Home Articles The rift in the lute: Rhino habitat in the Kaziranga National Park,...

The rift in the lute: Rhino habitat in the Kaziranga National Park, India

Rubul Hazarika
Department of Geography, Gauhati University
Guwahati -781014, Assam, India
[email protected]

Dr. Anup Saikia
Department of Geography, Gauhati University
Guwahati -781014, Assam, India

The Kaziranga National Park (KNP) lies in the heart of Assam, India along the southern banks of the Brahmaputra River. A World Heritage site since 1985, the KNP is home to the world’s largest population of Indian rhinos. Some 430 square kilometers, it is dotted with swamps and shrouded with tall thickets of elephant grass and provides an ideal habitat for the great one-horned Indian rhino.

This study focuses on land use change in the KNP and the quantum of the Park’s land lost to erosion, particularly on its northern banks, on the basis of satellite data. Supervised classification and rule-based expert classifier are used to gauge land use change and assess habitat suitability respectively. An attempt is made to identify suitable habitats for rhinos in forest areas in the immediate vicinity of the KNP. Finally, ideal locations for the creation of animal crossings, such as bridges, in the Park are identified to prevent more animal casualties on the highway that slices through the KNP.

The KNP Habitat
The soils in the park represent both new alluvium and old alluvium formed by the alluvial deposits of the Brahmaputra River. The Brahamaputra River carries more discharge per unit area than any other river in the world and therefore the KNP like the rest of the Brahmaputra Valley, is annually replenished by alluvial deposits that support a diverse flora, including grass that the rhinos of the KNP subsist upon. The alluvial plains vary from 40 to 80 meters in elevation and are prone to inundation during floods. The park enjoys sub-tropical monsoonal climate characterized by heavy rainfall during summer. The annual average rainfall is in the order of 180 cm. Wetlands, which are locally called beels, represent the most common and integral features of the fluvial landscape of the Brahmaputra valley and some 191 wetlands dot the KNP. The vegetative cover of the park is characterized by woodland, short grass and tall grass. Grasslands predominate with tall ‘elephant’ grasses on the higher ground and short grasses on the lower ground surrounding the ‘beels’.

The revival of the Indian rhinos at the KNP is one of the most successful conservation stories in the world. From a population of a mere dozen in 1908, it has grown to 1500 after some 90 years. The rhino census conducted in Kaziranga in April 1999 recorded a population of 1552 compared with 1164 in 1993 (Kaziranga National Park, 2002). The estimated rhino population from 1966 to 1999 is depicted in Figure 1.

Figure 1: Rhino population in KNP (1966 to 1999)


  1. Vigne, L. and Martin, E 1991. Census, Oryx 25.
  2. Vigne, L. and Martin, E 1994. Census, Pachyderm 18.
  3. Vigne, L. and Martin, E. 1998. Census, Pachyderm 26.
  4. Talukdar, B.K. 2000. Census, Pachyderm 29.

The total area of the park is 42,996 hectares, and some 5,000 hectares was already lost due to erosion at the northern boundary by the Brahmaputra River (Kushwaha et al., 1986). The rhino population in KNP has recovered from very low numbers, but threats to this species are still significant. Only through continued and increased protection will its survival be ensured. The stabilization, extension and improvement of their habitat are necessary for conservation. The IUCN report (1997) puts the potential rhino carrying capacity of the park at 1500, but the current population has exceeded this number. Finding suitable areas to accommodate the growing population is required, but it is also necessary to undertake a conservation plan that will ensure a steady yet measured increase in the population of rhinos; in order not to compromise the presence of other animals and threaten the stability of the park as a whole.

Although rhinos are found almost everywhere in the park they are found in greater numbers in particular parts of the park where their general habitat requirements are better.

Habitat Modelling
Many studies to date have used remote sensing and GIS for wildlife habitat analysis and their suitability evaluation. For habitat assessment of elk (Brian et al., 1997; Bright, 1984), reindeer (George et al., 1977) and kangaroo (Hill and Kelly, 1987) remote sensing and GIS technologies were used extensively. Rees et al. (2002) used Landsat and ETM+ imagery for mapping of land cover change in a reindeer herding area of the Russian Arctic. Mongkolsawat and Thirangoon (1998) used satellite imagery and GIS to evaluate wildlife habitat suitability mapping, mainly for Asian elephants in Thailand. Similar studies have been carried out by Foley (2002), Wiersema (1998), Zhix et al. (1995) and Polce (2004). Pertaining to the Indian context the works of Roy et al. (1995), Porwal et al. (1996) and Kushwaha et al. (2000 and 2004) and Raut et al. (2000) are noteworthy.

