Abstract
Introduction:
The grey wolf (Canis lupus) is an important apex predator whose persistence in human-dominated landscapes is increasingly influenced by livestock depredation and negative human perceptions. This study investigated grey wolf diet composition, reported livestock depredation patterns, and community perceptions in Orakzai District, Khyber Pakhtunkhwa, Pakistan.
Methods:
Field surveys were conducted from August 2024 to July 2025 across 25 purposively selected sites. Grey wolf scats were collected and analysed using morphological characteristics and microscopic examination of prey remains. Structured household interviews were conducted with 123 respondents across 25 villages. Firth penalized logistic regression was used to assess factors associated with negative rather than neutral perceptions of grey wolves.
Results:
A total of 29 wolf scats yielded 30 identified prey occurrences representing 15 prey species. Wild prey accounted for 17 occurrences (56.67%), whereas domestic prey accounted for 13 (43.33%). Wild boar (Sus scrofa) was the most frequently detected prey species, followed by Asiatic palm civet (Paradoxurus hermaphroditus), while cattle (Bos taurus) were the most frequently detected domestic prey. Dietary diversity was H′ = 2.52, with prey species richness of 15 and evenness of 0.93. These findings represent prey occurrences detected in scats and cannot establish prey preference or selection because prey availability was not independently quantified. All 123 respondents reported at least one livestock attack attributed to grey wolves. Most reported attacks occurred at night (61.0%), followed by evening (35.0%), and October had the highest frequency of reported attacks (20.5%). Overall, 80.5% of respondents expressed negative perceptions of grey wolves. The reported number of grey wolves killed during the previous 5–10 years was positively associated with the odds of negative rather than neutral perception (OR = 2.495, 95% CI = 1.453–4.740, P < 0.001). Age, occupation, livestock holdings, wolf sightings, attack timing, and livestock injury were not statistically significant.
Discussion:
The findings highlight substantial interactions between grey wolves and livestock-dependent communities. They emphasize the need for evidence-based conflict mitigation, improved livestock protection, and conservation strategies informed by prey availability and local ecological conditions.
1 Introduction
The grey wolf (Canis lupus) is one of the world’s most widely distributed terrestrial carnivores and serves as an apex predator that can influence ecosystem structure and function through top-down effects on prey populations and trophic interactions (Ordiz et al., 2013). However, the strength and ecological expression of these effects vary with landscape context, particularly along gradients of human disturbance and management intensity. Evidence for strong trophic effects is more consistent in relatively intact or less intensively managed systems, whereas such effects may be weaker or more difficult to detect in highly modified landscapes. By influencing prey populations and their behaviour, wolves can contribute to ecological processes, although the magnitude of these effects depends on environmental and anthropogenic context (Kuijper et al., 2016). Despite their ecological importance, wolf populations have experienced substantial declines in many regions because of habitat degradation, persecution, fragmentation, and increasing anthropogenic pressures (Mech and Boitani, 2003). Historically, the grey wolf occupied much of North America and Eurasia and represents the largest extant member of the family Canidae (Paquet and Carbyn, 2003; Rehman et al., 2021). Although classified globally as Least Concern by the International Union for Conservation of Nature (IUCN), the species remains threatened in several parts of its range because of local population declines and persistent human-wolf conflict. In Pakistan, grey wolves occur across a range of mountainous and arid landscapes, particularly in northern and northwestern regions. Available studies indicate that wolf populations in Pakistan are influenced by habitat conditions, prey availability, livestock depredation, and human-wolf conflict (Khan et al., 2020; Ali et al., 2016; Bano et al., 2021).
Understanding predator diet is fundamental for interpreting predator ecology because dietary composition can provide insights into survival, reproduction, habitat use, and predator-prey interactions (Litvaitis, 2000). Diet composition can also provide information about the prey species detected within a predator’s range and the ecological conditions under which feeding occurs. As an adaptable carnivore, the grey wolf exhibits considerable dietary variation across its geographical range in response to prey availability, habitat characteristics, climatic conditions, human activities, and interactions with sympatric carnivores (Fritts et al., 2003). Previous studies have shown that wolves consume a broad spectrum of prey, including large ungulates, livestock, small mammals, rodents, lagomorphs, birds, insects, and occasionally fruits, depending on local ecological conditions and prey availability (Singh and Kumara, 2006; Jhala, 2003).
The cryptic behaviour, extensive home ranges, predominantly nocturnal activity, and naturally low population densities of grey wolves make ecological studies particularly challenging (Clark, 1999). Consequently, information regarding wolf ecology remains scarce throughout much of Asia compared with Europe and North America (Hendricks et al., 2016). In Pakistan, available studies of grey wolf ecology and human-wolf conflict have largely focused on northern mountain ecosystems, including Swat, Gilgit-Baltistan, and Azad Jammu and Kashmir (Khan et al., 2020; Mahmood et al., 2021; Ali et al., 2016). In contrast, relatively little is known about wolf ecology and human-wolf interactions in several under-studied mountainous districts of Khyber Pakhtunkhwa, where rugged terrain, livestock-based livelihoods, human settlements, and natural habitats occur in close proximity. These landscape and livelihood characteristics may influence prey availability, livestock exposure, and the nature of human-wolf interactions. Large carnivores frequently come into conflict with humans because they compete for shared resources, particularly domestic livestock (Ripple et al., 2014).
Human-wolf conflict is recognized as an important challenge for long-term wolf conservation (Ali et al., 2016; Jamtsho and Katel, 2019). Livestock depredation can impose substantial economic losses on rural communities whose livelihoods depend heavily on animal husbandry and may contribute to retaliatory killing of wolves through poisoning, shooting, trapping, or other lethal control measures (Singh and Kumara, 2006; Ripple et al., 2014). Such retaliatory actions can contribute to population declines and range contraction in areas where persecution is substantial (Bocci et al., 2017).
(Distefano, 2005) describes human-wildlife conflict as interactions between people and wildlife that result in negative impacts on human well-being or wildlife conservation. Among large carnivores, wolves are frequently involved in conflict because of their extensive home ranges and potential use of livestock where wild prey is limited (Krithivasan et al., 2009). Although attacks on humans are extremely rare, livestock depredation can contribute to negative perceptions toward wolves, particularly in livestock-dependent communities (Khan et al., 2019; Ali et al., 2016; Khan et al., 2020). Similar challenges have been documented across Pakistan, including Machiara National Park and Chitral, where livestock depredation has contributed to economic losses and persistent human-carnivore conflict (Dar et al., 2009; Ahmad, 2020; Treves and Karanth, 2003). The occurrence and intensity of livestock depredation can be influenced by multiple ecological and anthropogenic factors, including prey availability, habitat characteristics, seasonal livestock movements, vegetation cover, terrain complexity, predator abundance, and livestock husbandry practices (Bano et al., 2021; Din et al., 2019). During winter, heavy snowfall may force wolves to lower elevations, potentially increasing contact with livestock and local communities (Bocci et al., 2017). Consequently, understanding both the ecological dimensions of wolf diet and the socio-economic dimensions of human-wolf conflict is important for developing evidence-based conservation strategies that promote coexistence between wolves and pastoral communities.
Although grey wolves are recognized as important components of terrestrial predator communities, evidence concerning their ecology and interactions with people in human-dominated landscapes of Pakistan remains limited. In Orakzai District, comprehensive information on wolf dietary composition, reported livestock depredation patterns, and respondents’ perceptions are also scarce. Addressing these knowledge gaps is important for developing evidence-based conservation and conflict-mitigation strategies in this human-wolf conflict landscape. Therefore, the present study investigated the dietary composition of grey wolves through scat analysis, quantified patterns of reported livestock depredation and respondents’ perceptions using household surveys, and identified factors associated with perceptions toward grey wolves using statistical modelling. The findings provide baseline ecological and socio-economic information that can support evidence-based management and contribute to long-term human-wolf coexistence in northwestern Pakistan.
