“I don’t want you to be hopeful. I want you to panic. I want you to feel the fear I feel every day. And then I want you to act on changing the climate”– Greta Thunberg
Greta Thunberg is a 16-year-old Swedish schoolgirl, who is famously called as a climate change warrior. She has started an international youth movement against climate change and has been nominated as a candidate for the Nobel Peace Prize 2019 for climate activism.
According to a recent report by the Intergovernmental Panel (IPCC), climate change is seen as the top global threat by many countries. The effects of climate change is going to make 1 million species go extinct, warns a UN report. The Earth’s rising temperatures are fueling longer and hotter heat waves, more frequent droughts, heavier rainfall, and more powerful hurricanes. Antarctica is breaking. Indonesia, the world’s 4th most populous country, just shifted its capital from Jakarta because it’s sinking. Singapore’s worried investments are moving away.
Last year, Europe experienced an ‘extreme year‘ for unusual weather events. After a couple of months of extremely cold weather, heat and drought plagued spring and summer with temperatures well above average in most of the northern and western areas. The UK Parliament has declared ‘climate change emergency’ after a series of intense protests earlier this month.
More than 1,200 people were killed across South Asia due to heavy monsoon rains and intense flooding (in some places it was the worst in nearly 30 years). The CampFire, in November 2018, was the deadliest and most destructive in California’s history, causing the death of at least 85 people and destroying about 14,000 homes. Australia’s most populous state New South Wales suffered from an intense drought in 2018.
According to a report released by the UN last year, there are “Only 11 Years Left to Prevent Irreversible Damage from Climate Change”.
Addressing climate change: How ARTIFICIAL INTELLIGENCE (AI) can help?
As seen above, environmental impacts due to climate changes are clear, the list is vast and depressing. It is important to address climate change issues as they play a key role in the workings of a natural ecosystem like change in the nature of global rainfall, diminishing ice-sheets, and other factors on which the human economy and the civilization depends on.
In the recently held ICLR 2019 (International Conference on Learning Representations), Emily Shuckburgh, a Climate scientist and deputy head of the Polar Oceans team at the British Antarctic Survey highlighted the need of actionable information on climate risk. It elaborated on how we can monitor, treat and find a solution to the climate changes using machine learning. Also mentioned is, how AI can synthesize and interpolate different datasets within a framework that will allow easy interrogation by users and near-real time ingestion of new data.
According to MIT tech review on climate changes, there are three approaches to address climate change: mitigation, navigation and suffering. Technologies generally concentrate on mitigation, but it’s high time that we give more focus to the other two approaches. In a catastrophically altered world, it would be necessary to concentrate on adaptation and suffering. This review states that, the mitigation steps have had almost no help in preserving fossil fuels. Thus it is important for us to learn to adapt to these changes. Building predictive models by relying on masses of data will also help in providing a better idea of how bad the effect of a disaster can be and help us to visualize the suffering.
By implementing Artificial Intelligence in these approaches, it will help not only to reduce the causes but also to adapt to these climate changes. Using AI, we can predict the accurate status of climate change, which will help create better futuristic climate models. These predictions can be used to identify our biggest vulnerabilities and risk zones. This will help us to respond in a better way to the impact of climate change such as hurricanes, rising sea levels, and higher temperatures.
Let’s see how Artificial Intelligence is being used in all the three approaches –
Mitigation: Reducing the severity of climate change
Looking at the extreme climatic changes, many researchers have started exploring how AI can step-in to reduce the effects of climate change. These include ways to reduce greenhouse gas emissions or enhance the removal of these gases from the atmosphere.
In view of consuming less energy, there has been an active increase in technologies to use energy smartly. One such startup is the ‘Verv’. It is an intelligent IoT hub which uses patented AI technology to give users the authority to take control of their energy usage. This home energy system provides you with information about your home appliances and other electricity data directly from the mains, which helps to reduce your electricity bills and lower your carbon footprint. ‘Igloo Energy’ is another system which helps customers use energy efficiently and save money. It uses smart meters to analyse behavioural, property occupancy and surrounding environmental data inputs to lower the energy consumption of users.
‘Nnergix’ is a weather analytics startup focused in the renewable energy industry. It collects weather and energy data from multiple sources from the industry in order to feed machine learning based algorithms to run several analytic solutions with the main goal to help any system become more efficient during operations and reduce costs.
Recently, Google announced that by using Artificial Intelligence, it’s wind energy has boosted up to 20 percent. A neural network is trained on the widely available weather forecasts and historical turbine data. The DeepMind system is configured to predict the wind power output 36 hours ahead of actual generation. The model then recommends to make hourly delivery commitments to the power grid a full day in advance, based on the predictions.
