Deep learning can help track changes in forest cover

India is one of the ten most forest-rich countries in the world, with about 80.9 million hectares of trees covering it (nearly 25 percent of the entire country). However, between the 1890s and the 1990s, a combination of rapid economic development and overexploitation of resources led India losing almost 80 percent of its forest area. Now, as the country’s forests continue to disappear, a team of researchers led by the Ohio State University has developed a deep learning algorithm that, by using satellite data, could provide real-time monthly land use and land cover maps of India.

“Our work was done in an effort to help India’s government and industries improve the country’s attempts at forest sustainability,” said study lead author Ying Zuo, a graduate student in Earth Sciences at OSU.

This land monitoring system was trained by using data from Norway’s International Climate and Forests Initiative (NICFI), an enterprise of the Norwegian government aiming to reduce the destruction of tropical forests by providing high-resolution satellite images of the tropics. By combining this data with a global land cover map produced by Tsinghua University in China, the deep learning algorithm was able to develop more detailed maps of Indian forests.

“In an effort to combine two datasets into the same system, we resampled them into the same spatial resolution and aligned each pixel to create an image-labeled paired training dataset,” Zuo explained. “The process helps us to assimilate the two datasets so they can be used to train our deep learning model.” 

After training the model through these methods, the scientists processed ten base monthly maps of the area, from January to October 2022. These maps allowed researchers to detect seasonal shifts across India, including changes to barren land, the way monsoons affected crop land, and the distribution of forests in mountainous regions.

The experts concluded that it is vital for ecologists to more comprehensively investigate the seasonal impact of monsoons on India’s forest cover. “As the average temperature of our Earth increases, natural hazards will become a lot more frequent, so having these maps at our disposal benefits everyone’s understanding of how this problem affects life on Earth,” Zuo said.

In future research, the scientists aim to expand the timespan of these base maps over several years instead of just a few months to study other annual changes around the world, such as the effects of floods on forests.

“The characteristics of local forests and their surrounding habitats might likely be different in other regions. But with the help of more detailed datasets, our work could easily be used in areas of the world where detecting and alerting the public to forest degradation and its side effects need to become a priority,” Zuo concluded.

The study was presented at the annual meeting of the American Geophysical Union.

—

By Andrei Ionescu, Earth.com Staff Writer

Check us out on EarthSnap, a free app brought to you by Eric Ralls and Earth.com.