Fruit tree based agroforestry system has great roles to play in the livelihood improvement and it provides multiple contributions of household income and supplementary food for smallholder farmers. However, there is limited quantitative empirical evidence on the contribution of fruit based agroforestry system. Therefore, this study was initiated to examine fruit tree based agroforestry system and its contribution of household income for livelihood improvement in Dale District. Data was collected through a combination of focus group discussions, key informant interviews, household survey and field inventories. A total of 145 respondents from three kebeles were interviewed and data were analyzed by using descriptive statistics and econometric analysis. The results revealed that the status of fruit based agroforestry in the study area varies with land holding size. Fruit trees such as; Musa acuminate, Persea americana and Mangifera indica were the major types of tree species grown in the system in the study area. The contribution of fruit for poor, medium and rich households was 3166.8 Birr, 3713.8 Birr and 1380 Birr respectively. The fruit tree contributes 24.75% for poor HHs, 23.34% for medium HHs and 5.16% for rich HHs from the total income. The average income earned from fruit trees was 2754 Ethiopian Birr (ETB) per year. Besides, the result from the econometric analysis indicates that access to extension service, family size, land size, and the number of livestock influenced the income from fruit trees. Further studies of examining of the market value chain, areas of intervention along the chain and economic value of the fruit tree based agroforestry system including environmental function served by the system is needed to fully understand the contribution of fruit tree based agroforestry system and livelihood improvement.

Keywords: agroforestry, contribution, fruit, household income

agroforestry is an intensive land management system that seeks to optimize the benefits from the biological interactions created when trees and/or shrubs are deliberately combined with crops and/or livestock.¹ agroforestry involves the cultivation and use of trees in farming systems and is a practical and low-cost means of implementing many forms of integrated land management, especially for small-scale producers.² Agroforestry can reduce the risks associated with agriculture, small scale or large, and may also increase the sustainability of agriculture.³ Such systems can, therefore, make a significant contribution towards reducing poverty and resource degradation in Africa. For instance, agroforestry technologies such as fruit trees can provide a more diverse farm income and reduce food insecurity.⁴ Fruit tree based agroforestry system is one of agroforestry system which comprises combinations of plants to maximize the natural resource use efficiency and enhance total factor productivity.⁴ Fruit-tree-based agroforestry is highly popular among resource limited producers worldwide due to its relative pre-production phase of fruit trees, the high market value of their products and the contribution of fruits to household dietary needs.^5,6 Tropical fruit tree species diversities are abundant in homesteads and farms of fruit tree growers’ households and have a number of economic, social, cultural, aesthetic and ecological functions important to livelihoods. Fruit trees based agroforestry system has great roles to play in the livelihood of the farming community because of its multiple benefits. Some of the benefits are income generation, food, fuel, construction material, fodder and shading for shade loving crops.⁷Therefore, this study attempted to assess the status, contribution and factors in fruit based Agroforestry system for livelihood improvement of smallholder farmers in Dale District, Sidama Zone, Southern Ethiopia.

Description of the study area

Dale is one of the District in Sidama zone, Southern Nation, Nationalities, and peoples Region of Ethiopia. Dale is located at about 320 km from Addis Ababa and 45km from Hawassa, the regional capital. Geographically, it is located 6° 50’ 30’’ to 6° 39’ 30’’ North and 38° 17’ 0’’ to 38° 32’ 0’’ East. The district has an altitudinal range of 800 to 2600 meter above sea level (Figure 1). Mixed agriculture, Enset, fruit tree and crops are the main livelihood strategy of the district, which is characterized by subsistence production. Annual and perennial crops are predominant and rain-fed agriculture is mainly practiced. The most commonly cultivated crops in the study area are Enset, Coffee, Maize, Haricot Bean and some other root crops such as sweet potato is one of the most common staple foods and whereas Coffee, fruit and tree products are a source of cash crops. Fruits such as Avocado, Mango and Banana, are cultivated both for household consumption and income generation.

Figure 1 The map of the study area.

