Investigation of Woody Species Composition, Diversity, and Carbon Stock under Agroforestry Practices in Oromia National Region State, Central Ethiopia

. Agroforestry is a common practice in the tropics that is characterized by various activities such as parkland on cultivated lands or home garden agroforestry around homesteads. In Ethiopia, agroforestry is an ancient land use type that is practiced by smallholder farmers. Scattered trees in cropland or parkland trees and home garden trees are old agroforestry practices, and the most dominant practices exist in diferent parts of Ethiopia. Tey cover large areas of highland, midland, and lowland agro-ecologies of Liban Jawi District in West Showa of Oromia National Regional State, Central Ethiopia. However, woody species particularly in parklands are declining in many agricultural landscapes due to the overuses for fuel wood, charcoal production, and expansion of agricultural lands due to the degradation of nearby forests. In the study sites, comprehensive studies in terms of their composition, diversity, structure, and carbon stocks are still not well explored. In this context, we assessed woody species composition and diversity, structure, and carbon stocks because parkland and home garden agroforestry practices in three agroecologies of highland, midland, and lowlands in the district were not quantifed


Introduction
Agroforestry has been practiced for a long period of time in many parts of the world, and it is a new name for a set of old practices [1].In Africa, many smallholder farmers practice diferent agroforestry practices [2].Agroforestry practices are regarded as an ancient land use practice in Ethiopia [3].Among those practices, home gardens, scattered trees in cropland, boundary planting, and live fences are old agroforestry practices, and the most dominant practices exist in diferent parts of Ethiopia [4][5][6].Te types of agroforestry practices are parkland and home garden which are found in northwestern parts of Ethiopia [7].Parkland and home garden agroforestry practices are common practices by smallholder farmers in many parts of Ethiopia.Parkland also known as scattered trees in croplands is a common type of agroforestry practice in the tropics and characterized by well-known scattered trees on cultivated and recently fallow lands, which is developed as a result of crop cultivation on a piece of land that became a permanent activity [8].Parkland agroforestry practice is also common and covers a large area of agricultural lands in northern Ethiopia [9].Te scattered Faidherbia albida in cropland is the oldest form of indigenous agroforestry practice which is present in central and eastern Ethiopia [10,11].In other ways, parkland agroforestry is the type of tree-based agroforestry land use practice whereby a tree is grown associated with cereal crops such as maize, tef, sorghum, barley, and wheat [12].Different forms of homegarden agroforestry are reported in diferent parts of Ethiopia [13][14][15].In southern and southwestern parts of Ethiopia, home garden agroforestry was common and the dominant agroforestry practice [16].
In Ethiopia, tree species integrated into the cropland of smallholder farmers characterized by various forms of traditional agroforestry systems from diferent countries and universal occurrence [17].Many smallholder farmers retained and intercropping trees on their cropland in the central highland of Ethiopia.For example, Cordia africana (Lam) is being intercropped with maize in the subhumid agroecological zone of the country [18].Trees retained or planted on cropland do not have any specifc patterns; for instance, Faidherbia albida, Croton macrostachyus, and Ficus vasta (Forssk) are the most common tree species in parkland agroforestry [19].
Farmers grow or retain trees in their cropland to provide more products or services [8].Multipurpose trees grown in and scattered trees on cropland are parts of agroforestry practice [20,21].Te integration of tree species with annual crops is accounted as the modern way of increasing the production capacity of single land through diversifying the production with higher yields [12].Farmers are maintaining multipurpose trees in their cropland for diferent purposes like frewood, food, fodder, windbreak, or soil fertility [20].Other woody species, like Millettia ferruginea Hochst, Cordia africana, and C. macrostachyus, are useful for improving soil fertility and conserving soil moisture [15,22].According to Desalegn and Zebene [22], farmers maintained scattered trees in their farmland, mainly for their wood products.Cordia africana is one of the best-quality timbers in Ethiopia [23,24].
Farmers in Ethiopia practiced various agroforestry practices in their parklands and homesteads.Smallholder farmers in Ethiopia deliberately retain trees from natural regeneration or planted them in their cropland for diferent purposes of income generation, cultural, material uses, and economical, ecological, social, and biophysical conditions which have livelihood implications [25][26][27].Moreover, dispersed trees grown on parkland agroforestry characterize a large part of the Ethiopian agricultural landscapes [28].However, woody species on parkland agroforestry are getting declined in many agricultural landscapes due to the overuses for fuel wood, charcoal production, expansion of agricultural land, and degradation of nearby forests [29].
In Liban Jawi District, no scientifc information was assessed about woody species composition, diversity, structures, and carbon stock of woody species for parkland and homestead agroforestry practices.Tese issues encourage us to go for study and come out with scientifc information for the continuity of management of these agroforestry practices.Tis study aimed to assess woody species composition and diversity, assess woody species structure, and assess the carbon stock of woody species in the Liban Jawi District of Oromia Regional State, Central Ethiopia.Tis study answers the following research questions: (1) What is the composition and diversity of tree species in parkland and home garden?(2) What is the structure of tree species in parkland and home garden agroforestry practices?(3) What is the carbon stock of tree species in parkland and home garden agroforestry practices in Liban Jawi District?

