Understanding forest fires and their effect on carbon emissions

Burn indices help demarcate forest fire burn scars

Stunning frequency: In India, 5,20,861 active forest fire events were detected during 2003-2017.

“Growing up in Uttarakhand, forest fires were a common sight every year. I’ve woken up with smoke hovering around,” says Sandeep Bhatt from IIT, Roorkee, who spent years tracking these fires and has now reported, along with an international team, their effect on carbon emission and the ecosystem.

During 2003–2017, a total of 5,20,861 active forest fire events were detected in India, and according to the report of the Forest Survey of India, over 54% of the forest cover in India is exposed to occasional fire.

The study published in

Science of the Total Environment used remote sensing–based models to measure primary productivity over an area and also looked at burn indices, which help to demarcate the forest fire burn scars using satellite imagery.

Common index

“The normalized burn ratio is an effective burn index commonly used to identify burnt regions in large fire zones. In normal conditions, healthy vegetation exhibits a very high reflectance in the near-infrared spectral region and considerably low reflectance in the shortwave infrared spectral region. These conditions get dismantled and reversed if a fire occurs,” explains Srikanta Sannigrahi, the first author of the paper in an email to

The Hindu . He is a postdoctoral researcher at University College Dublin, Ireland.

Promising tool

He adds that the spectral differences between healthy vegetation and burnt forest areas can easily be identified and highlighted byremote sensing burn indices. It can be a promising tool for land resource managers and fire officials.

The team notes that the States of northeast India, Madhya Pradesh, Odisha, Chhattisgarh, Himachal Pradesh and Uttarakhand are the most fire-prone in India.

Previous studies using forecasting models and in-situ observations in western Himalaya have shown a sharp increase of carbon monoxide, nitrogen oxides and ozone during high fire activity periods. The current paper noted very high to high carbon emissions in the eastern Himalayan states, western desert region and lower Himalayan region.

They note that the occurrence of high fire intensity at the low altitude Himalayan hilly regions may be due to the plant species (pine trees) in the area and proximity to villages. Villages make them more susceptible to anthropogenic activities like forest cover clearance, grazing and so on.

Studies have shown that the sharp increase in average and maximum air temperature, decline in precipitation, change in land-use patterns have caused the increased episodes of forest fires in most of the Asian countries.

The team plans to further work on the prediction of forest fires with the support of advanced machine learning models and AI-based techniques.

“Identifying the forest fire hotspots and forecasting the fire location and time accurately is the need of [the hour]. We, therefore, seek many scholarly scientific contributions to mitigate this concurrent issue in a smarter way,” adds Prof. Sandeep Bhatt.