Introduction
The extensive central massifs in Ethiopian covers over 80% land area making it the largest continuous highland area in Africa. As a result, the fauna and flora in the Ethiopian highlands are unique that makes it one of the planets’ diversity hotspots (Freilich et al., 2014). Of the over 300 species of mammals recorded from the country 31 are endemic (Largen, 2001) and the high level of endemicity is attributed to the large extent of highlands. Over sixty percent of the recorded mammals are medium and large sized (Bekele and Yalden, 2014).

Following the priority plan of Ethiopian government (poverty reduction) and the fast growing economy of the country in recent decades, attracted more national and international investors in agricultural sector. More investments in agrarian economy demands more fertile lands in most remote wilderness areas. Such activities inevitably affect wildlife. To safeguard the under studied wildlife resources of the country, the Ethiopian Wildlife Conservation Authority and regional governments allocated wildlife conservation areas under different categories (e.g. 21 National or regional Parks, 11 Wildlife Reserves, 3 Sanctuaries, 22 Controlled Hunting Areas and 69 Important Bird Areas) (Young, 2010), yet many are left unrecognized. Despite unique floral and faunal resources, the present study area is grouped under the later categories, only recently, the Oromiya Forest and Wildlife Enterprise (OFWE) proposed it under controlled hunting area category. However, except for the few preliminary surveys conducted by OFWE, no attempt was made to systematically record the mammalian faunal diversity that is crucial both for the management (for assigning appropriate mammals for controlled hunting) (Wihart, 2004), or to document the faunal resource to recommend appropriate conservation strategies (Varman and Sukumar, 1995).

As also the experience of most protected area categories of Ethiopia, the medium and large-sized mammals of the ‘Baroye Controlled Hunting Area (BCHA)’ are facing challenges for the highest hunting pressure, habitat destruction for agricultural expansion, competition for resource with domestic animals and illegal encroachments. Medium and large-sized mammals are most vulnerable and intolerant for the slightest habitat changes, hence used as indicators for habitat quality and stability. The continual deteriorating trends of the status of wildlife in the protected areas are frightening. Unless serious management interventions are taken, the condition becomes very crucial when it comes to the new candidates to protected area categories like BCHA. In any cases, fast recording of the biological resources of the areas precedes and managerial and conservation measures. Therefore, the present study was aimed to record the medium and large-sized mammalian species of BCHA and to determine their association to the diverse habitats as a base line to catalogue the faunal diversity of the area.

1 Results
1.1 Species composition
During the present study, a total of 1 720 (965 and 755 during the dry and wet seasons, respectively) individual mammals distributed in 23 species, 13 families and 7 orders were recorded from Baroye Controlled Hunting Area (Table 1). Among these, only five species: Crested porcupine (Hystrix cristata), Vervet monkey (Chlorocebus aethiopis), Bush hare (Lepus fagani), Rock hyrax (Procavia capensis) and White tailed mongoose (Icheumia albicauda) were considered medium-sized and the remaining were large-sized mammals. The African civet (Civetticitis civetta) was indirectly recorded, using its scat. From all the recorded families, Bovidae contributed the highest (five species), followed by Cercopitheci (four species) and Felidae (three species). While Suidae was represented by two species, the remaining families includinf Hystricidae, Leporidae, Procaviidae, Oryctestidae, Hyaenidae, Canidae, Hyrpestidae, Mustelidae and Viverridae were represented by one species each (Table 1).

Table 1 Medium and large-sized mammalian species recorded from Baroye controlled hunting area, Illubabor Zone, southwest Ethiopia

Seasonal variations were observed in the mammalian species composition and number of individuals among habitats and between seasons. Accordingly, the highest number of species (17) was recorded in the riverine forest during the dry season and the least was from Grassland (10 species) during the dry season (Table 2). Likewise, the highest number of individual mammals (475) was recorded from the riverine forest during the dry season and the least (135 individuals) from the grassland habitat during the same season. Within habitat, the seasonal abundance of mammals was significantly vary for all habitats (grassland: χ² =26.7, 1 df, P<0.05; Woodland: χ²=46.3, 1 df, P<0.05 and Riverine forest: χ² =48.0, 1 df, P<0.05).

