Long-term trends in human fatalities from human–elephant conflict in Chhattisgarh, India

Abstract

Human-elephant conflict poses a major challenge to both wildlife conservation and human livelihoods across Asia. These conflicts often occur due to habitat fragmentation, agricultural expansion, and the depletion of natural food resources, increasing the frequency of interactions between elephants and humans. We investigated human fatalities and injuries over 23 years (2000–2023) in Chhattisgarh, India, focusing on temporal variations and the underlying drivers of conflict. Data collected from 19 forest divisions revealed a total of 828 incidents, including 737 human fatalities and 91 injuries, with Jashpur, Dharamjaigarh and Surajpur, emerging with the highest concentration of incidents. Most of these conflict incidents occurred during the monsoon. Conflict incidents are significantly influenced by proximity to natural and human-modified landscape features, with high-conflict zones near forest edges. The frequency of conflict incidents decreased with increasing distance from roads, crop fields, and built-up areas. Additionally, conflict incidents show a decreasing trend in areas with larger forest patches. Village-level analysis revealed that high-conflict areas have higher forest cover (~ 75%) and lower crop cover (< 25%). Prioritizing high-conflict villages, safeguarding elephant corridors, and implementing localized conflict mitigation measures in high-risk zones are essential to address the escalating human-elephant conflict (HEC) challenges in Chhattisgarh.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-13005-3.

Introduction

Human-elephant conflict (HEC) is among the major issues challenging both the conservation of wildlife and human livelihoods across the globe. In Asia elephants frequently come into contact with humans due to habitat fragmentation, expansion of agriculture, and a decline in natural food sources¹. These conflicts typically result in damage of crop, property and, more critically, human and elephant fatalities². The level of conflict differs spatially and temporally depending on a multitude of factors, including resource distribution, agricultural practices, seasonal climatic conditions, habitat connectivity, and overall changes in land use pattern^3–9. Presently, HEC has become a serious conservation concern in India. The country harbours the largest population of Asian elephants, with an estimated 25,000–30,000 individuals occupying approximately 163,000 km² of diverse habitats. Mainly distributed over the states of Karnataka, Assam, Kerala, Tamil Nadu, and Odisha. These forested states have human habitations and cultivation interspersed, making the situation most conducive to conflict¹⁰. In recent years, frequent human-elephant interactions have occurred, particularly in the east-central region, which includes Chhattisgarh, Odisha, and West Bengal. This conflict leads to significant human casualties and elephant deaths, creating an increasingly hostile environment for both humans and elephants¹¹. HEC results in an average of 450 human deaths annually across India (2009–2020), with the east-central region (including Chhattisgarh) suffering the highest number of fatalities​¹². India has taken significant steps to protect its elephant population while addressing issues associated with conflict. One of the most significant initiatives is Project Elephant, launched in 1992, which aims to protect elephants, conserve their habitats, and secure vital corridors essential for their movement. Besides, 33 Elephant Reserves have been established across 14 states, covering 80,777 km². Unlike traditional protected areas, these reserves take a landscape approach, ensuring that elephants can move safely even outside protected areas. Strong legal protections, such as the Wildlife (Protection) Act of 1972 and the Forest (Conservation) Act of 1980, help regulate habitat destruction and infrastructure development, making conservation efforts more effective. These combined strategies reflect India’s commitment to balancing conservation with coexistence¹². In recent years, HEC has increasingly become a significant issue in the Central Indian landscape, largely due to growing human activities and the elephants’ return to areas they had long abandoned. Here, we tried to investigate spatio-temporal patterns and drivers of HEC in Chhattisgarh, highlighting trends, hotspots, and landscape factors influencing conflict intensity.

Chhattisgarh, a state in central India, including Maikal Hills, part of the larger Satpura-Maikal range, Bailadila range and parts of the Vindhyachal range providing an important refuge for elephants, has a unique case that is evolving in many ways. Historically, this tract was the elephant homeland, and throughout the greater part of the 20th century, there was hardly any resident population⁹. The late 20th and early 21st centuries have marked a remarkable change with the re-appearance of elephants in the Chhattisgarh landscape. Though the state is dominated by dense forests and tribal populations, this place did not have much of an elephant population until the late century⁹. The reappearance of elephants in Chhattisgarh could be associated with deforestation and mining in Odisha and Jharkhand, which have forced them to seek new habitats¹³. The state’s forests provide natural corridors that facilitated this migration, settling resident elephant populations in places like Surguja, Jashpur, and Raigarh¹⁴. Over the years, agricultural expansions into forested areas, combined with mining and logging activities, fragmented these corridors, forcing elephants to venture into human-dominated landscapes¹. This has drastically altered the movement patterns and led to frequent human-elephant conflicts due to the presence of this mega herbivore¹⁰.

