Our natural resources are polluted, glaciers are melting, forests are depleting and wildlife is dying: it doesn’t take much else to prove that environmental dangers are real. Our efforts so far have clearly been inadequate in healing our ailing planet. Consider this: despite ongoing efforts, global temperatures are projected to rise by up to 3.2 C by 2100. In short, by abusing the environment, we are setting ourselves up for catastrophic consequences. 

As the scale and urgency of the impacts from our deteriorating natural environment grows, we have an opportunity to look at how AI can help transform traditional sectors and systems to address this looming crisis. AI is not a silver bullet, but its ability to provide meaningful solutions for environmental sustainability can’t be undermined. The possibilities to harness AI for the earth include creating energy-efficient buildings, developing new low-carbon materials, optimising supply chains, forecasting weather and climate, better monitoring of deforestation, and greener transportation. Read on for detailed use cases.

• AI-based prediction model for detecting Arsenic in drinking water: Researchers from IIT Kharagpur have developed an AI-based prediction model for detecting Arsenic pollution in drinking water. By using AI for geoscience, the study provides information for the location of safe groundwater, which is the primary source of potable water for most of India. Arsenic has been a menace in Eastern India especially along the banks of the Ganga for almost two decades now, putting millions of people at severe health risk. The researchers have delineated the high and low arsenic zones across the entire delta using artificial intelligence and quantify the number of people exposed. They have developed probabilistic models of arsenic occurrence, exposure and human health risk assessment within the delta region. While the predictive model framework would prove to be vital typically for the identification of drinking water sources in arsenic affected areas of West Bengal, it can be used in other parts of the country which are also suffering from severe groundwater pollutants. Read more...
• GeoAI platform is helping target brick kiln hotspots of air pollution: The GeoAI platform of UNDP India, in collaboration with Bihar State Pollution Control Board, is an innovative effort to ensure better air quality in the state. It will use Space Technology and AI to help monitor air pollution hotspots and control emissions from the brick kiln industry. Research shows that brick kilns cause 14% of the air pollution in the Bihar. The GeoAI platform demonstrated how environmental non-compliance can be detected from space using Artificial Intelligence. It significantly reduces the humongous task; the challenge of monitoring about 9,000 brick kilns in Bihar was brought down to less than 1,000. Read more...
• IoT devices to monitor rural drinking water supply: The Union Jal Shakti Ministry, as a part of its Jal Jeevan Mission (JJM), has introduced sensor-based Internet of Things (IoT) devices to monitor the supply of drinking water in more than six lakh villages. Pilot runs have been completed in five states of India in collaboration with Tata Community Initiatives Trust (TCIT) and Tata Trusts. These pilots have covered different types of sources including groundwater based borewells, springs in hilly areas, and surface water (river and dams). Several types of sensors have been deployed, including flow meters, groundwater level sensors, chlorine analysers, pressure sensors, pump controller to measure all the relevant aspects of water service delivery – quantity, duration, quality, pressure, and sustainability – in addition to providing operational efficiencies. This approach has not only allowed effective monitoring and management on-ground with a futuristic vision to ensure regular tap water to every home, but enormous gains in terms of operational efficiencies, cost reduction, grievance redressal. Read more...
• Using AI to predict floods and save lives: The Google Flood Forecasting Initiative uses machine learning to provide accurate real-time flood forecasting information and alerts to those in affected regions. This is made possible through AI and physics-based modelling which create accurate and scalable inundation models in real-world settings. In comparison to classical physics-based models, this morphological model improves accuracy by 3%, which can significantly improve forecasts for large areas, while also allowing for much more rapid model development by reducing the need for manual modeling and correction. First piloted in the Patna region of Bihar in 2018, Google’s flood forecasting initiative has been extended to the whole of India by 2020, covering 200 million people across more than 250,000 sq km. Read more...
• AI for Sustainable Water in Bengaluru: SmartTerra is developing an AI-powered operational intelligence platform to help water utility operators and cities transition from ad-hoc operations to predictive and efficiency-driven operations. Bengaluru’s water utility, Bengaluru Water Supply and Sewerage Board (BWSSB), serves nearly 1 million connections with a diverse customer base spread across varied neighborhoods. Most of these connections are metered by manually-read meters of various types and brands. The city not only faces a challenge in efficiently distributing water, but also has to deal with intermittent supply operations. The first challenge is real water losses because of physical leaks, overflows, etc. The second challenge, often unaddressed, is apparent or commercial losses. Long-term environmental and city sustainability will be made possible only by addressing both challenges. Read more...

Image by Bela Geletneky from Pixabay