On July 29 this year, the government announced its quadrennial tiger survey: 2,967 tigers nationwide and as many as 2,462, almost 83%, of them photographed. This is a jump from the previous survey in 2015 that counted 2,226 tigers and ostensibly photographed 1,635 — almost 73%. The latest photos haven’t been made public but officials pointed to photos taken for the previous survey to say that such a high share of tigers being photographed leaves little room for extrapolation or inaccuracy.
India’s tiger population is undoubtedly on the rebound thanks to better protection and monitoring over the last 15 years. But an investigation of each of the 1,635 and 61 additional photos in the 2015 report by The Indian Express puts a question mark on the counting process.
For, by well-established norms of wildlife population estimation, 221 of these “tiger photos” should not have been counted. This works out to 16% over-reporting — in other words, there is one paper tiger for every seven tigers in the data. The survey is done by the Wildlife Institute of India and the National Tiger Conservation Authority. Both, under the Ministry of Environment, have been doing these counts since 2006-07.
Consider the key findings of the investigation that also involved a survey of the country’s top wildlife biologists who work on tigers and were asked to evaluate the photos on different parameters:
Duplicates: As many as 51 photos were found to be duplicates — the same tiger.
Tiger count is up but official photos show that one of seven could be a paper tiger photographed twice; in some cases even thrice; photos repeated, and photos repeated but shown as that of different tigers in the data set.
Cubs: Population surveys typically disregard very young tigers because of their high mortality. The cut-off age for the all-India tiger estimation is variously described as 12-18 months in government reports. Using this yardstick, The Indian Express weeded out photos of cubs below 12 months in age — these added to 46.
Faulty counting by flanks: This is less obvious but very glaring. A tiger has unique stripes on its right and left flanks. As per the counting method, two facing camera-traps are installed that simultaneously photograph both flanks of a tiger when it passes through.
A unique tiger is determined by its two unique flanks. Sometimes, a camera may not go off and only one flank is captured. In such cases, these single-flank photos are compared with corresponding flanks in the data set to rule out repetition.
To illustrate: Say, in a forest, 10 unique tigers have been identified based on their complete photos that capture both flanks for each. Distinct from these 10, let’s assume that the cameras have also captured eight unique single-flank photos: say, five right and three left. Since any combination of the five unique right and three unique left flanks could, possibly, belong to the same tiger, the smaller set of single flanks should be discarded. So the total number of individual tigers should be 10 plus five — 15. But in the official survey, it’s been counted as 10 plus five plus three, 18.
Going by this statistical rule, 136 photos need to be discarded.
Unidentifiable: At least 49 photos were of either indiscernible stripes or material insufficient (only whiskers, tail etc) for identification. A specialised software, ExtractCompare, is commonly used to match tiger stripe patterns. This software requires clear reference points (head/shoulder, hip-joint/tail etc) to create a 3-D surface model and then display the most likely matches within a data set.
So, a photo depicting only the head or rump of a tiger may be useful, said experts, for recording a re-capture event but should not be counted as a unique individual. For example, a head-only photo cannot be used to confirm identity even if a single unique tiger in the data does not have its head in the frame.
In all, these 282 discarded photos bring the number of photographed individual tigers down from 1,696 to 1,414 — 221 less of the government’s claim of 1,635 tigers.
India switched from the highly subjective pugmark count method to camera-traps for estimating tiger numbers after The Indian Express exposed in a series of reports in 2005 the local extirpation in Rajasthan’s Sariska tiger reserve and the dismal state of conservation in a number of other reserves. The Tiger Task Force appointed subsequently by the then prime minister recommended a slew of reforms.
The first estimate with the new method in 2007 corrected the inflated tiger numbers from over 3,000 to 1,411. The next census reported a 21% jump to 1,706 in 2011. This was followed by a massive 30% increase in 2015 that now appears to be inflated. This July, the government celebrated a remarkable 33% rise in tiger numbers to 2,967. That report, including 2,461 photographed tigers, is awaited.
Not just photographs, a close analysis of the Government’s own reports show glaring discrepancies.
* Announcing an estimated 2226 tigers for India in January 2015, the government claimed that 1,540 unique tigers were photographed in a summary report released on the occasion.
* Weeks later, this was replaced by a second report that claimed 1,686 unique individuals identified from photos and DNA analysis.
* Curiously, months later, a third report claimed all “1,686 individual tigers” were identified from “30,922 usable photo-captures of tigers” through “computer aided comparisons of stripe patterns”, and annexed 1,696 photos.
There are inconsistencies in details of DNA analyses as well:
* The second report said that 85 of 419 samples collected from 12 sites across the country for DNA analyses were confirmed as tiger droppings and these belonged to 47 different tigers.
* The third report revised this to: 86 of 369 samples collected from 13 sites belonged to 51 different tigers.
An illustrative case of divergent data is from Bihar and its only tiger reserve Valmiki. The first report claims 28 unique tigers were photographed in Bihar. The second report brings it down to 21. But the third report has three versions: it carries 21 photos of individual tigers for Valmiki in its annexure but on page 159, it claims 23 were photographed. Then, on page 20, it gives the reserve’s total tiger count as 22.
Take Rajasthan. The first report claims 51 tigers were identified through photographs. The second report raises this to 63. The third report annexes 63 photos for the state but then claims only 48 individuals were photographed. In another part of the report, the lower range of the state’s tiger number is put at 39.