2 Materials and methods
2.1 Study area
The study was conducted in Orakzai District, Khyber Pakhtunkhwa Province, northwestern Pakistan (Figure 1). Orakzai is situated between 33°33′-33°54′ N latitude and 70°36′-71°22′ E longitude and covers an area of approximately 1,583 km². The district is administratively divided into Upper and Lower subdivisions, comprising four tehsils: Upper, Ismailzai, Lower, and Central. The landscape is predominantly mountainous, with elevations ranging from approximately 1,700 to over 3,000 m above sea level, characterized by rugged terrain, narrow valleys, and extensive forest patches.
Figure 1
The climate is characterized by cold winters with frequent snowfall during December-February and mild summers, where temperatures rarely exceed 30 °C. Annual precipitation averages approximately 910 mm, primarily influenced by the district’s mountainous topography and forest cover. Major valleys include Khanki, Mastura, and Khurmana, while approximately 1,318 acres of the district are covered by forest, providing suitable habitat for grey wolves and their prey species.
The landscape of Orakzai District comprises rugged mountainous terrain interspersed with human settlements, agricultural areas, grazing grounds, and natural vegetation. Livestock husbandry forms an important component of local livelihoods, creating regular spatial overlap between pastoral activities and wildlife habitat. Such overlap can increase opportunities for interactions between wolves and livestock, particularly where livestock are grazed or guarded in areas used by wildlife. However, the present study did not quantify landscape-level human disturbance, wild prey abundance, or wolf population density. Consequently, these factors are considered as potential components of the local human-wolf conflict context rather than as measured predictors in the present study. Published estimates of grey wolf abundance are available for some other regions of Pakistan, including Swat and Gilgit-Baltistan, but no population-density estimate has been reported specifically for Orakzai District to date (Khan et al., 2020; Abbas et al., 2013). Consequently, the present study provides information on wolf occurrence and human-wolf interactions in Orakzai but did not estimate wolf population size or density.
2.2 Study design
Field surveys were conducted between August 2024 and July 2025 across 25 sampling sites representing potential grey wolf habitat in Orakzai District (Table 1). Sampling sites were purposively selected based on previous evidence of grey wolf occurrence, including pugmarks and other wolf signs, previous livestock depredation records, reports from local residents, habitat characteristics, and information provided by wildlife personnel. Because sampling sites were purposively selected based on evidence of wolf occurrence and conflict, the sampling design was intended to characterize wolf presence and associated conflict patterns rather than provide an unbiased estimate of district-wide wolf abundance or prevalence.
Table 1
| S# | Sites Name | Geographical Coordinates | Elevation | Habitat characteristics | |
|---|---|---|---|---|---|
| Latitude | Longitude | (m) | Forest/plain/mountainous | ||
| 1 | Nanawar Ghar | 33°42.800 | 71°00.270 | 1898.38 | Forest |
| 2 | Landu ka top | 33°43.421 | 71°01.508 | 2028.39 | Forest |
| 3 | Jabar top | 33°40.791 | 70°56.181 | 2203.10 | Forest |
| 4 | Boghlaan | 33°44.037 | 70°59.131 | 1812.92 | Mountainous |
| 5 | Zaridar | 33°45.831 | 71°00.372 | 1711.97 | Mountainous |
| 6 | Serona Mishti | 33°45.105 | 70°50.881 | 1863.22 | Plain |
| 7 | Bahramzai | 33°45.557 | 71°02.355 | 1891.06 | Forest |
| 8 | Starsam | 33°42.335 | 71°07.951 | 946.79 | Plain |
| 9 | Feroz Khel | 33°47.787 | 71°03.332 | 1572.77 | Mountainous |
| 10 | Stori Khel | 33°49.775 | 71°05.080 | 1586.73 | Mountainous |
| 11 | Mandary Ghar | 33°45.660 | 71°01.662 | 1941.04 | Mountainous |
| 12 | Zaira | 33°45.144 | 71°10.492 | 1525.12 | Mountainous |
| 13 | Bazi | 33°38.419 | 71°19.484 | 1015.54 | Mountainous |
| 14 | Kuryaz | 33°42.458 | 71°03.222 | 1434.42 | Mountainous |
| 15 | Chappar Mishti | 33°40.106 | 71°06.823 | 927.04 | Mountainous |
| 16 | Wampara | 33°38.324 | 71°06.293 | 964.83 | Mountainous |
| 17 | Suleiman Khel | 33°40.973 | 71°01.667 | 1420.75 | Mountainous |
| 18 | Targonda | 33°41.570 | 71°03.159 | 1381.89 | Mountainous |
| 19 | Chan | 33°41.258 | 71°15.198 | 1174.21 | Plain |
| 20 | Petao mela | 33°40.880 | 71°16.944 | 994.64 | Plain |
| 21 | Naqi kalay | 33°40.864 | 71°17.010 | 988.53 | Plain |
| 22 | Amber sae | 33°40.704 | 71°16.956 | 1012.84 | Plain |
| 23 | Kamber talab | 33°40.736 | 71°16.962 | 1002.62 | Plain |
| 24 | Mansha talab | 33°44.993 | 71°00.399 | 1718.62 | Mountainous |
| 25 | Sae mela | 33°40.797 | 071°16.881 | 1017.38 | Mountainous |
| Total sites25 | |||||
Details of the 25 sampling sites surveyed for grey wolf (Canis lupus) scats in Orakzai District, Pakistan.
Bold formatting is used to distinguish table headings and total/subtotal rows; it does not indicate statistical significance.
At each selected site, four line transects, each 2 km in length, were established for systematic searches for grey wolf signs and scats, with an effective search width of 10 m (5 m on either side of the transect line). Thus, the nominal transect length was 8 km per site and 200 km across the 25 sites for a single complete survey round. Surveys were conducted monthly, with three consecutive field days per survey and approximately eight hours of fieldwork per day. The same transects were not necessarily surveyed with equal frequency throughout the study period. Sites and transects considered highly suitable, or where repeated wolf sightings or fresh signs were detected, were revisited in some cases three to four times to increase the probability of detecting wolf signs and scats. Thus, the transect design was standardized across sampling sites, whereas the realized sampling effort varied according to field conditions and evidence of wolf occurrence.
When multiple suspected wolf scats were encountered at the same or nearby locations, all were collected separately, assigned individual identification codes, and georeferenced using GPS to retain their spatial identity. A total of 29 grey wolf scat samples were subsequently identified and analysed for dietary composition.
2.3 Dietary analysis
2.3.1 Scat collection and laboratory analysis
Grey wolf (Canis lupus) scats were collected during monthly field surveys using line-transect searches across the selected sampling locations. Scats were identified in the field based on a combination of morphological and contextual characteristics rather than a single diagnostic feature. Identification criteria included scat length and width, overall cylindrical or cord-like form, tapered ends, colour ranging from dark or blackish to greyish depending on condition and diet, and the presence of undigested hair and bone fragments. Scats were also evaluated in relation to associated wolf signs, particularly the presence of wolf tracks near the deposition site. Because domestic dogs occur in the study landscape and can produce morphologically similar scats, identification relied on the combination of multiple characteristics, including size, tapered morphology, hair and bone content, and associated wolf tracks, rather than size alone. Samples showing characteristics inconsistent with wolf scats or lacking sufficient supporting field evidence were excluded from the dietary analysis. Field identification was supported by published criteria for grey wolf scat identification (Khan et al., 2020).
Scat freshness was assessed in the field based on visible characteristics such as moisture, colour, surface condition, and overall appearance. Relatively fresh and intact scats were prioritized for collection to reduce the likelihood of substantial weathering or decomposition affecting dietary remains. Each suspected wolf scat was photographed, individually labelled, and georeferenced using a GPS receiver before being transported to the laboratory. Scat length, mean width, and fresh weight were recorded before laboratory processing. Across the 29 analysed scat samples, mean scat length was 76.09 ± 25.80 mm, mean width was 20.52 ± 3.69 mm, and mean fresh weight was 9.04 ± 5.49 g (mean ± SD).
