What if we could predict natural disasters in advance and have suitable risk mitigation plans in place that could save lives of thousands? What if we are able to make policy level changes based on solid research for natural resource allocation?

All this and more is possible if artificial intelligence and deep learning systems are used, say researchers.

Mohd Anul Haq, researcher and Assistant Professor at NIIT University, Rajasthan, has been using AI and deep learning for the last four years in climate change projects involving glaciers, forest and river restoration. For him, using these disruptive technology has earned satisfactory results and more accurate results.

“I have been studying glaciers for the past 10 years but have been using AI only in the last four years,” Haq said. He studies different aspects of glaciers such as thickness to learn water stored in them and impact of atmospheric temperature changes.

However, measuring thickness of glaciers consumes significant time, effort, and money besides posing risk. Researchers use ground-penetrating radar (GPR) for this purpose and lug the sophisticated instrument, weighing about 15 kg, around.

“It is impossible to do such measurements on all the 15,000 Himalayan glaciers carrying the equipment. That would mean huge cost that these research projects incur and also risk associated with it,” Haq explained.

This is where AI and deep learning can come to scientists’ help. If deep learning is used, scientists need to carry out measurements only on three or four glaciers. The information thus obtained, together with satellite images dating back 30 years, can be used to train the algorithm to be intelligent. It can then be used to calculate the thickness with better accuracy. Haq said they were able to predict calculations with almost 95 per cent accuracy.

Haq is also involved in Sahibi River restoration project in Rajasthan, where the researcher is using AI and Geographic Information System to identify patches of land where possibilities of finding ground water source in the Sahibi River track for digging. This way it is done is more efficient and less time consuming considering the years it would take to complete the project, he said.

At Ghent University, Belgium researchers are studying the impact of climate change on the vegetation and carbon cycle using modelling. They use AI and advanced instrument terrestrial LiDAR for their research. IBM’s deep technology platform

Watson is now being used for weather forecasting through high performance computing under IBM’s Deep Thunder.

Glaciers are a major source of fresh water and knowing the water stored in them helps government and international agencies in better forecasting and planning. The same can be extended to other natural resources like forests and oceans.

According to a report prepared by World Economic Forum in collaboration with PwC and Stanford Woods Institute for the Environment, these disruptive technologies, when harnessed right, could help build solutions to overcome challenges such as climate change, biodiversity conservation and water security.

But the challenge is in the adoption. Haq said, “Unless we adopt this disruption all the research we do will go obsolete.” Researchers should equip themselves with contemporary technologies and implement them so that such results can also be used to sensitise the government to do better planning.

comment COMMENT NOW