Human–wildlife conflict in the surrounding districts of Alage College, Central Rift Valley of Ethiopia

Abstract The study was conducted between September 2018 and March 2019 to investigate the nature and extent of human–wildlife conflict (HWC) in the surrounding area of Alage College, the Central Rift Valley of Ethiopia, and to assess the perception of the local people to wildlife. For data collection, a total of 140 household (HH) heads were selected randomly for interviews from nine villages using structured and semi‐structured questionnaires. Moreover, focus group discussion, key informant interviews, and personal observation were carried out to obtain additional information. Descriptive statistics, Chi‐square test (2 tailed), one‐way analysis of variance, Pearson correlation coefficient, and Likert scale statements were used to analyze the data. Based on, 66 (47.1%) of the respondents, HH heads, the local people of the area experienced livestock predation leading to HWC. Whereas, (40.7%, n = 57) of the respondents perceived both crop damage and livestock predation as a cause of conflict. A total of 932.43 total livestock unit of livestock and 218 dogs’ losses were reported by HH due to predators over the last 5 years. Thus, the largest number of livestock (89.9%) and dogs (100%) attacks was happened due to spotted hyenas. Nearly half of the respondents (49.3%, n = 69) ranked warthogs as the primary crop raiders, while the majority of respondents (82.1%, n = 115) reported maize as a severely damaged cereal crop. More than half, (57.1%, n = 80) of respondents used different methods simultaneously to minimize damage caused by wild animals. About half, (48.6%, n = 68) of respondents had a negative attitude toward wildlife conservation. The level of education and amount of money imposed as a penalty for illegal grazing were affecting the local community's attitudes to wildlife conservation. Using effective methods to reduce damage and loss to crops, including improved livestock husbandry and creating better awareness to the local community could make the locals actor of conservation.


| INTRODUC TI ON
The existence of human-wildlife conflict (HWC) goes back to time immemorial (Amare, 2015;Anand & Radhakrishna, 2017). HWC occurs when the needs and behaviors' of wildlife affected human life negatively and vice versa (Yihune et al., 2009). Currently, it is a widespread phenomenon and challenge facing conservationists around the world (Acha & Temesgen, 2015;Mekonen, 2020;Mekuyie, 2014). It is a serious problem to those whose livelihoods are dependent on agriculture and livestock production Kumssa & Bekele, 2014;Teshome et al., 2017), and those peoples living in and nearby wildlife habitats (Gebeyehu & Bekele, 2009;Tufa et al., 2018). HWC impacts range from crop-raiding to livestock predation and human attack to other intangible social costs (FAO, 2015).
The rapid growth of the human population in developing countries and the requirement for more land for settlement and agriculture have lead to loss, degradation, and fragmentation of habitats inhabited by wild animals resulted in HWC (Acha & Temesgen, 2015;Berihun et al.,2016). The wild animals involved in HWC range from smaller (red locusts), non-human primates to large herbivores. They cause vast damage to local people's crops and properties, and large mammalian carnivores cause livestock depredation and a threat to human life (FAO, 2010;Tufa et al., 2018).
Most of the Ethiopian wild animal resources have been restricted to protected areas due to a dramatic loss in natural habitat or coverage over the last few decades (Berihun et al., 2016;Ketema, 2017). This is particularly difficult for large carnivores which required a wide home range (Lagendijk & Gusset, 2008). This forced wild animals to spend some part of their lifetime on human-dominated landscapes that are highly vulnerable to anthropogenic activities (Watson, 2010). Proximity to wildlife areas often creates conflict between humans and wildlife due to competition for shared and limited resources (Acha & Temesgen, 2015). This has led to HWC which has become a major threat for rural people to secure their household (HH) livelihood requirements (Mekuyie, 2014). As result, local people develop a negative feeling toward wildlife (Lagendijk & Gusset, 2008). This negative impact could have led to the clearing of vegetation to reduce the nuisance of wild animals and people stand antagonistic to wildlife conservation (Mojo et al., 2014).
In the Central Rift Valley of Ethiopia in particular the low land parts, wild animal habitats have been burned mainly due to charcoal production to be sold in the market and generate income resulting severe deforestation (Biazen, 2014). The high deforestation has resulted in a scarcity of resources for wild animals to fulfill their requirement of survival and production (Amare, 2015). The major conflicts that happened in human-dominated landscapes were due to the segregation of wild animals in their farmlands or settlement areas (Makindi et al., 2014). These results in retaliatory killing (Tufa et al., 2018) and aggravating the disappearance of wildlife inhabited in human-dominated areas (Masanja, 2014 were frequently seen (Derebe & Girma, 2020;Personal observation). However, there is no sufficient systematically studied information regarding HWC in the Alage area. Therefore, in this study, we investigated the impacts of HWCs on both humans and wildlife in the surrounding area of Alage. Specifically, we assessed (1) the types and extent of HWCs, (2) the main wild animals involved in HWCs, (3) the major driver causes of HWCs in the study area, and (4) people's attitudes toward wildlife conservation. Addressing these objectives will assist policymakers and conservationists in developing and implementing appropriate conservation policies that will aid in improving human-wildlife coexistence in human-dominated landscapes.