All scats that met the field-identification criteria were collected and treated as individual scat samples. Although multiple scats were occasionally encountered within the same sampling area, most analysed samples were spatially separated by approximately 800 m to 2 km. Each sample was individually labelled and georeferenced, and multiple scats encountered in the same area were not subsampled to a single specimen. Because individual wolves could not be identified from field observations, complete statistical independence among scat samples cannot be assumed. Therefore, dietary results were interpreted as detected prey occurrences in the collected scat samples rather than as independent observations of individual wolves or estimates of individual wolf diets.
In the laboratory, scat samples were soaked, washed, and sieved to remove the faecal matrix and recover identifiable dietary components. This procedure was based on the method described by Mukherjee et al. (1994) and was adapted for grey wolf scat to facilitate the recovery, identification, and dry-mass measurement of prey remains. Recovered materials were separated into six broad categories: hair, bones, plant/wood material, anthropogenic material, keratinous remains, and stones. Hair was used as the principal material for species-level prey identification through microscopic examination and comparison with reference characteristics. Bones, feathers, porcupine quills, and mammalian nails/claws were recorded at the broad material-category level because these remains were not sufficiently diagnostic for reliable species-level identification. These materials were therefore used to characterize the composition of recovered scat remains but were not assigned to individual prey species. Keratinous remains included bird feathers, porcupine quills, and mammalian nails/claws, whereas plant/wood material was recorded as an undigested component but was not taxonomically identified.
Recovered components were oven-dried and weighed, and their frequencies were recorded. Each distinct prey species identified within a scat was recorded as a separate prey occurrence; consequently, the total number of prey occurrences could exceed the number of analysed scats. Prey composition was quantified using relative frequency of occurrence (RFO), calculated as:
where ni is the number of occurrences of prey species i and N is the total number of identified prey occurrences (N = 30). RFO values represent the relative frequency of detected prey occurrences and are not estimates of biomass consumed.
The frequency of each recovered material category was calculated as the number of occurrences of that component divided by the total number of recovered material occurrences (N = 63), multiplied by 100. Because recovered components were weighed after oven-drying, their proportional contribution was expressed as relative dry mass rather than percentage volume. Relative dry mass was calculated as the dry mass of each component divided by the total dry mass of all recovered components (173.74 g), multiplied by 100.
2.3.2 Prey species indices
The diversity of prey species detected in grey wolf scats was assessed using the Shannon-Wiener diversity index (H′), prey species richness (S), and prey species evenness (E), following Steinmetz et al. (2021). Proportional occurrence (pi) for each prey species was calculated from the number of occurrences of that species relative to the total number of identified prey occurrences (N = 30).
The Shannon–Wiener diversity index was calculated as:
where pi represents the proportional occurrence of prey species i and ln is the natural logarithm.
Prey species richness (S) was defined as the total number of prey species identified from the analysed scat samples. Prey species evenness (E) was calculated as:
where H′ is the Shannon-Wiener diversity index and S is the total number of identified prey species.
2.4 Human-grey wolf conflict survey
A cross-sectional household survey was conducted to evaluate reported livestock depredation patterns, respondents’ perceptions, and factors associated with perceptions toward grey wolves. Data were collected from 123 households across 25 villages distributed within Orakzai District. The surveyed villages encompassed areas where grey wolf occurrence and human-wolf interactions, particularly livestock depredation, had been reported. Respondents were selected from households actively engaged in livestock husbandry and/or those with reported experience of grey wolf depredation during the previous five years because these households were considered most likely to possess relevant experience of livestock losses, wolf encounters, and associated human-wolf conflict.
Within the selected villages, households were identified using a combination of availability-based and snowball sampling. Households that were available and met the study criteria were approached during field visits, and participating respondents subsequently helped identify other eligible households with relevant experience of livestock husbandry and grey wolf interactions. This approach facilitated access to households with direct experience of human-wolf conflict in a geographically dispersed and locally difficult-to-access study area. Households were therefore not selected through random or probability-based sampling. Although the survey covered 25 villages distributed across the study area, the non-probability sampling design may have resulted in greater representation of households with livestock ownership or previous experience of wolf encounters and depredation. Consequently, the findings describe the surveyed households and should not be assumed to represent the perceptions of all residents of Orakzai District. A population-level sampling proportion could not be calculated because a complete sampling frame of livestock-owning or conflict-exposed households across the 25 surveyed villages was not available.
The questionnaire was administered through face-to-face interviews by the research team during household visits, with assistance from local wildlife staff. The researchers completed the questionnaire forms based on information provided directly by respondents. The questionnaire was prepared in English based on questions and themes used in previous published studies of human-wildlife conflict, livestock depredation, and attitudes toward large carnivores, with additional questions included to address the specific objectives of the present study. Although the questionnaire was written in English, interviews were conducted verbally in Pashto, the local language of the respondents. One respondent was interviewed per household and provided information representing the household’s experience and views.
Household interviews were conducted with male household representatives because female participation in household interviews was not locally permitted. Consequently, all 123 respondents were male. This gender imbalance is acknowledged as a limitation because the survey could not assess potential gender-related differences in perceptions or generalize the findings across genders. No refusals or incomplete questionnaires were recorded; all 123 participating respondents completed the questionnaire. The complete questionnaire is provided as Supplementary Material S1.
The questionnaire comprised three sections. Section 1 recorded demographic information, including age, gender, occupation, education, and locality. Section 2 recorded livestock ownership, knowledge and observations of grey wolves, wolf encounters, livestock depredation, timing and type of attacks, livestock categories affected, compensation, and respondents’ views regarding the causes of livestock attacks. Section 3 recorded respondents’ perceptions toward grey wolves, preferred conflict-management strategies, knowledge of grey wolf conservation, wolf-killing history, reasons for killing, the reported number of wolves killed during the previous 5–10 years, and views on peaceful conflict resolution.
Information on livestock depredation was based on respondents’ reports of livestock attacks that they attributed to grey wolves. These reported incidents were not independently verified through veterinary examination, carcass investigation, wildlife-authority records, or other forensic assessment. Predator attribution therefore represents respondents’ reported identification of the predator rather than independently confirmed wolf depredation. Respondents were asked about the timing of reported attacks, livestock categories affected, and their explanations for the occurrence of attacks. These data were used to characterize respondent-reported depredation patterns and their temporal distribution and were not treated as independently verified depredation events or as a standardized measure of depredation risk.
Respondents’ perceptions of grey wolves were assessed during face-to-face interviews. The questionnaire was administered manually by the research team, and respondents were asked to describe their perception of grey wolves. Responses were recorded as positive, negative, or neutral according to the respondent’s expressed view. Where respondents expressed neither a clearly positive nor a clearly negative view, the response was recorded as neutral by the interviewer. No respondent selected the positive category; therefore, for binary logistic regression, the observed responses were restricted to negative and neutral perceptions, with negative perception coded as 1 and neutral perception coded as 0. Reported wolf-killing history was recorded using a question asking whether the respondent or community had previously killed a grey wolf. The number of grey wolves reported as killed during the previous 5–10 years was recorded as a count variable ranging from 0 to 4. Although the questionnaire initially elicited respondents’ estimates of the number killed in their community, these reported numbers were cross-checked with information obtained from local elders and wildlife staff in Orakzai District before being used in the analysis. Response formats varied according to the question and included categorical yes/no responses, categorical multiple-choice responses, numerical counts, and the three-category perception response (positive, negative, or neutral). Questions concerning agreement with peaceful conflict resolution used ordered response categories, which were summarized as agreement (agree/strongly agree) and disagreement (disagree/strongly disagree) for descriptive reporting. All survey responses were entered into Microsoft Excel and subsequently analysed in R version 4.5.1 (R Core Team).
