Population Ecology and Possible Threats to Defassa Waterbuck (Kobus Ellipsiprymnus Defassa) in the Nechsar National Park, Ethiopia

Because of potential conicts with humans, we investigated the population structure of waterbuck (Kobus ellipsiprymnus defassa) in the Nechsar National Park, Ethiopia, from November 2019 to August 2020, using line transect methods. e estimated population was 118 and 104 in the wet and dry seasons, respectively, and averaged 111 (+8) or 0.23/km. e proportion of each age group averaged 31% adult males, 45% adult females, 12% subadult males, 11% subadult females, and 2% juveniles; the large number of female individuals suggests a healthy and increasing waterbuck population. e distribution of waterbuck sightings among habitat types diered by season (χ 15.97, df 3, p< 0.05). Dierent conservation measures should be taken to create a conducive environment for waterbuck and large park biodiversity.


Introduction
Large herbivores, especially in Africa, have signi cant ecological and economic value [1], but increasing human activities are leading to the destruction of many habitats [2]. Since 1970, there has been a marked increase in the number of protected areas being established [3], and it is crucial for conservation e orts that protected areas are well-managed by maintaining wildlife populations within ecologically sustainable limits [4][5][6].
Waterbuck, the largest of the genus Kobus, are one of the heaviest antelopes (250-270 kg) and di erent from other antelopes in having long legs and fur [7], as well as an a nity for water sources. eir range-wide population has been greatly reduced, but signi cant numbers survive in protected areas [8]. e Nechsar National Park is an Ethiopian protected area established in 1974 to conserve the endemic Swayne's hartebeest (Alcelaphus buselaphus swaynei) and other key species, as well as preserve its inherent beauty.
However, in recent decades, there has been an observed decline in the population of Defassa waterbuck (Kobus ellipsiprymnus defassa), which were once numerous in the park.
e population status of wildlife in protected areas is gradually declining due to grazing, invasion, growing agricultural demands, climate change, and management problems for biodiversity conservation [9], and humanwildlife con ict has been documented elsewhere in Ethiopia [10]. In and near the Nechsar National Park, a signi cant number of people have been observed collecting rewood and harvesting sh [11]. In addition, huge numbers of cattle were observed grazing together with wild animals inside the territory of the park. As a result, negative attitudes amongst the local Guji people regarding livestock grazing have led to frequent con icts between the farmers and the park managers in the area [12,13]. us, the current study aims to investigate and document the current population status and possible threats to Defassa waterbuck in the park.

Description of the Study Area.
e 514 km 2 Nechsar National Park (NSNP) is situated 510 km south of Addis Ababa in the center of the Ethiopian Rift Valley (5°51′ to 6°10′ N, 37°32′ to 38°48′ E) between two major southern lakes, viz., Abaya and Chamo [14] (Figure 1). Within the park, water bodies cover 78 km 2 , complemented by a diversity of forest, grassland, open woodland, and fresh water habitats [14,15], occurring at elevations ranging from 1,100 to 1,600 m [16]. Annual rainfall follows a bimodal pattern with short rains from March to May and a main rainy season from September to November. e average annual rainfall and temperature are 919 mm and 24°C, respectively [17].
According to NSNP [20], 5% of the households in Gamo Gofa Zone reside within the territory of the park. ese households, and others located at the periphery, access the park via illegal routes. Human activities that are believed to threaten the normal activities and habitats of waterbucks include fuel wood collection, charcoal production, illegal grass collection, poaching, overgrazing, deforestation, settlement, and illegal fire. To carry out their protection duty, protection staff undertakes the patrol throughout the day via vehicles and on foot where car access is not possible.

