Skip to main content

Susceptibility assessment of human–leopard conflict in Aravalli landscape of Haryana using geospatial techniques


Increasing global population pressure and related proliferations in demands for resources which eventually resulted in sensitive pressure on regions covering valued biodiversity. Human–wildlife conflict is one of the most common issues in conservation, encircling a huge diversity of circumstances and species. However, reasons of conflict are frequently more complex than predicted and issues which can affect human behaviour in these circumstances need to be implicit. The Leopard (Panthera pardus fusca) populations are being isolated in their micro habitat due to landscape fragmentation and increasing anthropogenic pressure over the India. Therefore, encounters between Leopards, livestock and humans are increasing in many areas, rising concerns about the costs of Leopards conservation. The present study aimed to assess of human–leopard conflict in the Aravalli landscape of Haryana (Gurugram, Mewat and Faridabad districts), India. The study was undertaken to investigate the ecological aspects of human–leopard conflict through spatial characteristics and dynamics of such conflict. In the present study, GPS field survey and mapping were done for the conflict’s sites in three districts of Haryana by the reference of newspaper articles, news reports and internet sources. Afterwards, GPS field survey was carried out to verify conflict sites in the landscape. To understand ecological aspects of human–leopard conflict, the result shows that forest cover in the Aravalli landscape has been decreased continuously from 1996 to 2018. It has been also recorded that barren rocky land has been reduced in the given time period. In contrast, agriculture and settlement have been increased dramatically from 1996 to 2018. The change matrix analyses show that forest cover and barren rocky land has been converted in agriculture land as well as human settlements. In the study area, it has also been found that water bodies have also been declined. For Susceptibility zonation, 10 parameters were selected and prepared by help of literature review and validated using linear support vector machine model. Maxent model was run for 30-m grain size and output suggested a susceptibility zone for human–leopard conflict in the Gurugram, Faridabad and Mewat. The study concludes that the human–leopard conflicts were mostly recorded in the area which has been converted from forest land to agriculture or settlements and they were found to be most susceptible. The human–leopard conflicts were also recorded at the edge of forested land in study area.


Indian leopards are isolated through nature. So, they are entering in the human occupancy because of declining in prey base, habitat loss and poaching. Being distinct from tigers that prefer forest zones free from human and cattle populations, the Indian leopard (Panthera pardus fusca) has adopt being resident near or to human occupied landscapes (Athreya et al. 2016; Areendran et al. 2020). Their preferred prey animals are like hare, wild pigs and deer have reduced due to habitat fragmentation, so as a result they prey on dogs, sheep, goats and young ones of cattle (Athreya et al. 2016). The cat, Panthera pardus fusca, declared as an endangered species is accorded under the highest protection in Indian Wildlife Protection Act (1972) (Deka et al. 2012) and also listed as “Vulnerable” in the IUCN Red List of Threatened Species (Stein et al. 2016). Usually, there is high demand for leopard’s pelt and bones in the international market and even also their bones are used as a substitute to tiger’s bones. In fact, Poaching, is developing as a key threat to their survival (Rattan 2015). Nowadays, it is quite impossible to find out a month without newspaper reports about the attack by a leopard on humans especially in the region in and around the Gurgaon (Haryana), which is followed by days of reportage of the hunt for the big cat, and mostly, finishes with the rescue or death of the unlucky animal. For each newspaper headline that from reports states the rescue of cute leopard cub to horrific instances, where villagers have beaten the animal to death, because it dared to sneak into their midst. India is certainly a habitation to test the big cat’s survival—population in large numbers, living in outskirts of protected areas, frequently terrifyingly close to settlements (Better India 2019). It is not necessary that if a leopard is spotted in a village cropland or city outskirts is looking for food, because there is not enough to eat in the forests. Leopards are found to be the most adaptable of all big or small cats, also are great survivors, and they are not precious about prey (Indian Express 2019).