Conflicts majorly lead to human fatalities and injuries as well as elephant fatalities. Several human deaths and injuries from elephant encounters strain the human-wildlife relationship¹⁵. Elephants damage entire buildings, houses granaries and infrastructure, human injuries and fatalities thus escalating the conflict¹⁰. Several factors have been identified to aggravate the HEC problem in Chhattisgarh. Large-scale agricultural expansion and development projects reduce the elephant habitats, increasing the encounters¹. However, none of these driving forces have been quantified at the landscape scale in terms of HEC management and mitigation. In recent years, the fragmentation of forests has disrupted the natural elephant movement and increased the chances of contact with humans¹⁰. Conflict mitigation strategies have been inadequate and ineffective, failing to prevent or reduce conflicts, which remain an ongoing challenge¹⁶. Similarly, HEC has significant socio-economic impacts on local communities, including psychosocial stress, displacement, and migration. The continuous danger posed by elephants causes psychosocial stress and trauma among affected communities¹⁷. If these conflicts persist, some families are forced to migrate , disrupting their social and economic stability¹⁸. One of the critical factors that influence HEC in Chhattisgarh is the rapid expansion of open-cast mining¹⁹. The expansion of mining has led to extensive deforestation and habitat fragmentation. As such, with more land being cleared for open-cast mining, elephants are deprived of their natural habitats and thus are compelled to enter human-dominated landscapes in search of food¹⁸. It is, therefore, common to have frequent encounters with elephants causing substantial property damage. The financial toll of these conflicts can be overwhelming, particularly for farmers having small landholdings who rely heavily on their land for both sustenance and income. The destruction of crops and property can leave them vulnerable, as their livelihoods are directly tied to the agricultural output, they are able to produce²⁰. This has in most cases led to loss of property, relegating such families to poverty, thereby creating a cycle of conflict and hatred for elephants in general²¹ .To summarize, forest fragmentation caused by mining, road construction, and urbanization has disrupted elephants’ migration routes, forcing them through human settlements and agricultural fields. This increased reliance on human resources has led to more frequent and lethal encounters, escalating the human-elephant conflict and further tension²². Another major cause of HEC in Chhattisgarh is the ineffectiveness of mitigation strategies. Though plans for establishing elephant corridors have been made, they are often ineffective or poorly implemented. These corridors are supposed to be protected routes that enable elephants to travel between patches of forests; however, human activities such as agriculture, settlements, and mining often encroach on these corridors, obstructing elephant movement. This encroachment reduces the available spaces for safe elephant passage, increasing the conflict. Furthermore, community awareness programs designed to foster coexistence are often narrow in scope and fail to address the needs of all communities. These programs are typically not comprehensive enough to engage the entire population in efforts to rescue conflict and promote sustainable coexistence²².

In the current study, we tried to understand the spatial and temporal dynamics of HEC in Chhattisgarh and identify key ecological and anthropogenic drivers influencing conflict intensity and distribution. The key objectives are as follows: (a) to analyse the temporal variation in human fatalities, and injuries across the study period, with a focus on identifying peaks and trends in conflict occurrence (b) to examine the spatial distribution of conflict hotspots to identify divisions and villages with high conflict intensity (c) Investigate seasonal and gender-based variations in human fatalities and injuries to understand demographic disparities in conflict impact (d) to determine the influence of landscape features such as proximity to forests, protected areas, roads, water bodies, croplands, and built-up areas on human fatalities. The outcome of this study would provide insights into the ecological and anthropogenic drivers of HEC to inform targeted conflict mitigation and management strategies in high-risk areas.

Results

Temporal pattern and seasonality

In 23 years, Chhattisgarh reported 828 HEC incidents, including 737 human deaths and 91 injuries. Fatalities peaked between 2016 and 2018, with the highest number of incidents recorded in 2019 (Fig. 1). Among the divisions, Jashpur (152 deaths,19 injuries) emerged as the most severely affected, followed by Dharamjaigarh (135 deaths, 20 injuries), Surajpur (107 deaths, 04 injuries), and Korba (64 deaths, 04 injuries; Fig. 2). Spatial analysis of conflict hotspots, derived from kernel density estimations, revealed concentration in these high-conflict divisions (Fig. 3). Seasonal variation in human fatalities by gender revealed that the fatality and injury rates were significantly higher in males (540 cases) compared to females (288 cases) (χ² = 76.69, df = 1, p = 2.2e-16). However, there was no significant association between gender and seasons (χ² = 3.04, df = 3, p = 0.38). Seasonal data indicated that monsoon accounted for the highest fatalities and injuries (χ² = 14.35, df = 3, p = 0.002; 205 males, 117 females) (Fig. 4). A total of 321 villages were affected by HEC, with Jashpur being the most impacted (66 villages), followed by Surguja (45 villages), Dharamjaigarh, and Balrampur (35 villages each) (Fig. 5).

Fig. 1.

Trends in human fatalities and injuries by elephants in the State of Chhattisgarh, India, over 23 years (2000–2023).

Fig. 2.

Division-wise patterns of human fatalities and injuries by elephants in the State of Chhattisgarh over the past 23 years (2000–2023).

Fig. 3.

Hotspots of human deaths by elephants in the State of Chhattisgarh, India from 2000–2023. The map created using kernel density tool in ArcGIS Pro version 3.0.0 (https://www.esri.com/enus/arcgis/products/arcgis-pro/overview).

