Grid array overlaid across Malenad landscape for tiger occupancy surveys (left). Traditional, presence-only surveys (center) underestimated tiger occupancy by 45% when compared to occupancy modeling (right), which estimated that tigers occupied 66% of the landscape.

A team of four WCS India Program field members are sweating it out in the rugged hilly terrain of Malenad. Walking neither too fast, nor too slow, they follow a trail, diligently observing and recording signs of tigers and other wildlife along the way. The solitary bark of an alarmed deer nearby instinctively makes them stop and listen. However, this survey entails recording physical evidence exclusively, so they move on looking for more.

Dr. Ullas Karanth observing tiger pugmarks during sign surveys in the Malenad landscape. (Photo credit: Kalyan Varma)

The trail falls within one of the 205 grid-cells (each measuring 72 square miles) plotted over the 15,000-square-mile Malenad landscape. Three additional teams simultaneously replicate their work along other trails within the grid-cell. The entire crew have been there five months, walking up to 10 miles a day, almost daily. Moving from one grid-cell to another, they race against time to cover the entire landscape, before the monsoon season sets in again.

Key questions that interest wildlife scientists include knowing where species are found and understanding what might influence these distribution patterns in space and time. Field surveys that rely on tiger signs and sightings, along with interviews of people, can be used to determine where tigers are found today and compare this to where tigers were found historically across India.

Tigers in tropical Asia are generally found across large forested landscapes. To understand tiger occupancy in the Malenad landscape, WCS in India divided it into equal-sized grid-cells. These grid-cells are then surveyed to determine how many have tigers and how many do not, recording indirect evidence such as feces, pugmarks, scrape marks on the ground and scratch/scent marks on tree trunks. By correctly identifying tiger signs during these surveys, we can ascertain whether or not the tiger occupies a particular grid-cell.

WCS field biologist mapping the areas sampled during occupancy surveys. (Photo: S. Gopinath/WCS India)

Since tigers live in large landscapes that are generally thousands of square miles in size, such grid-cells should ideally be large. It would be impractical and, in fact, impossible for field biologists to search every inch of a forest for tiger signs in such large cells. Our work has established that carnivores like the tiger generally use forest roads and trails extensively, marking their territories, guarding their home ranges and looking for mates. Therefore, searching along such roads allows us to collect sign information, and cover large areas in a logistically feasible manner.

A particularly tricky problem with sign surveys is that some grids may have tigers but signs of them are not detected. Concluding that there are no tigers would lead to flawed inferences and underestimation of the proportion of area occupied by tigers.

Patterns of tiger occupancy across India.

Habitat occupancy modelling, built on the same framework as capture-recapture models, addresses this problem. By increasing the number of surveys (four in our case) in each grid-cell, we increase the chances of detecting tiger signs. Additionally, if there is no detection despite multiple visits, it is possible to apply probabilistic models to estimate tiger presence even in such areas. By including information on signs of prey animals, human disturbance and protection/management efforts, we can estimate their influence on tiger distribution and also obtain relative probabilities of tiger presence in each grid-cell. Our surveys found that tigers occupy nearly 66 percent of the forests in Malenad landscape and that their distribution is influenced by the presence of forest cover, prey species and human disturbance.

We have also looked at historical tiger occupancy by collating more than 3,600 natural history observations, museum specimens and hunting records from India. Until the early 1900s, tigers were found all across India. However, in the past 100 years, their distribution had shrunk by almost 68 percent. These distribution surveys play a critical role in monitoring population trends and help focus conservation efforts.

Using habitat occupancy modelling for tigers (as well as elephants, dholes, leopards and sloth bears), we have reliably estimated what proportion of a large landscape they actually occupy and the patterns and drivers of their current distribution. Expanding such occupancy studies to combine data from multiple years can also yield useful information on local extinctions, colonization and range-shifts, which are crucial pieces of information to secure viable tiger populations for the future.