the findings reported by (Sanchez, 2001; Su et al., 2004).

Change in soil fertility status after incorporated and decomposed leaf litter
The changing of soil nutrient status resulted from decomposed and released nutrients of incorporated leaf litter in the study area. Retaining of these species on farmland provide such important production enhancing functions. Management of easily decomposing and soil fertility improving organic inputs are important to restore soil for better production. According to Petit-Aldana et al. (2012), trees contribute the maintenance of soil organic matter through the production and decomposition of litter and pruning residues. The changing of soil nutrient status resulted from decomposed and released nutrients of incorporated leaf litter in the study area.

The result revealed that tree species retained on farmland should have a multifunctional value that contributes to ecological, economic, and social benefits. Multipurpose tree species play important role in terms of soil fertility improvement through leaf litter fall and nutrient turnover. Decomposition and nutrient release in the leaf litter of tree species is determined by climatic conditions and the inherent characteristics of the leaf quality. Leaf litter of tree species with high nitrogen concentration and low carbon to nitrogen ratio decomposes faster than the species with reverse character. Cordia african leaf litter decomposed and releases nutrients in a short time than that of Terminalia brownie and Vitex doniana. Based on the result it is important to recommend farmland tree species should need better management in terms of tree-crop compatibility, leaf litter resource assessment, and harmonizing crop nutrient demand to enhance nutrient use efficiency.

Author Contributions
Belayneh Lemage and Asmelash Tesfaye set up the experiments; Belayneh Lemage carried out the experimentation in the respective location. Belayneh Lemage and Abebe Hegano analyzed the data and developed the manuscript. Belayneh Lemage contributed to data analysis and manuscript review.

Acknowledgements
The authors acknowledge the Southern Agricultural Research institute for financial support to conduct this experimentation and the Jinka Agricultural Research center for its good facilitation by providing important inputs to handle the research work accordingly. The collaboration of researchers was highly appreciated.

Competing Interests
The authors declare that there is no conflict of interest regarding the publication of this paper