Lies, damned lies and statistics: getting to grips with the all-India tiger estimation controversy

Imagine running a business in which you are totally unaware how much stock you have. This is the situation that faces those that manage India’s national tiger estate. Tigers, as a species, are notoriously difficult to see. They are solitary, secretive animals that wherever possible seek to avoid contact with humans. On my first visit to a tiger reserve I became only too aware of this fact as I returned from successive drives empty handed. When I say empty handed I don’t necessarily mean that my trip was a disaster as I was lucky enough to witness a range of species populating the dry forests of Central India, each a key source of tiger prey but, unfortunately the big prize eluded me. On returning to the lodge I spoke in depth to my hosts who assured me that the largest of the big cats was active across the route I’d been driven and, on day two, I was rewarded not with an actual tiger sighting but with a close-up view of a very large, very real pug-mark that had been left upon the dusty road we were travelling.

Sign at the exit of Bandhagarvh National Park " Perhaps you may not have seen me, but please don't be disappointed. I have seen you."
Although difficult to see, tigers are nevertheless present.

Its very size hinted at the scale and power of the animal that had left it and, just around the next corner a sign post comforted me over my lack of an actual sighting. Here, a cartoon tiger stared down at me, a speech balloon coming from its mouth, informing me that although ‘you may not have seen me, please don’t feel disappointed. I have seen you’.

When I arrived back at base that evening and was flicking through a local journal in the small library I read that, on average, only one trip in 14 actually resulted in a tiger sighting. Clearly, these are not the easiest animals to spot. So, consequently, they are also very difficult to count. The results of the latest All- India Tiger Estimation were announced in January of this year and received enthusiastically by those willing the tiger to rebound from the threat of extinction. As Union Minister for Environment and Forests, Prakash Javadekar revealed the long anticipated result of the 2014 estimation exercise, the response to his words were mixed. Euphoria in some quarters was balanced by incredulity in others. While tiger lovers celebrated the meteoric rise in numbers, experts were more cautious and, as the days passed, their concern grew as to the validity of what they had been told. New analyses soon appeared in the media to challenge the government’s figures and there emerged a view that the new Modi government may have been merely fuelling the feel good factor that had heralded their landslide victory of 2014. If not, then the plight of the tiger had improved to such an extent that the problem was now less one of will the tiger survive than one of how many more can our tiger reserves accommodate before saturation?

Well, the history of the tiger count has been long and controversial. It was following such estimates in the late 1960s and early 1970s that Project Tiger was conceived, in recognition of an alarming drop in tiger numbers that had bypassed most people. Suddenly, a creature that had always been considered ubiquitous in Indian forests was now becoming increasingly rare and, it was considered, that without remedial action, the tiger would soon enter the shadowlands of extinction. Enter the most powerful person in India, Prime Minister Indira Gandhi who, with the backing of global conservation NGOs, considered ways to better protect the Indian tiger. The result of this collaboration was the enactment of legislation (the Wildlife Protection Act, 1972) and the establishment of nine tiger reserves as a safe haven for the country’s remaining tigers, estimated to have fallen to around 2,400 individuals. The reserves were largely carved out of existing protected areas and, within their boundaries, management was to focus on the protection of the largest of the big cats. Over the years, the national estate of tiger reserves has risen exponentially and, today, they number 48 individual sites. While arguments persist as to the success of this approach, the tiger has nevertheless evaded extinction although its numbers have oscillated due to a range of external pressures. Left undisturbed, tigers are known to be a resilient and fecund species yet habitat fragmentation has proved to be an ongoing problem as has the depletion of prey species, much of which has been for the consumption of impoverished local communities.

Then, in the late 1980s and into the 1990s, the first signs of a poaching epidemic materialised and once again, Indian tiger numbers declined. Since then, every effort to count them have added to the argument as to the best way to estimate their numbers. The pug-mark method emerged as the early favourite, where plaster casts of the footprints of tigers were collected, analysed and individual incidences recorded.

tiger pugmark in the dust
Initial efforts to count tigers utilised the now discredited pug mark count approach.