Sampling techniques

In this study, a multi-stage sampling technique was employed. Dale District was purposively selected as the presence of fruit trees on their home gardens and three potential kebeles (Showoa, Ajawa and Dagiya) were selected purposively from the study area. A reconnaissance survey was undertaken on October 2016 in order to familiarize with the study area and to obtain an insight into the farming systems. Ten percent (10 %) of the total household were randomly selected from each kebele. Households (HHs) with different wealth status who practing homegarden agroforestry practice were selected by using stratified sampling techniques. The sample households were stratified based on the basis of wealth classes (poor, medium and rich) with the help of key informants (KI). The wealth stratification was used to get a representative sample from all wealth groups and to assess the status of fruit trees found in homegarden. Hence a total of 145 households were considered for the household survey in this study (Table 1). The questionnaire was comprised of both open and close-ended questions regarding household characteristics (e.g. family size, land holding size, age, sex, education, distance from the market, access to extension, income sources). The questionnaire also incorporated amount of fruit produce per year, income contribution from fruit production. The interview was conducted through face to face in either head of the HHs or member of the household who is familiar for fruit tree production. 

Methods of data analysis

The focus group and farm survey data were analyzed by using descriptive and econometric procedures. Quantitative data were analyzed by Statistical Package for Social Sciences (SPSS Version 16), STATA 9 software and Microsoft Excel 2007. 

Descriptive analysis

Descriptive statistics like mean, standard deviation, percentages and tabular analysis were used to examine and understand the socioeconomic situations of sampled respondents. 

Econometric analysis

The multiple linear regression analysis was used to explain the relationship between a dependent variable and one or more independent variables. It is also a technique that allows additional factors to enter the analysis separately so that the effect of each can be estimated.⁸ It is valuable for quantifying the impact of various simultaneous influences upon a single dependent variable. Further, because of omitted variables bias with simple regression, multiple regression is often essential even when the investigator is only interested in the effects of one of the independent variables.^9,10

The general formula of the multiple linear regression models is:

$y={\beta }_0+{\beta }_1{X}_1+{\beta }_2{X}_2+\dots {\beta }_k{X}_k+\epsilon$

Where y is the dependent/explained variable and x₁ . . . x_k are the independent/ explanatory variables. In the case of this study xi…xk, were age, experience, family size, land size, access to extension service, distance from market, number of livestock kept, off farm activity and Y is income earned from fruit based agroforestry and ε is error term. The equation can be written as,

$Income={\beta }_0+ {\beta }_{1}age+{\beta }_{2}exp +{\beta }_{3}familysize+{\beta }_{4}farmsize+{\beta }_{5}Ext+{\beta }_{6}marketdistance+{\beta }_{7}livestock+{\beta }_{8}OFF+, + \epsilon$

Regression coefficients ,  and….  are known as partial regression or partial slope coefficients. measures the change in the mean value of y, E(y), per unit change in X₁(age), holding the value of all other explanatory variables constant and β₀ is intercept of the model.

The multiple linear regression analysis/Ordinary Least Square estimations (OLS) was used to capture the cause and effect relationship between the dependent variable, gross income from fruit tree based agroforestry system and the independent variables.

Households’ characteristics

A total of 145 individuals’ (households) of which 90.3% male and the remaining 9.7% female were involved in the study. Households with different occupation were involved in the study.The majority of the households (95.9%) age ranged from 18 to 64 and some proportion of the households (4.1%) households’ age ranged above 65 years, showing that availability of more productive working.The result shows that almost all respondents are in the productive age category, whereas less than 5% of the respondents belong to age category beyond 65 years. Regarding the education level of the households, 26.8% were not attended formal education, 52.4% were attended primary 1^st cycle (grade 1-4), and 21.4% primary 2^nd cycle (grade 5-8) and the rest1.4% were high school (grade 9-10). The result shows that about three fourth of the fruit tree producers (75.2 %) attended formal education. But one fourth of the fruit tree producers (24.8%) did not attend formal education (Table 2).
Farming experience of fruit producer’s household head is an important variable considered in this study. The result shows that the sampled household heads have a minimum of 4 years of farming experience and a maximum of 50 years of experience in farming. The average year of farming experience was 21.64 years. The result shows that the respondents have good farming experience (Table 3). The average family size of fruit tree producers was 5.5 persons although it ranges between 2 to 12 persons with a standard deviation of 1.92 (Table 3). This shows that the household heads in the study area can be categorized in medium family size category. The result in Table 4 shows that distribution of the family size of respondents in three categories and 64.8% of respondents have a family size within the category 4 to 6 persons (medium family size) and 10.4 % of respondents have a family size of the category 1 to 3 persons(low family size). Whereas, about one fourth of the respondents (24.8%) in the study area have a family size of greater than 6 person’s which is a large family size category. The result shows that the majority of respondents have a family size of 4 to 6 family members. As farming is one of the labor intensive activities that need more labor, the implication is that households in the study area may not be constrained by labour and the family members can engage themselves in different farming activities and can increase the income from fruit. This result is more evident as the majority of the respondents in the study area are already identified as they belong in the productive age category as described above. This result is consistent with the finding of ,¹¹ which shows that 47% respondents have a family size of 5 to 7 persons and found that farming is very labor-intensive and tedious because it is done manually in developing countries and the family’s needs to have more members in order to provide sufficient labour to work on their farm land.