Methods of Data Collection.
A preliminary survey was conducted in the study area before starting the actual data collection. e survey helped to stratified study area into different habitat types, identify topography, climate, infrastructure, and approximate size of the core habitat of Defassa waterbuck in the study area.
For study purpose, the study area was divided into four blocks of habitat based on vegetation types. For observation and counting purpose, transects of varying lengths were laid in each habitat. e transects' placement in each habitat type (block) was determined by the size of the habitat type, and more importantly, by the animal's abundance (density) expected to be found in each habitat [21]. Moreover, transects running along roads, ridge tops, and stream bottoms where disturbance might be apparent were avoided. e four major habitat categories in the study area were as follows: Block 1 (grassland): this habitat type covered an estimated area of 57 km 2 , where a total of 13 transects of varying lengths were laid. Block 2 (wetland): this habitat type covered an estimated area of 21 km 2 , where nine transects of varying lengths were laid. Block 3 (ccattered vegetation with bush land): this habitat type covered an estimated area of 50 km 2 , in which eleven transects of varying lengths were laid. Block 4 (riverine forest): this habitat type covered an estimated area of 78 km 2 , where seven transects of varying lengths were placed.
An in-depth data collection was undertaken from November 2019 to August 2020 to cover both the dry and wet seasons. e line transect counting method was employed. Moreover, a direct observation of the Defassa waterbuck was carried out to collect the required data. e survey was conducted on foot at an average speed of 1 km/h in the riverine forest, scattered vegetation with bushland and wetland, and 2 km/h in the grassland habitat. A silent detection method was carried out where trained observers searched carefully using binocular lenses along the center of the transect line at all times and were assured that animals on the center line of the transect were seen with certainty [22]. During transect walking, the observers recorded the start and end time, start and end of GPS locations, and GPS ID. Each time the Defassa waterbuck is seen, the observers recorded the following: time using standard wrist watch; GPS location by GPS; herd size, sex, number of Defassa waterbuck seen, and age using binocular eye inspection; perpendicular distance by using a binocular reticule; transect length, and habitat type where the animal is observed. All measurements were taken at the animal's original location.
To ensure that appropriate sampling effort is made, each habitat was surveyed twice using distance sampling method in each season [23]. e survey was carried out early in the mornings between 6 : 00 and 9 : 00 a.m. and in the late afternoon from 16 : 00 to 18 : 00 p.m. by the time when the Defassa waterbucks were active [24]. During counting, trained scouts were assigned to each habitat to record the number of individuals, season, age, and sex of the animals observed. To avoid double counting, each member, and/or herd size, and specific activities of Defassa waterbucks in different locations were noted [22]. e sexes of waterbucks were identified based on relative body size, external genitalia, and presence or absence of horns. Moreover, the age groups of waterbucks were classified as less than one year for juveniles, one to six years for subadults, and greater than six years for adults based on the observed relative body size [25]. e total number of individuals, group size, age structure, and sex ratio in each habitat was recorded in both seasons [26]. e existing possible threats to the Defassa waterbuck were surveyed using a semistructured interview. Eighteen protection staff (guards) and three management staff (totally 21 interviewees) were purposively selected because they were believed to give accurate information about the existing threats to the Defassa waterbuck species, as they usually explore the park's region. e selected interviewees were gathered, and each one of them was interviewed about the aspects he/she believed to have a negative impact on the normal activities of the Defassa waterbuck. ey were asked about the existing anthropogenic activities and natural events operating in the park and how they negatively in uenced the Defassa waterbuck.

Data Analysis.
e data analysis was done using Distance software version 7.3 and SPSS version 26 [23] (Buckland et al.). Before undertaking actual data analysis, preliminary analysis was done to check the data normality and model selection.

Data Normality Test.
e Shapiro-Wilk's normality test (p > 0.05) [27,28] and the visual inspection of their histograms, normal Q-Q plots, and box plots revealed that the sampled animals were collected from a normally distributed population in each habitat, with a skewness of 0.618 , and B4 (Riverine forest), respectively. e p value of each habitat is within ±1.96, which shows that the studied animal is normally distributed. In addition, the homogeneity of variance of the sampled animals in di erent habitat types was computed using Levene's Test. e test result F (3, 126) 0.859, p 0.464 showed that there was no signi cant variation in the variance of the sampled animals in each habitat type. erefore, equal variances were assumed in each habitat.

Model Selection.
Model selection was done in Distance software to discriminate between various candidate models (the uniform cosine, the half-normal cosine, and hazard rate), and the model that best ts the collected data was  International Journal of Zoology selected. e value of Akaike's Information Criterion (AIC) and the shape criterion of the detection function computed for each model were used as the major criteria for this selection. e model selection analysis result revealed that the half-normal cosine model was the best fit among the other candidate models as it gave the lowest AIC value (Table 1), and normal detection function (Figure 2) relative to the other models. erefore, all the computation related to detection probability, density, and abundance of the Defassa waterbuck presented in the current study was done using the half-normal key with cosine adjustment.
Descriptive statistics was used to compare the population size of waterbuck, overall sighting of sex ratios, population status, and the mean numbers of individuals. A chi-square test was used to test the seasonal difference in population structure of the Defassa waterbuck. e variation of population size within each habitat type (block) was analyzed using chi-square and ANOVA across seasons in the study area.