Wildlife is flourishing in and around Aravalli’s of Haryana. According to the Wildlife Institute of India (WII) report, there is nearly fourfold increase in the population of big cat in the 126 sq. km forest area since 2012, when a similar exercise put the number of leopards in the region as eight. And also states that leopard habitat was estimated to be spread around 200 sq. km, primarily in Gurgaon and Faridabad districts in areas such as Ghamroj, Bhondsi, Raisina, Mangar, Gothda, Badkhal, Kotla, Kansali, Nimatpur, Khol, and Panchota. They have the tendency to come around the edge of the forest areas in search for easy prey. Now, with more and more roads and human settlements in Aravalli landscape, the human–leopard sightings and conflict cases have increased and most of the time results in loss of animal life (Habib et al. 2013).

While habitat and the availability of natural prey for Leopards continue to decline in Aravalli range, the number of domestic livestock raised near their habitat is steadily increasing. The inevitable result is that the Leopards start preying on the livestock (Hindustan Times 2019). Due to decline in availability of food and water leopards come in urban setting which results in human–leopards’ conflicts and ends up affecting the livestock or life of animal. According to the WII survey report 2017, there is a fourfold increase in leopard population and we can be evidences with the increase in cases of leopard’s sightings, conflicts in and around the Aravalli landscape in last 10 years and even continued in present time. The assessment should be done on the human–leopard conflicts in Aravalli landscape of Haryana as it requires suitable management practices that can be suitable for the area and its wildlife so as a result conflicts can be decreased which will directly increase assessment of conservation measures for leopards in Aravalli landscape (India Today 2016).

In this paper, we investigated the spatial and ecological aspects of human–leopard conflict in the Aravalli landscape predominated by real estate and agriculture. We assessed the ecological aspects by conducting in formal interview to double-check the truth behind the newspaper and reports. We also assessed the susceptibility zones using Maxent model with the help of various parameters and conclude some data that could be used to formulate management interventions to reduce future incidents in the region.

Materials and methods

Study area

The Study area includes Faridabad, Gurgaon, and Mewat which are 3 districts of Haryana State that are encountered with maximum leopard conflict cases. The total area of study area (Faridabad, Gurgaon, and Mewat) is 2383.88 sq. km which includes 213.92 sq. km of Faridabad, 758.8 sq. km of Gurgaon, and 1507 sq. km of Mewat. These 3 districts are located in Aravalli landscape of Haryana (Fig. 1).

Fig. 1

Study area map of Aravalli landscape in Faridabad, Gurgaon, and Mewat districts 

Identification of human leopard conflict locations

In the present study, the conflict location mapping is done by gathering information about the human–leopard conflict cases in newspaper articles, News reports, Internet. 200 articles were searched and then 20 articles were found relevant to the conflict cases and 35 incidences were recorded from the relevant articles. Location of the conflict is confirmed by ground truthing the incidences and individual interviews. For mapping of location is done using Google Earth Pro and ARC-GIS Software, KML and Shape file of Point location is prepared and then plotted in Study Area.

Preparation of land use land cover map

Landsat 4/5 TM (1996) and Landsat 8 OLI (2018) satellite data and supervised classification was used for preparation of land use land cover map of the study area. Change in Land use and land cover map change map was prepared using the change matrix function (Sahana et al. 2015, 2016). Change analysis is used to compare land use/land cover classification of two different years in which two image data sets are required (i.e., 1996 and 2018 data). Global digital elevation model (SRTM DEM 30 m) was collected from USGS earth explorer website ( For the present study following data source and software’s was used (Tables 1, 2; Fig. 2).