While the figures generated emphasised the continuing success of Project Tiger, It soon emerged that there was a political angle to the registration of numbers as the more tigers a reserve (and indeed the state within which the reserves were located) claimed, the more money provided by the government for their upkeep. Inflation of figures was widespread and experts were reduced to a guessing game of how many tigers remained. All anyone seemed to agree on was that the actual number inhabiting the reserves was far less than the official fiction suggested.

Arguments persisted into the 21st century, with critics of the status quo pointing to the availability of new technologies and statistical procedures available to wildlife managers in more accurately predicting tiger population numbers. It largely remained a talking shop, however, as by 2004 tiger numbers were reported to have exceeded 4,000 individuals. Few believed such hype yet the government clung on to the outmoded pug-mark approach, refusing to accept criticisms as to its efficacy. Then, in 2005 the Wildlife Institute of India revealed that, contrary to census returns suggesting a healthy population of tigers inhabiting the popular Sariska Reserve in Rajasthan, there was in fact no evidence at all to support the presence of any tigers whatsoever. The resulting crisis, and its recriminations, proved to be a landmark event in the management of tigers in India. The extirpation of the big cats from such a high profile reserve prompted a review of the whole national approach towards tiger conservation, including the ways in which tiger numbers were calculated. Finally, the controversial pug-mark count mechanism was being dismantled in favour of a more statistically robust index-calibration approach, utilising technological advances to more accurately predict the extent of, and trends in, the population of wild tigers in India.

Surely this advance was widely applauded across the conservation lobby? Not necessarily, as it soon emerged that while critics welcomed the departure from an outmoded and discredited counting technique, they feared that it had been replaced by an equally problematic approach. Prominent tiger expert K. Ullas Karanth argued that the new procedure adopted by government was based on nothing more than an obsolete statistical technique that had fallen from favour as long ago as the 1930s. Whatever the reality of the situation, this method has now been utilised within the last three All-India tiger estimation exercises in 2006, 2010 and 2014.

As the first results emerged from the 2006 estimation, it became evident that the extent of tiger declines across India had become critical. Based on a median of an upper estimate of 1,657 and a lower figure if 1,165 tigers, the most likely population size was stated as 1,411. The news was greeted with widespread despondency as the world’s media picked up the story, although the figure generated at least appeared to be more realistic than those that had been provided from previous counts. The Indian establishment, stung by what they viewed as a censure of their (mis)management of the national tiger estate, sought to engage more closely with experts outside their familiar networks in order to tackle the shortcomings that had contributed to such a decline. By 2010, the next estimation period, their efforts had seemingly been justified as a headline figure of 1,706 tigers was communicated to an expectant media. This, we were told, represented an increase of 16 percent. Tigers were clearly rebounding, the message suggested, India was finally back on track. But was it? Just as the 2006 estimation was based on a median figure, so too was its 2010 version. This time, the lower limit was 1,571 against a ceiling figure of 1,875.

Following a brief period of reflection, and not a little back patting from those in power, critics pored over the data and came up with a more measured reading of the trends. What if, they said, you consider the figures from a slightly different perspective, say the maximum number from 2006 against the minimum range for 2010? Say, for the sake of argument that these were indeed nearest to reality and what you find is the possible loss of a further 86 tigers in just four years. Hardly something to celebrate. Add to this the fact that the 2010 count included areas not employed in the 2006 estimation, a prime example being the Sunderbans which yielded a further 70 estimated individuals and, far from rebounding, tigers may well be continuing to decline. An even closer look into the data reveals that the reported rise in tiger numbers had occurred during a period of accelerating habitat loss and, taking the published figures at face value means that the 16 percent rise in tiger numbers was achieved across a national estate that had lost 22 percent of its territory. This implies that tiger densities in India had increased by a staggering, and less than realistic, 49 percent in just four years.

