Assessment of livestock depredation by hamadryas baboons (Papio hamadryas) in and around Borena-Sayint National Park, northern Ethiopia

Abstract

Background

As human populations expand and natural habitats become increasingly fragmented, the interface between agricultural activities and wilderness areas has become a focal point of ecological tension. Livestock depredation is a prevalent form of human–wildlife conflict, particularly in landscapes bordering protected areas. Beyond causing substantial financial losses for subsistence communities, these conflicts often generate negative attitudes toward wildlife, ultimately hindering long-term conservation efforts. Hamadryas baboon (Papio hamadryas) is distributed across diverse environments in Ethiopia, ranging from the lowlands to the highlands. Across its range, the species is known to prey on small livestock, particularly goats and sheep. Despite this, knowledge of human–hamadryas baboon interactions, especially those involving livestock depredation in the Ethiopian highlands, remains scarce. Therefore, this study aimed to gather data on the extent of conflict between hamadryas baboons and local communities in and around Borena-Sayint National Park (BSNP), northern Ethiopia.

Methods

We collected livestock depredation data from January to July 2018 through a questionnaire survey of 300 sample households in surrounding local communities. We used descriptive and inferential statistics to analyze local communities’ attitudes and perceptions toward hamadryas baboons’ livestock-depredating behaviour.

Results

Respondents in the study area primarily reported livestock depredation (74.6%) and drought (18.4%) as the main challenges in livestock rearing. Respondents estimated that hamadryas baboons caused damage to an average of 3.84 (± 0.36) livestock per household over five years in and around BSNP. Hamadryas baboons were reported to cause the greatest losses (average, range) to sheep (1.3, 0–5), lambs (1.1, 0–10), goat kids (0.7, 0–8), and goats (0.6, 0–4) per household over five years. Reported livestock attacks varied significantly between the wet and dry seasons. The vast majority of respondents (93.9%) reported that they did not kill hamadryas baboons in retaliation for livestock losses. Following livestock attacks, respondents primarily reported the incidents to park scouts (80.5%), doing nothing (11.8%), or attempting to scare the animals by making noise (6.0%). The intense conflict between hamadryas baboons and the local communities could seriously threaten the species’ survival unless appropriate management measures are implemented.

Conclusion

The results of this study provide essential information for developing management plans and implementing effective conservation strategies, and we recommend the adoption of community-based conflict mitigation approaches, including improved livestock husbandry and collaborative monitoring with park authorities, to reduce depredation and promote coexistence between local communities and hamadryas baboons.

Introduction

Human–wildlife conflict is a widespread global phenomenon that occurs when humans and wildlife harm or threaten one another in pursuit of their respective needs or interests [1–3]. This conflict is primarily driven by rapid human population growth, which fuels agricultural expansion and the subsequent encroachment into previously undisturbed ecosystems. As these natural habitats undergo extensive transformation, habitat loss and fragmentation confine wildlife to increasingly small, isolated patches [4]. Consequently, transition zones between protected areas and human-dominated landscapes are diminishing, forcing wildlife populations, particularly non-human primates, to live in closer proximity to humans [5–7]. This proximity increases competition for space and resources and often intensifies conflicts between people and wildlife in shared ecosystems [8].

For adaptable primates such as baboons, habitat modification and reduced availability of natural food resources often lead to reliance on anthropogenic resources, including cultivated crops and other human-derived foods [9, 10]. The intensity and nature of these interactions are influenced by multiple factors, including habitat fragmentation, seasonal fluctuations in food availability, proximity of agricultural fields and settlements to wildlife habitats, and the availability of alternative human food resources [5, 8, 11–13]. On some occasions, baboons may also attack small domestic animals, although such incidents are generally opportunistic [14]. These conflicts can impose substantial economic costs on local communities through crop damage, livestock losses, and the considerable time and effort required to guard farms and domestic animals [5, 8, 12–14].

Human–primate conflict encompasses not only direct economic losses but also important social dimensions. In landscapes where human activities overlap extensively with primate habitats, local perceptions and attitudes toward wildlife are shaped by individual experiences, livelihood dependence, and perceived conservation costs and benefits [15]. Perceived damages to wildlife can lead to harassment, injury, or retaliatory killing of animals, while simultaneously creating tensions between local communities and conservation authorities [5, 13, 15]. Ultimately, these chronic material impacts can extend beyond direct economic losses to generate broader social conflicts, creating tensions between protected area managers and local communities over conservation policies and practices that may hinder long-term conservation efforts. Therefore, understanding both the ecological and human dimensions of conflict is essential for developing effective management strategies and promoting sustainable coexistence between people and wildlife [5, 10, 11].

The hamadryas baboon (Papio hamadryas) inhabits both lowland and highland ecosystems of Ethiopia [16, 17]. Although the species is classified as Least Concern in the IUCN Red List, local populations face increasing threats from habitat encroachment, human settlement, agricultural expansion, and hunting [18]. The population of hamadryas baboons is severely fragmented in the northern highlands of Ethiopia, including the present study area (Borena-Sayint National Park; BSNP), as is the case for many other primates in the country [1, 19–22]. Previous studies in Ethiopia have documented various forms of human–baboon conflict, including crop damage, livestock losses, and negative community attitudes, in areas such as Simien Mountains National Park [23], Awash National Park [24], and community forests in the Gasera District of southeastern Ethiopia [25]. However, previous Ethiopian studies have largely focused on documenting conflict occurrence and associated damages, with limited attention to how variation in landscape characteristics and human encroachment may influence conflict intensity and local perceptions.