| Description of the study area
Alage is located in the Great East Africa Rift Valley, 217 km southwest of Addis Ababa and situated very close to Abijata and Shala Lakes, National Park, and west of Bulbula town at a distance of 32 km from Addis Ababa-Shashemene highway. Alage shares its boundaries with Oromia Regional State (by Adami Tulu Judo Kombolicha and Arsi Negele districts) and SNNPRS (by Alaba special district).
It is owned and managed by the Alage College management office.
Geographically the study site is located in a range between 7°35′00″ and 7°37′30″N latitude and 38°25′00″ and 38°27′30″E longitudes ( Figure 1). The area is characterized by bimodal rainfall distribution with average annual rainfall ranges from 700 to 900 mm and the average minimum and maximum temperature range between 6.8 and 34.5°C, respectively. It covers a total area of 29.46 km² lands with an altitudinal range from 1580 to 1650 ma bove sea levels. Its north, east, and northeast directions are bordered by the Jido River and the river served as the main water source of wild animals and livestock.

| Sampling design and data collection
Prior to data collection, 2 days reconnaissance survey was held to identify the Kebeles (the smallest administrative unit) with the highest incidences of HWC in the study area. Thus, a total of three Kebeles, namely; Alege-Gero Kebele (from Alaba special district), Naka Kebele (from Adami Tulu Jido Kombolcha district), and Alge-Delbtu Kebele (from Arsi Negele district) were selected purposively based on the highest incidences of HWC and proximity to the study area. Then, villages from each Kebele were stratified based on the distances category of near (<1 km), medium (1-5 km), and far (>5 km) from the study area. Following this, one village from each distance category was randomly selected. Therefore, a total of nine villages (three from each Kebele) were selected based on distance categories (near, medium, and far) of Alage. Accordingly, Boraa, Mansalega, and Rogedia villages were selected from Alege-Delbtu kebele, from Naka kebele; Naka, Giro and Halaqee, and confidence as well as good information, the interviews were held in respondents' homes (Holmern et al., 2004) whose age is ≥18 years old. Each respondent of the study villages was selected randomly by following a pattern of skipping two HH, and the third HH was interviewed (Mekuyie, 2014). To reduce exaggerating response, prior to each interview, the respondent was informed that the survey was independent of the government and that no compensation would be paid for any damage. We also assured them

| Data analysis
The data were analyzed using Statistical Package for Social Science (SPSS) software version-16. Chi-square test was used to compute the differences between types of conflict and wild animals involved, crop damage by wild pests, causes of conflict, mitigation measures, livestock grazing, and firewood collection along with surveyed villages. One-way ANOVA analysis of variance was applied to compute the mean differences of land owned, loss of livestock, and money paid as a penalty. Likewise, the Pearson correlation coefficient was used to test the relation between both the duration of grazing and firewood collection with distances and amount of money paid as a penalty with the duration of grazing, total livestock owned, and distances of villages from the study area. Mean values, ranges, percentages, and frequencies are also computed using descriptive statistics.
The attitudinal data contained 10 Likert Scale statements that were used to test the attitude of local people toward wildlife conservation. Each respondent responded to the 10 statements based on the five-point Likert Scale method ranging from Strongly Disagree to Strongly Agree. Simple weightings (1-5) were assigned to the response categories. The maximum weight of 5 was given for 'Strongly agree' and the minimum 1 was assigned for 'Strongly Disagree'. A weight of 2, 3, and 4 were given for the response categories of Disagree, Neither Agree nor Disagree (Neutral), and Agree, respectively. Thus, if a respondent gives 5 F I G U R E 2 Some of the wild animals inhabited in Alage; (a) Vervet Monkeys around human settlement, (b) Squiller in riverial habitat, and (c) Olive baboons in Acacia wooded grassland for all 10 statements, the maximum weight would be 50 whereas 10 would be the minimum weight when a respondent scores 1 for all 10 statements. Hence, the average of the sum scores of all 10 statements for each respondent would again range from 1 to 5.
This gives another distribution of the data in which the mean and standard deviation is used to differentiate respondents according to their level of attitude toward wildlife conservation following the procedure applied by Gebrelibanos and Assen (2013  Respondents also viewed the type of predators that killed livestock. According to respondents, Hyena and Common jackal were the main livestock predators in the present study (Figure 3).