Description of the Study Area.
Tis study was conducted in Liban Jawi District, West Showa Zone of Oromia National Regional State in Central Ethiopia.Liban Jawi District is located between 8 °50′ 0″ and 9 °10′ 0″ N and 37 °10′ 0″ and 37 °40′ 0″ E (Figure 1).It is located 47 km from Ambo, the capital city of West Showa Zone, and 175 km far away from Addis Ababa, the capital city of Ethiopia, in the west direction.Liban Jawi District is bounded by Toke Kutaye in the east, Celiya in the west, Jibat in the south, and Mida Kegn district in the north direction [30].

Selection of the Study Sites.
In this study, sites were classifed into highland, midland, and lowland agroecological zones based on their altitudinal ranges for data collection.A total of 15 rural kebeles (the lowest administrative area in Ethiopia) existed in three agroecology of Liban Jawi District.Among these, six administrative kebeles were purposively selected based on the existence of woody species diversity in parkland and home garden agroforestry practices.Among selected kebeles in that agroecology, two 2 International Journal of Ecology kebeles of Kombolach Gadi and Haro Mariam were selected from lowland agroecology, two kebeles Liben Gamo and Kombolach Oli were selected from the midland agroecology, and two kebeles Toke Irensa and Munyo Keshembel were also selected from highland agroecology.

Inventory of Woody Species.
To identify parkland agroforestry that had diverse woody species, a reconnaissance survey was carried out for two consecutive weeks on each selected kebele.After identifying parkland agroforestry woody species in each kebele, data on woody species were collected by walking on transect lines designed to lay plots.
For the woody species data collection, the frst plot was taken by using random sampling techniques in the parkland agroforestry practice.In parkland agroforestry practice, a total of 150 plots were taken in 50 m × 100 m (5000 m 2 ) plot areas.Te distance between each transect line was 500 m, and the distance between each plot was 400 m.In home garden agroforestry practices, 70 plots were taken in 20 m × 20 m (400 m 2 ) plot areas.

International Journal of Ecology
Te circumference at breast height (CBH) and height of each woody species in both parkland and home garden agroforestry were measured in all plots.Te DBH of each woody species per plot was measured by using a caliper.Te height of each woody species was measured by using clinometers.All woody species ≥10 cm CBH (circumference at breast height at 1. 37 m from the ground) and tree species were measured for both parkland agroforestry and home garden agroforestry practices, see [31][32][33].From the plots taken in both parkland and home garden agroforestry, all woody species and shrubs found in plots were identifed in local names, counted, and collected for further identifcation.Te scientifc name for each woody species found in both parkland and home garden was recorded in their local feld, and their scientifc names were identifed using the Ethiopia and Eritrea Flora books of volumes 2 to 7 foras of Ethiopia and Eritrea [34][35][36][37][38][39][40].

Woody Species Diversity Analysis. Diversity indices
provide important information about the rarity and commonness of species in a community.In the study areas, woody species diversity was analyzed using species richness (n), Shannon diversity index (H ′ ), and Shannon evenness index (E).
Shannon diversity index is calculated as where H ′ is the Shannon diversity index, Pi is the proportion of total abundance represented by i th species, and ln is the natural logarithm of pi Species evenness is used to measure the homogeneous distribution of tree species in sample plots.It has attempted to quantify unequal representation against a hypothetical community in which all species are equally common [39].Te evenness of a population was calculated by where E is evenness.
Te higher the values of evenness, the more even the species in their distribution within the given area [40]: It is species diversity under maximum equitability conditions, where S is the total number of species in the sample and ln is the natural logarithm of the total number of species in each community.
Te value of the evenness index falls between 0 and 1.