Table 1 Details of satellite  imagery used in this study
Table 2 Environmental and anthropogenic parameters used for Maxent modelling 
Fig. 2

Methodological framework for Susceptibility assessment of human–leopard conflict in Aravalli Landscape, Haryana

Human leopard conflict in Aravalli urban landscape

Wildlife is thriving in Haryana Aravallis after a new report has evidenced a sharp rise in the number of big cats and mammals in the Aravalli Landscape. According to the Wildlife Institute (WII) of India, five districts of Gurgaon, Faridabad, Mewat, Rewari and Mahendragarh are home to 31 leopards, 166 jackals and 126 hyenas, also some other wild species. Aravallis Landscape by its lush green jungles acts as a green barrier and a real protection against desertification. It behaves as a check to the spread of the Thar Desert towards eastern Rajasthan, Indo Gangetic plains, Haryana and Western Uttar Pradesh. Today, the forests in the Aravalli hills no longer efficiently act as a green barrier (Habib et al. 2017). The forests of Aravallis range become one of the most degraded forests in India, most of the indigenous plant species have disappeared (Habib et al. 2017). Also, there is drastic increase in the human–leopard conflict cases in Aravalli urban landscape from last 10 years and it will further continue to grow in if measures are not taken to conserve the wildlife and forest which would include direct and indirect sightings to contributing to claims the increase number of big cats. Photographs and reports from camera traps, report created by wildlife guards, sightings of newborn and some months old cubs in Gurgaon, Faridabad and Mewat, regions and records of pugmarks near water holes by forest officials. In 10 years, the wildlife department confirmed more than 35 leopard’s sightings in the different parts of Gurgaon, Faridabad, and Mewat. Also, there are so many villagers who have claimed to see many more Sohna, Raisina Hills, Mangar, Faridabad, Asola Bhatti, Manesar and other common places for Leopard sightings (Times of India 2019) (Table 3).

Table 3 Details of human–leopard conflict cases in study area

Methodology for preparation of layers

A total of 10 variables was used for preparation of human leopard conflict susceptible zonation’s maps. NDVI is a technique which uses the simple band rationing calculated from the reflected visible and near-infrared light reflected by earth feature (Sahana and Patel 2019). Good vegetation has the tendency to absorb maximum of the visible light that incursions it and reflects the near-infrared light. Likewise, sparse vegetation reflects the visible light more in contrast to near-infrared light (Sahebjalal and Dashtekian 2013; Dou et al. 2019). Mean NDVI for two one season (October–November) for the time period of 1996–2018 is calculated using a simple tool in Arc-GIS called raster calculator (Sahana and Ganaie 2017).

The population density map was calculated by the population data for tehsil in census data 2011. Area of the tehsil was calculated by the help of tehsil shape file digitised from the georeferenced toposheet. Population density was calculated using formula:

$${\text{Population density}} = {\text{total population}} \div {\text{total area}} \times 100.$$

SRTM DEM provides the elevation of any area which acts as a significant geographical indicator. SRTM DEM (30 m) has been extract from four tiles by the help of mosaicking and then subset image has been created using the shape file of study area. Subsequently it has been used for preparation of slope percentage. For elevation, there is a simple slope tool (spatial analyst) available Arc GIS tool box (Sahana et al. 2018b). Data for roads, railway, water bodies, settlements was downloaded from open source Diva-GIS and then using multiple Ring Buffer the distance from these parameters are created as buffer layers. Bioclimatic variables play an important role in species habitat. So, two main variables from Bioclimatic data were chosen these are average mean temperature (BIO12) and average precipitation (BIO1) (Fig. 3).

Fig. 3

Parameters used for preparation of susceptibility zonation modelling (a, b)

