Woody Species Composition, Structure, and Regeneration Status of Gosh-Beret Dry Evergreen Forest Patch, South Gondar Zone, Northeast Ethiopia

. Owing to its variable topographic features, Ethiopia is endowed with rich biological resources. However, nowadays, these vital resources, mainly forests, are declining alarmingly, largely, due to agricultural expansion and energy consumption. Tis study was conducted at Gosh-Beret forest with the objective of investigating the status of forest species. Fifty-one main plots, spaced at a 100m interval, were laid on north-south oriented transects. Within the main plots, fve subplots, at each corner and center, were set to collect data for juveniles. In each plot, individuals of each woody species were identifed and recorded. Percent cover abundance of each woody species per plot was genuinely estimated, which was later converted into the modifed Braun–Blanquet scale. For each mature woody species, diameters at breast height (DBH ≥ 2cm) were measured. Hierarchical cluster analysis was performed to identify community types. Te Shannon–Wiener diversity index and Sorensen’s similarity coefcient were used to compare the species diversity and composition among communities, respectively. Te structure and regeneration status of the forest species were analyzed using structural parameters and size-class ratios, respectively. A total of 52 woody species distributed in 35 families were recorded. Fabaceae, Euphorbiaceae, and Asteraceae were the most dominant families with 4 species each. Te overall species diversity of the forest was 2.6, and fve community types were generated from cluster analysis. In the study forest, frequency and density of species decrease with increasing frequency and density classes. Likewise, density of individuals in each class decreases as DBH classes increase. Te total basal area of the forest was 19.81m 2 ha − 1 , and the forest was at fair regeneration status with species having small IVI values and few/no seedlings. Terefore, immediate conservation measures are required to save species with small IVI values and few/no seedlings.


Introduction
Ethiopia is characterized by a range of variable topographical features; a huge latitudinal spread and an immense altitudinal range resulted in great variation in climate and soils [1]. Tis variety led to the emergence of habitats that are suitable for the evolution and survival of various plant species [2]. Tis makes Ethiopia as one of the few countries in Africa where virtually major types of natural vegetation are represented, ranging from thorny bushes and tropical forests to mountain grasslands. According to [3], the vegetation of the country is very heterogeneous and estimated to contain 6025 diferent species of higher plants, of which about 10% are endemic.
Te vegetation of Ethiopia is, thus, segregated into diferent plant communities that show spatial and temporal variations across landscapes resulting from diferences in environmental gradients. Terefore, the evaluation of variation of environmental variables is essential to understand the factors governing the distribution of species [4]. Variations in the patterns of plant communities with variations in environmental variables were demonstrated in [5]. Such information is relevant to undertake appropriate conservation measures. Besides, examination of patterns of vegetation structures could provide valuable information about their regeneration status that could help devise conservation strategies [6]. Te vegetation structure is maintained by the process of regeneration. Plants regenerate via soil seed banks, seedling banks, and vegetative parts [7]. Many plant species possess combinations of these that widen the ecological range and degree of persistence under fuctuating environmental conditions [8]. Physical and biotic factors (competition, herbivory, and disease) as well as disturbance regimes infuence regeneration processes, and thereby, determine the abundance of species.
Tough forest resources are vital for all lives on earth, it faces several challenges that lead them to diminish rapidly and alarmingly to several folds [9,10]. Te causes for forest loss are many and interconnected. Associated with geometric population growth, energy requirement and demand for agricultural lands increased dramatically that attributed to the major decline of forests. About 94% of the country's fuel demand relies on biomass alone [11], of which trees and shrubs contribute the largest proportion. According to [12], the natural forests of the country have been cleared within the last 30 to 50 years as the demand for energy (for charcoal and frewood), food, and fodder has increased.
Likewise, Gosh-Beret forest, dry evergreen Afromontane forest patch, found in the matrix of agricultural lands and settlements, is facing pressure from agricultural land expansion and fuel wood collection. Tus, the status of this natural forest is dawdling through time as a consequence of the increase of local people pressure for the satisfaction of their daily life. Studying the status of forests is crucial to clearly visualize the anthropogenic activities as well as environmental factors afecting the vegetation of the area. Te lack of such basic information is one of the serious problems that afect conservation of forests [13]. To take appropriate conservation measures, basic scientifc information regarding the status of the forest is, thus, required. Terefore, the objectives of this study were to (a) assess woody plant species composition, (b) identify plant community types, (c) analyze the patterns of the vegetation structure, and (d) evaluate the regeneration status of the forest.