Structural Analysis.
Te basal area is the cross-sectional area of woody stems at breast height.It measures the relative dominance (the degree of coverage of a species as an expression of the space it occupies) of species in an area.It is the area outline of a plant near the ground surface and is expressed in m 2 •ha −1 [41,42].Te basal area was calculated for each woody species in parkland and home garden agroforestry practices with CBH ≥ 10 cm, respectively: Where Π � 3.14 BA is the basal area (m 2 ) and DBH is the diameter at breast height (cm).

Dominance �
total basal area of species sampled area × 100.garden agroforestry using a nondestructive method.Particularly, biomass carbon estimation was carried out using the adopted method of Pandya [43].International Journal of Ecology

Volume of Woody Species in Parkland and Home
Garden Agroforestry.Te height and diameter of the tree at the breast height of each species were converted into biovolume as follows: where D is the diameter (cm) and h is the height of the tree (m).

Aboveground Biomass (AGB) of Woody Species.
AGB is calculated by multiplying three biovolumes with wood density: where ρ is the wood density for woody species and "TBV" is the tree biovolume.
For his study, wood species density from the global wood density database was used [44].

Belowground Biomass (BGB) of Woody Species in
Agroforestry.According to Macdicken [45], the standard method for estimation of BGB could be obtained as 20% of AGB.Te belowground biomass was calculated by multiplying the AGB by 0.2 factors of the root-to-shoot ratio.Te synthesis of global data and the conservative ratio of root-toshoot biomass value were 5 : 1 [46]:

Woody Species Composition and Diversity.
A total of 80 species belonging to 53 families and 62 genera were recorded from the study area under agroforestry (Table1).From the identifed species, 45 were found in parkland agroforestry (Table2) and 35 species were found in home garden agroforestry (Table 3).Among the identifed species in the home garden agroforestry, 16 species were shrubs and 19 species were trees (Table 4).About 17 species were common for both parkland and home garden agroforestry.Among the total families identifed from both agroforestry practices, 29 families were identifed in parkland agroforestry (Table 2) and 24 families from home garden agroforestry (Table 5).Fabaceae and Moraceae were the most dominant families in parkland agroforestry practice (Table 2).Rosaceae and Rutaceae were the most dominant followed by Fabaceae and Lamiaceae families in home garden agroforestry practice (Table 5).About 16 families were common for both parkland and home garden agroforestry practices in the study of agroecologies.Te species richness, Shannon diversity index, and Shannon evenness of woody species in both parkland and home garden agroforestry are indicated in Table (2).Results showed that species richness in parkland agroforestry practice was 25, and the species richness in home garden agroforestry was 13.Species richness of midland, highland and lowland agroecology in parkland agroforestry was 38, 20, and 18, respectively, and in the home garden agroforestry was 20, 10, and 9, respectively.Te highest Shannon diversity index was computed in midland (3.4), but the lowest Shannon diversity index was in lowland (2.3) parkland agroecology practice (Table 6).In parkland agroforestry practice, the overall mean Shannon diversity index was 2.8 and Shannon evenness was 0.54.Meanwhile, in home garden agroforestry, the overall mean Shannon diversity index was 2.53 and Shannon evenness was 0.47.Te Shannon diversity index of midland, highland, and lowland in parkland agroforestry was 3.4, 2.7, and 2.3, respectively.In the meantime, the Shannon evenness of midland, highland, and lowland in parkland agroforestry was 0.62, 0.54, and 0.45, respectively.Te Shannon diversity index of home gardens in midland, highland, and lowland was 3.2, 2.3, and 2.1, respectively, whereas Shannon evenness of midland, highland, and lowland in the home garden agroforestry was 0.51, 0.48, and 0.47, respectively (Table 6).