Parameters validation

Linear support vector machine (LSVM) Model is used to validate the importance of a parameter in all characteristics to identify which class (or group) it belongs to. The accuracy of human–leopard conflict susceptibility zone mapping depends on selected model and most importantly on the selected parameters which are responsible for the conflict (Lee and Talib 2005; Pradhan 2013; Sahana et al. 2018a, 2020; Sahana and Sajjad 2017). So, the predictive capability of factors causing human–leopard conflict should be verified before the execution any susceptibility model (Jadda et al. 2009; Kayastha 2015). In this study, LSVM model was used which is given by Guyon and Elisseeff (2003) for assessing the prediction ability of human–leopard conflict causative factors in the study area. LSVM can be expressed as

$$g\left( x \right) = {\text{Sgn}}\left( {w^{T} a + b} \right),$$

where \(w^{T}\) is the weight matrix assigned for human–leopard conflict causative parameters, a = (a1, a2……a14) vector inputs of human–leopard conflict conditioning factor, b is the offset from the origin of the hyper-plane. The human–leopard conflict conditioning factor ith with the weight \(w_{i}\) close to 0 has a smaller effect on the prediction than the one with larger values of \(w_{i}\).

Susceptibility zonation modelling

The Maxent software is based on the maximum entropy approach for modelling species niches and distributions and also probability of risk. It requires a set of parameters (e.g., environmental or anthropogenic) and georeferenced incidence localities, the model expresses a probability distribution, where each grid cell has a predicted suitability and susceptibility conditions for the species (Young et al. 2011).

Maxent model is one of the best effective presence-only data models. It is a machine-learning method that can be used for estimation of species distribution by finding the likelihood distribution of Maximum Entropy for a zone under specified set of environmental and anthropogenic parameters (Behdarvand et al. 2014). Maxent model which is Maximum Entropy Models also evaluates the differences between the values of the observations from Parameters and the background consisting of the mean observations of parameters over the entire study area, as sampled from a big number of species presence points. One of Maxent’s most important characteristic is the capability to fit extremely complex response roles by combining several functions like linear, quadratic, product, threshold, and hinge. It can suitable jagged and suddenly discontinuous responses that cannot be modelled in even the best flexible regression techniques, such as generalized additive models. It is also familiar for over fitting by a process called regularization, which is a process that precludes the algorithm from matching the data too strictly (Chen et al. 2015).

After mapping, weightage was given to each parameter according to the susceptibility zonation with the help of several experts present in WWF-India, New Delhi. The weightage is given out of 5 score. 5 is given to most susceptible class of each parameter and 1 is given to least susceptible class of each parameter (Table 5).

Results and discussion

Land use land cover change and human–leopard conflict

The result shows that the conflicts sites are found around the Aravalli hills present in the study area and also Gurgaon witnessed most of the human–leopard conflict and then Mewat witnessed a slight low case followed by Faridabad with least conflict cases. It shows that there is a tremendous increase in human–leopard conflict cases which indicates the need to conserve the flora and fauna of Aravalli Landscape of Haryana (Fig. 4).

Fig. 4

Map of human–leopard conflict sites in Gurgaon, Faridabad, and Mewat (Aravalli Landscape)

The results of land use/land cover analysis of Aravalli Landscape are given below in Table 4. The data given in the table represent the area of each class of land use/land cover of Aravalli landscape for two different years (1996 and 2018) and seven classes (water, agriculture fallow, open land, vegetation, agriculture, settlement, low vegetation) (Figs. 5, 6).

Table 4 Statistics of the area under different land use land cover classes in Gurgaon, Faridabad, Mewat District of Haryana
Fig. 5

Land use/land cover maps of Gururgram, Faridabad, and Mewat of Haryana, 1996 (a) and 2018 (b)

Fig. 6

Change analysis of Aravalli Landscape (Gurgaon, Faridabad and Mewat) during 1996–2018

The weightage given to each classes of susceptibility parameters are represented in Table 5. The weightage was used to claculate the human–leopard conflict susceptibility model.Change analysis for 2 years in the Aravalli landscape, gives information about the changes in two different years in Table 6 and also represents the conversion of one land use/land cover class to other for two different years (1996 and 2018).