Within the fields of RS and GIS, the approach taken by Expert Classifier is closer in terms of similarity to GIS analysis than to traditional classification. Erdas Imagine’s Expert Classifier is one of the advanced classification methods to deal with different types of problems from basic land cover mapping to resource management. As has been pointed out, an expert system is a computer application that solves a specific problem or makes a decision based on a series of rules, conditions or hypotheses defined by an expert in a given field (Jordan, 2000). Expert systems (or hypothesis testing) and/or rule-based system have been used with a high degree of success in different studies, including those undertaken by Stefanov et. al., (2001), Karanja, (2002), Civco et. al. (2002), Schadt et. al. (2002) and Jacquin et. al. (2005). These illustrate the increasing use of remote sensing data combined with GIS techniques for wildlife and conservation research.

In the present study, GIS and RS data were used to assess habitat loss and locate suitable areas within the park and its immediate vicinity for possible future extension.

Database and Methodology
Imageries of six different periods from different sensors with varied spatial resolutions were used for this study. These imageries were from Landsat TM and ETM+ with 30 m spatial resolution, IRS 1C LISS III with 23.5 m spatial resolution and ASTER with 15 m resolution.

A Survey of India Topographic map sheet on 1:50000 scale of 1971 was used to extract roads and settlements along with the satellite imageries, while a Digital Elevation Model (DEM) was obtained from Shuttle Radar Topography Mission (SRTM) data. Twenty known points, collected with a GPS on January 25, 2004, were used as the training set for the supervised classification of the images. Data on rhino population, causalities of human-rhino conflicts, rhino’s habitat requirements and other related data were also collected from secondary sources and direct field observation. Since the available satellite datasets for the years 1987, 1988 and 2004 were not covering the entire area of the park, mosaics were prepared. Images from 1987 and 1988 TM data were used for 1988 and for 2004, 2003 IRS data and 2004 Aster data were used. Finally, data pertaining to six periods, i.e. October-December 1988, January 1994, December 1999, December 2001, December 2002 and January-February 2004 were used to acquire a subset according to the boundary of the park including River Brahmaputra. While data preparation, geo-referencing, mosiacing and re-projecting were performed in Erdas Imagine 8.7, ArcGIS 9.1 was used in database creation and spatial analysis.

Supervised classification was conducted for each image by using both parametric (Maximum likelihood) and non-parametric (Feature space) decision rules in Erdas Imagine 8.7. Information from the GPS points and the topomap were used to identify the landuse and generate the training sets. The park was classified into five broad classes: woodlands, grasslands, scrublands, water and sand, and for each class, the area was added to the attribute table.

Figure 2: Flowchart of main steps to obtain suitable habitat

A subset of the mainland of the park excluding Brahamputra River was then created for each year using the vector layer. The resulting land use maps were analyzed and attribute values were compared to detect the changes and were used as inputs in the expert classification. The created signatures were evaluated for separability and contingency. Accuracy assessment was done for each classified image with the help of 100 randomly generated points throughout the classified image using ‘stratified random’ distribution parameters.

The suitable areas for the rhinos were evaluated using Erdas Imagine’s Expert classifier for each year. The expert classifier has three main components, namely hypothesis, rules and conditions. The hypotheses represent the output and the intermediate classes and the rules define the hypotheses based on the input data sets through different combinations of conditions. The input data set consists of user-defined variables and includes raster imageries, vector coverages, spatial

Figure 3: Flowchart of main steps to obtain suitable habitat

models, external programs, and simple scalars. A rule is a conditional statement, or list of conditional statements, about the variable’s data values and/or attributes that determine a hypothesis. Multiple rules and hypotheses can be linked together into a hierarchy that ultimately describes terminal hypotheses. Confidence values associated with each condition are also combined to provide a confidence image corresponding to the final output which is a classified image. A hypothesis forms a classification based on the truth of one or more rules. (Source: Erdas Imagine Expert Classifier Overview)

Table 1 shows the hypothesis, rules and conditions with the confidence value for each rule. After setting the rules and defining the confidence levels, expert classification for each year was performed. Changes in the habitat were then determined by comparing the six outputs.

Table 1. Hypothesis, rules and conditions used in the classification

Finally, to determine suitable areas both inside and outside the park a modified version of the same decision tree was applied to the classified images of 2003 covering a large area including outside the Park. In this step, road buffer was not included, only the road feature was included, considering the creation of animal crossings over roads. Human settlement and the agricultural fields were taken as a single class without a buffer. The best suitable and suitable classes were combined to make one suitable class whereas the areas outside the one km buffer of water, with all other requirements was considered as less suitable. Since rhinos can travel up to 3 kms from the water bodies for their food, only the minimum requirements for the rhino habitat were considered to estimate suitable areas outside the park for the expansion and creation of corridors and animal crossings over-bridges.

Results and Analysis:

Grasslands occupy the maximum area, followed by scrublands and woodlands. The presence of grasslands makes the KNP the ideal home for Indian rhinos. Encouragingly, there has been a perceptible increase in the area under grasslands (Table 2) during 1988 to 2004.