Population Estimate of Waterbuck in NSNP.
e population of Defassa waterbucks in the Nechsar National Park was estimated to be 111, with 104 and 118 individuals in the dry and wet seasons, respectively (Table 2). e average number of waterbucks recorded throughout the study period was 16 ± 5.31 individuals, with 17 ± 8.12 and 15 ± 2.5 individuals in the wet and dry seasons, respectively. e average population density was estimated to be 0.23/km 2 . e populations of waterbucks in each habitat type show no significant variation in both seasons (two-ways ANOVA; F � 2.12, p > 0.05).

Age Structure of the Defassa Waterbuck.
Of the 15 individual waterbucks recorded during the dry season, 11 were adults, 3 were subadults, and 1 was a juvenile. Similarly, out of the 17 waterbucks recorded during the wet season, 12 were adults, 4 were subadults, and 1 was a juvenile. e age structure of the waterbuck population was dominated by adults during both seasons. On average, 76% were adults, 22% were subadults, and 2% were juveniles. ere was no significant difference among age groups (χ 2 � 0.32, df � 2, p > 0.05) during the dry and wet seasons. Proportionally, adults of both sexes have the highest number, while juveniles have the least number (Table 3). Even if the number of adult females was higher than juveniles of unknown sex during both seasons, there was no significant variation in sex ratio during the wet season (χ2 � 2.94, df � 4, p > 0.05).

Distribution of the Defassa Waterbuck.
e distribution of waterbucks in each habitat type was almost uniform, but juveniles and subadults were mostly found in grassland and scattered vegetation ( Table 2) (χ 2 � 4.02, df � 6, p > 0.05).
erefore, there was no significant variation between age groups and habitat types during both wet and dry seasons. Moreover, the distribution of waterbuck is highly associated with the presence of permanent water sources (Table 4).

Sex Ratio of Waterbuck in Wet and Dry Seasons.
Among all the ages and sexes of the Defassa waterbuck, the maximum and minimum sex ratios were 1 : 1.94 adult male to adult female and 1 : 0.02 adult to juveniles, respectively, in both seasons. erefore, there was no significant variation in sex ratios both in the dry and wet seasons (Table 3).

reats to Defassa Waterbucks in the Nechsar National
Park.
e survey results revealed that the major threats to the existing wildlife population in general and the Defassa waterbuck in particular are anthropogenic activities. e frequent human activities and their relative intensities are observed to have a negative impact on the normal behavior of the Defassa waterbuck. is is summarized in Table 5. e interview results revealed that fuel wood collection, illegal hunting (poaching), illegal grass collection, and overgrazing were the most frequently observed threats with destructive effects on waterbucks and wildlife population, as well as on the biodiversity in general.