Fig. 4.

Seasonal patterns of human fatalities and injuries (male and female) by elephants in the State of Chhattisgarh, India from 2000 to 2023.

Fig. 5.

Number of villages affected by conflict (human deaths and injuries) by elephants across the divisions in the State of Chhattisgarh, India, from 2000 to 2023.

.

Village-level analysis of ecological and anthropogenic variables

A comprehensive examination of ecological and anthropogenic variables at the village level revealed distinct patterns across varying conflict intensity categories. Among the variables analysed, road density was the only factor that showed a statistically significant difference, with higher road densities observed in non-incident zones compared to conflict zones (Kruskal-Wallis χ² = 14.87, df = 2, p = 0.0005).

Although other variables did not show statistically significant differences, the observed trends are noteworthy. High-conflict areas exhibited a wider range of built-up percentages, while medium-conflict zones had the lowest built-up percentages, and low-conflict zones showed greater variability. Crop cover was lower (< 25%) and less variable in high-conflict areas, whereas medium- and low-conflict zones displayed broader variation. Forest cover tended to be higher in high-conflict zones (~ 75%), with medium-conflict areas showing the lowest median forest cover and low-conflict zones exhibiting greater variability. Water availability was generally low in high-conflict areas and moderate in medium- and low-conflict zones. Mining percentages were lower in high-conflict areas and moderate in medium- and low-conflict zones. The overall trends, provide useful ecological context for understanding the spatial patterns of Human-Elephant Conflict across the landscape (Fig. 6a – f).

Fig. 6.

(a–f) Comparison of variation of variables (Built-up, Crop cover, forest cover, water, road and mines) across different incident types and non-incidence villages of the State Chhattisgarh from 2000–2023.

Key predictors of HEC

HEC patterns are strongly influenced by the interplay of natural landscapes and human-modified surroundings. Probability of conflict decreases with increasing distance from water bodies (β = -0.142, p < 0.001), roads (β = -0.543, p < 0.001), cropland (β = -1.802, p < 0.001), built-up area (β = -0.408, p < 0.001), mines (β = -0.133, p = 0.07) and elephant reserve boundary (β = -0.836, p < 0.001). In contrast decreases with increasing distance to forest patches (β = 1.487, p < 0.001) and protected areas (β = 0.233, p < 0.001). Additionally, conflict incidents show a decreasing trend in areas with larger forest patches (LPI) (β = -0.313, p < 0.001; Fig. 7; Tables 1 and 2).

Fig. 7.

Predictor variables (Z-scored) influencing Human-Elephant Conflict (Human Fatalities and Injuries) in the State of Chhattisgarh, India from 2000–2023.

Table 1.

Summary statistics loglikelihood (LogL), degrees of freedom (df), Akaike information criteria (AICc), relative support for hypothesis (Δ AICc), Akaike weights (Wi) of candidate regression model explaining HEC in Chhattisgarh.

Model description	LogL	df	AICc	ΔAICc	Wi
dw + dr + df + der + dmn + dpa + lpi + dc + db + ed	− 763.16	11	1548.49	0	1.00
dw + dr + df + der + dmn + dpa + lpi + db + ed	− 834.87	9	1687.86	139.36	0.00
dw + dr + df + der + dmn + dpa + lpi + db + ed	− 834.20	10	1688.54	140.04	0.00
dr + df + der + dmn + dpa + lpi + db + ed	− 836.74	9	1691.60	143.10	0.00
dw + dr + df + der + dmn + lpi + db + ed	− 839.63	9	1697.38	148.89	0.00
dr + df + der + dmn + dpa + ed	− 863.38	7	1740.84	192.34	0.00
dr + df + der + dmn + dpa + lpi + ed	− 863.24	8	1742.58	194.08	0.00
dw + df + der + ed	− 911.32	5	1832.68	284.18	0.00
dr + df + dmn + db + ed	− 945.62	6	1903.29	354.79	0.00
dr + dpa + dc + ed	− 976.10	5	1962.25	413.75	0.00
dw + dr + df + dpa + lpi + ed	− 990.93	7	1995.93	447.44	0.00
der + db + ed	− 995.64	4	1999.31	450.81	0.00
dw + der + dmn + lpi + ed	− 995.06	6	2002.17	453.67	0.00
dw + df + dmn + ed	− 1029.53	5	2069.10	520.60	0.00
dw + dpa + lpi + db + ed	− 1057.12	6	2126.29	577.79	0.00
dw + dr + lpi + ed	− 1060.64	5	2131.33	582.83	0.00
intercept-only model (null model)	− 1147.85	1	2297.70	749.20	0.00

Table 2.

Summary statistics of the best fit model with the lowest aicc.