Efect of Agroecology on Structure of Woody Species.
Results show that agroecology afected the IVI mean value of woody species under both parkland and home garden agroforestry practices (P < 0.05).But agroecology did not afect the mean value of woody species under both parkland International Journal of Ecology

Basal Area of Selected Woody Species in Home Garden
Agroforestry.Azadirachta indica, F. albida, C. lusitanica, and C. anisata were the species with the highest basal area, and J. procera, P. africana, Sesbania sesban, and Millettia ferruginea were woody species in home garden agroforestry in the highland of the study sites (Tables 13 and 14). A. indica, Calpurnia aurea, C. africana, and C. Macrostachyus woody species were with the highest basal area, and O. europaea subsp.cuspidata, J. procera, Podocarpus falcatus, and Spathodea campanulata were with the lowest basal area      16).Total basal areas of home garden woody species in highland, midland, and lowland agroecology were 7.2 m 2 •ha −1 , 23.2 m 2 •ha −1 , and 6.3 m 2 ha 1 , respectively (Tables 14-16).12).

Diameter Class Distribution of Woody Species in
Parkland Agroforestry.Woody species in parkland were classifed into six diameter classes based on their stem thickness.Tese classes included 20-50 cm, 51-80 cm, 81-110 cm, 111-140 cm, 141-170 cm, and 171-200 cm diameter as shown in graphical representation.Distribution of diameter classes between 111 and 140 cm had the highest number of species in parkland agroforestry followed by 81-110 cm diameter classes (Figure 2).In this class distribution, the most dominant woody species that existed were V. etbaica, N. congesta, C. macrostachyus, V. abyssinica, P. africana, F. albida, P. falcatus, and O. europaea,.
Te diameter classes' distribution between 171 and 200 cm has the lowest number of tree species in parkland followed by 20-50 cm diameter class distribution.Te biggest tree species which had the highest diameter DBH were found sparsely in this diameter class distribution in parkland agroforestry.C. macrostachyus, V. abyssinica, P. africana, C. africana, F. vasta, C. africana, and F. albida were the most dominant tree species in this class distribution.Te diameter classes' distribution between 51 and 80 cm had a medium number of tree species followed by a 141-170 cm diameter class distribution.Tese were dominated by V. abyssinica, O. europaea, P. africana, Nuxia congesta, F. sur, F. albida, P. falcatus, and E. capensis woody species in parkland agroforestry.

Diameter Class Distribution of Woody Species in Home
Garden Agroforestry.In this study, only the diameter class distribution for woody species was presented.Species in the home garden were classifed into six diameter classes based on their stem thickness.Tese classes include 10-30 cm, 31-60 cm, 61-90 cm, 91-120 cm, 121-150 cm, and >151 cm diameter as shown in graphical representation.Diameter  3).

Carbon Stock Estimation.
Te aboveground and belowground carbon stocks and total carbon storage of woody species in both parkland and home garden agroforestry were calculated in the study area.

Discussion
4.1.Woody Species Composition and Diversity.In this study, a signifcant amount of species belonging to various families and genera was recorded from study agroecology in the parkland and home garden practices of agroforestry.For example, our study indicated that a higher number of species and families were recorded than the result reported by Molla and Kewessa [50] in Dellomenna of the Oromia region in southeast Ethiopia.Te reason for the highest species richness in midland might be due to the availability of suitable conditions for growing diferent woody species, while lowland has low species richness because many woody species cannot easily adapt or tolerate the condition of lowland agroecology.Tis result is in line with Gochera et al. [51] who reported that species richness was higher in the  International Journal of Ecology midland followed by highland and lowland.However, our study was in disagreement with Tefera [52] who reported that lowland has high species diversity followed by highland and midland agroecology.Te Shannon evenness diversity of midland species in parkland and home garden was higher than in highland and lowland agroecology.Tis indicated that woody species in midland agroecology were sparsely growing or retained in both parkland and home garden agroforestry practices.Te higher the Shannon evenness index, the more even distribution within the given area [53], and the lower evenness index dominates one or few species in the community.A report by Motuma et al. [54] revealed that the Shannon diversity index of woody species in crop felds was 2.22, and Guyassa et al. [55] revealed also that the woody species diversity index in cropland was 2.32 in Ethiopia.In the present study, the value of Shannon diversity in parkland agroforestry was higher when compared with [54,55] reports.