Description of the Study Area.
Te study was conducted at Gosh-Beret dry evergreen Afromontane forest patch in Guna Begie-Midir district, South Gondar zone, Amhara region, Northeast Ethiopia. Kimir-Dingay, the capital of the district, is found 695 km north of Addis Ababa, the capital of Ethiopia ( Figure 1). Te district is boarded by Meketewa in the north, Ebinat in the northwest, East Estie in the south, Lay Gayint in the east, and Farta district in the west.
Meteorological data obtained from the National Meteorological Services Agency (from 2010 to 2019) showed that the mean annual maximum and minimum temperatures were 25.6°C and 6.8°C, respectively, and the mean annual rainfall was 1546 mm. Te study area obtained high rainfall between June and October and low rainfall from March to May. Generally, the study area has a unimodal rainfall pattern, which gradually increases in the periods between February and June, reaches at the highest peak in July and August, and then falls down from September to January ( Figure 2).
According to the Guna Begie-Midir District Agricultural Ofce (2019), the study area is generally characterized by mountainous topography with a cool to moderate climate, locally known as Dega (70%). Te forest area is featured by the slope facing, largely, north to south side with the fat summit, and it is crossed by a big river, Hamus Wonze, which drains into the Ribb Dam.
Like other parts of the country, large parts of the district were covered by continuous vegetation until recently. However, vegetation has been destroyed by anthropogenic activities such as agricultural expansion. Te depletion of vegetation is very rapid, leaving the area with remnant of small patches of forests and shrub lands encircled by large tracts of cultivated lands. Gosh-Beret is one of the dry evergreen montane forest patches characterized by diferent trees and shrubs interspersed with climbers and herbs with few wildlife inhabiting the forest (such as apes, monkeys, bushbuck, klipspringer, antelopes, common fox, hyenas, rabbit, porcupine, and various kinds of birds). Despite destruction of the vegetation around the study area, Gosh-Beret Forest is still, relatively, maintained for long time due to the protection of the forest by the government. Currently, Gosh-Beret forest covers about 435 hectares of land.
Te inhabitants of the study area are from Amhara ethnic groups and speak Amharic language. Te majority of the people are orthodox Christians (99.6%), and the major land use types in the area are farmland (45.3%), pasture (14%), forest with other vegetation types (16%), rural settlement (7.4%), degraded land (2.5%), and other landforms (14%). Human population growth resulted in a shortage of farm and grazing lands. Te economy of local people is predominantly based on subsistence agriculture. Te local people are involved in the collection of forest products (fuel wood and construction material) for domestic consumption and income generation. Tey partly depend on medicinal plants, most of which are harvested from the wild, to fulfll their healthcare needs.  International Journal of Forestry Research 3

Vegetation
A systematic random sampling technique was used to collect vegetation data. Following [14], eight parallel line transects, spaced at 100 m intervals, were laid across the forests in a north-south direction using a Suunto compass. A total of 51 sample plots, with a size of 20 m × 20 m, were laid on the line transects at every 100 m interval. To collect data on seedling and saplings, fve subplots of 2 m × 2 m, one at each corner and one at the center of the main plot, were laid, as used in [15].
In each plot, all woody species were counted and recorded by their local or scientifc names or by codes depending on the familiarity of the species. Woody plant species which were difcult to identify in the feld were collected, coded, pressed, and dried properly using plant presses and brought to the University of Gondar Herbarium where taxonomic identifcation was performed. Any woody plants having circumferences ≥6.3 cm were measured at 1.3 m above the ground so as to calculate diameter at breast height (DBH) following [16][17][18], and later, the circumference was converted into DBH using the formula DBH � C/π, where C is circumference and π is pi. Individual plants with DBH <2 cm were counted as juveniles, and those with height up to 1.5 m were counted as seedlings, whereas those between 1.5 m and 3 m height were counted as saplings, as used in [19]. In each subplot, density of size classes (seedlings, saplings, and adults) was calculated in order to determine the regeneration status of forest species. Te ground cover, in percent, for each woody species in each plot, was estimated genuinely and was later converted into coverabundance values using the modifed 1-9 Braun-Blanquette scale [20,21]. Environmental gradients of each plot were measured as follows: the altitude and the position were measured by using Garmin GPs 60, the aspect was measured using a compass, and the slope was measured using Suunto clinometers.