Predictor	Beta_Coefficient	Std_Error	Z_value	P_value & significance
(Intercept)	− 0.408	0.079	− 5.144	0.00***
Distance to waterbodies(dw)	− 0.142	0.072	− 1.974	0.05*
Distance to road (dr)	− 0.543	0.087	− 6.177	0.00***
Distance to forest (df)	1.487	0.119	12.493	0.00***
Distance to elephant reserves (der)	− 0.836	0.084	− 9.947	0.00***
Distance to mines (dmn)	− 0.133	0.073	− 1.824	0.07
Distance to protected areas (dpa)	0.233	0.075	3.084	0.00**
Largest Patch Index (lpi)	0.313	0.078	4.02	0.00***
Distance to cropland (dc)	− 1.802	0.186	− 9.641	0.00***
Distance to built-up (db)	− 0.408	0.075	− 5.382	0.00***
Edge Density(ed)	0.098	0.073	1.352	0.18

Discussion

In Chhattisgarh, various factors influence the overall pattern of HEC, particularly leading to incidents of human fatalities and injuries. Developmental activities including infrastructure projects, agricultural expansions, deforestation have caused substantial habitat loss for elephants, which in turn has intensified the occurrence of HEC^10,23,24. Over the years, Dharamjaigarh, Surguja, and Jashpur have become HEC hotspots, mirrors patterns observed in the neighbouring state of Odisha, where deforestation and urbanization are linked to higher rates of conflict²⁵. As elephants migrate into human-dominated areas, the likelihood of conflict increases, often resulting in agricultural damage and heightened risks for local communities²⁶. In regions like Nepal, Sri Lanka, and Kenya, the expansion of farmlands has led to habitat fragmentation, making elephants more likely to raid crops and come into conflict with humans^27–29. HEC incidents in Chhattisgarh show seasonal variations, often peaking during the monsoon season. This trend is consistent with observations from North Bengal, where a rise in crop-raiding incidents during the monsoon season, is attributed to the abundance of crops that draw elephants into human settlements³⁰.

Our analysis of HEC in Chhattisgarh highlights the intricate relationships between ecological conditions and human-driven activities. Villages with low conflict levels tend to have low infrastructure percentages, indicating that areas with less human interference are less prone to HEC. However, high-conflict villages are often located near regions with moderate infrastructure development, suggesting that elephants may avoid highly developed built-up areas with high human activity and noise pollution, but still encounter conflicts near transitional zones. Previous studies also reported that roads and human settlements disrupt elephant migration routes and fragment their habitats, exacerbating the frequency and severity of HEC¹¹. Forest cover also plays a pivotal role in influencing conflict levels; while intact forests act as buffers, fragmented forests often lead to high-conflict zones due to elephants entering human settlements in search of resources. These findings emphasize the importance of minimizing habitat fragmentation to maintain elephant corridors and reduce HEC intensity³¹. Water availability is another key factor; areas with ample water sources tend to have fewer conflicts, as elephants are less likely to enter human settlements when water is readily available in their natural habitat. This is consistent with findings from other regions, where water scarcity in conflict zones is a key driver behind elephant movements into agricultural and urban areas, increasing the potential for conflict^11,32. Human-altered landscapes, including agricultural lands and urban areas, also contribute to conflict. Conflict often occurs when there is intensive agriculture, particularly nutrient-enriched crops such as rice and maize, attracting elephants to the area to raid crops³³. Our study found a positive association between the percentage of crops in conflict areas and the intensity of the conflict, as elephants are drawn to these areas in search of food. Other regions have reported similar findings where agricultural land has been the key factor that pushes elephants into human settlements³⁴.

Proximity to the forest is strongly associated with higher conflict occurrences. Elephants prefer habitats near forests and water sources, leading to increased interactions with human settlements close to these areas²⁴. Increasing distances from roads, crop fields, and built-up areas are linked to lower conflict levels. Human-modified landscapes, such as infrastructure and agricultural expansion, disrupt elephant habitats and migration routes. Elephants tend to avoid areas with high road density and human activity, which act as deterrents to movement³⁵. Water bodies had a weak but significant effect on conflict, as elephants are drawn to them during water scarcity, particularly in dry seasons. This is consistent with findings identifying proximity to water as a key factor in elephant habitat suitability⁷. Larger continuous habitat patches (indicated by a higher Largest Patch Index, LPI) showed a negative relationship with conflict. Intact habitats reduce edge effects and human-elephant interactions, mitigating conflict³¹. Edge density showed a positive but non-significant relationship with conflict, indicating that boundary complexity may influence elephant movement patterns³⁶.Mining activities were weakly associated with conflict, suggesting their impact is less significant compared to roads or agricultural expansion. Mining contributes to habitat degradation but has a less direct correlation with human-elephant conflict³⁷.