2.5 Statistical analysis
Descriptive statistics were used to summarize dietary and household survey data. Continuous variables are presented as means ± standard deviations, whereas categorical variables are presented as frequencies and percentages. Dietary composition was summarized using relative frequency of occurrence (RFO), and dietary diversity, prey species richness, and evenness were calculated as described in Section 2.3.2.
To examine factors associated with respondents’ perceptions of grey wolves, a bias-reduced (Firth) logistic regression model was used because of sparse observations in some response categories. Perception was initially recorded as a three-category nominal response (positive, negative, or neutral); because no respondent selected the positive category, the regression analysis was restricted to negative and neutral perceptions, with negative perception coded as 1 and neutral perception coded as 0. Candidate explanatory variables were selected a priori based on the study objectives and questionnaire data and included respondent age, occupation, total livestock holdings, grey wolf sightings, reported attack timing, livestock injury, and the number of grey wolves reportedly killed during the previous 5–10 years.
Age was treated as a continuous variable; occupation was grouped into livestock-dependent occupations (shepherds and farmers) and other occupations; total livestock holdings were log-transformed as log(1 + total livestock) because of right-skewness; wolf sightings were coded as observed versus not observed; attack timing was grouped as nighttime versus non-nighttime because of sparse observations; livestock injury was coded as any injury versus none; and the number of wolves reportedly killed during the previous 5–10 years was retained as a continuous count ranging from 0 to 4. Variables with invariant or near-invariant responses were excluded from the regression model, and wolf-killing history was not entered simultaneously with the number of wolves killed because these variables represented overlapping information. Odds ratios (ORs), 95% confidence intervals (CIs), and P-values were reported, with statistical significance assessed at α = 0.05. Because the household survey was cross-sectional, associations were interpreted as statistical relationships rather than causal effects.
3 Results
3.1 Dietary composition of grey wolf
Analysis of the 29 grey wolf scat samples yielded 63 recorded occurrences of recovered material. Hair was the most frequently recovered component (30 occurrences; 47.62%), followed by bones (13; 20.63%), keratinous remains (9; 14.29%), plant/wood material (5; 7.94%), anthropogenic material (4; 6.35%), and stones (2; 3.17%) (Table 2). The total dry mass of recovered material was 173.74 g. Hair contributed the largest proportion of the total dry mass (128.20 g; 73.79%), followed by bones (22.19 g; 12.77%) and anthropogenic material (15.82 g; 9.11%). These measurements describe the frequency and relatively dry-mass composition of materials recovered from the analysed scats and should not be interpreted as estimates of biomass consumed or the proportional contribution of prey species to the diet.
Table 2
| Recovered material | Frequency (n) | Dry mass (g) | Relative Frequency (%) | Relative mass (%) |
|---|---|---|---|---|
| Hair | 30 | 128.20 | 47.62 | 73.79 |
| Bones | 13 | 22.19 | 20.63 | 12.77 |
| Plant/wood material | 5 | 5.06 | 7.94 | 2.91 |
| Anthropogenic material | 4 | 15.82 | 6.35 | 9.11 |
| Keratinous remains | 9 | 2.24 | 14.29 | 1.29 |
| Stones | 2 | 0.23 | 3.17 | 0.13 |
| Total | 63 | 173.74 | 100.00 | 100.00 |
Frequency and dry mass of recovered material from grey wolf (Canis lupus) scat samples in Orakzai District, Pakistan.
Bold formatting is used to distinguish table headings and total/subtotal rows; it does not indicate statistical significance.
3.2 Prey species composition
A total of 30 prey occurrences representing 15 prey species were identified from the 29 analysed grey wolf scat samples (Table 3). Eight wild prey species accounted for 17 recorded occurrences (56.67%), whereas seven domestic prey species accounted for 13 occurrences (43.33%). Wild boar (Sus scrofa) was the most frequently detected wild prey species (6 occurrences; 20.00%), followed by Asiatic palm civet (Paradoxurus hermaphroditus) (3; 10.00%). Desert hare (Lepus tibetanus) and house mouse (Mus musculus) each accounted for two occurrences (6.67%). Among domestic animals, cow (Bos taurus) was the most frequently detected species (4 occurrences; 13.33%), followed by sheep (Ovis aries) and donkey (Equus asinus) (2 occurrences each; 6.67%). Wild and domestic prey therefore contributed substantially to the detected prey occurrences, with wild prey occurring somewhat more frequently. These relative frequencies describe prey occurrences detected in scat samples and should not be interpreted as estimates of prey preference, prey selection, prey availability, or biomass consumed.
Table 3
| Prey type | Prey species | No. of occurrences (n) | Relative frequency of occurrence (%) |
|---|---|---|---|
| Wild prey | |||
| Wild boar (Sus scrofa) | 6 | 20 | |
| Asiatic palm civet (Paradoxurus hermaphroditus) | 3 | 10 | |
| Desert hare (Lepus tibetanus) | 2 | 6.67 | |
| House mouse (Mus musculus) | 2 | 6.67 | |
| Black rat (Rattus rattus) | 1 | 3.33 | |
| Rhesus macaque (Macaca mulatta) | 1 | 3.33 | |
| Indian crested porcupine (Hystrix indica) | 1 | 3.33 | |
| Jungle cat (Felis chaus) | 1 | 3.33 | |
| Subtotal | 17 | 56.67 | |
| Domestic animals | Goat (Capra hircus) | 1 | 3.33 |
| Sheep (Ovis aries) | 2 | 6.67 | |
| Cow (Bos taurus) | 4 | 13.33 | |
| Donkey (Equus asinus) | 2 | 6.67 | |
| Domestic cat (Felis catus) | 1 | 3.33 | |
| Domestic dog (Canis lupus familiaris) | 1 | 3.33 | |
| Buffalo (Bubalus bubalis) | 2 | 6.67 | |
| Subtotal | 13 | 43.33 | |
| Total | 30 | 100.00 | |
Prey species composition and relative frequency of prey occurrences identified from grey wolf (Canis lupus) scat samples in Orakzai District, Pakistan.
Bold formatting is used to distinguish table headings and total/subtotal rows; it does not indicate statistical significance.
3.3 Dietary diversity
The dietary indices calculated from the 30 recorded prey occurrences identified from 29 grey wolf scat samples yielded a Shannon-Wiener diversity index (H′) of 2.52, prey species richness (S) of 15, and prey species evenness (E) of 0.93 (Figure 2). These values indicate that multiple prey species were detected in the analysed scat samples and that prey occurrences were distributed relatively evenly among the detected species. Because prey availability was not independently quantified in the study area, these indices describe the diversity and distribution of detected prey occurrences rather than prey selection or preference relative to environmental availability.
Figure 2
3.4 Human-grey wolf conflict
A total of 123 household respondents were interviewed across the study area. Respondents represented several occupational groups, including farmers (33.3%), self-employed individuals (30.1%), shepherds (13.0%), government employees (6.5%), and students, private employees, and members of the armed forces (5.7% each). All respondents were male. Respondents collectively reported 1,628 livestock, comprising goats (1,091; 67.0%), sheep (319; 19.6%), cattle (173; 10.6%), donkeys (35; 2.2%), and buffalo (10; 0.6%) (Table 4). Percentages represent the proportion of the total number of livestock individuals reported by respondents (N = 1,628), rather than the proportion of households owning each livestock type; therefore, they sum to approximately 100%.
Table 4
| Livestock | Number | Percentage of all recorded livestock (%) |
|---|---|---|
| Goat | 1,091 | 67.01% |
| Sheep | 319 | 19.59% |
| Buffalo | 10 | 0.61% |
| Donkey | 35 | 2.15% |
| Cow | 173 | 10.63% |
| Total | 1,628 | 100% |
Livestock composition reported by surveyed households in Orakzai District, Pakistan.
Bold formatting is used to distinguish table headings and total/subtotal rows; it does not indicate statistical significance.