Understanding the local people’s perceptions, concerns, and priorities is essential for effective conservation planning and for establishing collaboration with the local community, thereby facilitating strategies that support both the conservation goals of BSNP and the livelihoods of the surrounding communities [26, 27]. BSNP provides a valuable setting for examining these relationships because it is characterized by substantial human settlement around the park and varying levels of habitat fragmentation. The park is bisected by a major river, creating distinct ecological conditions on either side. The portion of the park located in Borena District retains relatively continuous vegetation cover and lower levels of human disturbance, whereas the Mehal-Sayint side is characterized by greater habitat fragmentation and higher levels of human encroachment. By evaluating how these contrasting landscapes alter human dimensions of conservation, this study explores how landscape-level variation shapes both wildlife impacts and local social conflicts.

The primary objective of our study was to investigate livestock-related conflict involving hamadryas baboons and local communities’ perceptions towards hamadryas baboons surrounding the BSNP across two distinct districts (Borena and Mehal-Sayint). Specifically, we aimed to: (1) Characterize the baseline demographic and socioeconomic characteristics of the respondents living near BSNP, (2) Evaluate local communities’ attitudes and perceptions regarding whether hamadryas baboons caused livestock losses, including the specific types of livestock vulnerable to opportunistic exploitation and the extent of damage. (3) Investigate the social dimensions of this interaction by assessing the occurrence of retaliatory killing and the perceived efficacy of local mitigation strategies. Given the distinct landscape configurations, we hypothesized that the frequency of opportunistic livestock attacks by hamadryas baboons would vary between the two districts, with higher levels of interaction expected in the more fragmented Mehal-Sayint District than in the relatively contiguous Borena District. We further examined how these ecological differences influence local community perceptions and social tolerance toward the species.

Methods

Description of the study area

This study was conducted in BSNP located in the South Wollo Zonal Administration of Amhara Regional State, northern Ethiopia, within the geographical coordinates between 10^o50’45.4” 10^o53’58.3” latitude and 38^o40’28.4” 38^o54’49” longitude (Fig. 1). BSNP currently spans 4,375 ha and encompasses diverse topographic features, with altitudes ranging from 2,188 to 3,732 m above sea level. The study area extends across two major agro-climatic zones: Weina Dega (cool, sub-humid) and Dega (cold, humid), based on the ambient temperature and moisture regimes [28]. BSNP comprises natural forests and high-altitude grasslands, harboring diverse flora and fauna. The BSNP is confined to a mountain ridge in a highly degraded, eroded, and isolated ecosystem within the South Wollo Zone. It forms part of the Eastern Afromontane Biodiversity Hotspot, which is noted for exceptional species richness and endemism, yet also faces high human population pressure [29].

Fig. 1.

Map of the study area, Borena-Sayint National Park in northern Ethiopia

The present study area experiences a distinctive bimodal rainfall pattern, with a main wet season extending from June to September, and a short rainy season from April to May. Average annual rainfall is about 917 mm (SD ± 39.78, range: 655–1,165 mm), while the mean annual temperature is 16.7 °C (SD ± 1.05) [16]. The floristic composition of BSNP comprises 354 plant species across 265 genera and 95 families, including trees, shrubs, and herbs [28]. The study area also supports at least 23 medium to large-sized mammal species [30]. Its combination of striking landscapes, diverse flora and fauna, remarkable caves, and the rich cultural heritage of local communities makes BSNP highly suitable for ecotourism development [28].

The Denkoro River flows from east to west through BSNP, dividing the park into two distinct portions that lie within the Borena and Mehal-Sayint Districts, with the larger portion occurring in the Borena District. Vegetation patterns in BSNP reflect varying levels of anthropogenic influence, with more intensive human activities, habitat fragmentation, and encroachment observed on the Mehal-Sayint District side compared to the relatively better-preserved vegetation on the Borena District side [28]. Approximately 8,290 individuals residing in 1,190 households across 35 village clusters border BSNP in Borena and Mehal-Sayint Districts. Most of these villages are located in the Borena District and occur at varying distances from the park boundary. The primary sources of livelihood for these communities include agriculture, traditional livestock husbandry, honey production, and rain-fed farming on degraded hillsides, supplemented by small-scale irrigation schemes [28]. BSNP is surrounded by farmland in all directions, and the local livestock population includes cattle, sheep, goats, donkeys, horses, mules, and poultry; however, sheep constitute the primary livestock species.

Sampling design and data collection

We employed a comparative mixed-methods research design combining quantitative questionnaire surveys with qualitative focus group discussions to investigate human–hamadryas baboon conflict, as manifested in livestock depredation, in communities surrounding BSNP. The mixed-methods approach allowed quantitative assessment of the magnitude and patterns of conflict while simultaneously exploring local perceptions, attitudes, and experiences through qualitative inquiry. The qualitative and quantitative components were integrated during interpretation to facilitate triangulation and provide a more comprehensive understanding of the conflict [5, 15, 31, 32].