In Chhattisgarh, efforts must be particularly targeted to minimize HEC especially in the severely affected villages. Priority should be given to high-conflict villages such as: Amandon (45 incidents), Chhal (42 incidents), Dharamjaigarh (36 incidents), Sajbahar (32 incidents), Katghora (28 incidents), and Surajpur (24 incidents). In addition, in the medium-conflict village category are Kunkuri (20 incidents), Kudmura (20 incidents), Korba (16 incidents), Godhi Khurd (15 incidents), and Mahasamund (12 incidents), which require urgent attention to prevent escalations of the same. Finally, low-conflict villages (total number of villages: 310) have fewer incidents; these are integral parts of broader strategies to ensure long-term success and prevent new outbreak areas. Mining activities, especially in areas like Korba, have worsened habitat degradation, leading to landscape fragmentation and an increase in human-elephant interactions. The expansion of mining operations not only leads to the destruction of critical habitats but also brings greater human intrusion into ecologically sensitive areas, further heightening the risk of conflict³⁸.These disruptions are compounded by agricultural activities in the vicinity of forested regions, as farmlands with crops like rice and maize serve as major attractants for elephants. Thus, a multi-pronged strategy is essential. The escalating human-elephant conflict reflects the growing pressures of habitat fragmentation and unprotected corridors, as observed across the east-central region³⁹. With rapid infrastructure expansion and mining disrupting traditional elephant pathways, communities find themselves at increasing risk of tragic encounters. Safeguarding and restoring these critical corridors are not just about conserving wildlife but also about protecting lives and fostering coexistence between humans and elephants³⁹.The intensity of HEC in Chhattisgarh varies across the state, and therefore, mitigation strategies may need to be tailored accordingly. At present, the Forest Department, along with other stakeholders including non-governmental organizations, is engaged in activities such as fencing and raising awareness, Rapid Response Team, community led real time elephant movement. However, no systematic study has yet been conducted to assess the effectiveness of different deterrent methods. Herein, we tried to recommend specific mitigation strategies for highly prone HEC villages in the state are as follows:

Early warning systems

Implementing early warning systems is crucial for preventing unexpected encounters between humans and elephants. Technological solutions, such as sensor-based alerts and mobile applications, can inform communities of nearby elephant movements. For instance, in the state of Assam (India), the “Gajah Alert” system has been developed to notify villagers of elephant presence, enabling them to take precautionary measures. Chhattisgarh forest department has developed an AI-based system ‘Elephant Tracking and Alert App (Hathi Mitra App)’, which monitors elephant movements and alert villages within 10 km. Preliminary assessments suggest that while the app has improved the speed of information dissemination in certain areas Gariaband and Dhamtari (Udanti Sitanadi Tiger Reserve), Mahasamund, Baloda Bazaar, Surguja, Surajpur, Jashpur, and Balrampur within the Surguja division, Dharamjaigarh Forest Division. Its effectiveness remains inconsistent across the elephant distribution range within the state. The success of the app is largely concentrated in regions with better mobile network coverage and higher smartphone penetration, particularly in parts of northern Chhattisgarh. In contrast, remote villages with poor connectivity and low digital literacy continue to face challenges in accessing timely information. Additionally, implementation barriers such as limited community training, lack of regular updates, and low integration with ground-based early warning systems have constrained the app’s potential impact. Some community members still rely heavily on traditional methods of communication (such as drum beating or word-of-mouth alerts), which can delay the response time. Therefore, both the Centre and State governments should implement strategies for the execution and monitoring of an early warning system in high-conflict villages such as Amandan, Chhal, Mahasamund, Korba, and Surajpur.

Community-based patrols and conflict response teams

The state has already implemented the ‘Hathi Mitra’ program, where local villagers are trained to monitor elephant movements and alert communities about potential threats. This initiative has been particularly effective in districts like Mahasamund, where almost 60–70 villages experience regular elephant activity. Engaging local communities in monitoring and managing elephant movements can significantly improve the effectiveness of HEC mitigation. Establishing community-based patrols and training local teams to collaborate with forest departments in high-conflict villages, as mentioned above, will further strengthen these efforts.

Alternative cropping practices

Implementing agricultural practices that discourage elephants can substantially reduce crop-raiding incidents in agriculture-dependent villages such as Amanson, Chhal, Sajbahar, Kudmura, and Godhi Khurd, where elephant incursions are frequent. Cultivating less palatable crops like chili, garlic, and citrus has been recommended as a strategy. The adoption of such alternative crops has become a part of broader efforts to reduce crop losses due to elephant raids. This approach not only helps protect farmers’ livelihoods but also fosters coexistence. However, the effectiveness of these crops in the local landscape, along with their yield and the income they generate for farmers, needs to be thoroughly evaluated.

Physical barriers

The use of physical barriers, including solar-powered electric fences and bio-fencing with thorny plants, serves as a deterrent to elephants entering agricultural areas. These should be installed in Amandon, Chhal, Sajbahar, Kudmura, Godhi Khurd and Mahasamund. These places are situated near fragmented forests. Studies indicate that non-lethal electric fences around villages have been widely implemented to mitigate HEC. However, the placement of electric fencing must be carefully considered and should not obstruct elephant corridors, as this could escalate conflicts in new areas where communities have no prior experience living with elephants. Local authorities, in collaboration with all stakeholders, must first study elephant movement patterns before installing a fence.

Compensation and insurance schemes

Providing timely compensation for losses due to HEC is crucial for maintaining community support for conservation efforts. This must be strengthened in high conflict villages like Dharamjaigarh, Katgohra, Surajpur, Kunkri and Korba. Implementing efficient compensation mechanisms for crop damage, property loss, and human casualties helps to reduce the economic burden on individuals. The Ministry of Environment, Forest and Climate Change (MoEF&CC) has introduced crop insurance schemes for farmers in high-conflict zones. Therefore, at both the state and district levels, strategies need to be strengthened to ensure timely action.