Table 5 Weightage given to each class of susceptibility parameters
Table 6 Land use/land cover Change analysis of Gurgaon, Faridabad, and Mewat District of Haryana

It is observed that 0.24% of dense vegetation has been converted to settlements, 1.30% of vegetation has been converted to agriculture, 4.15% of Sparse vegetative has been converted to settlement and 4.15% of sparse vegetation area is converted to agriculture in years 1996–2018. After observations from change analysis, hence, it is concluded that maximum vegetation and low vegetation is converted into settlement and agriculture. Also, it is found that maximum conflict cases are happened in those particular converted areas.

The change analyses also shows that the water bodies present inside the vegetation areas are reduced in area and some of them are disappeared from some particular which can be become a reason for Leopard to come out in search of water and which leads to interaction with human and conflict. Also, most of the cases happened around NH-48 and Gurgaon-Faridabad road, areas, where human settlements are increasing in the form of Agriculture and settlements. Due to high speeding vehicles on highways, and traffic on roads, leopards mostly get diverted from their original route and enters the human encroachment which leads to accidents and conflicts.

Validation of parameters and the significance in human–leopard conflict

The average merit (AM) of human–leopard conflict conditioning parameters and their capability and validity for conflict susceptibility mapping was evaluated through LSVM model. Settlement buffer has the highest effect on conflict (AM = 8.7) which is followed by Mean NDVI (AM = 8.1), population density (AM = 7.9), LULC (AM = 7.2), road buffer (AM = 6.9), river buffer (AM = 6.1), elevation (AM = 5.2), rail buffer (AM = 3.3), Slope (AM = 1.9), average mean temperature (AM = 1.9) and average mean precipitation (AM = 1.9). Noticeable variations in influence were witnessed in the selected causative factors. The same contributing factors were used by many other researchers (Süzen and Doyuran 2004; Sahana and Sajjad 2017; Wu et al. 2007; Naha et al. 2019; Kshettry et al. 2017) (Fig. 7).

Fig. 7

Graph showing effect of parameters on conflict and their significance in Human–leopard conflict

Modelling susceptibility zonation human–leopard conflict

The Maxent model was run at 30-m grain size. Total 11 environment and anthropogenic parameters were used for susceptibility zonation and it is shown in Table 3. Jackknife graph (Fig. 8) of parameter importance displays data for training gain of each parameter used and the model was run in isolation and compares it to the training gain with all other variables. The values showing here are the average of replicate runs.

Fig. 8

Jackknife of regularized training gain for Panthera pardus

Elevation layer has contributed the most in predicting human–leopard conflict susceptibility the test gain is shown in Fig. 9 as many researchers found that elevation has a considerable contribution in both the training and test samples (Abade et al. 2014; Kshettry et al. 2017). The parameters that decreases most of the gain when elevation is absent is population density, which, therefore, appears to have the most information that isn’t present in the other parameters both in training and testing samples.

Fig. 9

Receiver operating characteristic (ROC) curve of  the susceptibility model

The receiver operating characteristic (ROC) curve for the same data is shown in Fig. 9, which is also averaged above the replicate runs. The specificity is well-defined using expected area, rather than true directive. The average test AUC for the replicate runs is 0.87. The value of ROC 0.87 shows that the model’s performance is better than random. Several researchers found that closer the value of ROC is to 1, the better the model has been performed (Adhikari et al. 2012). It shows that the model for present study has predicted the human–leopard conflict susceptibility with more than 87% spatial accuracy.

Figure 10 shows the test omission rate and predicted omission as a function of the cumulative threshold, averaged over the replicate runs. The omission rate must be close to the predicted omission on test samples because of the description of the cumulative threshold. The lines with orange and blue shading surroundings on graph represent variability.

Fig. 10

Average omission and predicted omission for human–leopard conflict with 30 m grain size

Figure 11 shows the susceptibility zonation map which predicts the red zones for high susceptibility for human–leopard conflict and green towards low susceptibility for conflicts.