We selected participants from the local communities adjacent to BSNP using a systematic random sampling method [9, 32]. We interviewed a total of 300 individuals from the 1,190 households distributed across all villages bordering BSNP, representing both the Borena and Mehal-Sayint communities. We determined the sample size using Cochran, W. G [33]. formula. To ensure the accurate sample representation from both districts, we allocated respondents proportionally according to the number of households in each district, resulting in 220 respondents from Borena District and 80 from Mehal-Sayint District. Since official, pre-existing household registry lists were unavailable at the village level, we selected participants using a spatial systematic random sampling method directly in the field [9, 32]. We divided the total estimated household count (1,190) by our target sample size (300) to determine the sampling interval (k = 4). We used a sampling interval of four, whereby every fourth household was selected for interview. We began sampling from one end of each village and continued systematically until the required number of households was reached. Within each selected household, we selected the primary household head as the principal respondent. If a household head was absent or declined to participate, we moved to the next immediate household, and then we started the spatial count interval from that point [9, 32]. Of the 300 respondents, 239 were males, and 61 were females. The higher proportion of male respondents reflects the traditional socio-cultural structure of rural Ethiopian households, where men typically serve as the primary heads of households and the primary decision-makers in livestock management and are therefore more available and considered more knowledgeable about livestock-related issues [34].

We employed a questionnaire containing both open- and closed-ended questions to explore the different dimensions of human–hamadryas baboon conflict, with a focus on livestock depredation [1, 5, 19, 23, 35]. All interviews were conducted by the Principal Investigator (HI) and lasted approximately one hour. Furthermore, the questionnaire addressed demographic and socioeconomic factors influencing conservation management [24, 25, 31]. We prepared the questionnaire in English and translated it into the local language (Amharic) to facilitate communication during the interview process. Before the main survey, we pre-tested the questionnaire with 37 randomly selected individuals representing diverse ages, sexes, and socioeconomic backgrounds to evaluate its clarity, content validity, and applicability. Based on feedback from the pre-test, minor revisions were made to improve the wording and sequencing of questions before administering the questionnaire to the study participants [24, 35]. We excluded these pilot responses from the final analysis. Following this preliminary survey in January 2018, we conducted the actual data collection from February to July 2018 using a questionnaire administered to selected households. Before each interview, we informed respondents about the purpose of the study and assured them that their responses would be treated confidentially and used solely for research purposes.

The semi-structured questionnaire explored ten core dimensions of human–hamadryas baboon conflict, with particular emphasis on livestock-related impacts. Specifically, we asked respondents (1) the occurrence and precise extent of perceived domestic animal losses, (2) the specific livestock species targeted during opportunistic foraging events, (3) the measures employed by respondents to prevent losses, (4) the prevalence and frequency of retaliatory killing of hamadryas baboons by community members, (5) perceived livestock losses over the past five year recall period, (6) whether respondents believed hunting baboons was legal or illegal, (7) explicit community awareness of the protected legal status of hamadryas baboons, (8) their perceptions of the benefits and costs of living near the park, (9) overall conservation attitudes toward the long-term survival of hamadryas baboons, and (10) who they believed should be responsible for compensating livestock losses [5, 13, 15, 25, 31, 36]. We selected the multiannual recall period to capture relatively infrequent and potentially seasonal livestock attack events that may not be adequately represented within shorter recall periods (e.g., 12 months). To minimize recall error, our primary approach during data collection was to explicitly instruct and encourage respondents to provide accurate information based strictly on their own direct, personal experiences and observations. However, no formal recall-bias reduction techniques, such as temporal anchoring or independent verification of reported events, were employed. Then, we assessed perceived livestock losses rather than independently verified depredation events; consequently, the data may be subject to recall bias.

We also employed focus group discussions (FGDs) to triangulate/ complement and validate the questionnaire survey data [5, 32, 35]. We organized two group discussions to obtain first-hand information regarding local experiences with human–hamadryas baboon conflict and attitudes toward conservation practices [5]. We used purposive selection criteria to select representative community members, local elders, village elders, scouts, and BSNP administrators. During the discussion, the Principal Investigator facilitated the discussion by introducing key themes emerging from the questionnaire survey and encouraging participants to discuss their experiences and perspectives. We recorded notes throughout the discussions and subsequently organized them into thematic categories for qualitative interpretation and triangulation with the questionnaire findings.

Data analysis

We analyzed the data using SPSS version 29.0 for Windows (Evaluation Version). All statistical tests were two-tailed with a 95% confidence interval, and we set the significance level at P = 0.05. We used descriptive statistics, expressed as percentages and frequencies, to summarize variables such as socioeconomic status, types of domestic animals reared, and perceived livestock loss. We compared categorical responses using chi-square tests (simple – among all respondents; independent – between respondents from Borena and Mehal-Sayint Districts). We used the nonparametric Mann–Whitney U test to compare differences in the number of livestock owned per household and the number of livestock attacks attributed to hamadryas baboons between the two districts.

Additionally, we employed a multiple linear regression model to test factors determining the intensity of livestock depredation by hamadryas baboons using district, residence distance measured using GIS from the village center, and numbers of goats and sheep as predictor variables. We also used the model to test the natural right of hamadryas baboons existing in and around BSNP. To assess the factors associated with domestic animal losses across communities, we utilized frequency distributions, and cross-tabulations. For questions that allow multiple responses, we conducted Multiple Response Analysis. We presented statistical summaries and percentage values in tables and figures. We analyzed information obtained from focus group discussions qualitatively through thematic content analysis. We summarized and used emerging themes and recurrent viewpoints to validate and enrich the interpretation of the questionnaire data [5, 32].