Thus it can be concluded that the family size of the household head is an important variable in smallholder agriculture notably in agroforestry. This is in line with the notion that large family size is normally connected with higher labor resources, which would enable a household to accomplish various agricultural tasks particularly during peak seasons.¹² The average land holding of the sample respondents was about 0.30 ha, 0.90 ha and 1.83 ha for poor, medium and rich HHs respectively. This means that the average land holding size for poor and medium HHs in the study area is below the national average of 1.18 hectare.¹³ This shows that the rich group has the largest land holding and the poor have the smallest in the study area (Table 5). Regarding the proportion of the households home garden size, the result shows that 46.9%, 38.6% and 14.5% of the total land holding in small, medium and large respectively were allocated to home garden agroforestry practices (Table 6). Accordingly, compared to their total landholding size, rich and medium households owned relatively larger landholding size than the poor households; they preferred production of annual crops rather than fruits. The result is agreed with¹⁴ who reported that small farms allocate a large share of their land for home gardening compared to those with medium and large size farms in Western Amhara for reason of economies of scale, households with large farm sizes overridingly concentrate on annual crops like maize that can satisfy their demand in the absence of fruits. The dominant source of income in the study area is from agricultural activities. However, the ofarm activities have a great potential to provide additional incomes during the slack season to rural households. The result in Table 7 shows that (97.2%) of respondents in the study area had to access to extend service whereas, only (2.8 %) had no access to extend the service. This implies that the DAs reached the majority of the respondents. According to the survey result, the development agent (DA) help farmers by providing training on crop cultivation and harvesting, livestock production and management, land management, visiting field activities and other services. The farmers were attending training in Farmers Training Center (FTC) in their respective kebele. Therefore access to extension service for farmers is more recommendable to increase the dual benefit from the agroforestry system, economic benefit and environmental benefit. This is supported by¹⁵ who revealed that farmers who had access to extension service from the government through the office of agriculture including technical support, training and to some extent provision of improved planting material had improved production from the whole system and improved their livelihood.

Status of fruit tree based agroforestry (FTBAF) practice in the study area

Mixed homegarden, locally called as “Gattae”/Qaie, is an essential part of the food production system in the study area. Distribution of the fruit tree species varied across the wealth categories. The result in Figure 2 shows that the majority of the rich household 96% have grown Persea americana, 95% have grown Musa acuminate and 57% Mangifera indica indicating that they were the most recorded fruit tree species in the study area. Whereas, in the medium wealth class households, the most recorded fruit tree species are Persea americana which is grown by 96% of the respondents, Musa acuminata by 91% of the respondents and Mangifera indica by 55% of the respondents. Whereas, for the poor households the results the total fruit tree species recorded were Persea americana by 91%, Musa acuminate 80% and Mangifera indica 50%. The result revealed that in all wealth categories the fruit trees such as; Musa acuminata, Persea americana and Mangifera indica are frequently represented by more individuals. While Psidium guajava, Prunus persica, Citrus medica, Annona senegalensis, Citrus sinensis, Citrus aurantifolia and Casimiroa edulis are the less frequent fruit tree species (Figure 2). The overall result shows that the three fruit tree species mentioned above were the most recorded ones in comparison to the other fruit tree species. The possible reason is that most of the time households prefer high value fruit tree species and good marketability, such as Musa acuminata, Mangifera indica and Persea Americana.¹⁶

Figure 2 Proportion of fruit tree producers/households with respect to the type of fruits and wealth category (N=145).