Data Analysis
2.3.1. Community Classifcation. Hierarchal cluster analysis was carried out based on the abundance of 48 species and 51 sample plots using R statistical software to group the vegetation into plant community types. Te similarity ratio was used to determine the resemblance function, and Ward's method was used to minimize the total within group mean square or residual squares [22]. Te community types which were identifed from the cluster analysis was further refned in a synoptic table, and species occurrences were summarized as synoptic cover-abundance values. Dominant species (where the community named) of each community type was identifed based on their higher synoptic values.

Species Diversity and Similarity among Community
Types. Species diversity of the communities was computed by the Shannon-Wiener Diversity Index (H′) using the R statistical program: where H ′ � Shannon Diversity Index, s � the number of species, P i � the proportion of individuals, and ln � natural logarithm to base n Evenness was calculated using the formula: where J � evenness, H ′ max � H ′ maximum � lns, and s is the number of species [20] Floristic similarities among diferent plant communities and other similar Ethiopian forests were calculated by employing Sørensen's similarity index [20,23] using the formula Ss � (2a)/(2a + b + c), where Ss � Sørensen's similarity coefcient, a � the number of species common to both samples, b � the number of species found only in sample 1, and c � the number of species found only in sample 2.

Vegetation Structure.
Te frequency, density, diameter at breast height, basal area, and Importance Value Index were used to analyze the vegetation structure of woody species.
Frequency (F) was computed as follows: (number of plots in which a species occur/total plots laid) × 100, where the value of each species was, then, sorted in increasing order and grouped into fve frequency classes ( Diameter at breast height (DBH) was calculated from the circumference of each adult woody species using the formula: D � C/π, where D � diameter at breast height, C � circumference, and π � constant with value 3.14.
DBH was, then, classifed into six DBH classes (in cm): Te basal area (BA) was calculated from the value of DBH using the formula:

International Journal of Forestry Research
where d denotes the diameter at breast height in meters and π is a constant � 3.14.

Regeneration Status.
Regeneration status of trees and shrubs was analyzed using density ratios between age classes (seedlings, saplings, and adults). Regeneration status of the forest was stated (labeled) based on the density ratio classes recommended by [24] as "good" if presence of seedling-> sapling > adult trees, "fair" if presence of seedling-> sapling < adult trees, and "poor" if a species survives only in the sapling stage, but not as seedlings.

Species Accumulation Curve.
During vegetation sampling, the number of species accumulates with the increasing area sampled. However, the number of newly added species decreases as the sampling efort continues. Such a trend of species accumulation could be represented graphically using a species accumulation curve. A curve labeled of (or in other cases a curve approximately reaching an asymptote) indicated that no new species would be added if the sampling efort is further continued [25]. Tis could prove that representative samples have been taken from the study area. Te species accumulation curve in Figure 3 is labeled of, indicating that further sampling efort will not bring change in the species number, which proved that representative data were collected from the study forest.

