A preliminary survey of medium and large‐sized mammals from Lebu Natural Protected Forest, Southwest Showa, Ethiopia

Abstract This study was conducted to determine the species composition and diversity of medium and large‐sized mammals from Lebu Natural Protected Forest, Ethiopia. Surveys were conducted to record mammals through direct observation and indirect evidence from three habitat types, namely: natural forest, bushland, and riverine forest. A total of 15 mammalian species were recorded. The species recorded were Papio anubis, Chlorocebus aethiops, Tragelaphus scriptus, Canis aureus, Crocuta crocuta, Panthera pardus, Procavia capensis, Colobus guereza, Sylvicapra grimmia, Orycteropus afer, Helogale parvula, Hystrix cristata, Lepus fagani, Potamochoerus larvatus, and Phacochoeus africanus. A total of 223 records of observations were compiled. About 74% of these records (N = 167) were obtained from direct sight, whereas the rest was recorded through indirect evidence. The dominant order recorded was order Primates (57.4%) followed by order Artiodactyla (17.5%) while the least record was order Lagomorpha (1.34%). The species richness varied across the stratified habitat types. However, there is no significant difference in Shannon–Wiener Index values between the habitat types. The species diversity of the study area was H′ = 2.119. The present study area is of great potential area for the conservation of the species. Long‐term detailed studies should be carried out for effective conservation and management initiatives in the study area.


| 12323
QUFA And BEKELE loss to mammals (Bernardo & Melo, 2013;Kasso & Bekele, 2014;Wale, Kassie, Mulualem, Tesfahunegny, & Assefa, 2017). To overcome such enormous pressure from the mammals, conserving and managing them in and outside protected areas is a must among the nations of the world.
Mammals are the most diverse and successful group of animals having approximately 5,416 extant species on the globe (Geleta & Bekele, 2016;Reale, Fonseca, & Uieda, 2014). About 320 species of mammals exist in Ethiopia of which 55 are endemic (Lavrenchenko & Bekele, 2017). Ethiopia possesses wide geographic, topographic, and climatic variations (Tefera, 2011). The variety of conditions created in a given ecosystem that harbors diversified-habitats that serves as home to a large number of endemic mammal species (Bantihun & Bekele, 2015;Yalden & Largen, 1992). A basic requirement for determining the status of species is surveying mammals (Keeping & Pelletier, 2014). Mammal inventories are essential tools to efficiently forward conservation strategies and management practices (Legese, Bekele, & Kiros, 2019). There are several previous studies conducted in Ethiopia based on mammals. Yalden (1988) investigated small mammals of Bale Mountains National Park; Yalden and Largen (1992) reviewed the endemic mammals of Ethiopia and Yalden, Largen, Kock, and Hillman (1996) identified Ethiopian and Eritrean mammal fauna in their review and recognize provisional totals of 277 terrestrial and 11 marine species. Woldegeorgis and Wube (2012) surveyed mammals of the Yayu forest in the southwest Ethiopia; Kasso and Bekele (2014) investigated threats to mammals on fragmented habitats around Asella Town, Central Ethiopia and Geleta and Bekele (2016) surveyed medium and large-sized mammals in Wacha Protected Forest, Western Ethiopia. Wale et al. (2017) also assessed the threats to wildlife and its relative severity from Eastern Ethiopia Protected Areas; Kasso and Bekele (2017) assessed the diversity, abundance and distribution of mammals in fragmented remnant forests around Asella Town, Ethiopia and Atnafu and Yihune (2018) investigated the species composition and relative abundance of medium and large mammals in Mengaza communal forest, East Gojjam, Ethiopia. However, still there is a need to conduct studies in fragmented forest of the country.
Even though studies conducted on mammals, mainly targeted National Parks and sanctuaries (Kasso & Bekele, 2014), the survey on fragmented forest and scrubs is scanty. Knowledge regarding the conservation status of Ethiopian mammals is fragmentary (Saavedra et al., 2009) as a vast area remains biologically unexplored due to a major habitat block within the country. A complete inventory of mammals on different ecosystem types of Ethiopia does not exist and is not well documented (Tefera, 2011). Documents on the status and trends of mammals in various protected areas are needed if they are to be protected and managed. Among various fragmented protected forests in Ethiopia, Lebu Natural Protected Forest is one of them, believed to harbor different mammalian species. The main objective of the present study was to determine the species composition, diversity and relative abundance of medium and large-sized mammals from Lebu Natural Protected Forest, Southwest Showa, central Ethiopia. Lebu Natural Protected Forest (between 8°28′42″N and 38°39′24″E) is found in the Southwest Shoa Zone, Oromia Regional State, Ethiopia.

