June 5th is World Environment Day – Can AI Enhance Environmental Protection and Sustainability Efforts?

Impactful AI and machine learning use cases emerge and evolve daily as the technologies begin to permeate all our lives. As we acknowledge World Environment Day, can AI also provide the progress we need to protect the planet?

World Environment Day 2024 focuses on land restoration, desertification, and drought resilience as billions of hectares of land are degraded. Land restoration can increase carbon storage, slow climate change, and save endangered species. It’s not just focused on restoration, but also the drivers of land degradation, drought, and desertification: the climate change and damage to the lush green and blue globe we call home. 

Can AI Protect the Environment and Stop Climate Change?

The United Nations Environment Programme (UNEP) led an article on AI with “We can’t manage what we don’t measure,” in 2022 to discuss AI’s value in handling the data crucial to managing environmental challenges of climate change, nature, and biodiversity loss, pollution and waste. 

Programme coordinator David Jenson says AI’s role ranges from designing energy efficient buildings and monitoring deforestation to optimizing energy deployment, adding:

“This can be on a large scale – such as satellite monitoring of global emissions, or a more granular scale – such as a smart house automatically turning off lights or heat after a certain time.”

Jenson’s comment illustrated how users of AI at any scale might be able to make a difference. AI has now evolved into publicly available generative AI and rafts of AI-powered tools for efficiency and productivity in every process and industry. 

AI for Climate Data and Response

AI’s applications for data collection, collation, analysis, interpretation, modeling, and forecasting are vast. The UNEP launched its AI-enhanced World Environment Situation Room (WESR) in 2022 for earth observation, recording such impacts as CO2 atmospheric concentration, methane emissions, changes in glacier mass, and sea level rise and to leverage the data into governments, classrooms, and boardrooms. The goal of WESR is to be a “mission control centre for planet earth,” says Jenson. 

The UNEP has also co-founded the GEMS Air Pollution Monitoring platform with IQAir. This platform aggregates air quality data from 25,000 stations in over 140 countries to provide insights on real-time air quality. Both platforms may help accelerate environmental action at scale.

The World Economic Forum (WEF) says AI’s power to process data and help humans make decisions is transforming industries and that it can contribute to the fight against climate change by tracking and predicting issues and improving agriculture. 

Scientists at the University of Leeds in the UK have an AI that can map and measure icebergs melting 10,000 times faster than humans. Space Intelligence is a company mapping over a million hectares of land to measure deforestation rates and the amount of carbon stored in forests. 

Google DeepMind is working with the non-profit Climate Change AI to create a wish list of datasets to advance global AI solutions for climate change. Google is also developing tools for weather forecasting and increasing the value of wind energy. 

Calculating and Mitigating Environmental Footprints

Jenson says AI is fundamental to calculating the environmental footprints of products through their lifecycles and supply chains to encourage businesses and consumers to make more informed decisions. 

“This kind of data is essential for sustainable digital nudging on e-commerce platforms such as Amazon.com. Shopify or Alibaba.”

Waste produces methane which is responsible for around 16% of greenhouse gas (GHG) emissions. Software company Greyparrot has built an AI system that has tracked over 32 billion waste items to help waste processing and recycling facilities recover and recycle more material. 

Environmental organization The Ocean Cleanup is using AI to create maps of ocean litter in remote locations so waste can be gathered and removed. The process is more efficient than previous efforts, which required ships and planes. 

Decarbonization, Emission Reduction, and Energy Efficiency

Reducing emissions is critical to limiting the effects of climate change and achieving net zero. Eugenie.ai is emissions tracking using satellite imagery and data from machines and processes to help companies track and reduce emissions by as much as 30%.

In Brazil, AI and drones are working together to calculate the number of seeds and then drop them to reforest hills around Rio de Janeiro, 100 times faster than humans. 

Where industry is responsible for around 30% of emissions, buildings are responsible for as much as 39%, 28% of which is for heat, cooling, and power. Mortar.io, out of the Google for Startups Accelerator, is using AI to plan carbon reduction for thousands of buildings quickly, performing automated digital audits to help companies progress to net zero. The company is also planning an AI chatbot to automate energy audits and deliver a tailored experience to customers. 

AI’s analytical and response speed may have a key role in the energy sector. Not only can AI predict weather events that impact renewable power generation and demand, but it can also optimize power flows in the grid to balance power supply during peak and off-peak periods.

Another company from Google’s accelerator, AgroScout, is using AI to monitor crop development, which can be affected by climate change impacts. AI can help detect pests and diseases early, allowing farms to treat them more precisely and reduce agrochemical use. It can also help ensure farm land and the water it needs is used effectively.

Agricultural equipment manufacturer John Deere has also developed crop sprayers that can differentiate plants from weeds to only spray where required to reduce pesticide use and build healthier soil. 

These are just some examples of how AI is being used to protect the environment, foster sustainability, and mitigate and remediate climate change. There are hundreds more, and given the current status quo, hopefully, all of them will have a positive impact.

AI and the Environment – The Largest Pitfall

Whilst AI and machine learning could have a huge impact in the battle to prevent climate change, it’s critical to remember and mitigate AI’s own environmental cost. The global ICT sector generates a substantial volume of emissions, data centers use significant amounts of water for cooling, and e-waste alone is predicted to hit 75 million metric tonnes by 2030. Google’s DeepMind was used back in 2016 to collect data and model data center cooling scenarios and discovered an optimal setup that reduced energy usage by 40%. 

The International Energy Agency (IEA) says:

“Emissions from data have grown only modestly despite rapidly growing demand for digital services, thanks to energy efficiency improvements, renewable energy purchases by information and communications technology (ICT) companies, and broader decarbonization of electricity grids in many regions.”

However, the IEA adds that for the data centers and transmission networks to “get on track with Net Zero,” emissions must drop by half by 2030. 

Hence, it’s vital that AI developers and users consider, reduce, and otherwise optimize the carbon footprint of their leverage of this transformational technology.