Floristic Composition.
A total of 52 woody species representing 35 families were identifed (Table 1). Compared to other forests (except Wanzaye and Yemrehane Kirstos Church), Gosh-Beret forest has lower species richness ( Table 2). Te reasons for variations in species richness in the study area might be due to excessive anthropogenic disturbances and disparity in conditions for regeneration.
Fabaceae, Euphorbiaceae, and Asteraceae were speciesrich families (4 species each), followed by Lamiaceae (3 species) and Rosaceae, Oleaceae, Sapindaceae, Loganiaceae, Capparidaceae, and Rutaceae (2 species each), whereas the remaining families had 1 species each. Te dominance of Fabaceae and Asteraceae agreed with the report in the Flora of Ethiopia and Eretria. In addition, the dominance of Fabaceae is mentioned in other similar previous studies [18,28]. Tis indicated that the dominance of these families might be due to the adaptation potential to wider agroecologies in various climates with efcient pollination and dispersal mechanisms.
Species in Asteraceae, for instance, are morphologically endowed with parachute-like structures. Tese are adapted for air dispersal and are capable of being blown high up, especially during strong winds [32]. Increased chances of successful establishment of the diaspores after arrival in various areas (refugia) could have allowed the species and genera of these dominant families to establish successful populations. Tis is often possible through physiological and/or genetic adaptation such as increased power of germination as well as ability to withstand the vagaries of extreme climatic conditions and biotic competition for scarce resources, especially during the initial colonization of virgin habitats [33].
Of the total plant species identifed, 38.5% were trees, 46.1% were shrubs, and the remaining 15.4% were liana species. High numbers of shrubby species were also mentioned by some previous studies [26,27]. Lower number of tree species (high number of shrubby species) in the forest could indicate that the forest is either protected recently or still under high pressure of anthropogenic factors.
Endemic species accounted for 5.8% of the total species recorded (Table 3). Te endemic species proportion is lower than the proportion reported in the Flora of Ethiopia and Eretria (10%), Gedo (10.64%) [34], and Kumuli dry evergreen Afromontane (7%) [27]. Tis might be due to the smaller size of the forest and the less topographic heterogeneity of the study area.

Overall Species Diversity and Equitability of the Forest
Species. Te overall species diversity (H′) and evenness of the study forest were found to be 2.6 and 0.66, respectively. Shannon-Weiner diversity is high when it is above 3.0, it is medium when it is between 2.0 and 3.0, and it is low when it is smaller than 2.0 [20]. Tus, the overall species diversity of the forest is medium. Te species evenness value ranges between 0 and 1. When it is 0, the area is dominated by single species, and when it is 1, species are evenly distributed in the area [20,35]. Te result indicated that the species in the study forest are little over 50% evenly distributed, i.e., had medium evenness. Tus, the study forest has medium diversity and evenness.
In addition, compared to other dry evergreen Afromontane forests (Table 4), the diversity and evenness of the forest were low. Tis might be due to local anthropogenic disturbance of the forest for diferent purposes (fre wood,   [20] that would imply the need to conserve the forest to restore its foristic diversity as well as to reduce human pressure.

Plant Community
Types. Cluster analysis of foristic data generated fve diferent community types (Figure 4).

Maytenus arbutifolia-Bersama abyssinica Community.
Tis community type was distributed between the altitudinal ranges of 2325 and 2413 m.a.s.l. It was represented by 14 plots and 30 associated species (Table 5). Most of the plots of the community were situated in the north-west direction. Teclea nobilis and Albizia schimperiana were the dominant tree species. Calpurnia aurea, Vernonia myriantha, and Solanecio gigas were the dominant species in the shrubby layer, whereas Entada abyssinica was the dominant liana ( Table 6).
Since sample plots were close to the surrounding village, they were exposed to human impacts such as collecting frewood, grazing, and charcoal production. Most of the plots, thus, were under the infuence of human encroachment that might relatively result in lower species diversity (2.51), species evenness (0.74), and species richness (30) of the community type.

Euphorbia candelabrum-Acokanthera schimperi
Community. Species in this community were spread between 2386 and 2535 m.a.s.l. altitudinal ranges. Eleven plots represented the community with 34 associated species (Table 5). Most plots were situated in the northwest direction. Teclea nobilis and Albizia schimperiana were the dominant tree species, whereas Calpurnia aurea, Carissa spinarum, Vernonia myriantha, and Dovyalis abyssinicum were common in the shrub layer. Dregea rubicunda, Entada abyssinica, and Clematis simensis were common lianas. Hibiscus macranthus and Prunus africana were indicator species of this community ( Table 6). Relative to community 1, the infuence of human activities is reduced as plots in this community were far from the surrounding village, resulting in a relatively higher species diversity (2.76), species evenness (0.78), and species richness (34).