| Study area description
It is located in Sodo Dachi Woreda; about 110 km away from Addis Ababa, the capital city of Ethiopia ( Figure 1). The total area of the forest is around 32 hectares bounded by Gara Molcha Kebele to the north, Kerchufa Kebele to the east, Suten and Tiya Town to the south, and Cheeka Kebele to the west. The area was designated by the Wordea's Natural Resource Department as a protected area.
F I G U R E 1 Map of Sodo Dachi Woreda, Harooma Kebele, Oromia Regional State, Ethiopia A climatic data for the study area of ten consecutive years were obtained (ENMA, 2018). The study area has unimodal rainfall distribution, a long rainy season from June to September and a dry season from November to February. The average mean monthly maximum temperature is 26.85°C, and the average mean monthly minimum temperature is 12.05°C. The mean monthly rainfall of the area varies between 3.5 mm (December) and 346.0 mm (July), while the average mean monthly rainfall of the area is 139.08 mm.
The study area was stratified into three habitat types based on the types of vegetation structure and land cover. These habitats, include: natural forest, bushland, and riverine forest. The dominant or abundant plant species identification was made in situ based on the researchers' previous knowledge of the plant vernacular name and a field guide book. Then, within these strata plant specimens were randomly sampled, observed, identified into taxonomical group.
Besides, a field guide book of useful trees of Ethiopia and the Flora of Ethiopia and Eritrea, written by Bekele (2007) and Phulips (1995), was used to compare the morphological features of plant found in their natural habitat. Accordingly, across the entire study area plant growth forms such as trees, shrubs, and herbs were identified. The identified trees from Lebu Natural Protected Forest were included;

| Methods of data collection
For this study, the study area was stratified into three patches of habitats based on the vegetation structures and topography of the landscapes. These habitats, include: natural forest (6 km 2 ), bushland (2 km 2 ), and riverine forest (8 km along). Diurnal line-transect is a well-recognized and cost-effective methodology for surveying medium and large vertebrates in tropical forests and savannas (Effiom, Nuñez-Iturri, Smith, Ottosson, & Olsson, 2013;Haugaasen & Peres, 2005;Ogutu, Bhola, Piepho, & Reid, 2006). It is one of the best methods for estimating abundance of relatively large and conspicuous mammals (Krebs, 2006). So, combining diurnal line-transect with indirect surveys (including fresh tracks, feces, hair, horns, burrows and digging) can enhance the detectability for many mammal species, contributing to maximize the species lists (Larsen, 2016).
A survey was conducted for seven consecutive days when the activities of mammals were more active: early morning from 6:00 to 10:00 and from 16:00 to 19:00 (Legese et al., 2019). During data collection, the observers were walking on foot along the line transect established and directly count all the individuals sighted with their respective species using unaided eyes and binocular. A total of 12 randomly laid transect line (four of each habitat type) were established to count the sighted mammals. Also, the researcher and two field assistants walk in a transect line and recorded indirect evidence of the animal signs in each sampled area and identified. Surveys were conducted during the dry season from October to December, 2018. Besides, information on lists of species, the altitude, coordinate systems, and vegetation features was recorded by Garmin 72 Geographic Positioning System when the accuracy level read below 10. The distribution of mammals found in Lebu Natural Protected Forest was based on the presence or absence of the species in the habitats categorized. Finally, rarefaction curve can be computed to compare the mean of species richness among the three habitat types from the study area (Colwell, Mao, & Chang, 2004).