Amount of fruit produced in fruit tree based agroforestry

Among the total volume of fruits production, Banana comprises 44.39% and followed by Avocado 32%, Mango 7.01%, Papaya 3.9% and other fruits accounts (12.7%) (Table 8) and (Figure 3). The result indicates that Banana, Avocado and Mango were the three most dominant fruits produced by households in the study area.

Figure 3 Fruit tree production in percent in the study area.

The contribution of FTBA for income of smallholder farmers

The contribution of fruit for poor, medium and rich households was 3166.8 Birr, 3713.8 Birr and 1380 Birr respectively in the study area. The fruit tree contributes 24.75% for poor HHs, 23.34% for medium HHs and 5.16% for rich HHs from the total income. The income that farmers earn from fruit production was investigated and the result shows that most respondents plant different types of fruit trees in the study area both for income generation and consumption. The amount of income earned from fruit tree based agroforestry system depends on the type of fruits. The average income earned from fruit trees was 2754 Ethiopian Birr (ETB) per year. This indicates that fruit tree production in the agroforestry system plays a great role in generating additional income and contributes to the livelihood of smallholder farmers in the study area. The finding was in line with¹⁷ who reported that home garden agroforestry contribute various benefits to households and elsewhere in Ethiopia.¹⁸ The poor and medium households got relatively higher annual income from fruit production than rich househoulds in the study area.

The mean annual net income from fruit trees was highest for medium followed by poor households. The income difference among wealth categories could be due to the variation of fruit trees and the size of land to plant a given fruit tree species.¹⁹ This was perhaps due to more dependency of the rich HHs in crop farming or related activities than fruit production. This finding was supported by what was reported by²⁰ in India. Fruits contribute to livelihoods through income generation and as a safety-net for consumption and income smoothing²¹ and fruits play an important role especially during a time of famine and other stress as food, nutrition and cash income.²² In the study area households used fruits for both household consumption and as a source of income. However, according to the discussants, household consumption of fruits was higher in rich wealth categories due to their diverse livelihood sources than the poor. Contrarily, poor households usually deliver most of the fruit they produced to the market since they consider it as their main source of income. According to the survey result, about 89.98% of incomes were earned from on-farm activities and remain 10.02 were from off-farm activities (Table 9).