Among the 107 respondents who provided a location for their wolf observation, hillsides were most frequently reported (73%), followed by pastures (17.8%), valleys (7.5%), and grazing areas (0.9%) (Figure 3). All 123 respondents reported at least one livestock attack attributed to grey wolves involving their household or community, consistent with the targeted sampling of households with livestock husbandry and/or previous experience of wolf depredation. Reported attacks were concentrated at night (75; 61.0%), followed by evening (43; 35.0%) and morning (5; 4.1%); no daytime attacks were reported (Figure 4). Among the 122 respondents who provided a month for the reported attack, October had the highest frequency of reported attacks (25; 20.5%), followed by November (17; 13.9%) (Figure 5). These values describe respondent-reported attack patterns and should not be interpreted as a standardized livestock-depredation rate because livestock exposure time and the number of livestock at risk were not systematically recorded for each household.
Figure 3
Figure 4
Figure 5
Among the 121 respondents who provided information on the livestock affected by reported attacks, goats were the most frequently selected single livestock category (52; 43.0%), followed by cattle (12; 9.9%), sheep (3; 2.5%), and donkeys (3; 2.5%); 51 respondents (42.1%) reported that all listed livestock categories had been affected. Respondents most frequently attributed attacks to reduced availability of natural prey (58; 47.9%) or livestock being a preferred food source (45; 37.2%), followed by livestock vulnerability (12; 9.9%) and livestock not being adequately gathered (6; 5.0%).
3.5 Perceptions of grey wolves and conservation attitudes among surveyed households
Perceptions toward grey wolves were predominantly negative among the surveyed respondents. Ninety-nine respondents (80.5%) expressed negative perceptions, while 24 (19.5%) expressed neutral perceptions; no respondent reported a positive perception (Figure 6). Overall, 88 respondents (71.5%) reported that they or members of their community had previously killed grey wolves, whereas 35 (28.5%) reported no previous wolf killing. The number of grey wolves reported as killed during the previous 5–10 years ranged from 0 to 4. The reported counts were cross-checked with information provided by local elders and wildlife staff in Orakzai District. Thirty-five respondents reported no wolves killed, 39 reported one, 40 reported two, seven reported three, and two reported four wolves killed. Among the 87 respondents who provided a reason for wolf killing, 79 (90.8%) cited livestock protection, six (6.9%) cited self-defense, and two (2.3%) reported another reason.
Figure 6
Respondents expressed varied preferences regarding wolf management. Increasing wild prey was the most frequently selected option (46; 37.4%), followed by translocation of wolves (37; 30.1%) and lethal control (35; 28.5%). Increased livestock guarding was selected by four respondents (3.3%), one respondent (0.8%) selected doing nothing, and none selected improvement of livestock-pen structures. Among the 121 respondents who provided a response regarding peaceful conflict resolution, 91 (75.2%) agreed or strongly agreed that human-apex predator conflict could be resolved peacefully, whereas 30 (24.8%) disagreed or strongly disagreed.
3.6 Factors associated with respondents’ perceptions
Because no respondent reported a positive perception, logistic regression contrasted negative perception (coded as 1) with neutral perception (coded as 0). A bias-reduced (Firth) logistic regression model was used to account for sparse observations in the survey data. After adjustment for age, occupation, total livestock holdings, wolf sightings, attack timing, and livestock injury, the reported number of grey wolves killed during the previous 5–10 years was positively associated with the odds of negative rather than neutral perception (OR = 2.495, 95% CI = 1.453–4.740, P< 0.001; Table 5; Figure 7). The other predictors were not statistically significant, including age (OR = 1.019, 95% CI = 0.981–1.061, P = 0.333), livestock-dependent occupation (OR = 0.849, 95% CI = 0.309–2.324, P = 0.748), log-transformed total livestock holdings (OR = 1.063, 95% CI = 0.599–1.939, P = 0.834), wolf sightings (OR = 0.950, 95% CI = 0.164–4.077, P = 0.948), nighttime attacks (OR = 1.745, 95% CI = 0.650–4.765, P = 0.268), and livestock injury (OR = 0.422, 95% CI = 0.156–1.129, P = 0.085). Thus, the number of wolves reportedly killed during the previous 5–10 years was the only predictor showing a statistically significant association with negative perceptions after adjustment for the other variables in the model. The predicted probability of negative perception increased with the reported number of wolves killed (Figure 8). Because the survey was cross-sectional and the wolf-killing counts originated from community-reported information that was subsequently cross-checked with local elders and wildlife staff, this association should be interpreted as a statistical association rather than evidence of a causal relationship between wolf killing and respondents’ perceptions.
Table 5
| Predictor | OR | 95% CI | P |
|---|---|---|---|
| Age | 1.019 | 0.981–1.061 | 0.333 |
| Livestock-dependent occupation | 0.849 | 0.309–2.324 | 0.748 |
| log(1 + total livestock) | 1.063 | 0.599–1.939 | 0.834 |
| Wolf sighting | 0.950 | 0.164–4.077 | 0.948 |
| Nighttime attack | 1.745 | 0.650–4.765 | 0.268 |
| Livestock injury | 0.422 | 0.156–1.129 | 0.085 |
| Number of grey wolves reported killed during the previous 5–10 years | 2.495 | 1.453–4.740 | <0.001 |
Firth penalized logistic regression model of factors associated with negative perceptions of grey wolves in Orakzai District, Pakistan.
OR = odds ratio; CI = confidence interval; P = P-value. The model assessed factors associated with negative rather than neutral perceptions of grey wolves using Firth penalized logistic regression. Odds ratios greater than 1 indicate higher odds of negative perception, whereas odds ratios less than 1 indicate lower odds, relative to the relevant reference category or per-unit increase in the predictor, as applicable. Bold values indicate statistically significant associations at P < 0.05. Interpret odds ratios alongside their 95% confidence intervals and P-values. These associations should not be interpreted as evidence of causation.
Figure 7
Figure 8
4 Discussion
4.1 Diet composition and dietary diversity
In the present study, analysis of 29 grey wolf scats yielded 30 identified prey occurrences representing 15 prey species. Wild prey accounted for 17 occurrences (56.7%), whereas domestic prey accounted for 13 occurrences (43.3%). Wild boar was the most frequently detected wild prey species, while cattle were the most frequently detected domestic prey species. These findings indicate that both wild and domestic prey formed substantial components of the detected dietary assemblage in Orakzai District, with wild prey occurring somewhat more frequently.
Previous studies from Pakistan have documented both wild and domestic prey in grey wolf diets, although their relative occurrence varies among study areas (Khan et al., 2020; Din et al., 2019; Mahmood et al., 2021). Dietary composition can vary considerably among human-dominated and pastoral landscapes depending on wild prey communities, livestock availability, livestock husbandry practices, seasonal movements of herds, habitat characteristics, and opportunities for scavenging. For example, Mohammadi et al. (2019) reported a high contribution of cattle to wolf diet in Hamadan Province, Iran, illustrating that domestic prey can constitute a substantial component of wolf diets under particular local conditions. In northern Pakistan, Din et al. (2019) also documented variation in wolf and snow leopard prey use in relation to livestock depredation and prey availability. Interactions with sympatric carnivores may further influence resource use and scavenging opportunities, particularly where human activities alter prey communities and food availability (Bocci et al., 2017). Seasonal changes in livestock movements, prey distribution, and husbandry practices may also modify the accessibility of wild and domestic prey (Singh and Kumara, 2006).
In Orakzai District, these ecological and anthropogenic factors may influence the relative availability and accessibility of wild and domestic prey to grey wolves. However, wild prey abundance, livestock density, carcass availability, livestock husbandry practices, free-ranging dog abundance, wolf density, and interactions among sympatric carnivores were not independently quantified in the present study. These factors should therefore be considered potential explanatory factors rather than measured determinants of the observed dietary composition. The substantial contribution of domestic prey should be interpreted as a pattern observed in a livestock-producing landscape rather than as evidence of prey preference or selection. Determining the relative importance of wild prey and domestic livestock would require future studies incorporating independent measures of prey availability, livestock exposure, and wolf population characteristics.