| Data analysis
Data were analyzed using descriptive statistics and species diversity index. Mammalian species diversity of the study area was computed using Shannon-Weaver Index of diversity: where H′ is the Shannon index of diversity, Pi is the proportion of individuals of a species in a sample and ln = Natural logarithm (Shannon & Weaver, 1949). Simpson index of diversity was followed (1 − D) using the formula: J = H′/H′ max, where H′ is the observed index of diversity and H′ max = ln(S); S = the number of species in each habitat; ln = Natural logarithm was computed to determine the evenness and dominance among the mammalian species. Finally, relative abundance was used to compute for each species occurrence in the study area.

| RE SULTS
The study compiled 223 observational records, belonging to 15 species and grouped into seven orders and 11 families (Table 1).
Among the seven orders identified, order Carnivora and order Artiodctayla were represented by four species each and the other orders have one species except order Primates represented by three species. Based on trophic guilds, mammal species recorded such as Canis aureus, Crocuta crocuta, Panthera pardus, and Helogale parvula were carnivores and the remaining majority were herbivores. The majority of the recorded mammal species has terrestrial habits. The number of individual observations recorded and there the relative frequency of each mammal species was presented in Table 2. Papio anubis has the highest relative frequency of 33.12%  However, species P. pardus was assured of its presence from the local villagers.
The results of the present study showed that of the 223 total observations, 61.4% (N = 137) was recorded in the natural forest, 28.25% (N = 63) in the riverine forest and 10.3% (N = 23) in the bushland habitats ( Figure 3). The diversity in abundance of records within and between the habitat types were given in Table 4. Also, the diversity of mammal species in each habitat type and of the study area were determined (Table 5). The highest species richness was re-