Graph showing a constant rise since 2006 using the new counting approach
The headline data for the 2014 tiger count was eagerly picked up by the world media.

Let us now move on another four years and consider a little more closely the latest All-India tiger estimation results released in January 2015. The official estimate was revealed to be 2,226 (representing a 30 percent increase in numbers) based upon a minimum limit of 1,945 and an upper range of 2,491. Whilst the previous assessment was restricted to tigers inhabiting protected areas, the 2014 version sought to expand its reach to include all available tiger habitat across the country. The value of this type of exercise, however, is its longevity in providing data over an elongated period that can be compared over time to identify key trends and patterns emerging and to use this management information to better guide policy and action. Constantly changing its terms of reference negates much of its value and, in an expensive process like the tiger estimation, this represents a lost opportunity in the fight to save the tiger.

If the 2010 count had been greeted with a certain amount of derision and disbelief, this time the response was far more incredulous and antagonistic. Karanth referred to the estimation as ‘ridiculous and unscientific’ and, with colleagues set out to discredit the validity of the index-calibration approach in a paper published in ‘Methods in Ecology and Evolution’. Now comes the difficult part. The arguments put forward in this debate are based on complex mathematical and statistical applications, issues of which I have little grasp of. For reasons of necessity, as well as choice, I have therefore attempted to simplify the argument to a basic description of the index-calibration method and the ways in which the protagonists deviate in terms of assessing its value and accuracy. As a technique, index-calibration relies on the measurement of animals over a relatively small territory using a range of methods including camera trap technology.

night vision shot of a tiger face on.
Caught in a trap. Camera trap technology adopted as part of the all-India tiger estimation toolbox

The accumulated data from this stage is then related to more easily available indicators, such as animal track counts by means of calibration. The resulting index can then be used in order to extrapolate numbers over an extended area, in this case the whole of India.

Returning to the study, it goes on to argue that the adopted approach represents a poor way in which to predict actual tiger numbers, and that more advanced, applicable methodologies exist. The subtext here is important as, it appears to me, the crux of the argument is that the Indian government, by clinging on to control every facet of tiger conservation, are missing a key opportunity by their refusal to engage with the wealth of experience and expertise that exists within the non-government sectors. Indeed, a cursory reading reveals the fact that the authors of the article are not even disputing the fact that tiger communities may have rebounded in India, but rather the extent of the rise. Moreover, it points out that management information is only of use if it is accurate and trustworthy. Rather than engaging in a reasoned discussion to justify their stance, the government however have embarked instead on a crusade to force their detractors to retract their criticisms of the generated numbers. This has created an impasse in which both sides are firmly entrenched, compromising the opportunity for informed debate in order to agree on the best way ahead. Here, the prospect of wider partnerships in the immediate future appear at best unlikely.

An image of a tiger with digital mesh superimposed and a closeup showing the stripe pattern
tiger stripes are unique and software now exists to distinguish between individuals

As Lisa Newton noted in her 2005 book on ‘Business ethics and the natural environment’ the resulting lack of meaningful engagement tends to result in the “polarisation of the people best qualified to solve environmental problems” of this kind. Unless such a impasse can be bridged and as long as the protagonists rely, in the immortal words of Mark Twain, on ‘lies, damned lies and statistics’ to justify their divergent positions, business as usual will ensure that, in a little under four years, we will again be revisiting the same arguments. Nothing will have changed.

2 thoughts on “Lies, damned lies and statistics: getting to grips with the all-India tiger estimation controversy”

  1. I not to mention my pals ended up reading the excellent recommendations located on your web blog and so immediately developed an awful feeling I had not thanked the site owner for those tips. Most of the people are actually as a result passionate to see them and have now absolutely been making the most of these things. Many thanks for indeed being simply considerate and also for getting certain high-quality ideas millions of individuals are really needing to be aware of. My personal sincere regret for not saying thanks to earlier.

Leave a Reply

Your email address will not be published. Required fields are marked *