Albizia schimperiana-Entada abyssinica Community.
Species in this community were distributed in altitudes ranging from 2370 to 2530 m.a.s.l. Fifteen plots, with a total of 45 species, were clustered in this community (Table 5). Te majority of plots were laid in southwest and northwest directions. Along with the dominant species used to name the community, Teclea nobilis and Euphorbia candelabrum were the common tree species in the community. Te dominant shrub species were Carissa spinarum, Maytenus arbutifolia, and Calpurnia aurea. Entada abyssinica was the dominant climber (liana) species. Ficus sur and Myrsine africana were indicator species of this community ( Table 6).
Tis community had 2.81 species diversity, 0.73 species evenness, and the highest species richness (45), which might be due to furthest distances of plots from the village, and most of the plots were found in gentle slopes where soil erosion is low. In addition, local people believed that this part of the forest is the habitat of wild animals including hyenas. Moreover, some plots in this community were found on sloppy and inaccessible sites where human and grazer infuences are reduced, which agreed with the work in [17] who stated that sloppy nature of sites, which are not easily accessible to disturbance through selective cutting and grazing, maintains better species composition and richness.

Vernonia myriantha-Laggera tomentosa Community.
Species in this community were distributed in altitudes ranging from 2395 to 2515 m.a.s.l. It was represented by the smallest number of plots (6) and 30 associated species (Table 5). Te majority of the plots were oriented to southeast directions. Relative to other community types, this community was the most diverse (2.98), with relatively equitable representation of each species (0.87). Tis community is, thus, expected to be stable and healthier. Tis might be due to reduced human and livestock pressure as plots were found in inaccessible and sloppy areas. Te dominant tree species were Albizia schimperiana, Croton macrostachyus, Teclea nobilis, and Olea europaea subsp. cuspidata. Jasminum abyssinicum and Phytolacca dodecandra were common climbers (lianas). Tis community was dominated by shrubby species such as Calpurnia aurea, Maytenus arbutifolia, Clausena anisata, and Dodonaea angustifolia ( Table 6). Selective cutting of tree species for diferent purposes and ongoing ecological succession of the component species might be the reason for the dominance of shrubby species.

Juniperus procera-Acanthus sennii Community.
Species in this community were distributed in higher altitudes ranging from 2490 to 2560 m.a.s.l. Five plots with 23 associated species were included in this community (Table 5). Te majority of the plots were oriented to the southeast and south directions. It has the second highest species evenness value (0.83) and the least species richness (23). Te least species richness might be attributed to agricultural land encroachment as plots were located around crop lands. Te dominant tree species included in the community were Euphorbia candelabrum, Acacia abyssinica, and Teclea nobilis. Carissa spinarum, Calpurnia aurea, and Maytenus arbutifolia were common species in the shrub layer. Entada abyssinica was the common climber in the community (Table 6).

Species Composition Similarities among Communities.
Te overall similarity coefcient among communities ranges from 63 to 84% (Table 7). Sorensen's similarity coefcient was used to detect similarities among plant communities. Based on this, similarity in species composition slightly varied among communities. Relatively, the highest similarity was observed between communities 1 and 2 (84%), which might be due to close proximity of the communities or exposure to similar environmental factors that led to similar adaptation. Te least similarity was observed between communities 1 and 5, 2 and 5, and 3 and 5 (63%). Tis might be due to diferences in altitudinal gradients or environmental or anthropogenic factors, leading them to have relatively lower similarities. 8 International Journal of Forestry Research   P1  P8  P2  P37  P38  P6  P23  P22  P3  P4  P9  P7  P20  P21  P19  P35  P28  P29  P30  P44  P5  P17  P24  P31  P32  P41  P42  P14  P43  P45  P16  P10  P12  P15  P13  P11  P39  P33  P27  P36  P18  P34  P46  P40  P25  P26  P47  P48  P50  P49 Table 8). Te frequent occurrence indicates the successful reproduction and adaptation of a species to the area [4]. Dispersion of species was classifed into 5 frequency classes ( Figure 5). Te trend showed a higher number of species (66.7%) in frequency class 1, followed by a continuous decrease in the number of species up to frequency class 4 from which the species number started to increase till frequency class 5. Frequency indicates the homogeneity and heterogeneity of a given vegetation in which the high number of species in higher frequency classes and the low number of species in lower frequency classes show similar species composition (higher homogeneity), while the large number of species in lower frequency classes and the small number of species in higher frequency classes indicate higher heterogeneity (Lambrecht, 1989). Tus, from the result, it can be said that the vegetation of the study forest is more or less heterogeneous. Similar results were reported by previous studies [17,38]. Heterogeneity of vegetation might be attributed to habitat preferences among species, species characteristic for adaptation, degree of disturbance, and availability of suitable conditions for regeneration [39].
Species that contributed the highest number of individuals ha −1 was Carissa spinarum (809.72 ha −1 ), followed by Calpurnia aurea (570.58 ha −1 ), Teclea nobilis (519.48 ha −1 ), and Maytenus arbutifolia (515.24 ha −1 ) (Table 8), respectively. Te highest number of individuals ha −1 could mean that the species is under good reproduction and recruitment despite the presence of trampling, grazing, and other anthropogenic disturbances. Te density distribution of species in the density classes showed more or less zig-zag patterns ( Figure 6). A similar result was also reported by the authors of [17] who pointed out that such a pattern showed selective cutting of species for diferent purposes.