| D ISCUSS I ON
During the present preliminary survey of large and medium-sized mammals from Lebu Natural Protected Forest, a total of 15 species were identified from 223 total observational records. These mammal species were grouped into seven orders and eleven families.
Similarly, Geleta and Bekele (2016) Total  SC  FP  TC  BH  SC  FP  TC  BH  SC  FP  TC  BH   Common jackal  -2  ---4  ---2  --8   Spotted hyaena  -3  3  --3  3  --3  --15 Olive baboon - F I G U R E 3 Mean number of species richness computed by rarefaction curve among the three stratified habitat types The orders of the mammalian species recorded in the present study were inconsistent with the study conducted in Yayu forest by Woldegeorgis and Wube (2012) as seven orders of mammalian species were recorded here in Lebu Natural Protected Forest whereas only four of the orders namely: Primates, Artiodactyla, Carnivora, and Lagomorpha were observed there. This might account for agronomic practice in the buffer zone, such as some economic activities like collection of forest resources, including coffee that makes Yayu forest accommodating fewer wildlife species than Lebu Natural Protected Forest. The majority of the recorded mammal species has terrestrial habits while very few of them were arboreal. The proportion of large to medium-sized mammals was more than half during the surveys in the study area. According to the conservation concern as per the IUCN (2016) Red List of Threatened species, the leopard, P. pardus global population is decreasing and is listed as a vulnerable species (Stein et al., 2016).
The dominant order in relative frequency record observation (relative abundance) was recorded by order Primates (57.4%) followed by order Artiodactyla (17.5%). From the primate groups, olive baboon has the highest relative frequency 33.12% (N = 74), the most dominant species in the area and P. larvatus having 5.83% (N = 13) was the dominant recorded from order Artiodactyla. These results were consistent with the study conducted in the Mengaza communal forest (Atnafu & Yihune, 2018). This similarity might be due to the availability of a variety of resources in both areas in which most herbivores depended on and particularly olive baboons are well adapted to feed on a variety of food items (Geleta & Bekele, 2016).
From order Carnivora, the highest relative frequency was recorded for C. crocuta 6.73% (N = 15) and the least was for H. parvula 1.79% (N = 4). The low frequency of observation for carnivores might be due to their nocturnal habits, avoidance of their visualization as they are shy and the inaccessibility of the night survey in the study area.
The majority of the species recorded in the present study area has diurnal habits in their activity patterns. This result was inconsistent with the result obtained by Alves et al. (2014) in which out of a total of 239 records, 75% (N = 178) were obtained from footprints.
Olive baboon (P. anubis) was the most abundant species, both in the natural forest habitat type and in the present study area. Similarly, Girma, Mamo, and Ersado (2012) (Mullu & Solomon, 2016). Also, this species might account to use the natural forest to escape from the local people attack to prevent their crops-ride from wildlife.
The second most abundant in the study area and the highest abundant within the riverine forest was C. aethiops. Similarly, Legese et al. (2019) reported that this species was abundant in riverine forests. This might account for the presence of dense riparian vegetation which enables the species to shelter with and easily accesses the product of the trees such as fruits. Besides, Legese and co-authors suggested that the presence of this species at the edge of riverine forest account with a system of feeding on crops from a shorter distance.
The species richness varied across the habitat types stratified.
The mean species richness computed showed that the natural forest has the highest species richness than the rest two habitat types. Similar results were obtained by Geleta and Bekele (2016) as openness of the habitat which might have resulted from habitat loss and fragmentation leading to the species to be easily detected in the natural forest. In the present study, sparsely vegetated natural forest enhanced the detection of mammals. However, a limitation posed by the massive rocky gorge prevented us from making intensive searching throughout the entire course of the natural forest. This might be a reason for the low mammal species record in the area.
Species index of diversity showed that there is a variation in species diversity among the habitats. For instance, bushland forest has the highest species diversity (H′ = 1.96) while the least species diversity was recorded from the natural forest (H′ = 1.56). Similarly, studies conducted by Kasso and Bekele (2017) showed that variation in the number and abundance of mammal species among different habitats is related to the quality of the habitat and preference of the species. Our finding is inconsistent with the study conducted by Geleta and Bekele (2016), in which they obtained higher species diversity in the natural forest. These variations might be due to human pressure on Lebu Natural Protected Forest before it was protected. The overall species index of diversity of the study area showed minimal species richness (H′ = 2.119). However, there is no variation between the habitat types and number of mammalian species recorded. Different possible factors contributed to this. It might be due to lower survey period and habitat fragmentation.

| CON CLUS IONS
The present study area has a representative species of medium and large-sized mammals. Notable previous history of human influence such as intensive deforestation for agricultural expansion, charcoal production, fuelwood collection, and grazing by livestock resulted in alteration of the natural forest in the study area. This finding showed that attention should be given to the varieties of mammal species to avoid any aspect of human pressures on the protected forest. Therefore, joint conservation practice with the local community should be initiated to conserve and enhance the welfare of mammals that occur in the area. In so doing, the entrance of people and their livestock for grazing in the natural forest must be prevented. Long-term comprehensive assessment of mammals needs to be documented and provision of knowledgebased conservation and management initiatives must be given in the area.

ACK N OWLED G M ENTS
We thank the Sodo Dachi Wordea's Natural Resource Department for allowing us to research in the Lebu Natural Protected Forest. We thank Teshome Adugna and Jenberu Bayessa for their assistance during data collection in the field. We also thank the Department of Zoological Sciences, Addis Ababa University for their invaluable support. Finally, we thank the two anonymous reviewers for their comments on the manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no competing interests in conflicts.

AUTH O R CO NTR I B UTI O N S
Chala Adugna Qufa and Afework Bekele conceived, designed the study data collection. Chala Qufa conducted fieldwork, analysis, write the manuscript and revised the whole document. Afework Bekele designed the survey method, edit the manuscript and revised the final version of the main document for submission for potential review. All authors contributed to the writing of the manuscript and approved the submitted version.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data used in this study are archived in the Dryad data repository