AI for Environmental Sustainability

by department of educationNovember 14th, 2021


Are AI and sustainability the perfect match? According to the United Nations’ report, 79 % of listed 17 Sustainable Goals can be met with Artificial Intelligence in the next years. The question now is how easy it would be to accomplish this, amidst of all challenges.

Life has evolved over several years on this beautiful planet. Ever since the evolution, two most complex systems have grouped into a bonding relationship – Human culture & living world. Since then human wellbeing is closely linked to the health of environment. However, the rapid industrial growth and over consumption of natural resources has led to a global concern of a sustained and balanced environment. According to the United Nations World Commission on Environment and Development, environmental sustainability is improving quality of human life without compromising the needs of future generations. [1] A recently published paper called” Tackling Climate Change with Machine Learning” [17] offers up to 13 areas where machine learning could be deployed including Energy systems, Transportation, climate prediction, Agriculture, etc.

What about AI in the Ocean? 

Improving Ocean mapping
It’s a no brainer that 70% of earth surface is covered by Oceans itself and yet only 5% is mapped. Can AI be an ocean saver? Oceans provide a rich biodiversity of flora and fauna along with supporting life systems of humans. However, global warming has caused rising sea levels and unpredictable tsunamis and floods in the last century. Fundamentally, AI can accelerate our ability to observe ocean dynamics. Data can help understand the ocean temperature, rising sea levels and health of oceans in general. Researchers at IMT Atlantique’s team are using Microsoft Azure to build the 3d models of ocean for analysis and reconstruction of ocean atmosphere by using satellite remote sensing data. [2]

Restoring Healthy marine life (Acoustic pollution)
Restoring a healthy marine life can also play a huge role in reducing poverty and increasing food sources in coastal areas. Acoustic pollution from nearby ports disturbs Cetaceans (dolphins, whales) which are highly sensitive to sound pressure levels. Efficient ML techniques could be employed to detect marine mammal proximity and better construction decisions in real-time [3].

AI in the Field of Agriculture

Enhancing healthy plants
Rising population needs better quality crops. Fortunately, various ML techniques are being used to improve the quality of crops, identify bacteria’s, vermin’s and produce best yield. A Munich based start-up Treesense uses sensor data on trees to predict when trees need to be irrigated and understand the world of plants through various ML techniques. Another start-up Descartes labs uses satellite images and cloud computing to discover how good and healthy the crops are. [4].
AI can also help reduce both fertilizer and water, all while improving crop yields. Trace Genomics uses information of bacteria and Fungi in soil to provide customers insights into soil health. Other companies here include Blue River Technology and Harvest CROO Robotics [5].
Monitoring Peatlands
Peatlands account for the largest source of sequestered carbon on Earth, though only 3% of Earth’s land area are covered by these wetland ecosystems. Remote sensing data was used by [16] along with ML to extract features and estimate thickness of peat and assess the carbon stock of tropical wetlands.

AI and Energy Systems 

Building Energy efficient Systems 
How can/will AI help produce clean and affordable energy? Researchers at RMIT University, Australia have created a ML model that predicts the Power-conversion efficiency of materials to design next-gen organic cells. [6] Another area where AI has set foot in is in the development of magnetic confinement tokamak reactors that promise clean fusion energy. Scientists at Princeton Plasma Physics Laboratory have introduced Fusion Recurrent Neural Network (FRNN) to forecast disruptive events in Tokamaks [7].

Improving intelligent Grid Systems 
Several organizations have started using AI to build intelligent grid systems through optimizing efficiency, cutting costs and carbon pollution generation. Companies such as Foghorn systems have come up with Automated Flare stack monitoring system that uses Computer vision to detect flame and smoke from flare stacks. [8]

AI in the Field of Transportation

Reducing transport activity
ML can be used in Freight consolidation (bundling shipments together) to reduce the number of trips by optimizing complex interaction of shipment sizes, modes and service requirements ML can be used to forecast road traffic and understand which links passengers use the most. This in return help in reducing GHG emissions. 