Factors that influence the contribution of FTBAF for household’s income

The results reveal that four of the eight explanatory variables included in the analysis such as extension service, family size, land size and number of livestock kept were statically significant, whereas, farming experience, education, distance from market and household age were not significant (p<0.05) (Table 10). This implies that the four predictors which significantly explain variations had an impact on the household’s net income in the study area than others. Increase in size of these predictors brought about an increase in the household’s annual net income at magnitudes indicated by their respective coefficients and thus contributing to livelihood improvement.Consistent with the prior expectation farm size was found to be positively associated with income generated from agroforestry practice. Therefore, those who have large farm size get high income than those who have a small size. The coefficient value of 8.579 indicated that other factors held constant when the farm size increase by one unit the income generated from FTBAF increase by 8.579 ETB. This is because when there is a large size of land there is more diversification of components, which increases the income from the system. This finding revealed that the increase of components of agroforestry can increase the income of the system, but the diversification of component is directly affected by the size of the farm.^23,24 Livestock holding by the household as measured by Tropical Livestock Unit (TLU) is also found to influence the income from agroforestry system. The result in table 10 indicates that the possession of livestock positively influences the income from agroforestry system. In line with the prior expectation, it is positively associated with income generated from fruit-tree based agroforestry system. The coefficient value of 1.273 indicatedthat other factors held constant when the number of livestock increases by one unit, the income generated from FTBAF increase by 1.273 ETB. In a ruralarea, livestock dung is also the major source for soil fertility. Thus, farmers who have more livestock can get more dung, which increases soil fertility and increase the yield of the agroforestry system. Therefore, the number of livestock has a positive impact on income generated from agroforestry system. This result supported by a study conducted in Nepal livestock is a major source of income, manure for agricultural crop and power for ploughing and the number of tree species per household increased with the number of livestock units.²⁵ Family size is statistically significant and positively associated with the income generated from the fruit-tree based agroforestry system. The coefficient value of 5.1624 indicated that other factors held constant when the family size increase by one unit the income generated from FTBAF increase by 5.1624ETB. This positive impact may be due to the nature of farm activity, which is labour intensive, and that needs more family labour. The households who have more family size are favorable to supply more family labour. The finding was consistent with the study carried out by²⁶ in Nigeria and other studies elsewhere in Ethiopia, large family size has a positive impact on farm income.²⁷ The regression result indicated that extension service is statistically significant. Consistent with the prior expectation it is positively associated with the income generated from the fruit-tree based agroforestry system. The coefficient value of 3.1008 indicated that other factors held constant when the farmers have access to extension service it increasethe income generated from FTBAF by 3.1008 ETB. This is because extension service is a means to deliver the message that comes from the research center and development agencies at a different time and enhance the accuracy of implementation of the technology which is important for improving farming activities. Extension services are also important to expand the knowledge and skills of farmers to increase income. Similar study by.²⁸ from Tigrai indicated that provision of technical advice on farming issues such as what to produce, how to produce and when to produce to facilitating credit availability and input supplies and even to the provision of market information and capacity building training to farmers and other study report.²⁷ shown that membership in extension service program is positively associated with total crop production.

The regression result indicated that extension service is statistically significant. Consistent with the prior expectation it is positively associated with the income generated from the fruit-tree based agroforestry system. The coefficient value of 3.1008 indicated that other factors held constant when the farmers have access to extension service it will increase the income generated from FTBAFS by 3.1008 ETB. This is because extension service is a means to deliver the message that comes from the research center and development agencies at a different time and enhance the accuracy of implementation of the technology which is important for improving farming activities. Extension services are also important to expand the knowledge and skills of farmers to increase income. Similar study by ²⁸ from Tigrai indicated that provision of technical advice on farming issues such as what to produce, how to produce and when to produce to facilitating credit availability and input supplies and even to the provision of market information and capacity building training to farmers and other study report²⁷ shown that membership in extension service program is positively associated with total crop production.

The study has revealed that fruit tree based agroforestry system contribute several products. Ffinancial analysis showed that the fruit tree based agroforestry practice is more profitable land use system than mono cropping land use system. Farm size, family size and number of livestock found to be the most important factors that affect the practice of fruit tree based agroforestry system. Further studies of examining of the market value chain, areas of intervention along the chain and economic value of the fruit tree based agroforestry system including environmental function served by the system is needed to fully understand the contribution of fruit tree based agroforestry system for household income to livelihood improvement small holder farmers.

We would like to extend our great appreciation to Dale farmers’, District and local experts for devoting their precious time to provide us with information and sharing their knowledge for this study. We also would like to thank the Ministry of Agriculture and Natural Resource and Hawassa University, Wondo Genet College of Forestry and Natural Resource.

The authors declare that there is no conflict of interest regarding the publication of this article.