The dietary indices further indicated a Shannon-Wiener diversity index (H′) of 2.52, species richness (S) of 15, and evenness (E) of 0.93. These values indicate that multiple prey species were detected and that their occurrences were relatively evenly distributed among the detected species. However, because prey availability was not independently quantified, these indices describe the diversity and distribution of detected prey occurrences rather than prey selection or preference. The broad prey spectrum therefore indicates diversity in the detected dietary assemblage rather than demonstrating that wolves selected particular prey according to their availability.
4.2 Dietary composition and spatial sampling limitations
Spatial dietary variation could also not be evaluated reliably in the present study. Although wolf scats were individually georeferenced during collection, only 29 analysed scats yielding 30 identified prey occurrences were available, and scats were distributed among the subset of the 25 surveyed sites. Furthermore, individual wolves could not be identified, preventing inference about population- or site-specific diets. Consequently, the present results should be regarded as a study-area-level description of detected prey occurrences rather than evidence of dietary specialization among wolf populations. Future studies combining site-specific scat sampling with independent measurements of wild-prey and livestock availability would provide a stronger basis for evaluating spatial dietary variation and prey selection.
4.3 Human-grey wolf conflict and livestock depredation
Understanding human-wolf conflict requires consideration of the ecological, social, and economic factors that shape interactions between wolves and people (Younus et al., 2018). Livestock depredation is an important component of human-wolf conflict in many pastoral and livestock-producing landscapes and can impose economic costs on households that depend on livestock for their livelihoods (Ripple et al., 2014). Din et al. (2019) reported a relationship between livestock numbers and wolf depredation in the Hindu Kush and Himalayan mountains, with an average annual loss of 0.58 livestock animals per household.
In the present study, all 123 surveyed respondents reported at least one livestock attack attributed to grey wolves involving their household or community. This figure reflects the targeted sampling of households with livestock husbandry and/or previous experience of wolf depredation and should therefore not be interpreted as a district-wide depredation prevalence or rate. Research conducted by Bano et al. (2021) in Misgar Valley documented 310 livestock losses between 2014 and 2019, of which grey wolves were reported as responsible for 54%. Sheep, goats, domestic yaks, and cattle were the principal livestock categories affected. In the present study, goats represented the largest proportion of livestock owned by respondents (67.0%) and were also the most frequently reported single livestock category affected by wolf attacks (43.0% of respondents providing information on affected livestock). However, because goats also constituted the largest proportion of livestock owned by respondents, these data do not allow relative depredation risk among livestock species to be estimated.
Reported attacks occurred predominantly at night (61.0%), followed by evening (35.0%), while October had the highest frequency of reported attacks (20.5% of respondents providing a month of attack). The concentration of reported attacks during nighttime may be related to increased vulnerability of livestock when animals are confined or less actively guarded. However, this interpretation should be considered cautiously because livestock husbandry practices, livestock confinement, guarding intensity, and wolf activity were not independently monitored during the present study. The temporal pattern nevertheless identifies livestock protection as a potentially relevant component of local human-wolf conflict management.
Approximately 80% of respondents expressed negative perceptions toward grey wolves. Because the survey was conducted among households situated within a livestock and human-wolf conflict context, this proportion should not be generalized to the wider population of Orakzai District. Similar negative attitudes have been reported in other parts of Pakistan where compensation schemes and effective conflict-mitigation measures are limited (Dar et al., 2009; Ali et al., 2016). The present findings therefore indicate a substantial human-wolf conflict context among the surveyed households, particularly in relation to reported livestock depredation. However, the study does not establish a causal pathway between livestock losses, negative perceptions, and wolf killing.
4.4 Community perceptions and factors associated with negative perceptions
The household survey identified a positive association between the reported number of grey wolves killed during the previous 5–10 years and negative perceptions of grey wolves. Respondents reporting a greater number of wolves killed had higher odds of expressing negative rather than neutral perceptions after adjustment for age, livestock-dependent occupation, total livestock holdings, wolf sightings, attack timing, and livestock injury (OR = 2.495, 95% CI = 1.453–4.740, P< 0.001). In contrast, age, livestock-dependent occupation, livestock holdings, wolf sightings, nighttime attacks, and livestock injury were not statistically significant predictors of negative perception.
This association should not be interpreted as evidence that wolf killing itself causes negative perceptions. Because the survey was cross-sectional and the wolf-killing counts were based on community-reported information that was cross-checked with local elders and wildlife staff, the observed relationship may reflect reciprocal interactions between perceptions, conflict experiences, and lethal control, or other unmeasured socioeconomic and environmental factors. The finding may therefore indicate a broader relationship between the intensity of human-wolf conflict and negative perceptions rather than a direct causal effect of wolf killing on attitudes.
4.5 Study limitations
Several limitations should be considered when interpreting the findings of this study. First, dietary information was derived from scat analysis and therefore represents detected prey occurrences rather than direct observations of prey consumption. Scat analysis also cannot reliably distinguish between active predation and scavenging. In addition, prey availability was not independently quantified; consequently, the observed dietary composition cannot be used to assess prey selection, preference, or opportunistic use of domestic livestock. Scat identification was based on morphological and contextual characteristics rather than molecular confirmation, and domestic dogs can produce morphologically similar scats. Although multiple field characteristics and associated wolf signs were considered, some uncertainty in species identification cannot be completely excluded.
Second, the dietary dataset was relatively small, comprising 29 analysed scats and 30 identified prey occurrences. Although scats were georeferenced during collection, individual wolves could not be identified, and the available site-level dietary information was insufficient to reliably evaluate spatial dietary variation or specialization. Digestibility differences among prey items may also influence their representation in scats. Future studies using larger and spatially stratified scat samples, molecular confirmation where feasible, and independent measurements of prey availability would provide stronger estimates of dietary composition and potential prey selection.
Third, the household survey used availability-based and snowball sampling and targeted households involved in livestock husbandry and/or reporting previous wolf-attributed depredation. The sample was therefore non-probabilistic and was not designed to represent all households or residents of Orakzai District. Because all respondents reported at least one wolf-attributed livestock attack within the sampling context, the survey data cannot be used to calculate a standardized district-wide depredation rate. Livestock exposure, time at risk, and independently verified livestock losses were not systematically recorded. Information on past livestock attacks was based largely on respondents’ recollection and could therefore be affected by recall or reporting bias. Although reported wolf-killing counts were cross-checked with local elders and wildlife staff, historical information may still be subject to uncertainty because the events occurred during the previous 5–10 years and were not based on a standardized contemporaneous monitoring system. Fourth, all respondents were male because of local cultural restrictions, which may have limited representation of other household perspectives and prevented assessment of potential gender-related differences in perceptions. The study was also confined to Orakzai District, and its findings may not be directly generalizable to other regions with different ecological conditions, prey communities, livestock systems, and patterns of human-wolf interaction.
Finally, the perception model included the survey variables considered relevant to the study objectives, but potentially important ecological and husbandry-related factors, including wild-prey availability, livestock-guarding practices, enclosure characteristics, seasonal wolf movements, and wolf population characteristics, were not directly measured. The cross-sectional design also limits interpretation of the regression results to statistical associations rather than causal relationships. Future studies incorporating systematic prey-availability surveys, standardized livestock-loss monitoring, probability-based community sampling, and longitudinal data would provide a stronger basis for understanding the ecological and socioeconomic drivers of human-wolf conflict.
4.6 Conservation and management implications
The dietary results demonstrate that both wild and domestic prey occurred in the analysed scats. These findings should be interpreted as patterns of detected prey occurrence rather than evidence of prey preference or selection.