Fig. 11

Susceptibility zonation map for human–leopard conflict in Gurugram, Faridabad, Mewat (Aravalli Landscape)

Table 7 represents the estimates of relative contributions of the all the parameters used in the Maxent model. For first assessment, in each repetition of the training algorithm, the increase in normalized gain is added to the contribution of the corresponding variable according to their weightage. Likewise, for the second assessment, for each parameter in turn, the values of that variable on conflict locations and background data are randomly permuted. As along with the parameter jackknife, parameter contributions should be understood with courtesy when the predictor parameters are connected and correlated. As evident from the Tables 16, the percent contributions of the different parameters vary for each parameter. But average mean temperature, distance from railways signifies the environmental as well as anthropogenic factors accounts for 99.6 and 90.7% of the aspect liable for conflict risk distribution. Separately, from population density, distance from road, distance from settlement, distance from water bodies, average precipitation, slope, elevation, mean NDVI (1996–2018) which is a representation of vulnerable area for conflicts and also plays a substantial role in the assessing the susceptibility zones for the human–leopard conflict in the study area.

Table 7 Percentage contribution of parameters

The response curves in Fig. 12 demonstrations the logistic prediction value changes by change in the environmental as well as anthropogenic parameters, Elevation which has the highest contribution to the gain as observed in jackknife shows higher predicted susceptibility values between 0 and 433 m and lesser susceptibility will be towards 433 m. Population density has higher values of predicted suitability at around 462–4097 sq. km and lower towards 426 m which is making sense that nearer distance to more human population are highly vulnerable area. The trend which is shown for the distance from railways, settlements and water bodies follows same trend and which shows that susceptibility is constant for all values. However, Mean NDVI of 1996 and 2018 has higher values from − 0.206 to 0.235 which demonstrates that higher density of forest or dense forest areas contribute in the suitable but there are no such dense forest patch present is area overall susceptibility is seen towards low to high density of NDVI. Lower percentage value of slope values shows the higher predicted susceptibility.

Fig. 12

Response curves of the susceptibility modelling


Based on the present, some main issues necessitate consideration to reinforce the utility of spatial human–leopard conflict susceptibility risk modelling as a tool for mitigating the leopard attacks in study area. First, conflict susceptibility modelling should be assimilated into long-term monitoring to measure and assess the impacts of related efforts on conflict management. As part of this process, risk models must be updated with recent data to account for behavioral feedbacks and changes in carnivore–livestock interactions. Second, innovative approaches for displaying, sharing and applying results from risk models should be used to reach new people among locals, stakeholders, and policymakers. Third, outreach with officials should be highlighted so that susceptibility maps can notify large-scale decisions and mitigation for conflict management. Finally, there is great opportunity for more conflict susceptibility zone modelling, mostly in the ranges of high potential livestock attack in India. Efforts can be established for systematic conflict susceptibility modelling efforts in those high conflict susceptible regions should be highlighted for economic and logistical support, especially since human–carnivore conflict is currently causative to the rapid of greatly endangered carnivores such as the leopards.

This study reviews to validate that spatial conflict susceptibility modelling which can be serve as a useful tool for managing on ground decision making also about wherever to implement prevent the conflict. Especially in circumstances, where authorities regularly collect material on livestock deaths as part of compensation programs, conflict susceptibility modelling can use existing data to offer supplementary vision into the spatiotemporal forms and socio-ecological drivers of human–wildlife conflict. The use spatial conflict susceptibility modelling as a valuable tool in the conservation and management, the technique will endure to improve the efficiency of mitigation efforts for reducing livelihood and animal life losses and also strengthening wildlife conservation.


  1. Abade L, Macdonald DW, Dickman AJ (2014) Assessing the relative importance of landscape and husbandry factors in determining large carnivore depredation risk in Tanzania’s Ruaha landscape. Biol Cons 180:241–248

    Article  Google Scholar 

  2. Adhikari D, Barik SK, Upadhaya K (2012) Habitat distribution modelling for reintroduction of Ilex khasiana Purk, a critically endangered tree species of northeastern India. Ecol Eng 40:37–43.

