Faulty Counts May Have Hurt India Tigers, Experts Say

Pallava Bagla in New Delhi
for National Geographic News

August 7, 2003

Decades of counting India's wild tiger population by studying pug (paw) marks in the earth have come to nil. Indian and United States researchers have concluded that the technique is misleading. The data collected in this way has led to wrong estimates of the size of the population of the country's wild tigers and, as a result, to "poor conservation practices," the experts say.

The findings of the investigation has already spurred India's Ministry of Environment and Forests (MOEF) to launch a new 1.1 million dollar (U.S.) hi-tech tiger habitat and population monitoring mechanism, which not only hopes to gather data by using India's remote-sensing satellites but also endeavors to prepare a geographical information system (GIS)-based tiger atlas for the entire country.

India's traditional tiger census method, and the numbers it has produced, has long been a matter of debate in conservation circles. Using the pugmark technique, the current official tiger population in the wild in India is placed at 3,624— not one more or less.

The method used to count the numbers of this elusive, nocturnal animal that survives because of its brilliant camouflage, has always been controversial, but this is the first time a scientific paper establishes exact reasons for criticism.

Recognized as the last sanctuary for the wild population of the Royal Bengal Tiger, India today has 27 designated Project Tiger Reserves that, according to official estimates, support 1,576 tigers.

Considered a jewel in the crown of the Indian conservation movement, Project Tiger had a budget of seven million dollars (U.S.) last year. In a rare admission that things were not right, T. R. Baalu, India's environment minister, recently said the tiger population in the country had declined from 3,836 in 1997 to 3,642 in 2002.

India has traditionally estimated tiger numbers by laboriously collecting and then comparing impressions of tiger footprints, based on the assumption that each tiger pugmark is unique.

Now, writing in the latest issue of the British journal Animal Conservation, published by the Zoological Society of London, a nine-member team of scientists categorically states that "three decades of tiger monitoring has basically failed in India, despite being backed by massive investments and best of intentions."

The Indo-American team of wildlife biologists that made the finding includes top names like K. Ullas Karanth, carnivore ecologist from the Wildlife Conservation Society, New York; A. J. T. John Singh, wildlife biologist from the Wildlife Institute of India (WII), Dehradun; John Seidensticker, wildlife biologist at the Smithsonian National Zoological Park, Washington, D.C; James D. Nichols, bio-statistician at the U.S. Geological Service, Maryland; Eric Dinerstein, chief scientist World Wildlife Fund-US, Washington, D.C.; Charles McDougal, anthropologist of Tiger Tops, Nepal; and Valmik Thapar, a well-known tiger expert.

They suggest that conservation of India's wild tigers would be better served by adopting sampling-based approaches rather than trying to fix exact numbers.

The authors of the paper say there is deficiency of science in the conservation practice in India, especially in the procedure universally adopted in the country for estimating tiger populations, which is the three-decade-old pugmark census approach. This approach involves taking plaster casts or tracings of pugmarks and then, based on data collation and analysis, exact tiger numbers are calculated.

The investigating scientists say the pugmark approach is nonsense, since of the 300,000 square kilometers of tiger habitat in India only an "unknown fraction" is searched intensively.

They also say that pugmarks are very hard to locate in regions where the land is hard or rocky, while finding suitable pugmarks in an area rich in tigers like the Sunderbans is near impossible because of the squashy mangrove conditions.

The WCS' Karanth says "the discrimination ability of the pugmark approach completely breaks down when data from different substrates is pooled in."

In an empirical study, the team took 32 tiger pugmark tracings from two different substrates of four captive tigers in the Mysore zoo. When shown to India's tiger census experts in a blind test, Karanth says, none of them was able to segregate the individuals, while the estimates of the total number of animals ranged from 8 to 23 individual tigers."

This test categorically "demonstrates that the present [pugmark] census-based paradigm does not work," said the investigating scientists. And "ecological data on tigers from field studies do not support the results generated from pugmark census."

Rajesh Gopal, an expert on herbivore biology and director of Project Tiger, New Delhi, defends the well-entrenched official method of tiger census, however. "As wildlife managers we have to be practical and pragmatic in adopting any method, and since the pugmark approach is in tune with the local conditions it will continue," he said, although he added that a panel of scientists is already working on refining the pugmark approach.

In a new three-year initiative costing about 1.1 million dollars (U.S.) in collaboration with the Wildlife Institute of India (WII), Dehradun, Project Tiger hopes to develop a GIS-based tiger habitat and population evaluation system which Gopal claims will be able to generate "ecologically audited maps for all tiger areas."

Officials at the Project Tiger Directorate in New Delhi say that the new system will indicate the trend of tiger population in the wild. It will also give an idea as to how tiger populations are spaced and what is the connectivity among them.

The new system will give an indication of the state of tiger habitat—if it is increasing, decreasing, or stable. If it is decreasing, it will show which are the areas where the decrease has occurred and why.

The system is also expected to facilitate the study of the status of tigers' wild prey. It will devise specific monitoring and estimating protocols for obtaining reliable data from the field and will store this information so that it is readily accessible for effective field conservation, policy, and management decision-making.

Above all, it is hoped that the new system will help in disseminating the census, habitat evaluation, and monitoring protocols to field personnel through regional training workshops and manuals.

Lead researcher for this new initiative, Yadavendradev Jhala, a wildlife biologist from WII, admits that "tiger census conducted in India does not follow the protocol of sampling design or standard data quality control," but adds that some statistical models are being refined to identify individual tigers from their pugmarks. Once this is done, he said, it will "take the art to being a science."

Melvin Sunquist, an expert on tiger ecology at the Department of Wildlife Ecology and Conservation, University of Florida, Gainesville, said: "We need to get beyond the idea that we can get total counts of animals. It is one thing to count the number of cows in a pasture and be fairly certain that the number is accurate, but quite another to think that it is possible to count animals that are active at night and rarely seen…We don't need total counts, we need accurate estimates. To get these estimates we can sample, just as political pollsters sample the general population, to estimate how many people will vote for a candidate."

Gopal says the new system will "serve as a monitoring tool for the tiger and its habitats" but will also serve to monitor the forests, their extent and the threats to them. "In effect, it would monitor the entire wilderness biodiversity resource for which the tiger serves as a flagship," he said.