Awareness and capacity Building

Educating communities about elephant behaviour and safe practices is fundamental to reducing HEC incidents. Awareness programs and capacity-building initiatives equip local communities with the knowledge and tools necessary to coexist with elephants safely. This is particularly important in areas like Dharamjaigarh, Katgohra, Surajpur, Kunkri, and Mahasamund where conflict incidents have heightened over the years. Educating local communities to actively participate in monitoring and managing elephant movements can significantly enhance the effectiveness of HEC mitigation efforts.

Elephant corridors and habitat restoration

One of the key strategies in mitigating HEC is restoring degraded habitats, which can provide both shelter and food for elephants. While this requires significant resources and long-term investment, habitat restoration will be crucial in addressing HEC, especially in regions like central India, where elephants have significant seasonal movements. This effort should be prioritized around Katgohra, Kudmura, Korba, Chhal, and Kunkuri, as these villages are situated near key elephant corridors, as outlined in Table 3.

Table 3.

The following key elephant corridors, as identified in the “right of passage” and “elephant corridors of india” report, are critical for mitigating human-elephant conflict.

Corridor name	Connected areas	Conflict villages
Badalkhol-Tamor Pingla Corridor	Badalkhol Wildlife Sanctuary–Tamor Pingla Wildlife Sanctuary	Goreyapipar, Jashpur, Kunkuri
Lemru Elephant Reserve	Korba District, Chhattisgarh	Katghora, Banhar
Guru Ghasidas-Tamor Pingla Corridor	Guru Ghasidas National Park–Tamor Pingla Wildlife Sanctuary	Jashpur, Goreyapipar
Raigarh Forest Corridor	Raigarh District	Raigarh, Gharghoda, Dharamjaigarh, Chhal
Dharamjaigarh Corridor	Dharamjaigarh Forest Region, Chhattisgarh	Dharamjaigarh, Chhal, Banhar
Jashpur-Kunkuri Corridor	Jashpur and Kunkuri Forest Areas	Jashpur, Kunkuri

In the context of Chhattisgarh’s Human-Elephant Conflict landscape, the feasibility of early warning systems (EWS), such as the Hathi Mitra App and SMS-based alerts, is moderate to high in areas with reliable mobile connectivity. However, in remote and forest-fringe villages, poor network coverage and limited digital access remain significant barriers. To improve feasibility, hybrid systems combining digital alerts with community-based traditional warning methods (such as watch towers, firecrackers, or local volunteers) are recommended.

In terms of cost implications, mobile-based EWS have relatively low setup and maintenance costs compared to physical barriers like solar or electric fences. However, effective implementation requires investment in training, maintenance, and regular updates to ensure continued community participation and system reliability. Electric and solar fencing, while effective in reducing crop depredation, has proven to be expensive both in terms of installation and upkeep. Costs often exceed ₹5–7 lakh per kilometre, with additional recurring costs for maintenance and repair. In many instances in Chhattisgarh, damaged fences have not been repaired due to budgetary constraints, reducing their long-term effectiveness.

Regarding long-term sustainability, the success of both EWS and fencing interventions depends heavily on community ownership, consistent financial support, and institutional commitment. Community-managed fences and participatory monitoring teams have shown more durable outcomes in similar settings. Moreover, integrating HEC mitigation with broader land-use planning and habitat management strategies would enhance long-term viability.

Methods

Collection of HEC occurrences

HEC data from 2000 to 2023 were collected from 19 Divisional Forest Offices in Chhattisgarh, detailing incident locations, division name, date of the incidents, human fatalities and injuries (with gender). Surveys conducted in high-conflict villages, in collaboration with forest officials, to collect more accurate and detailed information on the incidents.

Land use land cover, HEC Spatial patterns, and forest fragmentation

HEC records include 828 documented incidents of human deaths and injuries between 2000 and 2023. The incidents were grouped into five-year intervals: 2000–2005, 2006–2010, 2011–2015, 2016–2020, and 2021–2023. Each incident was classified by type (death or injury), gender of the affected individuals, year, season, forest division. A kernel density estimation approach was used to map conflict hotspots. This estimate was calculated through ArcGIS at a cell size of 200 m to be able to get an accurate representation of the spatial distribution of geospatial. The output of this was stratified into three levels of intensity: high (more than 20 incidents), medium (11–20 incidents), and low (1–10 incidents). Moreover, the frequency of occurrences was further calculated to examine the temporal and spatial patterns of HEC at village levels within their corresponding forest divisions.

A LULC map of Chhattisgarh was generated for the years 2000, 2005, 2010, 2015, 2020, and 2024, using satellite images from Landsat 5 TM and Landsat 8 OLI). Landscape fragmentation was analysed through key landscape metrics calculated using FRAGSTATS (v4.2). The LULC maps were reclassified into forest and non-forest categories using ArcGIS Spatial Analyst tools. To capture the critical landscape features, important metrics such as Patch Density (PD), Edge Density (ED), and Largest Patch Index (LPI) were computed. A 7 km moving window analysis was done based on average elephant movement, generating a continuous surface that ensured ecologically meaningful results⁴⁰. This analysis provided a comprehensive understanding of the relationship between land cover changes, fragmentation, and HEC incidents.