Results

Demography and socio-economic features of the respondents

Respondents’ ages were categorized into five groups, with the largest proportion (24.7%) aged 60 or older, followed by 22.3% aged 41–50 and 21.7% aged 51–60. The lowest proportions were observed in the 18–30 (16.3%) and 31–40 (15%) age groups (independent χ² = 4.36, df = 4, P > 0.05). The majority of respondents (53.3%, n = 160) were illiterate, while 25.3% (n = 76) and 5.3% (n = 16) completed primary and secondary education, respectively. Respondents with informal education accounted for 16% (n = 48). There was no significant difference in educational levels between the two districts (independent χ² = 4.59, df = 3, P = 0.204). The livelihoods of most respondents (82.3%, n = 247) were primarily based on subsistence farming, including both crop cultivation and livestock rearing. The remaining respondents relied solely on crop production (15.5%, n = 34) or exclusively on livestock rearing (1.4%, n = 4). There was no significant difference between respondents in the two Districts in livelihood activities (independent χ²: = 0.18, df = 3, P = 0.916).

The overall average (± SD) number of livestock per household was 9.02 ± 1.45 for respondents from Borena District and 9.85 ± 1.71 for respondents from Mehal-Sayint District. The Mann–Whitney U test revealed a highly significant difference in the number of sheep owned per household between the two Districts (U = 6529, Z = -3.49, P = 0.001). Respondents situated within the Borena District landscape matrix maintained noticeably larger flocks (Mean = 5.21 ± 3.28 sheep) compared to those located in the more fragmented Mehal-Sayint District (Mean = 3.66 ± 3.13 sheep). In contrast, no significant difference was observed in the number of goats owned per household between the districts (U = 9706.5, Z = 1.69, P = 0.092). Poultry and beekeeping were also practiced in the present study area. The respondents utilized different locations to graze their livestock. The largest numbers of respondents (52.4%, n = 155 and 45.6%, n = 136) grazed their livestock in private grazing fields and near the BSNP boundary, respectively. A small proportion grazed their livestock within the BSNP boundary (1.4%, n = 4), and only two individuals used communal grazing fields. Significant differences were observed among the grazing locations and between residents of the two districts (simple χ² = 275.22, df = 3, P < 0.001; independent χ² = 14.27, df = 2, P < 0.05).

Livestock depredation by hamadryas baboons

The problems the respondents face in their domestic animal rearing are indicated in Fig. 2. Multiple response analysis revealed that respondents at the study sites mainly faced depredation of domestic animals by predators from the BSNP (74.55%) followed by drought (18.4%). A small proportion of respondents (3.5%) reported experiencing theft of domestic animals. In addition, 2.5% of respondents’ domestic animals were affected by disease. Almost 1% of the respondents did not face any problem with their domestic animal rearing. There was no significant difference (P > 0.05) between the Borena and Mehal-Sayint Districts regarding the problems encountered by the respondents in livestock rearing.

Fig. 2.

Major constraints to domestic animal rearing reported by respondents in Borena and Mehal-Sayint districts

The majority of respondents, 82% (n = 246) from both Districts, experienced livestock depredation by wild animals residing in and around BSNP (Table 1). 18% of the respondents (n = 54) reported no incidents of domestic animal depredation (simple χ² = 122.8, df = 1, P < 0.05). Domestic animal depredation was significantly higher in Borena District (85%) than Mehal-Sayint District (59%) (independent χ² = 5.03, df = 2, P = 0.025).

Table 1.

Livestock depredation attributed to hamadryas baboons and reported trends over the previous five years in the study area

Districts	Livestock depredation n (%)
	Presence	Absence	Trends in the past 5 years
			Increasing	Decreasing
Borena	187 (85)	33 (15)	176 (94.12)	11 (5.88)
Mehal-Sayint	59 (73.75)	21(26.25)	55 (93.22)	4 (6.78)
Total	246 (82)	54 (18)	231 (93.9)	15 (6.1)

In the past five years, the trend of domestic animal depredation by hamadryas baboons was aggravated in both Districts (Table 1). Among respondents who experienced livestock depredation, 93.2% in Mehal-Sayint District and 94.1% in Borena District reported that livestock losses increased in recent years. Respondents reported that they perceived an increase in the number of hamadryas baboons in the area over the recent five years (simple χ² = 189.6, df = 1, P < 0.001; independent χ² = 0.063, df = 1, P = 0.802).

The average number of reported livestock losses attributed to hamadryas baboons was 3.84 ± 0.357 animals per household per five-year period (Table 2). The most attacked livestock were sheep (1.3 ± 0.078), lambs (1.12 ± 0.083), goat kids – young of goat (0.7 ± 0.08), and goats (0.59 ± 0.071). Lambs (1.27 ± 1.02), goat kids (0.76 ± 0.101) and sheep (1.5 ± 0.094) were more affected by hamadryas baboons in Borena District than Mehal-Sayint, 0.73 ± 0.12, 0.53 ± 0.115, 0.76 ± 0.118, respectively (Fig. 3). On the other hand, loss of goats (0.62 ± 0.145) was higher in Mehal-Sayint District than Borena (0.58 ± 0.081). According to respondents from Mehal-Sayint District, they did not experience any loss of poultry by hamadryas baboons. A Mann-Whitney U test revealed that the loss of domestic animals was statistically different between the two Districts during the attack on lambs (U = 6987, Z = -2.95, P = 0.003) and sheep (U = 6102.5, Z = -4.27, P < 0.001). However, the losses on goat kids (U = 8235, Z = -1.07, P = 0.284) and goats (U = 8235, Z = -0.08, P = 0.938) were not statistically different between Borena and Mehal-Sayint Districts.

Table 2.

Summary of depredation on domestic animals by hamadryas baboons in the study area (number of animals/ household/ five years)

Domestic animals	Livestock losses by hamadryas baboons (Mean ± SE)
Poultry	0.13 ± 0.045
Lambs	1.12 ± 0.083
Goat kids	0.7 ± 0.080
Sheep	1.3 ± 0.078
Goats	0.59 ± 0.071
Total	3.84 ± 0.357

Fig. 3.

