Bushland habitat
Bushland is locally named “Berwsirt”. Its altitude ranges from 2825 to 2891 m asl. This habitat type comprises deciduous and evergreen shrubs mainly Dedonaea viscose intermixed with grasses and other herbaceous vegetation forming the understory. Herb species consumed by the geladas such as Trifolium temnense and Plantago lanceolate are rarely found (DWANRDO, 2016).
Sampling design and data collection
The study area was stratified into five dominant habitat types based on the dominant vegetation types (Figure 2). The area coverage of each habitat type was as follows; grassland (336 ha, 24.45%), wooded grassland (226 ha, 16.45%), natural forest (dry evergreen Afromontane forest) (243 ha, 17.69%), plantation forest (251 ha, 18.27%) and bushland (318 ha, 23.14%). Proportional to the area coverage of each habitat type, each habitat was further subdivided in to counting blocks. Grassland and bushland habitats were sub divided in to 6 and 5 blocks respectively, whereas plantation forest, natural forest and wooded grassland were subdivided into 4, 4 and 3 blocks respectively. Each habitat type was subdivided into counting blocks using natural boundaries and marks such as mountains, cliffs, springs, valleys, etc.
Data collection was carried out from August 2017 to February 2018 covering both dry and wet seasons. The wet season data collection was carried out from August 2017 to October 2017 and the dry season data collection was carried out from December 2017 to February 2018.
Total count method was employed to census the population of gelada in the study area. Total count method has been widely practiced to census the population of geladas over its ranges (Beehner et al. , 2008; Adem, 2009; Ayalew, 2009; Abu, 2011; Kifle et al., 2013; Moges, 2015; Goshme and Yihune, 2018; Girmay & Dati, 2020), due to their behaviour to live in open habitat that increases visibility and due to their highly mobile nature. The total census was carried out in each habitat type on foot simultaneously from a suitable vantage points or by moving along the habitat through involving well trained people in each habitat types to avoid double counting problems (Fashing and Cords, 2000; Mekonen, Bekele, Hemson & Alemu 2010). Since the simultaneous counting of the individuals geladas demands the involvement of groups of people, in each of the five habitats and counting sub blocks; people from the locality were selected and trained. A total of 40 people were involved in the data collection. One day theoretical and two days practical (on site) training was given to the trainees on the methodology of counting geladas, age and sex identification techniques as per following Beehner et al. (2008). All trainees’ received a secondary school education and had prior experience about the species identification and its behavior. A census was conducted only when gelada start moving to the cliff edge and no individual remained on the cliff-edges following Hunter (2001). All gelada population counting was conducted on the same day and at the same time during early morning (7:00 to 11:00 a.m.) following Beehner et al. (2008) and Mekonen et al. (2010).
Determination of age and sex structure was carried out according to the procedures of Beehner et al. (2008). The sex and age of individuals of gelada were identified based on body size and developmental characteristics of individual gelada in the troop, this was done using 7×40 Binocular and with the naked eye through direct observation of individuals of geladas. Adult gelada males were identified by their visible manes and their size about twice bigger than that of adult gelada females. Sub adult males were defined as males equal in size to adult females gelada, but with partially developed mane. Sub adult female geladas were distinguished from adult females due to the smaller body size (Beehner et al., 2008). Young geladas sex identification was found difficult, so were recorded as juvenile with no sex discrimination. In addition, the group composition and size of the group were recorded before treating the group in to respective age and sex categories and recorded on a separate date sheet as: adult males, adult females, sub-adult males, sub-adult females and juvenile (young). A group of gelada is a multilevel social order that consists of individuals of gelada usually composed of a leader male, several adult females, and their offspring (Gippoloti and Hunter, 2017). Those individuals seen within a distance of < 50 m from the nearby group was recorded as members of the same group. One-male units (OMUs) group is defined as a group of gelada that consist of a leader male, several adult females, and their offspring. Whereas, all-male units (AMUs) group is defend as group of gelada that composed of 2–15 males (Gippoloti and Hunter, 2017). A band of gelada is defined as group of geladas that is made up of two to 27 reproductive units and several all-male units (Gippoloti and Hunter, 2017).
To determine the distribution pattern of gelada intensive ground inspection was conducted in each habitat type and in each block during the continuous daily counting time periods following Groves (2005). The GPS location (coordinate points), group size, altitude and habitat type were recorded. During the survey 6 to 8 coordinates points of all non-over lapping range groups of geladas were recorded.
Data analysis
Statistical Package for Social Science (SPSS) version 16 computer software program was used for all statistical analysis. All statistical tests were two tailed with 95% of confidence intervals and the rejection level were (P ≤ 0.05). Mann- Whitney- U test was used to test the population count among wet and dry seasons. Chi square test was used to test the association of population density with season (among dry and wet seasons). One way ANOVA was used to test the variations in population structure (age and sex structure) among seasons. During the total count identifying the group of gelada and counting the number of individual that form the group was a base for estimating the population density of gelada in the study area (Bocian, 1997).
\begin{equation} \text{Population\ Density}=\frac{Number\ of\ individuals\ that\ were\ counted\ in\ the\ area\ldots\ldots\ldots Eq\ 1\ }{\text{Total\ area}}\nonumber \\ \end{equation}
The population density of each habitat type in both wet and dry season was calculated and expressed as the number of individuals per km2.
Non overlapping ranges of groups of baboons were mapped using coordinates collected using GPS and using the help of Arc GIS software. In addition, the population count at each dominant habitat type was used to describe the distribution pattern. To determine the distribution pattern of wild animal populations, different researcher used the relative frequencies of observation of the animal in each habitat type (Yazezew, Mamo & Bekele, 2011; Abie and Bekele 2017; Goshme and Yihune 2018).