Diameter at Breast Height (DBH) Distribution.
A total of 1586.75 individuals ha −1 of woody plant species (DBH ≥2 cm) were recorded from the forest. Te DBH distribution was classifed into 6 DBH classes. Most of the individuals were distributed in the lowest DBH classes, showing a continuous decrease towards higher DBH classes. Such a distribution of individuals represents an inverted J-shape curve (Figure 7). A similar pattern was reported by previous     [17,28,38,41]. Tis implies a good reproduction and recruitment status of the species in the forest [43].

Population Structure for Some Selected Woody Species.
Te analysis of the population structure of the selected woody species based on density of DBH resulted in four patterns (Figures 8(a)-8(d)). Te frst pattern was inverted Jshape. Tis pattern was also reported by several previous similar studies [28,30,41,43]. Such a structure showed good reproduction and recruitment [43] (Figure 8(a)). Plant species showing this pattern were Teclea nobilis, Capparis tomentosa, Dombeya torrida, Entada abyssinica, and Calpurnia aurea.
Te second pattern was a bell-shaped distribution (Figure 8(b)) where lower classes have a lower number of individuals followed by an increase in the number of individuals towards middle classes and then a progressive decrease towards higher DBH classes. Te woody plant species included in this pattern were Albizia schimperiana, Nuxia congesta, Euphorbia candelabrum, Juniperus procera, Ficus sur, Allophylus abyssinicus, Ekebergia capenis, Olea europaea, and Croton macrostachyus. Similar results were shown in [30,43]. A bell shape follows a Gaussian distribution pattern indicating poor reproduction and recruitment of species, which might be associated with the overharvesting of seed-bearing individuals or the presence of only some seed-bearing individuals [43].
Te third pattern consisted of those species occurred only in lower DBH classes (classes 1 and 2). Tis type of population structure was also reported in [19] that suggested the good reproduction and bad recruitment status of the species. Te species in this pattern include Acokanthera schimperi, Acanthus sennii, Capparis tomentosa, Clausena anisata, Clematis simensis, Dodonaea angustifolia, Maytenus arbutifolia, Bersama abyssinica, and Osyris quadripartita (Figure 8(c)). Te last (4 th ) pattern was represented by only Ficus sur (Figure 8(d)) that occurred only in the medium DBH classes (10.1-20 cm and 20.1-30 cm). Tis may be due to low regeneration status of the species and removal of the matured individuals of the species for construction materials.
3.5.5. Basal Area. As calculated from DBH data, the total basal area (BA) of adult woody species of Gosh-Beret forest was 19.81m 2 ha −1 . Te normal value of the basal area in Africa is expected to be between 23 and 37 m 2 ha −1 [23]. Tis could mean that the basal area of the study forest is lower than the normal value.
Te basal area of Gosh-Beret forest was also compared with the basal area of other 5 dry evergreen Afromontane forest patches in Ethiopia (Table 9). Te basal area of Gosh-Beret forest was higher than that of Yegof and Amoro forests, but it was lower than that of Tara Gedam, Alemsaga, and Gedo dry evergreen Afromontane forests. A lower basal area value would mean that the study forest is composed of shrubby and young tree species that revealed that the forest is either protected lately or still under anthropogenic pressure. Albizia schimperiana accounted for the highest basal area value (8.99 m 2 ha −1 ), followed by Euphorbia candelabrum (2.23 m 2 ha −1 ), Croton macrostachyus (1.79 m 2 ha −1 ), Carissa spinarum (1.6 m 2 ha −1 ), and Calpurnia aurea (1.07 m 2 ha −1 ) ( A high IVI value indicates that the species is most successful in regeneration and pathogen resistance and grows in the shade, and in comparison with other species, these species are least preferred by browsing animals and seed predators and have high attraction of pollinators that facilitate seed dispersal within the existing environmental conditions of the study area [44]. Tus, these species are the most dominant ones in the forest [39]. Such species may not need immediate conservation measures, but they need regular monitoring. On the other hand, species with a low IVI value would mean they are at the risk of local extinction that needs priority of conservation measures.