Average reported loss of domestic animals attributed to hamadryas baboons in Borena and Mehal-Sayint districts during the wet and dry seasons (January 2013 to December 2017

Adult male hamadryas baboons were the primary predators of the majority of domestic animals (95.2%), followed by sub-adult males (2.6%). Adult females and young individuals accounted for 1.7% and 0.4% of the reported losses, respectively. Most respondents (98.7%, n = 296) reported that no family members had been attacked by the baboons; however, four respondents from Borena District reported that family members had been attacked. Incidents of attacks on family members differed significantly (simple χ² = 284.213, df = 1, P < 0.001).

Monthly attack on livestock by hamadryas baboons

Multiple response analysis revealed that domestic animals were frequently attacked by hamadryas baboons in different months (Table 3), including December (14.4%), January (16.45), February (11.35%), and March (10.7%). The least frequent attacks were reported in June (4.8%) and October (4.9%). A Chi-Square Goodness-of-Fit test confirmed that the observed monthly damage frequencies varied significantly (simple χ² = 55.69, df = 11, P < 0.001). However, a Chi-Square test of independence revealed that the temporal distribution of conflict incidents was not significantly associated with district location (independent χ² = 17.48, df = 11, P = 0.094). On the other hand, an evaluation of broader seasonal trends pooled across both study sites varied significantly (simple χ² = 10.05, df = 1, P = 0.002). Additionally, a Chi-Square test of independence revealed a highly significant association between seasonal livestock losses and district location (independent χ²= 11.35, df = 1, P < 0.001).

Table 3.

Summary of monthly attacks on domestic animals by hamadryas baboons in Borena and Mehal-Sayint Districts (January 2013 to December 2017)

Months	Responses (%)
	Borena District	Mehal-Sayint District	Average
January	13.7	19.2	16.45
February	10.8	11.9	11.35
March	10.2	11.2	10.70
April	7.5	8.5	8.00
May	6.8	6.5	6.65
June	7.3	2.3	4.80
July	11.2	3.5	7.35
August	11.2	3.5	7.35
September	2.1	0.4	1.25
October	3.0	6.8	4.90
November	5.1	8.5	6.80
December	11.1	17.7	14.40
Total	100	100	100

Retaliation against livestock attacks by hamadryas baboons

The majority of respondents (n = 231, 93.9%) did not kill hamadryas baboons in retaliation for domestic animal loss (Table 4). On average (± SE), the respondents reported in Borena and Mehal-Sayint Districts, there were killings of 0.04 ± 0.017 and 0.11 ± 0.036 hamadryas baboons, respectively, throughout their existence in the area (F = 3.961, df = 1, P = 0.047). Similarly, respondents observed on average (± SE), 0.11 ± 0.04 and 0.24 ± 0.07 killings of hamadryas baboons in Borena and Mehal-Sayint Districts, respectively, in retaliation for livestock loss. Regarding mutilation (physical injury) of hamadryas baboons in retaliation for the loss of domestic animals, the vast majority of respondents (n = 242, 98.4%) did not cause injury to the animal (Table 4). Few of the respondents who experienced loss of domestic animals caused injury to hamadryas baboons (simple χ² = 230.26, df = 1, P < 0.001; independent χ² = 5.81, df = 1, P = 0.016).

Table 4.

Reported retaliatory actions following livestock losses attributed to hamadryas baboons in Borena and Mehal-Sayint districts (January 2013 to December 2017)

Categories		Districts, n (%)		Total/ Average
		Borena	Mehal-Sayint
Killing of the hamadryas baboons by respondents
	Yes	6 (3.2)	9 (15.3)	15 (6.1)
	No	181 (96.8)	50 (84.7)	231 (93.9)
Killing of the animal by non-respondents
	Yes	11 (5)	13 (16.25)	24 (8)
	No	209 (95)	67 (83.75)	276 (92)
Injuring the animal
	Yes	1 (0.5)	3 (5.1)	4 (1.6)
	No	186 (99.5)	56 (94.9)	242 (98.4)

The typical responses of the respondents after the attack of hamadryas baboons on their domestic animals are shown in Table 5. Multiple response analysis showed that most respondents reacted to hamadryas baboons’ attacks on their livestock by reporting the incidents to BSNP scouts (80.5%). A smaller proportion of respondents did nothing (11.8%), while 5.95% responded by making noise. None of the respondents reported using fire, building fences, or hiring labor as preventive measures. A chi-square test of independence revealed no statistically significant association between the choice of response strategy and the study district (independent χ² = 3.84, df = 5, P = 0.57). in the typical responses of respondents following an attack on their domestic animals by hamadryas baboons.

Table 5.

Respondent-reported strategies for preventing livestock attacks by hamadryas baboons in Borena and Mehal-Sayint districts (January 2013 to December 2017)

Reported strategies	Response (%)
	Borena District	Mehal-Sayint District	Average
Report to Park scouts	84.1	76.9	80.50
Kill the hamadryas baboons	1.0	1.5	1.25
Poison the hamadryas baboons	0.5	0.0	0.25
Noise making	5.8	6.1	5.95
Fire	0.0	0.0	0.00
Build fence	0.0	0.0	0.00
Hire labour	0.0	0.0	0.00
Dogs	0.5	0.0	0.25
Do nothing	8.2	15.0	11.80
Total	100	100	100

Factors determining the intensity of livestock depredation by hamadryas baboons

The Multiple Linear Regression Model for the factors that determine the loss of domestic animals caused by hamadryas baboons explains 21.3% of the variance. However, the likelihood ratio goodness-of-fit test fitted the model (P = 0.002). This R² value indicates that while the model captures a highly significant explanatory signal, a proportion of the variance is driven by high natural behavioral variability in hamadryas baboons’ foraging strategies or other unquantified local conditions. We found the district to be an important factor in determining livestock damage by hamadryas baboons (Table 6). There was a difference between the two districts in the incidence of hamadryas baboon attacks on domestic animals.