Conflict-management interventions should prioritize practical measures for reducing livestock losses and minimizing opportunities for wolf-livestock encounters. Respondents expressed diverse management preferences, including increasing wild prey populations (37.4%), wolf translocation (30.1%), and lethal control (28.5%), whereas only 3.3% selected increased livestock guarding and none selected improvement of livestock-pen structures. These responses indicate that there was no uniform community preference for a single management strategy. The relatively high support for translocation and lethal control highlights the importance of addressing local concerns about livestock losses while considering conservation objectives.
The preference for increasing wild prey should be interpreted as a reported community preference rather than evidence that wild-prey depletion is responsible for livestock depredation. Because wild-prey abundance and availability were not independently quantified, the present study cannot determine whether livestock use is related to insufficient wild prey, greater accessibility of domestic animals, livestock husbandry practices, or other ecological and anthropogenic factors. Any intervention aimed at increasing wild prey should therefore be preceded by assessment of prey populations, habitat carrying capacity, and potential ecological consequences.
Preventive livestock-management measures should be evaluated according to local feasibility rather than recommended as universal solutions. Improved nighttime enclosures, livestock guarding, supervised grazing, and other husbandry measures may reduce opportunities for predator attacks, but their effectiveness will depend on household resources, livestock-management practices, and local environmental conditions. Where households lack the resources required to implement preventive measures, targeted financial or material assistance may be necessary. Compensation or livestock-insurance schemes may also help reduce the economic consequences of livestock losses and potentially discourage retaliatory killing, although their effectiveness depends on programme design, verification procedures, equitable implementation, and integration with preventive measures (Dickman et al., 2011; Nyhus, 2016). Similar approaches have been proposed for reducing grey wolf-livestock conflict in northern Pakistan, including active herding, improved livestock management, education, and compensation mechanisms (Khan et al., 2019). Such interventions should be developed with local communities and evaluated for their effectiveness rather than assumed to be universally effective.
The association between the reported number of wolves killed during the previous 5–10 years and negative perceptions also highlights the importance of addressing the broader human-wolf conflict context. However, because this relationship was cross-sectional and based on community-reported information that was cross-checked with local elders and wildlife staff, it should not be interpreted as evidence that wolf killing causes negative perceptions. Conflict mitigation should therefore combine livestock-protection measures, community participation, and conservation interventions rather than relying solely on lethal or translocation-based responses.
Overall, management strategies in Orakzai should be evidence-based, locally appropriate, and developed through participatory approaches. Future work should integrate systematic assessments of wild-prey availability, standardized monitoring of livestock losses and exposure, evaluation of livestock husbandry practices, and broader community-based sampling. Such information would provide a stronger basis for designing interventions that reduce livestock losses and retaliatory killing while supporting long-term coexistence between local communities and grey wolves.
5 Conclusion
Analysis of 29 grey wolf (Canis lupus) scats yielded 30 identified prey occurrences representing 15 prey species. Wild prey accounted for 17 occurrences (56.7%), whereas domestic prey accounted for 13 occurrences (43.3%). Wild boar was the most frequently detected wild prey species, while cattle were the most frequently detected domestic prey species. The dietary assemblage had a Shannon-Wiener diversity index (H′) of 2.52, species richness (S) of 15, and evenness (E) of 0.93. Because prey availability was not independently quantified, these results describe the composition and diversity of detected prey occurrences; whether the observed patterns reflect prey preference, selection, opportunistic use, or differences in prey availability cannot be determined from the present data.
The household survey documented a substantial local context of human-grey wolf conflict, with respondents reporting livestock attacks most frequently at night and during October. Perceptions of grey wolves were predominantly negative, with 80.5% of respondents reporting negative perceptions and 19.5% neutral perceptions. A total of 71.5% of respondents reported that they or members of their community had previously killed grey wolves, most commonly to protect livestock. In the multivariable Firth logistic regression model, the reported number of grey wolves killed during the previous 5–10 years was positively associated with negative perceptions (OR = 2.495, 95% CI = 1.453–4.740, P< 0.001), after adjustment for age, occupation, total livestock holdings, wolf sightings, attack timing, and livestock injury. The cross-sectional design and reliance on community-reported information, despite cross-checking with local elders and wildlife staff, means that this relationship should be interpreted as an association rather than evidence of causation.
Overall, the findings demonstrate the need to address the ecological and socioeconomic dimensions of human-grey wolf conflict simultaneously. Conservation and conflict-management efforts should prioritize practical livestock-protection measures, strengthen community participation, and improve understanding of wild prey populations and their availability. Because respondents expressed diverse management preferences, including increased availability of wild prey, wolf translocation, and lethal control, interventions should be locally appropriate, evidence-based, and developed through participatory approaches. Future research should incorporate systematic assessments of wild-prey availability, standardized monitoring of livestock losses and exposure, and broader community-based sampling to improve understanding of the drivers of conflict and perceptions. The baseline information generated by this study can support future ecological research, conflict mitigation, and conservation planning aimed at reducing livestock losses and retaliatory killing while promoting coexistence between local communities and grey wolves in northwestern Pakistan.
Statements
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.
Ethics statement
The study involving human participants was conducted in accordance with the relevant institutional ethical requirements and approved by the Board of Studies of the Department of Zoology and the Directorate of Advanced Studies (DAS), PMAS Arid Agriculture University Rawalpindi, Pakistan, under Notification No. PMAS-AAUR/DAS/148, dated 31 January 2024. Participation in the household interviews was voluntary, and informed consent was obtained from all participants before data collection. The field surveys involving wildlife were conducted in accordance with the applicable wildlife research permissions and regulations.
Author contributions
SH: Methodology, Data curation, Validation, Investigation, Writing – review & editing, Software, Conceptualization, Writing – original draft. MJ: Validation, Writing – review & editing, Data curation, Investigation. TarM: Validation, Conceptualization, Data curation, Writing – review & editing, Supervision. TahM: Writing – review & editing, Formal Analysis, Software.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Acknowledgments
The authors sincerely thank supervisor Tariq Mahmood, the Mammalian Ecology and Conservation Laboratory (MECL), Department of Zoology, Wildlife and Fisheries, PMAS-Arid Agriculture University Rawalpindi, and the wildlife staff of Orakzai District for their valuable academic, field, and logistical support. We are grateful to DFO Kiramat Shah and Luqman Ullah for their cooperation and assistance during field surveys, and to our lab mates, especially Muqaddas Javed and Saira Akbar, for their support throughout the study.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. Open AI was used only for grammatical corrections, language refinement, and consistency of selected manuscript text, and was not used for data collection, data generation, statistical analysis, or interpretation of the study results.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcosc.2026.1955584/full#supplementary-material
References
1
AbbasF.-I.RooneyT. P.MianA. (2013). Grey wolf in Gilgit-Baltistan, Pakistan: distribution, abundance, and persecution. Canid. Biol. Conserv.16, 18–24
2
AhmadO. (2020). Assessment of human-carnivore conflict in Chitral, Pakistan. Ukrainian. J. Ecol.10, 61–65.
3
AliU.MinhasR. A.AwanM. S.AhmedK. B.QamarQ. Z.DarN. I. (2016). Human-grey wolf (Canis lupus Linnaeus 1758) conflict in shounther valley, district neelum, azad jammu and kashmir, pakistan. Pak. J. Zool.48, 861–868.
4
BanoR.KhanA.MehmoodT.AbbasS.KhanM. Z.ShedayiA. A.et al. (2021). Patterns of livestock depredation and human-wildlife conflict in Misgar Valley of Hunza, Pakistan. Sci. Rep.11, 23516. doi: 10.1038/s41598-021-02205-2
5
BocciA. S.KhanM. Z.MoriE. (2017). Sympatric snow leopards and Tibetan wolves: coexistence of large carnivores with human-driven potential competition. Eur. J. Wildl. Res.63, 92. doi: 10.1007/s10344-017-1151-0
6
ClarkT. W. (1999). Carnivores in Ecosystems: The Yellowstone Experience (New Haven, CT, USA: Yale University Press).