Structure of Woody Species.
In this study, results indicated that midland agroecology for woody species had the highest basal area followed by lowland and highland agroecology.Te results were similar to Gochera et al. [51] who reported that woody species in parkland agroecology had the highest basal area in lowland followed by midland and highland agroecology.Results showed that the basal area midland woody species was found to be the basal area of woody species in order of midland > highland > lowland.Tese results were in agreement with the reports of Latamo and Wondmagegn [24] who reported that a higher basal area International Journal of Ecology was obtained in midland agroecology as many numbers of trees with the higher DBH class.Wood species that had the highest basal area provided multiple benefts like shade and improved soil fertility, fodder, and fuel wood for a household in the study area.Asfaw et al. [56] reported that C. africana had signifcantly higher nutrients in the topsoil underneath its canopy.Tis improves soil fertility and is used for soil and water conservation purposes in the farmland.According to Senbeta et al. [57], woody species had the lowest importance value which may be due to adverse environmental conditions and selective disturbance of humans for resource uses and computation in the study sites.P. africana, F. albida, and C. macrostachyus had the highest importance value index in the three agroecologies of parkland agroforestry.Woody species which had the highest importance value index was the most dominant in the three agroecologies.Tis study was similar to Shibru and Balcha [58] who reported that woody species with the highest IVI are the dominant species in given vegetation.Woody species which had the highest importance value provide shade, timber production, construction material, fuel wood, and as livestock feeds that were benefted by local farmers in study areas.C. africana is one of the best-known indigenous wood species that was used for quality timber in Ethiopia [23,24].
Te highest importance value species in the home garden were ecologically and socioeconomically important for local farmers.Tose woody species were shown in the results of the three agroecologies of the study areas.Mekonnen et al. [15] reveal that woody species with the greatest importance value index were the most dominant species of particular vegetation.Farmers obtained benefts such as fuel wood, livestock fodder, construction material, shade, soil fertility improvement, and human food consumption from important species in their home gardens.According to Gindaba et al. [59], tree species such as C. macrostachyus whose leaves decompose rapidly could supply nutrients in the short term for uptake nutrients by crops.
In this study, results showed that with the increase of total biomass in woody species, there was an increase in carbon dioxide sequestration in woody species in the study agroforestry practices in diferent agroecologies.Carbon dioxide sequestered in the highland was higher followed by midland and lowland agroecologies for woody species.On the other hand, the carbon stored in highland woody species was higher followed by midland and lowland agroecology woody species.Te aboveground, belowground, and total biomass of woody species in the home garden was higher in midland woody species followed by highland and lowland woody species in the study area.It indicates that the midland agroecology of home gardens accumulated a high amount of total biomass followed by highland and lowland agroecologies.In the midland agroecology, woody species had the highest carbon storage and carbon dioxide sequestration followed by woody species of highland and lowland.
Under both agroforestry practices, the largest DBH of tree species in the study area were scarce.Te reason could be due to selective cutting of the trees for diferent uses.Tis study is in agreement with the fnding of Tamirat [60], who reported that the proportion of individual woody species in higher diameter classes was smallest in parkland agroforestry.Te most dominant species in the home garden were small trees because some tree species might be regenerated and other species sprouted were from old trees.Trees and shrubs greater than 151 cm diameter classes had the biggest size and smallest in number followed by 121-150 diameter classes and 91-120 cm diameter classes.In these diameter classes, species C. africana, C. macrostachyus, and F. albida were the most dominant species.Te number of individual species decreased with the increase in the distribution of diameter classes in home garden agroforestry.In this case, the results are in line with Misgana et al. [61] who reported that the total number of woody species in each DBH class decreased with the increasing diameter classes.

Carbon Stock of Woody Species.
Diameter at breast height is commonly used for aboveground biomass estimation because it is simple to be repeatedly measured with high accuracy and generally follows commonly acknowledged forestry conventions [61].Te higher total biomass of woody species in the highlands for parklands was observed than in midlands and lowlands.Tese results were in contrast to reports by Tsedeke et al. [62] who reported that the mean biomass in midland was higher than in lowland agroecology.Te average aboveground, belowground, and total biomass of woody species in highland agroecology was higher followed by midland and lowland agroecology.Te reason could be likely highland agroecology dominant with evergreen woody species due to this reason ABG of woody species in the highland was signifcantly higher than in other agroecology.Te AGB depends on the height and diameter of woody species because the AGB increases with an increase in the diameter and height of the tree.Te relationship between height and diameter is also related to species, climate, soil characteristics, region, and even tree diversity [63].BGB of woody species measured from AGB, due to this AGB of woody species increases with BGB in the parkland study area.Te AGB of woody species increases with increasing altitude.Tis study was in disagreement with Misgana et al. [61] who reported that the aboveground biomass of vegetation decreased with increasing altitude.Te total biomass of woody species in agroecology was higher in highland than in midland and lowland agroecology, because highland agroecology is always dominated by evergreen woody species.