Improving vehicle efficiency
Efficient engines improved aero-dynamics and hybrid electric engines are the need of hour to reduce the energy consumption of vehicles. Lai et al. [10] uses computer vision to detect aerodynamically inefficient loading on freight trains.

AI based climate prediction

Clouds and aerosols
Deep neural networks could be combined with thermodynamic knowledge to understand the uncertainty in clouds. Bright clouds block sunlight and cool the Earth; dark clouds catch outgoing heat and keep the Earth warm. Gentine et al, trained a DNN to emulate these processes and understand cloud simulation for a fraction of cost [17].

Storm Tracking 
Climate change is real and is the ultimate concern for this generation. Deep learning-based models can help understand the climate data for future weather forecasting and for a smarter disaster response. [8] A start-up Jupiter Intelligence uses ML to make climate predictions by computing localized flood and temperature risk scores. Researchers have used deep learning to classify [12], detect [13] and segment [14] cyclones and atmospheric rivers, as well as tornadoes [15] in historical climate datasets.


The vastpotential of AI can be used to fill gaps and identify high impact problems that we see in day to day lives. Are we as Entrepreneurs, engineers, scientists, investors, leaders and most important as an individual ready to exploit the immense applications of AI to restore, protect and preserve our beautiful environment? Then stay curious and complete the circle!


[1] Marni Evans. “What is Environmental
Sustainability”, The balance small business, July 07, 2020.
[2] Microsoft Reporter. “Saving the seas: how AI is helping to protect our oceans”,
July 02, 2019.
[3] Snowdrop solution. “Can AI Save Our Oceans? Let’s Start With The Data”,
September 19, 2019.
[4] Technostacks. “Role of Machine Learning in Modern Agriculture”
[5] Glenn Gow. “Environmental Sustainability And AI”, Forbes, August 21, 2020
[6]  Nastaran Meftahi. “New Machine learning
program to speed up clean energy generation”, RMIT Australia, November
09, 2020
[7] William Tang. “AI and Deep Learning Accelerate Efforts to Develop Clean,
Virtually Limitless Fusion Energy, Princeton Plasma Physics Laboratory,
April 22, 2019
[9] Celine Herweijer. “8 ways AI can help
save the planet”, World Economic Forum, January 24, 2018
[10] Y-C Lai, et al.Machine vision analysis of the energy efficiency of intermodal
freight trains. Proceedings of the Institution of Mechanical Engineers, Part F:
Journal of Rail and Rapid Transit, 221(3):353–364, 2007.
[11] IPCC. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and
III to the Fifth Assessment Report of the Intergovernmental Panel on Climate
Change [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. 2014
[12] Yunjie Liu, et al. Application of deep convolutional neural networks for
detecting extreme weather in climate datasets. International Conference on
Advances in Big Data Analytics, 2016.
[13] Evan Racah, et al. ExtremeWeather: A large-scale climate dataset for semi-supervised
detection, localization, and understanding of extreme weather events. In
Advances in Neural Information Processing Systems 30, pages 3402–3413. 2017.
[14] Thorsten Kurth, et al. Exascale deep learning for climate analytics. In
Proceedings of the International Conference for High Performance Computing,
Networking, Storage, and Analysis, SC ’18, pages 51:1–51:12, Piscataway, NJ,
USA, 2018. IEEE Press.
[15] Valliappa Lakshmanan, et, al. An objective method of evaluating and devising
storm-tracking algorithms. Weather and Forecasting, 25:701–709, 2010
[16] Budiman Minasny, et al. Open digital mapping as a cost-effective method for
mapping peat thickness and assessing the carbon stock of tropical peatlands. Geoderma,
 313:25–40, 2018.
[17] David Rolnick, et al. Tackling Climate Change with Machine Learning. arXiv:
1906.05433v2 [cs.CY] , 5 Nov 2019