7
DarN. I.MinhasR. A.ZamanQ.LinkieM. (2009). Predicting the patterns, perceptions and causes of human-carnivore conflict in and around Machiara National Park, Pakistan. Biol. Conserv.142, 2076–2082. doi: 10.1016/j.biocon.2009.04.003
8
DickmanA. J.MacdonaldE. A.MacdonaldD. W. (2011). A review of financial instruments to pay for predator conservation and encourage human-carnivore coexistence. Proc. Natl. Acad. Sci.108, 13937–13944. doi: 10.1073/pnas.1012972108
9
DinJ. U.NawazM. A.MehmoodT.AliH.AliA.AdliD. S. H.et al. (2019). A transboundary study of spatiotemporal patterns of livestock predation and prey preferences by snow leopard and wolf in the Pamir. Global Ecol. Conserv.20, e00719. doi: 10.1016/j.gecco.2019.e00719
10
DistefanoE. (2005). “ Human-wildlife conflict worldwide: a collection of case studies, analysis of management strategies and good practices,” in Food and Agriculture Organization of the United Nations (Rome). Available online at: FAO Corporate Document Repository, http://www.fao.org/documents.
11
FrittsS. H.StephensonR. O.HayesR. D.BoitaniL. (2003). “ Wolves and humans,” in Wolves: Behavior, Ecology, and Conservation. Eds. MechL. D.BoitaniL. (Chicago, IL: University of Chicago Press), 289–316. doi: 10.7208/9780226516981-016
12
HendricksS. A.CleeP. R. S.HarriganR. J.PollingerJ. P.FreedmanA. H.CallasR.et al. (2016). Re-defining historical geographic range in species with sparse records: implications for the Mexican wolf reintroduction program. Biol. Conserv.194, 48–57. doi: 10.1016/j.biocon.2015.11.027
13
JamtshoY.KatelO. (2019). Livestock depredation by snow leopard and Tibetan wolf: implications for herders’ livelihoods in Wangchuck Centennial National Park, Bhutan. Pastoralism9, 1–10. doi: 10.1186/s13570-018-0136-2
14
JhalaY. V. (2003). Status, ecology and conservation of the Indian wolf Canis lupus pallipes Sykes. J. Bombay. Natural History. Soc.100, 293–307.
15
KhanM. R.MahmoodT.FatimaH.AkrimF.AndleebS.HamidA. (2020). Distribution, diet menu and human conflict of grey wolf Canis lupus in Mahoodand Valley, Swat District, Pakistan. Pak. J. Zool.52, 179–191.
16
KhanT. U.LuanX.AhmadS.MannanA.KhanW.KhanA. A.et al. (2019). Status and magnitude of grey wolf conflict with pastoral communities in the foothills of the Hindu Kush Region of Pakistan. Animals9, 787. doi: 10.3390/ani9100787
17
KrithivasanR.AthreyaV.OddenM.LinnellJ. D. C. (2009). Patterns of human–large carnivore conflict in India and implications for conservation. Conserv. Soc.7, 1–12.
18
KuijperD. P. J.SahlénE.ElmhagenB.Chamaille-JammesS.SandH.LoneK.et al. (2016). Paws without claws? Ecological effects of large carnivores in anthropogenic landscapes. Proc. R. Soc. B. Biol. Sci.283, 20161625. doi: 10.1098/rspb.2016.1625
19
LitvaitisJ. A. (2000). “ Investigating food habits of terrestrial vertebrates,” in Research Techniques in Animal Ecology: Controversies and Consequences. Eds. BoitaniL.FullerT. K. ( Columbia University Press, New York, NY, USA), 165–190.
20
MahmoodT.IsmailS.AkrimF.FarooqM.MunawarN.KhanM. R. (2021). Diet composition of grey wolf (Canis lupus) varies seasonally in Deosai National Park, Gilgit-Baltistan, Pakistan. Pak. J. Zool.54, 2267–2275.
21
MechL. D.BoitaniL. (2003). Wolves: Behavior, Ecology, and Conservation (Chicago, IL: University of Chicago Press). doi: 10.7208/9780226516981
22
MohammadiA.KaboliM.SazatornilV.López-BaoJ. V. (2019). Anthropogenic food resources sustain wolves in conflict scenarios of western Iran. PloS One14, e0218345. doi: 10.1371/journal.pone.0218345
23
MukherjeeS.GoyalS. P.ChellamR. (1994). Standardisation of scat analysis techniques for leopard (Panthera pardus) in Gir National Park, Western India. Mammalia58, 139–144. doi: 10.1515/mamm.1994.58.1.139
24
NyhusP. J. (2016). Human–wildlife conflict and coexistence. Annu. Rev. Environ. Resour.41, 143–171. doi: 10.1146/annurev-environ-110615-085634
25
OrdizA.BischofR.SwensonJ. E. (2013). Saving large carnivores, but losing the apex predator? Biol. Conserv.168, 128–133. doi: 10.1016/j.biocon.2013.09.024
26
PaquetP. C.CarbynL. N. (2003). “ Gray wolf,” in Wild Mammals of North America: Biology, Management, and Conservation. Eds. FeldhamerG. A.ThompsonB. C.ChapmanJ. A. ( Johns Hopkins University Press, Baltimore, MD, USA), 482–510.
27
RehmanE. U.DinJ. U.AhmadS.HameedS.ShahK. A.MehmoodT.et al. (2021). Insight into occupancy determinants and conflict dynamics of grey wolf (Canis lupus) in the dry temperate zone of the Hindukush Range. Global Ecol. Conserv.25, e01402. doi: 10.1016/j.gecco.2020.e01402
28
RippleW. J.EstesJ. A.BeschtaR. L.WilmersC. C.RitchieE. G.HebblewhiteM.et al. (2014). Status and ecological effects of the world's largest carnivores. Science343, 1241484. doi: 10.1126/science.1241484
29
SinghM.KumaraH. N. (2006). Distribution, status and conservation of Indian gray wolf (Canis lupus pallipes) in Karnataka, India. J. Zool.270, 164–169. doi: 10.1111/j.1469-7998.2006.00103.x
30
SteinmetzR.SeuaturienN.IntanajitjuyP.InrueangP.PrempreeK. (2021). The effects of prey depletion on dietary niches of sympatric apex predators in Southeast Asia. Integr. Zool.16, 19–32. doi: 10.1111/1749-4877.12461
31
TrevesA.KaranthK. U. (2003). Human–carnivore conflict and perspectives on carnivore management worldwide. Conserv. Biol.17, 1491–1499. doi: 10.1111/j.1523-1739.2003.00059.x
32
YounusS.NazerS.AltafM.ManzoorI.SafeerB.YasrubS. (2018). Study of human and Asiatic jackal (Canis aureus) conflict from Bagh District, Azad Jammu and Kashmir, Pakistan. J. Wildlife. Ecol.2, 1–10.
Summary
Keywords
Canis lupus, community perceptions, conservation, diet composition, grey wolf, human-wolf conflict, livestock depredation, Pakistan
Citation
Hussain SA, Javed M, Mahmood T and Mehmood T (2026) Diet pattern and human grey wolf conflict in Orakzai District, Pakistan. Front. Conserv. Sci. 7:1955584. doi: 10.3389/fcosc.2026.1955584
Received
01 August 2026
Revised
01 October 2026
Accepted
05 October 2026
Published
04 November 2026
Volume
7 - 2026
Edited by
Katharina Kasper, Polish Academy of Sciences, Poland
Reviewed by
Kinga Magdalena Stępniak, University of Warsaw, Poland
Svenja Capitain, University of Veterinary Medicine, Austria
Updates
Copyright
© 2026 Hussain, Javed, Mahmood and Mehmood.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Syed Arif Hussain, arifain028@gmail.com
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.