Class name Area in km2
1988 1994 1999 2001 2002 2004
Grasslands 148.58 135.12 109.09 134.05 192.51 186.89
Water 35.80 53.68 39.44 27.28 27.10 31.52
Scrublands 82.41 55.27 121.02 91.23 57.07 70.73
Woodlands 128.91 150.79 126.38 140.28 114.43 100.02
Sand 1.74 0.37 0.12 0.30 0.23 0.17
Total 397.44 395.23 396.06 393.14 391.34 389.33

Table 2: Land Uses in Kaziranga National Park

Apart from grasslands, all other land use types decreased. In 1994, there were decreases in the area of the grasslands and the scrublands, while area of the woodlands and water increased. On the other hand, from 1994 to 1999, area in all other land uses decreased except in scrublands. Burning of grasslands for the management purpose is one of the possible reasons for this change. In 2002, the area of the grasslands increased, but from 2002-04 there was a slight decrease. At the same time, as the grasslands were increasing from 1999 to 2002, the scrublands and woodlands areas were decreasing. During 1988-2004, sand showed the least change.

Gain was observed in grasslands category, as an area covering 38.31 km2 was converted to grasslands from 1988 to 2004. The increase of grasslands was manifested in corresponding decreases in other categories. It was observed that during this period, 28.89 km2 of woodlands, 11.68 km2 of scrublands, 4.28km2 of water and 1.57 km2 of sand were lost or converted to other categories.

YearsGrasslandsWaterScrublandsWoodlands Sands
1988-1994 -13.46 17.87 -27.14 21.88 -1.37
1994-1999 -26.02 -14.24 65.75 -24.41 -0.25
1999-2001 24.95 -12.16 -29.79 13.90 0.18
2001-2002 58.46 -0.17 -34.16 -25.85 -0.07
2002-2004 -5.62 -5.62 13.66 -14.41 -0.06
1988-1999 -39.48 3.64 38.61 -2.53 -1.61
1988-2001 -14.53 -8.53 8.82 11.38 -1.43
1988-2002 43.93 -8.70 -25.34 -14.48 -1.50
1988-2004 38.31 -4.28 -11.68 -28.89 -1.57

Table 3: Landuse changes in KNP (1988-2004)

The following graph (Figure 4) shows the trend in total land area change from 1988 to 2004. The total land area estimated in 1988 was 397.44 km2, a decrease from 395.23 km2 in 1994. There was about 0.83 km2 area added in 1999, but there was a gradual decrease in land area onwards. A loss of 2.92 km2 was observed between 1999 and 2001, while 1.80 km2 was observed between 2001 and 2002. In 2004 the area estimated as 389.33 km2, indicating that there was a loss of about 2.01 km2. The park lost a total of 8.11 km2 (0.41%) from 1988 to 2004. The main reason of the land area changes is the erosion of the River Brahmaputra. Although the loss is not so alarming, when considered in the context of a growing rhino population and a high density of the other animals it does indeed make this loss significant.

Figure 4: Changes in Total land area of KNP

In terms of habitat suitability, the conditions in the KNP remained more or less the same. While significant gains were made in the most suitable category, a loss in the area of the next suitable category was recorded (Table 4)

Class name Area in km2
1988 1994 1999 2001 2002 2004
Most Suitable 103.15 94.84 81.32 96.07 131.22 130.72
Suitable 144.33 142.19 164.80 156.39 120.88 118.41
Moderately Suitable 57.90 50.18 51.86 50.01 49.75 48.22
Unsuitable 55.88 54.12 58.48 63.16 62.23 60.24
Water 35.56 53.44 39.08 26.91 26.74 31.20
Total 396.81 394.76 395.54 392.54 390.82 388.80

Table 4: Changes in Suitability in the KNP

Beyond the KNP boundary there are suitable areas in the southern and south-western areas. Although these areas are less suitable, considering their relatively higher elevation they can be used by animals during the monsoon seasons when floods inundate low lying areas of the KNP. Further they could also serve as a corridor to reach the more suitable areas. As the national highway passes along the southern boundary of the park, certain areas are identified as potential sites for the creation of over-bridges for animal crossings (Figure 5). The western and south-western parts of the area along the park boundary have been identified as the suitable areas for the expansion of the park.

GIS and RS methodologies are useful in habitat suitability studies and have been fruitfully employed to delineate landuse-landcover change in the KNP. While landuse changes are not substantial enough to be alarming, seen in the context of growing wildlife density, including rhinos, there is cause for concern and preventive steps will need to be taken sooner or later. In fact, the IUCN placed the KNP’s rhino carrying capacity at 1500, however the current population has exceeded this number, and thus finding suitable areas to accommodate the growing rhino population is more than an academic exercise.