Table 2.Seasonal abundance (number of individuals counted) and distribution of mammals among   different habitats in Baroye controlled hunting area, Illubabor Zone, southwest Ethiopia

1.2 Mammalian species richness and evenness
Among the three habitat types, riverine forest supported the greatest diversity of mammals (H’ =2.37) during the dry season followed by woodland during wet season (H’=2.25). The highest species evenness was obtained from woodland during wet seasons (J=0.85) and riverine forest during dry season (J=0.84) (Table 3).

Table 3 Diversity indices of medium and large sized mammals in different habitat types during dry and wet seasons

1.3 Relative abundance of mammals
Among the 23 species of mammals recorded, African buffalo (S. caffer) was the most abundant species (contributing 24.24% and 22.07% individuals during the wet and dry seasons, respectively). With 21.72% (during the wet) and 20.73% (dry season), Olive baboon (P. anubis) was the second most abundant mammal, while Leopard (P. pardus) (with 0.13% and 0.52%) and Lion (P. leo) (0.26% and 0.41%) during the wet and dry seasons, respectively, were the least abundant species (Table 4).

Table 4 Relative abundance of medium and large sized mammalian species recorded in the study area during dry and wet seasons

1.4 Occurrence of mammals
Of the 23 species of mammals recorded for the Baroye Controlled Hunting area, during this study, 9 species (39.13%) were common, 9 species (39.13%) were uncommon and 5 (21.74%) species were rare (Table 5).

Table 5 Occurrence of mammals in Baroye controlled hunting area, Ilubabor Zone, southwest Ethiopia

1.5 Species similarity
Among the three habitat types, the highest mammalian species similarity was observed between woodland and riverine forest both during the dry (SI =0.67) and wet seasons (SI =0.66) and followed by the species between grassland and woodland habitats (with SI values of 0.64 and 0.65 during the wet and dry seasons, respectively). However, with the SI value of 0.48 (during the wet) and 0.31 (during the dry season), mammalian species were least similar between grassland and riverine habitats. 

2 Discussion
The two season survey may not enough to completely record all mammalian resource in such a complex habitat like the Baroye Controlled Hunting Area. The 23 species of medium and large sized mammals during the present study, however, gives good picture for how rich the area was. In some instances, the mammalian species recorded for the BCHA was comparable to the mammalian diversity recorded from some well-established national parks in the country with similar ecosystem (e.g. Ayele, 2008 (unpublished); Girma et al., 2012; Woldegeorgis and Tilaye, 2012; Chane and Yirga, 2014; Gonfa et al., 2015).  This indicates long-term studies by extending the study period and the sampling area may identify additional mammalian species.

The high diversity of mammalian species in the riverine forests and woodland during the dry season, during this study, was expected because these are areas with fresh grass and other leafy plants closer to the perennial river mouths and wetlands associated with the major rivers. Unless the mid-dry season bush fire cleared the debris of dry grasses, finding sufficient fresh grass is difficult for mammals. Most species of mammals recorded for the present study area require water in daily basis (e.g. Yirga, 2008; Redfern et al., 2003), however, accessing surface water in grassland and part of woodland habitats was difficult during the dry season. Such tendency of mammals to favor one habitat over the others following the change in the abundance and quality of resources was also repeatedly reported (Smith, 1992; Mekonen et al., 2011; Yimer and Yirga, 2013). The over flooding of the major rivers and thick undergrowth, observed in the study area, during the wet season probably made the riverine and the majority of woodland habitats less hospitable for mammals.

In abundance, the three primate species, the Olive baboon (Papio anubis), Vervet monkey (Chlorocebus aethiopis) and Colobus monkey (Colobus abyssinicus) stood from the second to fourth level, respectively, in both seasons. They particularly favored the riverine and woodland habitats and Colobus totally absent from grassland. From their high reproductive success, the diversified foraging behavior (Johnson et al., 2012) and for the abundance of alternative prey species for carnivores, the abundance of these species was highly predicted. Studies from different localities, in Ethiopia, reported the abundance of these primate species in riverine and woodland habitats (Girma and and Bekele, 2008; Fetene et al., 2011; Woldegiorgis and Tilaye, 2012).

