Smallholder agriculture is the dominant sector of the Ethiopian economy that provide over 85% of the total employment and foreign exchange earnings and approximately 55% of gross domestic product (GDP) with 85% of the population living in rural area (Abera, 2010). Tef (Eragrostis tef) is an indigenous and one of the major cereal crops in Ethiopia and covered an area about 2.8 million ha (22.95%) followed by maize (16.91%), sorghum (14.85%) and wheat (13.33%) during the main season  of   2015/2016  (CSA,  2016).  It  has  16.76%  of share from total production of cereal (88.68%) in the country. However, in semiarid area of Ethiopia, crop productivity in general and tef productivity in particular have been affected by recurrent drought and low soil fertility (Kidane and Tesfaye 2016). Water and soil fertility are the most limiting and determining factors for low crop productivity and instability.

Nevertheless, agroforestry system which is the integration of trees on farmlands with annual crops has provided   low  cost  (Yengwe  et  al.,  2018),  sustainable opportunity for soil fertility enhancement (Tanga et al., 2014) and has increased agricultural productivity (Nair, 1989). Trees as the main components of agroforestry can play a substantial role to modify the microclimate, enhance soil fertility and crop productivity of the area of its canopy influence by adding nitrogen through nitrogen-fixation and recycling nutrients through litter-fall or biomass-transfer (Umar et al., 2013; Bishaw and Abdu, 2003; Kohili et al., 2008; Young, 1997; Berhane and Agajie, 2006). Despite its importance, the integration of tree with annual crops can have both positive and negative effect on the function of the agro-ecosystem. For instance, both the negative and positive effects of tree on soil moisture content have been reported by various scholars in the country (Hailu et al., 2000; Hailemariam et al., 2010) as well as elsewhere outside the country (Akpo et al., 2005; Raddad and Luukanen, 2007).

Acacia species are the commonest tree species that has been grown in agricultural land (Birhane et al., 2018). Acacia seyal, is one of the tree species widely used in agroforestry practices in Ethiopia in general and in the study area in particular (Endale et al., 2017; Hailemariam et al., 2018). The farmers had left the tree on their farm boundaries, marginal areas, and inside farmlands in order to fulfill their wood requirement and to generate extra income.

A. seyal Del belongs to the family Fabaceae and subfamily Mimosoideae and it has the common name such as Wachu in Amharic, and white galled acacia and white whistling torn in English (Tesemma, 2007). A. seyal is common in many parts of Africa, especially north of the equator, from 10 to 12 degrees. It also occurs in Eastern and Southern Africa (Orwa et al., 2009). In Ethiopia, it is also found on seasonally flooded black-cotton soil, in river valleys and wooded grassland of dry and moist Weyna Dega agro-climatic zones (Bekele and Tengnäs, 2007).

A. seyal used for firewood, charcoal, poles, posts, medicine (bark, gum), fodder (leaves, twinges, bark, flowers and pods), bee forage, nitrogen fixation, soil conservation, shade, windbreak, gum, tannin, dye (McAllan, 1993). A. seyal forms either pure stands of different densities or mixed stands associated with other species (Mohammed, 2011).

In the study area, there are no scientific evidences on the effect of scattered trees on the microclimate, soil fertility and crop productivity. Yigardu (2002) has been studied on the biomass and composition of tree species on-farm lands. Therefore, the objective of the study is to investigate the effect of A. seyal (Del) on soil bulk density, soil moisture content and yield of tef in parkland agroforestry system of semiarid area of Amhara.

Description of the study area

The study was conducted in Sirinka  catchment  which   is  found  in Habru district (Wereda), North Wollo zone, Amhara region. It is geographically located in between 11°41.2' - 11°47.7' N and 39°31.4' - 39°37.6' E. The catchment comprises three adjacent kebeles (the smallest administrative unit of Ethiopia), namely: Gerado, Sirinka and Goshuweha. The research was conducted at Sirinka and Goshuweha kebeles based on the presence of good distribution of the tree. The district is intermediate lowland agro ecologies that range from 1000 to 2400 m above sea level. The climate of the study area is generally characterized by arid and semiarid climate. The mean minimum and maximum daily temperature falls between 13 and 26°C and the mean annual rain fall varies from 750 to 1000 mm with bimodal distribution, the main rainy season lies between June and September and the short rainy season occurring from March to April but the amount and distribution of the rain is fluctuated (Figure 1). The district characterized by rugged and undulated topography. The western part makes a chain of hill that border the high land escarpment of the country from rift valley low lands. According to the soil study conducted by Sirinka Agricultural Research Center (SARC), the two dominant soil types found in the study area are utric vertisol and vertic cambisol.

Soil bulk density (g/cm³) was determined by measuring the dry mass of soil per unit of volume of the core (98.13 cm³). The following formula was used to calculate the bulk density of the soil:

Data analysis and interpretation

All statistical comparison used SAS software window version 9 (SAS Institute, 1999). The quantitative data that were obtained from laboratory were analyzed by two way analysis of variance (ANOVA) to test variations between distances, depths and their interaction. Least significant different test (LSD) was used for mean comparison when differences were significant at 5% of probability level. Simple correlation analysis was also carried out by SAS to examine magnitudes and direction of relationships between selected soil physical properties and biomass and grain yield of tef.