Table 6.

Test results of multiple linear regression model for the factors determining the intensity of livestock losses by hamadryas baboons (January 2013 to December 2017)

Predictor variables	B	t – value	P - value
District	-0.485	-4.22	0.000
Number of sheep	0.021	0.186	0.853
Number of goats	0.185	1.601	0.114
Residence distance	-0.165	-1.487	0.141
Constant		4.531	0.000

Respondents’ attitudes and perceptions towards hamadryas baboons

Community perceptions of threat and well-being

A substantial portion of respondents (82.3%) perceived hamadryas baboons as a threat to their long-term livelihood. However, only 17.73% of respondents perceived that the hamadryas posed no threat to human survival, a difference that was highly significant across districts (χ2 = 17.52, df = 1, P < 0.001). We constructed a binary logistic regression model to isolate which socio-demographic factors independently predicted whether a respondent viewed the hamadryas baboons as an active threat (Table 7). The overall model provided a highly significant fit to the data (Nagelkerke R² = 0.13, P < 0.05), correctly classifying 82.3% of the binary outcome positions. The analysis also revealed that geographic location was the sole significant structural driver of community threat perception (Wald χ² = 11.07, df = 1, P < 0.05). Respondents from the Borena District were significantly less likely to perceive hamadryas baboons as a threat to their survival compared to those in Mehal-Sayint. Conversely, individual socio-demographic traits, including participant sex, age, and formal educational status, were not significant predictors in the model.

Table 7.

Binary logistic regression model examining factors influencing respondents’ perceptions of hamadryas baboons as a threat to local livelihoods (outcome variable: perceived threat; 1 = Yes, 0 = No)

Predictor variables	B	SE	Wald χ2	df	P – value	Exp (B)	95% CI for Exp (B)
							Lower	Upper
District (Ref: Mehal-Sayint)	-2.453	0.737	11.066	1	0.001	0.086	0.02	0.365
Sex (Ref: Female)	-0.006	0.418	0.000	1	0.988	0.994	0.438	2.254
Age (Ref: 18–30 years)	0.061	0.133	0.210	1	0.647	1.063	0.819	1.38
Educational status (Ref: Illiterate)	-0.147	0.197	0.556	1	0.456	0.864	0.587	1.27
Constant	1.317	1.175	1.257	1	0.262	3.732

Attitudes toward primate population management

When assessing local tolerance thresholds and management preferences, many respondents (54.3%) favored reducing the local population of hamadryas baboons, while 28% preferred the complete disappearance of the species from their locality. This structural leaning toward population suppression was statistically significant (simple χ² = 280.83, df = 4, P < 0.001). This negative attitude was similarly independent of geographic location, displaying highly consistent distribution patterns between the two study districts (independent χ² = 16.93, df = 4, P < 0.05).

Perceptions of conservation legality

Despite displaying low ecological tolerance and perceiving significant livelihood risks, local communities exhibited high awareness regarding the protected legal status of the hamadryas baboons. Nearly all respondents (98.3%) were aware that catching, trapping, or hunting hamadryas baboons is prohibited under national conservation laws, with only 1.7% reporting otherwise (χ² = 1.85, df = 1, P = 0.174). Furthermore, 91.3% of respondents did not perceive local communities as a threat to the long-term survival of hamadryas baboons, whereas a small minority (8.7%) believed that human activities posed a threat to the species’ persistence in and around BSNP (independent χ² = 3.32, df = 1, P = 0.068).

Bivariate analyses of conservation acceptance

The views of local communities regarding the natural right of hamadryas baboons to exist in the study area are summarized in Table 8. Overall, 89.2% of respondents agreed that hamadryas baboons possess a natural right to exist, while 10.8% disagreed. However, this acceptance of conservation was significantly shaped by spatial and gender boundaries. By geographic location, respondents in Mehal-Sayint (97.5%) showed a significantly higher level of agreement regarding the animals’ right to exist than those in the Borena District (80.9%; independent χ² = 12.90, df = 1, P < 0.001). By gender, male respondents (88.7%) also significantly differed from female respondents (72.1%) in their endorsement of the hamadryas baboons’ right to co-exist in the landscape (independent χ² = 10.63, df = 1, P = 0.001). Conversely, no statistically significant differences (P > 0.05) were observed across age cohorts or formal educational categories.

Table 8.