Variables influencing human casualties

To investigate the factors contributing to HEC, spatial data were analysed for variables such as distances to forests, croplands, built-up, roads, waterways, protected areas, elephant reserves, and mining areas. Fragmentation metrics like Patch Density (PD), Largest Patch Index (LPI), and Edge Density (ED) were computed to assess landscape structure and fragmentation. The selection of these variables, along with the underlying hypotheses are as follows-.

The current study investigates the role of landcover features in influencing HEC by calculating the proximity of conflict points to various landcover types using ArcPro 3.0.0. Distances from built-up areas, croplands, forests, waterbodies, and mining areas were measured based on classified land cover data. The a-priori hypotheses suggest that proximity to built-up areas may decrease HEC due to intense human activity and habitat loss, whereas closeness to croplands and waterbodies is expected to increase conflict due to crop damage and water needs, particularly in dry seasons. Forest proximity is also linked to increased conflict, as elephants may emerge from forested habitats into nearby settlements. Areas near mines and quarries may exhibit higher conflict likelihood due to habitat disruption caused by extraction activities (see Table 4). The study also incorporates landscape metrics—patch density, edge density, and largest patch index—using Patch Analyst using Fragstat to evaluate habitat fragmentation and structural configuration. Higher patch and edge density are anticipated to escalate HEC by increasing the number of habitat-human interfaces, disrupting movement corridors, and fragmenting elephant ranges. Conversely, a larger largest patch index could indicate better habitat integrity, potentially reducing conflict; however, if these large patches are fragmented, the risk of conflict may rise as elephants are forced into human-dominated landscapes.

Table 4.

Variables and a-priori hypotheses for analysing factors influencing Human–Elephant conflict in chhattisgarh.

Feature	Variable	Description and source	A-priori hypothesis
Landcover	Distance from Built-up Areas(dba)	Distance between conflict points and built-up areas is calculated using the Near Table tool in ArcPro 3.0.0. Built-up areas are extracted from classified land cover data.	Proximity to built-up areas may negatively impact HEC due to higher human activity and potential habitat loss for elephants.
	Distance from Cropland(dc)	Distance between conflict points and croplands is calculated using the Near Table tool in ArcPro 3.0.0, based on classified land cover.	Proximity to cropland is expected to increase HEC due to crop damage and competition for resources.
	Distance from Forest(df)	Distance between conflict points and forests is calculated using the Near Table tool in ArcPro 3.0.0, based on classified land cover.	Proximity to forest may increase HEC as elephants may venture into human settlements in search of food and habitat.
	Distance from Waterbodies(dw)	Distance between conflict points and waterbodies is calculated using the Near Table tool in ArcPro 3.0.0. Waterbodies are extracted from classified land cover.	Proximity to waterbodies may increase HEC, particularly during the dry season when elephants move closer to human settlements in search of water.
	Distance from Mines and Quarries(dmn)	Distance between conflict points and mines/quarries is calculated using the Near Table tool in ArcPro 3.0.0. and Google Earth Pro for digitization.	Mining activities significantly influence the occurrence of HEC by disrupting elephant habitats; areas close to mines may show a higher likelihood of conflict.
Anthropogenic	Distance from Roads(dr)	Distance between conflict points and roads is calculated using OpenStreetMap data and the Near Table tool in ArcPro 3.0.0.	Proximity to roads may elevate HEC due to increased human activity, habitat fragmentation, and infrastructure development.
	Distance from Protected Areas(dpa)	Distance between conflict points and protected areas is calculated using shapefiles from WII’s Elephant Cell and the Near Table tool in ArcPro 3.0.0.	Proximity to protected areas is expected to influence HEC, with potential increase in conflict near edges where human activities are more prevalent.
	Distance from Elephant Reserves	Distance between conflict points and elephant reserves is calculated using the Near Table tool in ArcPro 3.0.0.	Proximity to elephant reserves will impact HEC, as increased elephant movements around reserves may lead to more frequent conflicts near settlements.
Landscape Metrics	Patch Density (PD)	Patch density is calculated using the Fragstat based on landcover data from classified imagery.	Higher patch density, indicating more fragmented landscapes, may lead to increased conflict due to disrupted habitats and movement corridors for elephants.
	Edge Density (ED)	Edge density is calculated using the Fragstat, focusing on boundaries between different land cover classes (e.g., forest/cropland, forest/built-up).	Higher edge density may increase HEC by creating more interfaces between habitats and human settlements, where elephants are more likely to cross paths with humans.
	Largest Patch Index (LPI)	Largest patch index is calculated using Fragstat, measuring the size of the largest continuous patch of a particular land cover type (e.g., forest).	A larger largest patch index may indicate a more stable habitat for elephants, potentially reducing HEC; however, fragmentation of the largest patches could lead to more conflicts.