Soil texture

The result indicated that clay, silt and sand fractions were similar both for distance from the tree base and soil depths (Table 1). The textural classes of soil both under and outside the tree canopy and surface and subsurface soil have had clay texture. This result indicated that scattered A. seyal tree do not influence soil texture and since the soil depth was only 30 cm deep, it  may  be  too shallow to show textural differences. In addition, similarity in %clay, %silt and %sand fraction of the soil under different canopy position of the tree and soil depths might indicate the similarity of the parent material in the studied field. Because, size of soil particles is not influenced by soil management practices, rather it is predominantly by the parent material form which the soil formed. These findings are supported by other studies that observed no variation of soil texture both laterally and vertically under and outside the canopy of tree in Ethiopia (Hailu et al., 2000; Abebe et al., 2009) and in Kenya (Githae et al., 2011).

Soil bulk density (BD)

Soil bulk density was significantly influenced by both horizontal distances from the tree base and by soil depths (Table 1). Generally, in the present study the value of bulk density shows increasing trend from the tree base towards outside the canopy and from surface soil to subsurface soil (Table 2). It indicated that the soil under A. seyal canopy had lower bulk density as compared to that of outside the canopy, this may be due to organic matter build up under the canopy from litter fall and higher turnover of fine roots closest to the tree. Thus the accumulation of litter fall under the canopy buffered the soil against rain drop impact, wind erosion and associated compaction as evidenced by the lower bulk density (Shukla  et  al.,  2006).  The present study agrees with the observations of Hailu et al. (2000), who found that lower soil bulk density under Millettia ferruginea tree canopy than that of open area in Southern Ethiopia. Similarly, Jiregna et al. (2005) also found that the soil bulk density under the canopy of Commiphora africana is lower than that in the open area. Nevertheless, Hailemariam et al. (2010) have shown in his report that there was no significant difference between bulk density of the soil under and outside the canopy because of lack of differences in soil organic matter level.

Soil moisture content (SMC)

The moisture content of the soil influenced by distance from the tree base and soil depths showed declining trend as a function of distance from the tree base toward open lands and increasing trend from surface soil to the corresponding subsurface soil (Table 2). The results of the present study are in line with that of Hailu et al. (2000). In his findings, the moisture content of surface and subsurface soil under the canopy was also found to be higher than that of the corresponding surface and subsurface soil outside the canopy. Akpo et al. (2005) also found higher moisture content in surface soil under tree canopy than the corresponding  surface  soil  outside the canopy in Senegal. The higher moisture content under the tree canopy might be due to the mulching effect from litter layer and shade that reduce evaporation (Moody and Jones, 2000). Moreover, the higher soil organic matter under the canopy may be responsible for higher moisture content. In agreement to the present study the higher soil moisture content (SMC) was observed at subsurface soil than that of the surface soil under Acacia senegal agroforestry in clay soil of Blue Nile, Sudan, due to top soil dried much faster than subsoil during dry season (evaporation) and water extracted by crops (Raddad and Luukanen, 2007). In contrast, Hailemariam et al. (2010) observed that moisture content of the soil decreased towards tree trunk. 

Grain yield of tef

As shown in Table 3, the grain yield of tef did show significant difference among distances from the tree base. The higher grain yield of tef was observed at 2.5 m from the tree base, which is the place where the interface of tree canopy and the open land. Therefore, the grain yield increment may be due to the improvements of soil environment and minimum competition between the roots of  tree  and  associated  crop.  In  line  with  the   present study, Mubarak et al. (2012) in semi-arid tropics of Sudan reported that millet yield under Azadirachta indica and Balanites aegyptiaca were higher than the control by about 43%. Under Faidherbia albida in the highland of Tigray, Northern Ethiopia, Hadgu et al. (2009) found that barley yield was significantly higher at 1 m distance from the tree trunk compared to yields at 25 and 50 m, due to nitrogen fixation by F. albida and nitrogen supply by shedding its leaves during cropping season. However, in contrast with the present study, research conducted in Northern Ethiopia by Hailemariam et al. (2010) reported that grain yield of sorghum was not different among distances from the tree base, Osman et al. (2011) observed that the performance of cowpea and pearl millet intercropped under Parkia biglobosa did not differ among the four zones which was laid at different distance from the tree trunk. Bazie et al. (2012) observed that the grain yield of sorghum under pruned Phyllocrania paradoxa and P. biglobosa tree were five times higher than that of unpruned trees. These yield reduction are mainly attributed by the shade of the tree (Kohili et al., 2008) that reduce sunlight reaching soil and crops.

Biomass yield of tef

Fresh as well as oven dry biomass of tef in the present study were not affected by distance from the tree base or by different canopy positions of A. seyal tree. However, as shown in Table 3, the fresh biomass of tef shows increasing trend as a function of distance from the tree base towards open canopy but dry biomass of tef had not shown clear trends. In line with the present investigation Hailemariam et al. (2010) observed that biomass yield did not differ among three zones away from the tree. Bazie et al. (2012) found that straw dry matter of sorghum was higher under pruned P. paradoxa and P. biglobosa as compared to unpruned trees. This may indicate shading effects on dry matter production.

This study evaluated the possible impact of A. seyal trees on available soil water, soil bulk density and yield of tef (Eragrostis tef) grown under the tree canopy. From the result of this study, we can conclude that the presence of A. seyal tree on croplands had no effect on soil texture, but it improves soil moisture content and soil bulk density. In addition, the grain yield of associated tef crop is better at the interface of the tree canopy. Therefore, the incorporation of A. seyal tree on cultivated land could be an additional source of benefit for small holder farmers. However, the impact of crown management on yield and growth performance of associated crop and fine root distribution and dynamics along soil depths and horizontal distances should be further investigated.

The authors have not declared any conflict of interest.

The authors would like to express their deepest gratitude to Amhara Region Agricultural Research Institute (ARARI) for giving them the chance to pursue this study and financially supporting the expenses and field research of the study. They are also grateful to Sirinka Agricultural Research Center (SARC) for the study leave, laboratory facilities and transportation during the field work.