During this study, some mammals including Leopard (P. pardus), Lion (P.leo), African civet (C. civetta), honey badger (M. capensis) and black backed jackal (C. mesamolas) were least abundant. Information from local informants, however, the former two big cats were abundant and occasionally predate livestock, lions blamed most. As the consequence, there was high human-lion conflict in the area and many lions were persecuted each year. The reason for the present recode may be attributed to their cryptic nature and the demand for high home range. During this study the African civet was blamed for crop raiding, black backed jackals for predating sheep and goats, and honey badger for bee hive breaking. In Ethiopia, the least abundance records were common for these species in different localities (e.g. Chane and Yirga, 2014; Gonfa et al., 2015). 

Mammals in the present study area show no uniform distribution among the three habitats. Hence, their abundance significantly varies among habitats between seasons. More mammals shift from grassland to the riverine and woodland habitats during the dry season. Such seasonal movement in search of resources conditions is common in mammals (Yaba et al., 2011; Girma et al., 2012).

Regarding species similarity among the three habitat types of the study area, the highest species similarity was obtained from woodland and riverine forest both during the dry and wet seasons followed by grassland and woodland. The reason for the observed similarity may be because the resource and cover conditions of these two habitats are relatively similar to each other than either to the grassland habitat. The species similarity was relatively less between the riverine and grassland than the transitional woodland habitat. This record contradicts with the findings by Mengesha and Bekele (2008) and Gonfa et al. (2015) that report high mammalian species similarity between woodland and grassland habitat.

As short term conservation measures, delineating the mosaic borders, specifying appropriate species and setting law enforced quota for hunting may increase the importance of the area as a key wildlife area in the region. As long term conservation measures, however, the national wildlife conservation authority, based on the current and potential biological resources of the area, should design appropriate management plan to upgrade the current status to higher protected area categories with all the logistics and personnel are recommended to safeguard the ecosystem and the resources there in.

3 Materials and Methods
3.1 The Study Area
This study was conducted in Baroye Controlled Hunting Area, Oromia National Regional State, Metu District, southwest Ethiopia. BCHA is located between 36 P740000 and 770000 East and between 36 P928000 and 942000 North and covers an area of 335 km² (Figure 1). The altitude within the study area ranges between 1350 and 1811 m asl. The area is within the southwestern tropical forest best receiving about 1705 mm average annual rainfall (ranging between 1527 and 2015 mm). The average annual temperature of the area is 19.9 oC and shows little variation through the year (EMA, 2014). The area is characterized by rugged landscape dissected by perennial rivers including Sor, Geba, Offa and Chebel. Three habitat types Grassland (60%), Woodland (25%) and Riverine forest, are recognized in the study area, the latter being the smallest (15%). 

Figure 1 Map of the study area

3.2 Methods
Data to study the diversity, relative abundance, and habitat association of medium and large-sized mammals in Baroye controlled hunting area were collected from the three habitats through line transect survey (Varman and Sukumar, 1995; Sutherland, 2007). In rough proportion to the extent of the area, a total 38 line transect were arbitrarily established (18 on the grassland and 10 each on the woodland and riverine forests). The maximum length of transect in grassland was 3.2 km, 2.5 km for the woodland and 3.5 km for the riverine. Each transect was spaced at least by 2 km and the width of each was between 200 m (both for grassland and woodland) and 50 m (for the riverine). From the established transects, 11 (5 from the grassland, 3 each from the woodland and riverine habitats) were randomly sampled and permanently surveyed. Each transect was surveyed twice a month for three months in each season (from January to March, 2014, for the dry and between July and September, 2014, for the wet season). All line transects in a habitat were surveyed at the same time, two individuals per transect and twice in a day, between 06:00 and 10:00 am in the morning, and between 16:00 and 18:00 pm in the afternoon.

All mammals observed along transect lines were identified, counted, sexed and age categorized.  Observation was conducted with naked eyes or aided by binoculars (7x50 mm). Body size, pelage color, presence or absences of horn were used to determine sex and age (Kingdon, 1997; Yirga, 2008; Bekele and Yalden, 2014). Field guide to the African Mammals (Kingdon, 1997), Mammals of Ethiopia and Mammals of Eritrea and Ethiopia (Bekele and Yalden, 2014) were used for mammalian identification. Indirect records such as fecal droppings, calls, track survey for marks and prints, quills, holes, feeding signs were employed to record the presence or absence of mammals (Wilson et al., 1996; Silveira et al., 2003; Swihart, 2004). Indigenous people were also consulted for vernacular name, call and sign identifications.