Anthropogenic features were evaluated to assess their influence on conflict occurrence, with distance from roads, protected areas, and elephant reserves considered as key variables. Using OpenStreetMap and shapefiles from WII’s Elephant Cell, these distances were calculated in ArcPro. The hypothesis posits that roads may elevate HEC risks through increased human presence and habitat fragmentation. Edges of protected areas could become conflict zones due to the interface between wildlife habitats and human use zones. Proximity to elephant reserves is also expected to affect HEC, as elephants often move through adjacent landscapes, increasing the chance of encountering human settlements (Table 4).

Land Use Land Cover (LULC) data, including forest, crop, and urban land types over 23 years, were obtained from satellite imagery, converted to vector format, and matched with conflict locations for each relevant year. Using the “Generate Near Table” function in GIS, distances from conflict points to these landscape features were calculated. These distance measurements, along with fragmentation metrics, were then treated as predictor variables, while human fatalities and injuries served as the response variable. We applied Generalized Linear Models (GLMs) to identify the factors contributing to conflict, and for model selection we used R package “MuMIn”. The models were built based on theoretical assumptions, ensuring that variables with known relevance to HEC were included. Prior to model construction, all variables were standardized (z-transformed), and multicollinearity was checked (SF1). Each HEC event involving human fatalities/injuries was marked as 1, while pseudo-points (representing areas without conflict) were randomly assigned a value of 0. These pseudo-points were generated near actual conflict zones within the study area using ArcGIS Pro to maintain spatial accuracy. The factors influencing human fatalities in HEC events were analyzed using binomial logistic regression within a Generalized Linear Model (GLM) framework. This approach allowed for the examination of relationships between explanatory variables and the occurrence of fatalities in conflict events. Model selection was based on the Akaike Information Criterion corrected for small sample sizes (AICc), following Burnham & Anderson’s (2001)⁴¹ recommendation to prioritize models with AICc values ≤ 2 for ecological inference. Percentage of each environmental variables water, built-up, road density, forest, crop, mines percentage were calculated for each village. Furthermore, at the village scale, the Kruskal-Walli’s test was conducted to evaluate significant differences in the ecological and anthropogenic variables across villages classified as low, medium, and high conflict. The village boundaries were obtained from the ArcGIS Online, shapefile: Indian Administrative Layer 2024.

Study area

The state of Chhattisgarh, located in the central Indian region, is comprised of an area of about 135,192 km sq. (Fig. 8). The forest covers approximately 44% of the geographical area of the state. Being an important migratory route for elephants, though not traditionally critical elephant habitat, Chhattisgarh supports the corridors with habitats drastically reducing in the neighbouring states of Odisha and Jharkhand¹. ​As of 2022, the elephant population in Chhattisgarh was estimated to be approximately 450 individuals (MOEFCC). Northern districts of Surguja, Jashpur, and Raigarh connect Simlipal forests of Odisha and Dalma Wildlife Sanctuary of Jharkhand and act as crucial corridors that allow the easy movement of elephants into Chhattisgarh⁴². Forests in Chhattisgarh, primarily of Sal (Shorea robusta) and mixed deciduous types, carry resident and migratory populations of elephants. Critical corridors include the Hasdeo-Arand forests and the Mahanadi Wildlife Sanctuary, which are under severe threat from mining and deforestation⁴³. A western Chhattisgarh extension of the Satpura Range also sweeps into Madhya Pradesh, opening habitats for longer elephant movements. Habitat fragmentation forces elephants into human-dominated landscapes, thus further increasing the prospects of human-elephant conflict⁴² .The state’s ecological diversity is complemented by a relatively moderate climate and seasonal precipitation patterns, with rainfall typically ranging between 1,200 mm to 1,600 mm annually. This varying precipitation affects both the forest cover and the agricultural landscapes, making certain areas more prone to human-elephant encounters during the monsoon season when elephants may move into agricultural fields in search of food. Approximately 70–80% of the rural population in Chhattisgarh relies on subsistence agriculture. Commonly cultivated crops include rice (the dominant crop), maize, millets (such as kodo and kutki), pulses (black gram, green gram, pigeon pea), oilseeds (groundnut, soybean, mustard), and wheat⁴⁴.The population density of Chhattisgarh, which is around 270 people per square kilometre, also poses a challenge for conservation efforts, as human settlements are encroaching on these critical wildlife corridors. These pressures, exacerbated by habitat fragmentation, force elephants into human-dominated landscapes, increasing the likelihood of human-elephant conflict⁴².

Fig. 8.

Map of Chhattisgarh State, India, with forest cover and location of human deaths and injuries from 2000–2023. The map created using ArcGIS Pro version 3.0.0 (https://www.esri.com/enus/arcgis/products/arcgis-pro/overview).

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

BH, RKP - conceptualized the study. KR. ANG and AD conducted fieldwork and collected data. KR. ANG, AD, RKP, AN, BH, DM, PN, performed statistical analysis and prepared the original draft, while PN and DM reviewed and edited the manuscript. BH and RKP acquired funding and supervised the project. All authors read and approved the final manuscript.

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.