1. Molua EL. The economics of tropical agroforestry systems: the case of agroforestry farms in Cameroon. Forest policy and economics. 2005;7(2):199–211.
2. Leakey, Roger RB. Should we be growing more trees on farms to enhance the sustainability of agriculture and increase resilience to climate change. Special Report, ISTF News, USA: 2010.
3. Martin FW, Sherman S. Agroforestry Principles. 1992.
4. Thangata PH, Peter EH, Christina HG. Modeling agroforestry adoption and household decision making in Malawi. African studies quarterly. 2002;6(1):271–293.
5. Shamebo D. Banana in the southern region of Ethiopia (SRE). Bananas and Food Security: 1998;119.
6. Bellow J. Fruit tree based agroforestry in the western highlands of Guatemala Dissertation, University of Florida, Gainesville, USA: 2004.
7. General agro–forestry – silviculture and mixed farming solutions.
8. Baker SL. "Multiple Regression Theory." Multiple Regression Theory. 2006;1–15.
9. Gujarati DN. Basic Econometrics. 2nd ed. McGraw–Hill, Inc.;1998.
10. William HG. Econometric analysis 5th ed. Upper Saddle River, New Jersey 07458, 2003.
11. Adekunle OA, Adefalu LL, Oladipo FO, et al. Constraints to youths’ involvement in agricultural production in Kwara State, Nigeria. Journal of agricultural extension. 2009;13(1).
12. Croppenstedt A, Demeke M, Meschi MM. Technology adoption in the presence of constraints: the case of fertilizer demand in Ethiopia. Review of Development Economics. 2003;7(1):58–70.
13. Central Statistical Authority. Area and Production of Major Crops. Agricultural Sample Enumeration Survey. Addis Ababa, Ethiopia: 2008.
14. Fentahun MT, Hager H. Exploiting locally available resources for food and nutritional security enhancement: wild fruits diversity, potential and state of exploitation in the Amhara region of Ethiopia. Food Security. 2009;1(2):207–219.
15. Ashenafi M. Economic Analysis and Adoption Determinants of Fruit Tree Based Agroforestry Practice. In: Dilla Zuria Woreda, Gedeo Zone, Souther Ethiopia. Msc Thesis, Wondo Genet College of Forestry and Natural Resources, Hawassa University, Ethiopia: 2011.
16. Honja T. Review of mango value chain in Ethiopia. J Biol Agric Healthc. 2014;4(25):230–239.
17. Amenu BT. Home–Garden Agro–Forestry Practices and Its Contribution to Rural Livelihood in Dawro Zone Essera District. Journal of Environment and Earth Science. 2017;7(5):88–96.
18. Anshiso A, Woldeamanuel T, Asfaw Z. Financial Analysis of Fruit Tree Based agroforestry Practice in Hadero Tunto Zuria Woreda, Kembata Tembaro Zone, South Ethiopia. Research Journal of Finance and Accounting. 2017;8(3):72–80.
19. Soemarwoto O. Homegardens: A traditional agroforestry system with a promising future. In: Steppler HA, and Nair PKR, editors. Agroforestry: A Decade of Development, ICRAF, Nairobi: 1987. 157–170 p.
20. Mahendra DS. Small farmers in India: Challenges and opportunities: 2014.
21. Kalaba FK, Chirwa PW, Prozesky H. The contribution of indigenous fruit trees in sustaining rural livelihoods and conservation of natural resources. Journal of Horticulture and Forestry. 2009;1(1):1–6.
22. Akinnifesi FK, Sileshi G, Ajayi OC, et al. Contributions of agroforestry research and development to livelihood of smallholder farmers in Southern Africa: 2. Fruit, medicinal, fuelwood and fodder tree systems. Agricultural Journal. 2008;3(1):76–88.
23. Desta H. Traditional agroforestry homegarden component management and its contribution to household livelihood in Tembaro District, Southern Ethiopia. MSc Thesis, Wondo Genet College of Forestry and Natural Resources, Hawassa University, Ethiopia: 2012.
24. Tesfaye A. Diversity in homegarden agroforestry systems in Southern Ethiopia. PhD. Thesis. Wageningen University: 2005.
25. Khanal S. Contribution of agroforestry in Biodiversity Conservation and Rural Needs Fulfilment: A Case Study from Kaski District. Doctoral dissertation, Tribhuvan University: 2011.
26. Adekunle OA, Adefalu LL, Oladipo FO, et al. Constraints to youths’ involvement in agricultural production in Kwara State, Nigeria. Journal of agricultural extension. 2009;13(1).
27. Kebede W. The contribution of agroforestry based integrated development Project in sustaining the rural livelihoods:–The case of Tombiya Agroforestry Project in Woliso Woreda, West Shewa of Oromia Regional State, Ethiopia: 2011.
28. Goitom A. Commercialization of smallholder farming in Tigrai, Ethiopia: Determinants and welfare outcomes. MSc Thesis, the University of Agder, Kristiansand, Norway: 2009.