During this study, using body weight as a basic feature, Emmons and Feer (1997) mammals in the study area were categorized in to medium-sized (those between 2 and 7 kg) and large sized (those over 7 kg body mass) mammals. According to this classification, mammals such as small carnivores and primates, large rodents, hyraxes, and pangolins are grouped under medium and most diurnal primates, carnivores larger than a fox or house cat, all perissodactyla and artiodactyls categorized under large sized mammals. 

Shannon-Wiener diversity Index (H’) (H’=-∑Pi lnPi; where H’ is Shannon index of diversity, Pi is the proportional of individuals of species in a sample) was used to compute diversity of mammals in the study area (Krebs, 1999). The evenness index (J) (J= H’/ H’ max, where H’ is observed index of diversity and H’ max =ln(S), where S is the number of species in a sample) was computed to determine the number of individuals of the mammalian species between habitats and seasons (Krebs, 1999). Species similarity between seasons and among the different habitats was determined using Simpson’s similarity index (SI) (SI= 2C/A+B,) where C is common species in the habitats A and B. A is the number of species observed in habitat A and B is the number of species observed in habitat B (Simpson, 1949). Abundance of mammals was calculated as, Abundance = total number of individual species/ sample blocks (Brown, 1984). Chi-square (χ2) was calculated for each test to determine whether the differences were significant. A mammalian species was labeled as common if the probability of seeing it was 100% every time during each survey or indirect evidences recorded once a day,  uncommon if probability of seeing is more than   50% and/or evidence recorded once a week and rare if the probability of seeing it was less than 50% or only single recorded for the indirect evidences during the whole survey periods (Hillman, 1993).

Author’s contributions
Author 1 and 2 identified the problem and prepared the proposal. The same authors also carried out the field data collection though more weight is given to author 1. All the three authors participated in the preparation of the first draft of the manuscript. The final version of the manuscript is done by author 2 and 3. All the authors have read the final version. 

Acknowledgements
We thank Jimma University for the financial support. We are also grateful to administrative staffs of the Oromia Forest and Wildlife Enterprise, Metu branch and scouts of the controlled hunting area for their valuable guidance and cooperation during the time of data collection in controlled hunting area.

References
Bekele A., and Yalden D.W., 2014. The mammals of Ethiopia and Eritrea, Addis Ababa University Press, Addis Ababa, Ethiopia, pp.391

Brown J., 1984, The relationship between abundance and distribution of species, Am. Nat., 124: 255-279
http://dx.doi.org/10.1086/284267 

Chane M., and Yirga S., 2014. Diversity of Medium and Large-sized Mammals in Borena-Sayint National Park, South Wollo, Ethiopia, Int. J. Scie., 15(1): 95-106

Emmons L., and Feer F., 1997, Neotropical Rainforest Mammals. Chicago University of Chicago Press, pp.380

Fetene A., Mengesha G., and Bekele T., 2011, Spatial distribution and habitat preference of selected large mammalian species in the Nech Sar National Park (NSNP), Ethiopia, Nat. Sci., 9: 80-90

Freilich X., Tollis M., and Boissinot S., 2014, Hiding in the highlands: Evolution of a frog species complex of the genus Ptychadena in the Ethiopian highlands, J. Mole. Phyl. Evol., 71: 157-169
http://dx.doi.org/10.1016/j.ympev.2013.11.015 

Girma Z., Mamo Y., and Ersado M., 2012, Species composition, Distribution and Relative Abundance of large mammals in and around Wondo Genet Forest Patch, Southern Ethiopia, Asian J. Appl. Sci., 5(8): 538-551
http://dx.doi.org/10.3923/ajaps.2012.538.551 

Gonfa R., Gadisa T., and Habtamu T., 2015, Diversity, Abundance and habitat association of medium and large sized mammals in Dati Wolel National Park, Western Ethiopia, Int. J. Bio. Divers. Cons., 7(2): 112-118