| Major driver to causes of humanwildlife conflict
The majority, 53 (38%) of surveyed respondents thought that a major drivers cause of HWC were driven due to combined effects of anthropogenic activities. However, six percent of the respondents perceived the increasing wildlife population as a cause of HWC in the study area ( Figure 4). The perception of respondents about the major driving forces of conflict was statistically significant (χ 2 = 1,31,988, df = 8, p = .005) across the study villages.

| Mitigation measures
More than half of, 80 (57.14%) respondents used guarding, chasing, fencing, smoking, and scarecrow simultaneously to minimize damage caused by wild animals in the study area (Table 4).
In the case of livestock husbandry, 101 (72.1%) of HHs kept their livestock outside the home overnight. Of the remainder, 27.9% kept their livestock inside traditional enclosures ( Figure 5). The level of predation incident had no significant differences (χ 2 = 6.963, df = 4, p = .069) between HH who kept their livestock inside and outside enclosures' during night times.

| Resource dependences and a penalty for illegal grazing
The local community used the natural resources as livestock grazing (46%) and as sources of firewood (46.4%) for their house (

| Local people perception toward wildlife conservation
According to the present findings, a few (24.3%) of the respondents had a positive feeling while some (27.1%) respondents had a neutral feeling. On the contrary, nearly half (48.6%) of the respondents had an unfavorable feeling toward the conservation of wildlife (Table 6). Relatively uneducated people had high negative attitudes (χ 2 = 37.585, df = 4, p < .001). Besides, people who paid a high penalties (χ 2 = 20.888, df = 3, p < .001) would develop high negative attitudes toward co-existence as compared to those who did not yet.

| DISCUSS ION
As shown by the current results and reported studies (e.g. Mekuyie, 2014;Shanko et al., 2021;Yilmato and Takele  to chase out of crops because of their sophisticated social organization habits and capacity to climb surrounding trees. The intensity and vulnerability of crops being damaged by wild animal pests were varied depending on the type of crop planted (Mekonen, 2020) and its land coverage, and the type of wild animal involved in crop-raiding (Gobosho et al., 2015;Mc Guinness & Taylor, 2014). Farmlands in close proximity to wild animal habitats are known to be frequently visited by crop raiders and are more vulnerable to damage (Mamo et al., 2021;Nibret et al., 2017). Maize was the most preferred crop by crop raiders, followed by sorghum. The possible reasons could be due to its nutritional values (Raphela & Pillay, 2021), and large farm sizes in proportion to other crops in the area.
In the study area, the natural habitat of wild animals has recently been altered for livestock fattening and horticulture practices. The Ethiopia used guarding, chasing, fencing, scarecrow, and smoking to reduce crop damage and livestock predation (Gobosho et al., 2015).
Farmers in Kenya also used different crop protection methods simultaneously depending on the type of raiders involved (Musyoki, 2014 This is in line with the findings of Gezahagn et al. (2014) and Shi et al. (2010), who reported that restricting access to wildlife resources and enforcing punishments have a negative impact on the perception of the local people. Residents with formal education were more appreciated conservation objectives (Karanth & Nepal, 2012;Shibia, 2010). In contrast to the findings of Eshete et al. (2018), who found that even low levels of livestock predation might lead local people to develop negative attitudes toward wildlife, surprisingly, in the current study, a high level of predation has no significant negative impact on local people's perceptions. This could have significant implications for wildlife conservation in human-dominated areas.

| CON CLUS ION AND RECOMMENDATIONS
As indicated by the current study, where subsistence farming is a major income source of the HH, high levels of conflict could occur between humans and wildlife. Therefore, losing their property and human threats due to wild animals may sometimes lead the local people to kill wild animals intentionally and develop unfavorable perceptions toward wild animals. Moreover, habitat destruction for subsistence farming, overgrazing, and proximity to wildlife habitat were the major factors. Thus, combined anthropogenic factors and to some extent increase in wildlife in the area escalated the conflict. Poor livestock husbandry was attributed to the loss of large numbers of livestock by spotted hyenas and common jackals in the study area. Conservation education needs to be given rather than imposing inappropriate penalties. Livestock F I G U R E 7 Type of traditional enclosure used to keep livestock during night husbandry needs to be improved and the livestock enclosure needs to be well built. Further study is important to understand the abundance and diversity of fauna and enhance the coexistence of humans and wildlife in the area.

ACK N OWLED G M ENTS
We are grateful to the Ethiopian Ministry of Agriculture (ATVET Coordination office) for funding the study. We are also thankful to the local community for their kind responses and sharing their knowledge and to field assistants for their assistance of data collection to our inquiry.

CO N FLI C T O F I NTE R E S T
The authors declare that we have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are publicly available at https://doi.org/10.5061/dryad.02v6w wq4z.