    Article  Google Scholar 

  3. Areendran G, Sahana M, Raj K, Kumar R, Sivadas A, Kumar A et al (2020) A systematic review on high conservation value assessment (HCVs): Challenges and framework for future research on conservation strategy. Sci Total Environ 709:135425

    Article  Google Scholar 

  4. Athreya V, Odden M, Linnell JD, Krishnaswamy J, Karanth KU (2016) A cat among the dogs: leopard Panthera pardus diet in a human-dominated landscape in western Maharashtra, India. Oryx 50(1):156–162

    Article  Google Scholar 

  5. Better India (2019). Accessed on 21 Apr 2019

  6. Behdarvand N, Kaboli M, Ahmadi M, Nourani E, Mahini AS, Aghbolaghi MA (2014) Spatial risk model and mitigation implications for wolf–human conflict in a highly modified agroecosystem in western Iran. Biol Conserv 177:156–164

    Article  Google Scholar 

  7. Chen F, Du Y, Niu S, Zhao J (2015) Modeling forest lightning fire occurrence in the Daxinganling Mountains of Northeastern China with MAXENT. Forests 6(5):1422–1438

    Article  Google Scholar 

  8. Deka K, Athreya V, Odden M, Linnell J (2012) Chemical immobilization of Leopard Panthera pardus in the wild for collaring in Maharashtra, India. J Bombay Nat Hist Soc 109(3):153–157

    Google Scholar 

  9. Dou J, Yunus AP, Bui DT, Merghadi A, Sahana M, Zhu Z et al (2019) Assessment of advanced random forest and decision tree algorithms for modeling rainfall-induced landslide susceptibility in the Izu-Oshima Volcanic Island, Japan. Sci Total environ 662:332–346

    Article  Google Scholar 

  10. Guyon I, Elisseeff A (2003) an introduction to variable and feature selection. J Mach Learn Res 3:1157–1182

    Google Scholar 

  11. Habib B, Talukdar G, Sapra RK (2013) Status of wildlife in Haryana State. Census report 2013. Haryana Forest Department and Wildlife Institute of India, p 63

  12. Habib B, Talukdar G, Jain P, Bhasin A (2017) Mapping land use/land cover patterns in Aravallis Haryana with special reference to key wildlife species. Project Completion Report. Wildlife Institute of India, Dehradun and Haryana Forest Department, India

    Google Scholar 

  13. Hindustan Times (2019). Assessed on 20 Jan 2019

  14. India Today (2019). Assessed on 27 Jan 2019

  15. Indian Express (2019). Accessed on 21 Apr 2019

  16. Jadda M, Shafri HZ, Mansor SB, Sharifikia M, Pirasteh S (2009) Landslide susceptibility evaluation and factor effect analysis using probabilistic-frequency ratio model. Eur J Sci Res 33(4):654–668

    Google Scholar 

  17. Kayastha P (2015) Landslide susceptibility mapping and factor effect analysis using frequency ratio in a catchment scale: a case study from Garuwa sub-basin, East Nepal. Arab J Geosci 8(10):8601–8613

    Article  Google Scholar 

  18. Kshettry A, Vaidyanathan S, Athreya V (2017) Leopard in a tea-cup: a study of leopard habitat-use and human-leopard interactions in north-eastern India. PLoS One 12(5):e0177013

    Article  Google Scholar 

  19. Lee S, Talib JA (2005) Probabilistic landslide susceptibility and factor effect analysis. Environ Geol 47(7):982–990

    Article  Google Scholar 

  20. Naha D, Sathyakumar S, Dash S, Chettri A, Rawat GS (2019) Assessment and prediction of spatial patterns of human-elephant conflicts in changing land cover scenarios of a human-dominated landscape in North Bengal. PLoS One 14(2):e0210580

    Article  Google Scholar 

  21. Pradhan B (2013) a comparative study on the predictive ability of the decision tree, support vector machine and neuro-fuzzy models in landslide susceptibility mapping using GIS. Comput Geosci 51:350–365