Hillman J., 1993, Ethiopia: Compendium of Wildlife Conservation Information, Vol.1, Ethiopian Wildlife Conservation Organization, Addis Ababa, pp.454

Johnson C.M., Swedell L., and Rothman J.M., 2012, Feeding ecology of Olive baboons (Papio anubis) in Kibale National Park, Uganda: Preliminary results on diet and food selection, Afr. J. Ecol., 50(3): 367-370
http://dx.doi.org/10.1111/j.1365-2028.2011.01316.x 

Kingdon J., 1997, The Kingdom Field Guide to African Mammals, Academic Press, London, pp.488

Krebs C.J., 1999, Ecological methodology 2nd ed., Melen park, California: Addison–Wellesley education publisher Inc., 1-85

Largen M., 2001, Catalogue of the amphibians of Ethiopia, including a key for their identification, Trop. Zool., 14: 307- 402
http://dx.doi.org/10.1080/03946975.2001.10531159 

Mekonen T., Yaba M., Bekele A., and Malcolm J., 2011, Food selection and habitat association of Starck's hare (Lepus starcki petter, 1963) in the Bale Mountains National Park Ethiopia Asian, J. Appl. Sci., 4: 728-734

Mengesha G., and Bekele A., 2008, Diversity, Distribution and habitat association of large mammals of Alatish, North Gonder Ethiopia, Curr. Zool., 54: 20-29

Redfern J., Grant L., Biggs H., and Getz W., 2003, Surface water constraints on herbivore foraging in the Kruger National Park, South Africa, Ecol., 84: 2092-2107
 http://dx.doi.org/10.1890/01-0625 
Silveira L., Ja’como A.T., and Diniz-Filho J.A., 2003, Camera trap, line transect census and track surveys: a comparative evaluation, J. Biol. Cons., 114: 351-355
http://dx.doi.org/10.1016/S0006-3207(03)00063-6 

Simpson E.H., 1949, Measurement of diversity, Nature, 163: 688
 http://dx.doi.org/10.1038/163688a0 

Smith R., 1992, Elements of ecology, Harper Collins Publishers Inc, New York

Sutherland W.J., 2006, Ecological Census Techniques: a handbook, 2nd ed., Cambridge University Press, New York, pp.450
http://dx.doi.org/10.1017/CBO9780511790508 

Varman K.S., and Sukumar R., 1995, The line transect method for estimating densities of large mammals in a tropical deciduous forest: An evaluation of models and field experiments, J. Biosci., 20: 273-287
http://dx.doi.org/10.1007/BF02703274 

Vaughan A., Ryan M., and Czaplewaki N., 2000, Mammalogy, 4th ed., Saunders College Publishing, New York, pp565

Wilson D.E., Cole F.R., Nichols J.D., Rudran R., and Foster M.S., 1996, Measuring and Monitoring Biological Diversity, Standard methods for mammals. Smithsonian Institute Press, Washington, DC

Woldegeorgis G., and Wubie T., 2012, A survey on mammals of the Yayu forest in the southwest Ethiopia, SINET: Ethiop. J. Sci., 35(2): 135-138 

Wright S., Zeballos I., Domínguez M., Gallardo M, Moreno H., and Ibanez R., 2001, Poachers alter mammal abundance, seed dispersal, and seed predation in a Neotropical forest, Cons. Biol., 14: 227-239
http://dx.doi.org/10.1046/j.1523-1739.2000.98333.x 

Yaba M., Mekonen T., Bekele A., and Malcolm J., 2011, Food selection and feeding behavior of Giant mole Rat (Tachyoryctes macrocephalus, Ruppell, 1842) from the Sanetti plateau of Bale Mountains National park Ethiopia, J. Asia Appl. Sci., 4: 735-740
http://dx.doi.org/10.3923/ajaps.2011.735.740 

Yimer D., and Yirga S., 2013, Mammals of the Mazie National Park, Southern Nations, Nationalities and Peoples Regional State, Ethiopia, SINET: Ethiop. J. Sci., 36(1): 55-61

Yirga S., 2008, "Atibiwochu", Ethiopian Wildlife and Natural History Society, Addis Ababa, pp.55-61

Young J., 2010, Ethiopian protected areas. A 'Snapshot'. A reference guide for future strategic and project fund, EWCA, Addis Ababa, pp.46