Trusted Source Image

IIT Mandi researchers develop AI-based model to predict wildlife population changes

Press Trust of India | November 21, 2026 | 10:09 AM IST | 2 mins read

The mathematical modelling framework developed by IIT Mandi researchers uses historical population data to estimate reproduction, natural population changes and predator-prey interactions, helping forecast wildlife numbers for up to five years.

AI-based framework can forecast wildlife population trends for up to five years, says IIT Mandi (Image: Wikimedia Commons)
AI-based framework can forecast wildlife population trends for up to five years, says IIT Mandi (Image: Wikimedia Commons)

Researchers at the Indian Institute of Technology (IIT) Mandi have claimed to develop a mathematical modelling framework that uses artificial intelligence (AI) techniques to predict changes in wildlife populations. The model can help scientists understand how predator and prey populations influence each other and forecast their future numbers, the institute said on Friday in a statement. 

The research focuses on the wolf and moose populations of Isle Royale National Park, a 544-square-kilometre island in Lake Superior, Michigan, USA. Wolves are the primary natural predators of moose in the park, with the latter accounting for nearly 90 per cent of wolves' winter diet. Their populations have been monitored annually since 1959, one of the world's longest-running records of predator-prey populations. 

Model usage

Over the decades, both populations have witnessed significant ups and downs. For instance, wolf numbers declined sharply in the early 1980s following a disease outbreak, while the moose population declined in the mid-1990s. The research has been conducted by Dr Nitu Kumari, Professor at the School of Mathematical and Statistical Sciences, along with her PhD student, Anurag Singh. 

Also read At 7 IITs, 1,495 reserved SC, ST, OBC faculty posts vacant; IIT Kharagpur alone has 710, reveals RTI data

The researchers developed a model that uses historical population data to estimate how quickly moose reproduce, how wolf populations change naturally, and how predation affects the moose population. Using these estimates, the model can forecast population trends for up to five years. The researchers tested two different approaches to understand how wolves hunt moose. Their findings showed that the model based on the relationship between predator and prey numbers performed better than the alternative approach. It also successfully identified periods of unusual population decline in the historical data. 

For the period between 2020 and 2024, the model predicted an increase in wolf numbers and a decline in moose numbers, broadly matching the population trends subsequently observed in the park. According to Prof Kumari, this is the first study, to the best of her knowledge, to use a self-adaptive mathematical modelling framework combining physics-informed neural networks and transfer learning to study predator-prey population fluctuations. Prof Kumari recently gave an interview to a US-based news outlet, The Primary, whose article on the research is expected to be published online soon, the statement said.

MakeCAREERS360
My Trusted Source
Trusted Source ImageAdd as a preferred source on google

Follow us for the latest education news on colleges and universities, admission, courses, exams, research, education policies, study abroad and more..

To get in touch, write to us at news@careers360.com.