Erdas Imagine’s Expert Classifier is a useful tool to assess the suitability in areas and habitats such as the KNP and the methodology adopted in this study can well be replicated elsewhere and for other wildlife species.

Figure 5: Suitability Analysis outside the Park showing the potential areas for expansion and corridor and animal crossings (over bridges) creation

Dr. Norman Kerle of the ITC, The Netherlands provided substantive guidance during the writing of the main report in August 2005. His contribution is hereby gratefully acknowledged.


  • Choudhury, A. 2005. Threats to the Greater one-horned rhino and its habitat, Pabitora Wildlife Sanctuary, Assam, India. Pachyderm 38 pp 82-88
  • Civco, Daniel L. Hurd, James D. Wilson, Emily H. Song, Mingjun and Zhang, Zhenkui, 2002, A Comparison of Land Use and Land Cover Changes Detection Methods, ASPRS- ACSM Annual Conference and FIG XXII Congress
  • Hazarika, R. 2003. Asian Elephant Habitat Analysis and Suitability Modeling using Remote Sensing and GIS in Kameng-Sonitpur Elephant Reserve, India. Academic Output, CSSTEAP, India
  • Jacquin, A. Cheret, V. Denux, J. Gay, M. Mitchley, J. and Xofis, P. 2005. Habitat Suitability Modelling of Capercaillie(Tetrao urogallus) using earth observation data. Journal for Nature Conservation, Volume 13, 2005, pp161-169 (www.sciencedirect.com)
  • Jordan, L.E. 2005. Expert Systems help extract information from New Geospatial Data, www.eomonline.com/Common/Archives/2000jan/00jan_jordan.html, viewed on July 6, 2005
  • Kushwaha, S.P.S. Khan, A. Habib, B. Quadri, A. and Singh, A. 2004. Evaluation of sambar And muntjak habitats using geostatistical modeling, Current Science Vol.86, No.10, pp1390-1400
  • Kushwaha, S.P.S. 2002. Geoinformatics for wildlife habitat characterization, Map India, 2002 Mongkolswat, C. and Thirangoon, P. 1998. Application of Satellite Imagery and GIS to Wildlife Habitat Suitability Mapping, AARS-ACRS1998, (www.gisdevelopment.net)
  • Polce, Chiara. 2004. Application of Geographic Information Systems to assess the environmental suitability for reintroduction of Griffon vulture (Gyps fulvus) in Retezat National Park, Romania, IFA Report, ITC, The Netherlands
  • Ramirez, J.A. 2005. Habitat Suitability Mapping using Expert Knowledge and Simulation. M.Sc. Thesis, ITC, The Netherlands.
  • Status Survey and Conservation Action Plan of Asian Rhinos, 1997, IUCN/SSC Rhino Specialist Group, published by, IUCN, Gland, Switzerlands.
  • Schadt, S. Knauer, F. Kaczensky, P. Revilla, E. Wiegand, T. and Trepl, L. 2002. Rule-based assessment of suitable habitat and patch connectivity for the Eurasian Lynx, Ecological Application, Vol.12, No.5, pp 1469-1483 (www.esajournals.org)
  • Nico J, van Strien, 1985, The Sumatran Rhinoceros in the Gunung Leuser National Park, Sumatra, Indonesia, PhD thesis, Agricultural University, Wageningen
  • Talukdar, B.K. 2000. The current state of rhino in Assam and threats in the 21st century, Pachyderm 29 pp 39-47
  • Venugopal, J. 2001. Analysing rhino habiata suitability and its changes in chitwan district, Nepal using remote sensing and GIS, IFA Report, ITC, The Netherlands
  • Vigne, L. and Martin, E.B. 1994. The Greater one-horned rhino of Assam is threatened by poachers. Pachyderm 18 pp 28-42
  • Vigne, L. and Martin, E.B. 1998. Dedicated field staff continues to combat rhino poaching in Assam. Pachyderm 26 pp 25-39
  • Vasu, N.R. 2002. Periodic reporting exercise on the application of the world heritage convention )
  • Weiers, S. Bock, M. Wissen, M. and Rossner, G. 2004. Mapping and indicator approaches for the assessment of habitats at different scales using remote sensing and GIS methods. Landscape and Urban Planning 67 (2004) pp 43-65
  • Laurie, William Andrew, 1978, The Ecology and Behaviour of the Greater One-horned Rhinoceros, Phd thesis, University of Cambridge
  • Liu, X. Bronsveld, M.C. Skidmore, A.K., Wang, T. Dang, G. and Yong, Y. Mapping Habitat Suitability for Giant Pandas in Foping Nature reserve, China
  • WWF Species Status Report, Asian Rhino, 2002,
    Viewed on 08.07.05
    Viewed on July 6, 2005