    Article  Google Scholar 

  22. Rattan T (2015) Human-leopard conflict: big cats losing battle for survival. Retrieved from Accessed on 1 Mar 2019

  23. Sahana M, Ganaie TA (2017) GIS-based landscape vulnerability assessment to forest fire susceptibility of Rudraprayag district, Uttarakhand. India. Environ Earth Sci 76(20):676

    Article  Google Scholar 

  24. Sahana M, Patel PP (2019) A comparison of frequency ratio and fuzzy logic models for flood susceptibility assessment of the lower Kosi River Basin in India. Environ Earth Sci 78(10):289

    Article  Google Scholar 

  25. Sahana M, Sajjad H (2017) Evaluating effectiveness of frequency ratio, fuzzy logic and logistic regression models in assessing landslide susceptibility: a case from Rudraprayag district, India. J Mount Sci 14(11):2150–2167

    Article  Google Scholar 

  26. Sahana M, Sajjad H, Ahmed R (2015) Assessing spatio-temporal health of forest cover using forest canopy density model and forest fragmentation approach in Sundarban reserve forest. India. Model Earth Syst Environ 1(4):49

    Article  Google Scholar 

  27. Sahana M, Ahmed R, Sajjad H (2016) Analyzing land surface temperature distribution in response to land use/land cover change using split window algorithm and spectral radiance model in Sundarban Biosphere Reserve, India. Model Earth Syst Environ 2(2):81

    Article  Google Scholar 

  28. Sahana M, Hong H, Sajjad H (2018a) Analyzing urban spatial patterns and trend of urban growth using urban sprawl matrix: A study on Kolkata urban agglomeration, India. Sci Total Environ 628:1557–1566

    Article  Google Scholar 

  29. Sahana M, Hong H, Sajjad H, Liu J, Zhu AX (2018b) Assessing deforestation susceptibility to forest ecosystem in Rudraprayag district, India using fragmentation approach and frequency ratio model. Sci Total Environ 627:1264–1275

    Article  Google Scholar 

  30. Sahana M, Rehman S, Sajjad H, Hong H (2020) Exploring effectiveness of frequency ratio and support vector machine models in storm surge flood susceptibility assessment: A study of Sundarban Biosphere Reserve, India. Catena 189:104450

    Article  Google Scholar 

  31. Sahebjalal E, Dashtekian K (2013) Analysis of land use-land covers changes using normalized difference vegetation index (NDVI) differencing and classification methods. Afr J Agr Res 8(37):4614–4622

    Article  Google Scholar 

  32. Stein AB, Athreya V, Gerngross P, Balme G, Henschel P, Karanth U, Laguardia A (2016) Panthera pardus. The IUCN red list of threatened species 2016: e T15954A50659089. IUCN, Gland

    Google Scholar 

  33. Süzen ML, Doyuran V (2004) A comparison of the GIS based landslide susceptibility assessment methods: multivariate versus bivariate. Environ Geol 45(5):665–679

    Article  Google Scholar 

  34. Times of India (2019) Accessed on 5 May 2019

  35. Young N, Carter L, Evangelista P (2011) A MaxEnt model v3. 3.3 e tutorial (ArcGIS v10). Fort Collins, Colorado

    Google Scholar 

  36. Wu CH, Tzeng GH, Goo YJ, Fang WC (2007) A real-valued genetic algorithm to optimize the parameters of support vector machine for predicting bankruptcy. Expert Syst Appl 32(2):397–408

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Mehebub Sahana.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Yadav, N., Areendran, G., Sarma, K. et al. Susceptibility assessment of human–leopard conflict in Aravalli landscape of Haryana using geospatial techniques. Model. Earth Syst. Environ. 7, 1459–1473 (2021).

Download citation


  • Leopard (Panthera pardus fusca)
  • Human–leopard conflict
  • Land use/land cover
  • Susceptibility zones
  •  Aravalli landscape