Respondents’ views about the natural right of hamadryas baboons to exist in the study area

Categories		Natural right of hamadryas baboons, %		Chi-square association model
		Yes	No	χ2 - value	df	P - value
Districts (Ref: Mehal-Sayint)
Borena		80.9	19.1	12.903	1	0.000
Mehal-Sayint		97.5	2.5
Sex (Ref: Female)
Male		88.7	11.3	10.633	1	0.001
Female		72.1	27.9
Age (Ref: 18–30)
18–30		83.7	16.3	2.528	4	0.64
31–40		82.2	17.8
41–50		89.6	10.4
50–60		81.5	18.5
> 60		87.8	12.2
Educational status (Ref: illiterate)
Illiterate		85	15	0.077	3	0.994
Informal education		85.4	14.6
Primary		85.5	14.5
Secondary		87.5	12.5

Discussion

Baboons exhibit notable dietary flexibility and frequently exploit human-dominated matrices, which may be widely reported to be associated with damage to crops and small domestic animals [5, 24, 37, 38]. In the present study area, farmlands surround BSNP in all directions, particularly along the forested boundaries, while sleeping cliffs are situated mainly at the park’s edge. This spatial arrangement may have allowed regular movement of hamadryas baboons outside the park boundary to feed on standing crops, harvested yields, agricultural leftovers, and small livestock. This proximity creates recurrent overlapping zones of human-baboon interaction, which may increase the exposure of edge-dwelling baboon populations to localized hunting pressure [5].

Residents around BSNP primarily practice mixed farming, relying on both crop cultivation and livestock rearing for their economic subsistence. This socio-economic reliance on a combination of crop production and livestock husbandry for rural livelihoods mirrors patterns documented across several other conservation landscapes in Ethiopia [14, 19, 24, 37]. This shared agricultural dependency underscores a broader regional pattern whereby agricultural frontiers adjacent to protected areas remain highly attractive for farming due to fertile soils and access to water resources, despite the inherent risks of human–wildlife interactions [39]. Consequently, variation in land-use intensity and livestock husbandry practices along protected-area boundaries may substantially influence the nature and intensity of human–wildlife interactions across these shared ecotones [6, 37, 40, 41].

The concentration of attractive resource patches, particularly crops and livestock along the BSNP boundary, may be a primary factor driving the hamadryas baboons foraging outside the park, as these anthropogenic resources can provide highly rewarding foraging opportunities regardless of food availability within the protected area [7, 42, 43]. The widespread experience of livestock depredation by hamadryas baboons observed in this study, with sheep and goats bearing the most severe damage, comports with behavioral trends documented in other protected and community-managed ecosystems in Ethiopia, including the Simien Mountains National Park [23], Awash National Park [24], and Gasera District [25]. Rather than representing an isolated phenomenon, this pattern of livestock depredation reflects a widespread behavioral adaptation among hamadryas baboon populations across diverse Ethiopian agro-ecosystems, where the close juxtaposition of agricultural frontiers and protected-area boundaries increases opportunities for human–baboon interactions [37, 44].

Seasonal variation is widely recognized to play a key role in shaping the diet and foraging movements of many baboon species [16]. In the present study, reports of livestock attacks by hamadryas baboons were markedly more severe during the dry season than the wet season. This seasonal pattern may be associated with local livestock management practices, as domestic animals were often kept in Kuren, areas where harvested crops are stored and seeds are traditionally obtained, with limited protection while households focused on harvesting activities. In addition, the availability of preferred natural food resources, such as fruits, is lower in BSNP during the dry season [16], which may increase the tendency of hamadryas baboons to exploit alternative food sources outside their natural habitat.

At the spatial scale, reported sheep losses were significantly higher in the Borena District than in the Mehal-Sayint District. Local respondents attributed this spatial difference to the relatively greater abundance of hamadryas baboons along the Borena section of the park [45], a finding that contrasts with our initial hypothesis that livestock attacks would be more frequent in the more fragmented portion of BSNP. Collectively, these findings highlight the importance of seasonal resource dynamics and livestock husbandry practices in shaping human–hamadryas baboon interactions and suggest that improving livestock protection during high-risk periods could help reduce conflict between local communities and baboons.

Some respondents reported incidents of retaliatory actions against hamadryas baboons in BSNP following attacks on livestock. We documented these incidents through questionnaire surveys rather than direct field observations, with respondents indicating that they had observed evidence of human retaliation against baboons in agricultural fields and around household areas. Although the reported incidents could not be independently verified, the occurrence of retaliatory responses is consistent with findings from Awash National Park, where conflict over shared resources similarly resulted in targeted baboon mortality, and carcasses were directly observed within agricultural landscapes during field surveys [24]. This reliance on lethal intervention reflects a broader pattern observed in human-dominated landscapes across Africa, where problematic wildlife is often managed through retaliatory killing or localized population reduction [38]. The findings of the present study highlight the vulnerability of baboons to retaliatory killings and underscore the need for mitigation strategies that reduce conflict while promoting the species’ conservation.

During the group discussion and informal talk during the questionnaire survey, the local people reported that hamadryas baboons also consumed honey from the hives [45]. First, they snatch a portion of honey from the bee hives and throw it to the ground to repel the bees, and then feed on the honey. In addition, if the beehive is small in size, the baboons throw it down to the ground so that the honey is released from the hive for consumption [45]. Consequently, the local people usually cover their beehives with thorny plants several times to avoid the attack by hamadryas baboons (Ibrahin H. pers. obs.).

Our findings indicate that local communities surrounding BSNP generally hold negative perceptions toward hamadryas baboons, with most respondents favoring either a reduction in their numbers or their complete removal from the local landscape. These attitudes likely reflect the substantial impacts of crop foraging [5] and livestock depredation on subsistence livelihoods. Consistent with this pattern, 82.3% of respondents perceived hamadryas baboons as a significant threat to their livelihoods, a finding that aligns with reports from other regions where the species is often viewed negatively because of its foraging behavior and the associated economic losses [24, 46, 47]. However, a more nuanced pattern emerged regarding conservation attitudes. Despite these negative perceptions, most respondents (91.3%) stated that local communities did not wish to threaten the long-term survival of the hamadryas baboons. This suggests that, although many community members favor reducing hamadryas baboon numbers locally, they are generally reluctant to support indiscriminate lethal control or eradication. Thus, while resource competition generates negative attitudes, it does not entirely erode conservation tolerance, indicating potential for community-based conflict mitigation and long-term human–hamadryas baboon coexistence.