Regeneration Status of Gosh-Beret Forest Species.
Composition and density of seedlings, saplings, and adults would indicate the regeneration status of the forest species. Te density of adult individuals was greater than that of seedlings (0.97 : 1) and saplings (0.68 : 1), whereas the density of seedlings was greater than that of sapling (1.42 : 1). Tis implies that the forest is grouped under fair regeneration status. In some of South Gondar Zone Church forests, similar results were reported in [1], but some other authors [4,28,30,31] reported good regeneration status in their study forest patches. Te highest sapling density was recorded for Carissa spinarum (215.7 individuals ha −1 ), followed by Teclea nobilis (192.8 individuals ha −1 ) and Calpurnia aurea (142 individuals ha −1 ). Te highest density of saplings in Carissa spinarum might be due to special survival mechanisms having spines, and the plant by its nature cannot reach big trees that people do not use for charcoal and timber production.
Species that had poor regeneration status include Buddleja polystachya, Capparis tomentosa, and Dregea rubicunda. Some of the species that had fair regeneration status include Acokanthera schimperi, Allophylus abyssinicus, and Bersama abyssinica. Species that had good regeneration status include Maytenus arbutifolia, Acanthus sennii, and Croton macrostachyus. Tree species without juveniles need immediate conservation measures since they are at a higher risk of local extinction.

Conclusion and Recommendations
Te result of the study revealed that the forest harbors considerable numbers of plant species that act as in situ conservation sites for some endemic as well as indigenous species of the country. Te structural data analysis result, basal area, and DBH showed that the forest is composed of largely young individuals of tree and shrubby species, which revealed that either the forest is secondary forest or the forest is protected lately. Te result from the analyses of IVI and regeneration status indicated that there are species which have very small IVI values and or low numbers or no seedlings that lead to the conclusion that the species are at risk of local extinction. Te overall regeneration status of the forest showed that most woody species of the forest have fewer juveniles than adult plants that imply forest species are in fair regeneration status, which need management supervision.
Terefore, in order to ensure the sustainable utilization of the forest resources, the following recommendations are suggested: (i) Immediate conservation measures are required for those species with very small IVI values such as Hibiscus macranthus Myrsine africana, Prunus africana, Ritchiea albersii, and Clerodendrum myricoides and species with low numbers of seedlings or species which lacked seedlings though safety net programs of the country. (ii) Local people shall be encouraged about planting of indigenous and suitable exotic tree species in the agroforestry systems and waste lands in order to reduce the pressure on the natural forests via the green legacy initiative of the government (iii) Continuous awareness enhancement is required to develop a sense of ownership in the local people, which ensures the long-term maintenance of the forests. Tis can be done during social gathering events and religious ceremony.

Data Availability
Some of the data used to support the fndings of this study are available in the fgure fles (fgures and tables) and the remaining data (fgures and tables in Excel) are available from the corresponding author upon request.

Disclosure
Te research was performed as part of the employment of the authors, which is considered one of the regular works of the employees for the employer organization, University of Gondar.

Conflicts of Interest
Te authors declare that they have no conficts of interest.