Discussion
e population of Defassa waterbucks in the Nechsar National Park was estimated to be 111, with 104 and 118 individuals in the dry and wet seasons, respectively. e large number of waterbucks recorded in the wet season is in agreement with the result of the study conducted by Fetene et al. [15] that a large number of waterbucks were recorded in the wet season in the same study area. e possible reason for the reduction in population size of waterbucks during the dry season might be due to the fact that in the dry season, vegetation dries up, leading the animals to migrate and accumulate in smaller greener areas proximal to water sources and nutritious food, permitting less visibility and encounter rate for the Defassa waterbuck. e waterbuck requires an exceptionally high amount of water [29]. As a  result, their habitat preference and abundance are highly influenced by proximity to drinking water, as well as the presence of quality grasses associated with savannah ecosystems, which grow close to water sources [30]. e need for a greater water intake is due to the consumption of mostly protein-rich grasses [31]. According to Taylor et al. [29], waterbuck are more prone to dehydration in hot weather than the average allied group. Moreover, waterbuck have been described as selective feeders to a degree [32] where they may select more nutrient-rich grass species in the wet summer season that quickly lose their nutritive value and decrease in abundance in the dry season months. In addition, Van and Peter [33] reported that the abundance of herbivores during the wet season is associated with the presence of abundant nutritious food items. Because of the specific physiological as well as behavioral nature of the waterbucks, they were found more abundantly in the wet season where there was easy access to water and protein-rich grasses. Typical density estimates obtained by aerial surveys of areas where the species is reasonably common are in the order of 0.05-0.15/km 2 . Ground surveys have provided density estimates in the order of 0.4-1.5/km 2 in areas where the species is common [34]. e population dynamics of an organism can be reflected by information on sex ratio and age distribution for evaluating the viability of a species. As a result, the sex ratio and age structure of any population at any given time are indicators of the status of the population [22]. e sex ratio analysis of the Defassa waterbuck in the current study is in agreement with the study reported by Adane et al. [35]. e high number of female waterbucks recorded in this study shows that the waterbuck population has a chance to increase in the near future in the study area. However, the significant reduction in the number of juvenile waterbucks currently in the area indirectly signifies that the population is declining. e possible reason for this unbalanced sex ratio of the waterbucks might be the occasional nonselective illegal hunting by the local community, targeting mostly adult males. e reason for the low number of juvenile waterbucks might be because they usually hide in the dense, tall grasses, and shrubs of the plains and in the surrounding bushes, where they were not encountered during the survey. Moreover, they are not strong enough to run fast enough to escape from predators, so that they may get attacked [28,36,37]. Regarding waterbuck's population ecology, it has an average life span of 18 years and the recruitment takes place from 1 to 2.5 years of age. Its birth is all year round, but with its peak in the rainy season (September to November). e major threats to the Defassa waterbuck are anthropogenic activities being executed in the study area. e fast growing human populations puts most wildlife population on the verge of extinction. Human activities such as poaching, deforestation, overgrazing, and hunting are common in threatening and imposing quick declines on wildlife throughout their range [38]. e waterbuck population decline observed in the current study is intensified by poor management strategies being operated in the park. e government is not allocating enough funds to undertake an effective management system. Despite the promising scientific and economic value of wildlife, anthropogenic activities remain a growing threat to their populations, particularly in densely populated communities [3].  International Journal of Zoology

Conclusions
e existing waterbuck population was mostly found in large numbers in habitats with better food quality near permanent water sources. e present study documented the population estimate of Defassa waterbucks in the Nechsar National Park to be 111, with 104 and 118 individuals in the dry and wet seasons, respectively. Out of the total population of waterbucks, females were higher in number. However, the low number of juveniles indicated that the population of the Defassa waterbuck is declining. e existing waterbuck population was mostly found in large numbers in habitats with better food quality near permanent water sources. e survey revealed that anthropogenic activities are the major threats not only to the Defassa waterbuck, but also to the overall biodiversity of the study area. Because of these human activities, the population of waterbucks and other biodiversity appears to be in immediate threat. ere are many conservation gaps that would greatly affect the studied species negatively in the future, if effective conservation measures were not taken. e threats in the study area include human-induced habitat loss and degradation through deforestation, livestock grazing, especially in the open plain of the park. More grass species were removed from the area over time, which may have hastened the loss of biodiversity in the study area.
ough the population status of waterbuck in the study area is in good condition, the ongoing human activities are a growing concern for the population of the Defassa waterbuck and the biodiversity at large.

Recommendations
Based on the present study, the following recommendations are forwarded: (i) Intensive research is required to identify the root cause of the waterbuck's population decline in the study area. (ii) Local communities should be aware of the ecological, economic, and social values of wildlife so that they can stop threatening nature and appreciate the benefits of natural resources. (iii) An alternative source of livelihood needs to be implemented in order to reduce the dependency of the local communities on the park and its biodiversity. (iv) To reduce such threats, holistic conservation strategies incorporating the local communities need to be designed in order to create a sense of ownership for surrounding people.
(v) Local communities should be immediately involved in designing, planning, implementation, and evaluation of the wildlife conservation program.
Data Availability e data sets used and analyzed during the current study are available from the corresponding author upon a reasonable request.

Ethical Approval
is research was performed in accordance with the laws, guidelines, and ethical standards of Ethiopia, where the research was done.

Conflicts of Interest
e authors declared that they have no conflicts of interest. International Journal of Zoology 7