When comparing these attitudes across regions, the socio-ecological context must be carefully considered. For example, Biquand et al. [46] reported high levels of tolerance toward commensal hamadryas baboons in Saudi Arabia, where handfeeding by local communities and tourists was common. However, this provisioning-based system differs markedly from the situation around BSNP, where subsistence farmers experience direct economic losses from hamadryas baboon foraging behavior. Within Ethiopia, negative attitudes accompanied by retaliatory actions have similarly been reported among communities living near Trigni Forest toward anubis baboons [37] and around the Zegie Peninsula toward grivet monkeys [48]. Generally, these studies suggest that although direct economic impacts are important drivers of negative perceptions toward primates, the level of local tolerance and the severity of responses vary according to resource availability, land-use intensity, and local socio-cultural conditions.

Within the current study area, traditional active defense mechanisms, primarily guarding and shouting, were reported as the principal techniques used to reduce livestock depredation and crop foraging [5, 45]. The reliance on human guarding of both cultivated fields and domestic herds has been widely documented as a common mitigation strategy across regions to protect agricultural landscapes [13, 46, 47]. However, guarding is time-consuming and keeps people away from other activities, especially preventing children from attending school [40, 49]. Alternative non-lethal approaches, including the use of physical barriers and domestic dogs, were also reported by respondents as effective means of reducing livestock losses [10, 42, 46]. Given the seasonal decline in preferred natural food resources documented for hamadryas baboons in BSNP [16], reducing reliance on labor-intensive guarding may require a combination of improved livestock husbandry practices, strategic use of physical barriers, and habitat management measures that enhance the availability of natural food resources within the park. Such interventions, together with the establishment of buffer zones between agricultural fields and wildlife habitats, may reduce the attractiveness of human-modified areas and lower the frequency of direct interactions between hamadryas baboons and local communities.

Generally, our findings should be interpreted with caution because assessments of livestock losses were based on respondents’ perceptions rather than independently verified field observations. Although we instructed respondents to report only events based on their own direct experiences and observations, no formal recall-bias reduction techniques were employed. Consequently, the five-year recall period may have introduced recall bias. In addition, the predominance of male respondents may have introduced gender-related bias, as men and women can differ in their experiences, perceptions, and responses to wildlife-related conflicts. Furthermore, ecological factors such as food availability, habitat productivity, and baboon population density were not directly quantified. Consequently, the reported patterns should be viewed as descriptive associations rather than definitive evidence of ecological causation.

Conclusion

Hamadryas baboons in BSNP are reported by local communities to associate with substantial losses to both crops [5] and domestic animals, contributing to notable socio-economic challenges for surrounding subsistence households. Our findings suggest that sheep and lambs were the livestock categories most frequently reported as vulnerable to attacks attributed to hamadryas baboons. Seasonal variation in livestock attacks suggests that conflict intensity may be associated with fluctuations in natural food availability [16] and agricultural activities. Although most respondents did not report direct retaliatory killing, continued livestock losses may contribute to negative attitudes toward the animal.

Given that most respondents already report conflict incidents to park scouts, BSNP authorities have an opportunity to strengthen community-based monitoring and conflict mitigation programs. Improving livestock protection, enhancing supervision during high-risk periods, and increasing community participation in conservation initiatives may help reduce conflict. Future studies should integrate ecological monitoring, GPS telemetry, and direct behavioral observations to better understand the drivers of livestock attacks and human–hamadryas baboon conflict in the area.

Abbreviations

B

Unstandardized regression coefficient

BSNP

Borena-Sayint National Park

Exp(B)

Exponentiated regression coefficient (odds ratio)

FGDs

Focus group discussions

GPS

Global positioning system

mm

millimetre

n

number of respondents

SD

Standard deviation

SE

Standard error

SPSS

Statistical package for the social sciences

Author contributions

H.I, A.B and A.Me conceived and designed the study. H.I conducted data collection. H.I and A.Me performed data organization and analysis. H.I, A.B, D.Y, A.Mo, M.C, and A.Me interpreted the results and contributed to writing the manuscript.

Funding

This research project was supported financially by Primate Conservation, Inc. (PCI), and IDEA WILD provided essential equipment for field data collection. The funding organizations were not involved in the study design; the collection, analysis, and interpretation of data; the writing of the manuscript; or the decision to submit the article for publication.

Data availability

All data supporting the findings of this study are available within the paper.

Declarations

Ethics approval and consent to participate

We obtained ethical approval for this study from the Research Ethics Committee of Wollo University (approved in September 2017), and official research permission was obtained from the Ethiopian Wildlife Conservation Authority. We conducted all components of this research involving human participants in accordance with the ethical principles outlined in the Declaration of Helsinki. Before data collection, we explained the study’s objectives and data-use policies to all prospective respondents. Then, we obtained written or oral informed consent voluntarily from each participant via an Informant Consensus Form. Because all study participants were aged 18 years or older, consent from legal guardians or representatives was not required. Furthermore, all non-invasive observational protocols and data collection methods involving non-human primates adhered to the ethical standards established by the American Society of Primatologists.

Consent for publication

This manuscript does not contain any person’s data, and further consent for publication is not required/ applicable.

Competing interests

The authors declare no competing interests.