Researchers at the Indian Statistical Institute, Bengaluru, and the Wildlife Conservation Society (WCS) have come up with a new methodology that has the potential to revolutionise the way tigers and other big animals are counted over large geographical areas.
How many tigers are there in India?
According to the Ministry of Environment, Forests and Climate Change, in 2006, the number was estimated at 1,411 and in 2010 it was 1,706. In 2016, the world rejoiced when India announced that it had nearly 2,500 members of this critically endangered big cat.
However, there were many who were not sure of this.
The methods used to count tigers do not measure up to rigorous scientific scrutiny but the country’s big cat conservation body the National Tiger Conservation Agency (NTCA) brushed those objections under the carpet. The new method is aimed at making the tiger count robust and reliable.
New Method for Counting Big Cats
Estimating the number of big cats is a difficult process. For a long time, conservationists have been asking for a methodology that helps assess the number of tigers over a large area with a fair degree of accuracy.
The new method called ‘Bayesian Smoothing Model’ or BSM, can bridge the gaps, say the researchers. Their paper titled ‘Bayesian Methods for Estimating Animal Abundance at Large Spatial Scales Using Data from Multiple Sources’ is published in the current issue of the Journal of Agricultural, Biological, and Environmental Statistics.
So, in what way is this method more accurate than the ones used previously?
Dr K Ullas Karanth, Director for Science in Asia, WCS, and a co-author of the paper spoke to The Quint over an email interview, “When you want to estimate animal numbers over larger regions, the challenge is how to extrapolate reliable estimates derivable from small areas to wider regions where only cheaper and less reliable methods can be used.”
Dr K Ullas Karanth, Director, Wildlife Conservation SocietyA widely used method known as ‘index calibration’ tries such extrapolation using a simple-linear regression technique suggested years ago. However, because of the wide variations in data collected using such less reliable methods at wider scales, these index calibrations fail. Dr K Ullas Karanth, Director, Wildlife Conservation SocietySo, the current method of extrapolation of tiger numbers, based on index-calibration is fundamentally flawed. The BSM approach we developed, addresses explicitly the issue of ‘imperfect detection’, that is, not all animals present are detected, and better mimics or, models, the way in which animal count data are generated and collected. Thereby it presents a statistically more complex, but biologically more realistic way to deal with this long-standing problem.
In fact, this “issue affects not just tiger counting in India, but many other species in many other countries too. We used tiger data because I work on tigers,” says Dr Ullas Karanth.
Collecting Reliable Data
Scientists more often than not never rely on animal data collected using methods such camera trapping – taking photographs of animals using a remotely activated camera. However, these work fine and generate reliable results when used in smaller areas.
The cost of using these methods make scientists look for other traditional but weak markers such as tiger track counts by looking at paw marks when doing surveys in areas spread over thousands of square kilometers. This is a challenge for the scientists and the BSM has addressed this critical flaw.
How does the model work?
Prof Mohan Delampady, of the Indian Statistical Institute, Bangalore, and a co-author of the paper explains, “Typically, tigers are counted using good camera trap methods by identifying individuals in protected areas. In the remaining 80-90 percent habitat only counts of tiger tracks can be generated. BSM approaches the issue along with the serious problems involved in it by trying to integrate these two types of data, by explicitly laying out the mechanics of their connection, recognising the weaknesses in the track count data; it does not hide the problems involved.”
Both Dr Ullas Karanth and Prof Delampady are confident that BSM will work just fine and help the biologists get a clearer picture of the tiger count in India.
Mohan Delampady, Professor, Indian Statistical Institute, BengaluruWe did test it at a very large scale in our study: Across 33,000 sq km area or more than 15 percent of India’s tiger habitat. It is not a small scale pilot study and can be scaled up to a country-wide scale. If the government wants our help to do so, we will be happy to help.
Along with Prof Delampady and Dr Ullas Karanth, the others who co-authored the paper are Soumen Dey, Ravishankar Parameshwaran, N Samba Kumar, and Arjun Srivathsa.
Will the Authorities Wake Up?
The real question is whether NTCA (National Tiger Conservation Authority) and government will listen to the scientists? Dr Ullas Karanth does not sound too optimistic. “We had earlier critiqued and pointed out that the extrapolations used to generate tiger numbers beyond reserves did not work. These critiques have not been responded to scientifically. Now we have also shown a better alternative method, and even this can be refined further. We would be glad to help the Government to adopt it, but the motivation to abandon tradition and opt for better methods must be there.”
Tiger conservationists are aware of the way the government works but there is much more at stake than waiting for the authorities to wake up. It is up to tiger lovers to make NTCA adopt these methods. Until then, the numbers are up in the air.
(With scientists coming up with a new method for estimating the no. of tigers, The Quint debates whether the authorities should junk traditional methods of keeping a tab on the population of tigers. This is the view, you can read the counterview by NC Bahuguna here.)
(Deb Kumar Mitra is a Kolkata-based freelance science writer. This is an opinion piece and the views expressed above are the author’s own. The Quint neither endorses nor is responsible for the same.)
Join us on WhatsApp. Type “JOIN” and send to 9910181818.