Conclusions
Te present study identifed the availability of 45 woody species belonging to 29 families in three agroecologies of parkland and 35 woody species which were belonging to 24 families identifed in the three agroecologies for home garden agroforestry practice.Te three agroecologies in the study area had diferent woody species composition, richness, and diversity in both parkland and home garden practices.Midland agroecology had the highest species richness in both parkland and home garden followed by highland and lowland agroecology in the parkland International Journal of Ecology agroforestry practices implying the development of more woody species in farmers' farmlands in the midland.Te basal area of trees in midland agroecology both in parkland and home garden was higher in highland and lower in lowland agroecology.Midland agroecology was more dominated by the biggest tree woody species than by highland and lowland agroecologies.Te total biomass carbon stored in highland agroecology of parkland woody species was higher in midland and lowland agroecologies which showed that higher sequestering potential of CO 2 of woody species in the highland than the later agroecologies.Te improvement of woody species diversifcation and sustainability in parkland agroforestry practices needed high awareness of local farmers in the study area.Local farmers who managed woody species in their parkland and homestead can be aware of farmers who had low management abilities.As a result, the cut of trees may be carried out with sufcient regeneration of other woody species.Farmers in the study areas cut woody species only to obtain benefts like fuel wood, charcoal, and construction.However, farmers would have to understand the ecological, economic, and social efects before cutting woody species in the parkland and home garden.Communities have to also predict the environmental consequence of the destruction of woody species.Tus, to transfer the economic, ecological, cultural, and social benefts of woody species in parkland and homestead agroforestry practices to the next generation, farmers should conserve and manage woody species in the parkland and home garden.

Figure 1 :
Figure 1: Map of the study sites.

Figure 4 :Figure 5 :
Figure 4: Te aboveground, belowground, and total biomass of woody species for parkland and home garden practices.

Table 1 :
List of woody species in the study areas.

Table 2 :
Family of woody species collected from parkland agroforestry.

Table 3 :
List of woody species collected from parkland agroforestry in the study area.

Table 4 :
List of trees and shrubs collected from home garden agroforestry in the study area.

Table 5 :
Family of woody species collected from home garden agroforestry.

Table 6 :
Species richness (n), Shannon diversity (H ′ ), and evenness (E) among the three agroecologies of parkland and home garden in the study areas.

Table 7 :
Mean basal area and IVI of woody species in the three agroecologies under parkland and home garden agroforestry practices.Superscripts with similar letters are statistically not signifcant, but diferent letters are statistically signifcant.

Table 8 :
Selected species with the highest basal area under parkland agroforestry in diferent agroecologies.
Note.BA is the basal area, RF is the relative frequency, RDom is the relative dominance, RD is the relative density, and IVI is the importance value index.

Table 9 :
Basal area of parkland agroforestry.

Table 10 :
Parkland agroforestry woody species data collected from highland agroecology.

Table 11 :
Parkland agroforestry woody species data collected from midland agroecology.
computed for woody species of midland in home garden agroforestry (Table15).C. africana, C. macrostachyus, Persea americana, and S. Campanulata woody species were the species with the highest basal area, and S. sesban sinensis and C. aurea were the species with the lowest basal area (Table

Table 12 :
Parkland woody species data collected from lowland agroecology.

Table 13 :
Importance value index of selected woody species under home garden agroforestry.Note.BA is the basal area, RF is the relative frequency, RDom is the relative dominance, RD is the relative density, and IVI is the importance value index.

Table 14 :
Home garden basal area BA (m 2 ) in highland agroecology.

Table 16 :
Home garden basal in lowland agroecology.

Table 17 :
Carbon stored and carbon dioxide sequestered in home garden agroforestry.