Large industrial systems are the cause of 54% of global energy consumption. This high-level of energy consumption is the primary contributor to greenhouse gas emissions.
In 2016, Google’s ‘DeepMind’ was able to reduce the energy required to cool Google Data Centers by 30%. Initially, the team made a general purpose learning algorithm which was developed into a full-fledged AI system with features including continuous monitoring and human override. Just last year, Google has put an AI system in charge of keeping its data centers cool. Every five minutes, AI pulls a snapshot of the data center cooling system from thousands of sensors. This data is fed into deep neural networks, which predicts how different choices will affect future energy consumption. The neural networks are trained to maintain the future PUE (Power Usage Effectiveness) and to predict the future temperature and pressure of the data centre over the next hour, to ensure that any tweaks did not take the data center beyond its operating limits. Google has found that the machine learning systems were able to consistently achieve a 30 percent reduction in the amount of energy used for cooling, the equivalent of a 15 percent reduction in overall PUE.
As seen, there are many companies trying to reduce the severity of climate change.
Navigation: Adapting to current conditions
Though there have been brave initiatives to reduce the causes of climate change, they have failed to show any major results. This could be due to the increasing demand for energy resources, which is expected to grow immensely globally. It is now necessary to concentrate more on adapting to climate change, as we are in a state where it is almost impossible to undo its effects. Thus, it is better to learn and navigate through this climate change.
A startup in Berlin, called ‘GreenAdapt’ has created a software using AI, which can tackle local impacts induced both by gradual changes and changes of extreme weather events such as storms. It identifies effects of climatic changes and proposes adequate adaptation measures. Another startup called ‘Zuli’ has a smartplug that reduces energy use. It contains sensors that can estimate energy usage, wirelessly communicate with your smartphone, and accurately sense your location. A firm called ‘Gridcure’ provides real-time analytics and insights for energy and utilities. It helps power companies recover losses and boost revenue by operating more efficiently. It also helps them provide better delivery to consumers, big reductions in energy waste, and increased adoption of clean technologies.
With mitigation and navigation being pursued enough, let’s see how firms are working on futuristic goals.
Visualization: Predicting the future
It is also equally important to visualize accurate climate models, which will help humans to cope up with the aftereffects of climate change. Climate models are mathematical representations of the Earth’s climate system, which takes into account humidity, temperature, air pressure, wind speed and direction, as well as cloud cover and predict future weather conditions. This can help in tackling disasters. It’s also imperative to fervently increase our information on global climate changes which will help to create more accurate models.
A startup modeling firm called ‘Jupiter’ is trying to better the accuracy of predictions regarding climate changes. It makes physics-based and Artificial Intelligence-powered decisions using data from millions of ground-based and orbital sensors. Another firm, ‘BioCarbon Engineering’ plans to use drones which will fly over potentially suitable areas and compile 3D maps. Then, it will scatter small containers over the best areas containing fertilized seeds as well as nutrients and moisture gel. In this way, 36,000 trees can be planted every day in a way that is cheaper than other methods. After planting, drones will continue to monitor the germinating seeds and deliver further nutrients when necessary to ensure their healthy growth. This could help to absorb carbon dioxide from the atmosphere.
Another initiative is by a ETH doctoral student at the Functional Materials Laboratory, who has developed a cooling curtain made of a porous triple-layer membrane as an alternative to electrically powered air conditioning.
In 2017, Microsoft came up with ‘AI for Earth’ initiative, which primarily focuses on climate conservation, biodiversity, etc. AI for Earth awards grants to projects that use artificial intelligence to address critical areas that are vital for building a sustainable future. Microsoft is also using its cloud computing service Azure, to give computing resources to scientists working on environmental sustainability programs.
Intel has deployed Artificial Intelligence-equipped Drones in Costa Rica to construct models of the forest terrain and calculate the amount of carbon being stored based on tree height, health, biomass, and other factors. The collected data about carbon capture can enhance management and conservation efforts, support scientific research projects on forest health and sustainability, and enable many other kinds of applications. The ‘Green Horizon Project from IBM’ analyzes environmental data and predicts pollution as well as tests scenarios that involve pollution-reducing tactics. IBM’s Deep Thunder’ group works with research centers in Brazil and India to accurately predict flooding and potential mudslides due to the severe storms.
As seen above, there are many organizations and companies ranging from startups to big tech who have understood the adverse effects of climate change and are taking steps to address them. However, there are certain challenges/limitations acting as a barrier for these systems to be successful.
What do big tech firms and startups lack?
Though many big tech and influential companies boast of immense contribution to fighting climate change, there have been instances where these firms get into lucrative deals with oil companies. Just last year, Amazon, Google and Microsoft struck deals with oil companies to provide cloud, automation, and AI services to them. These deals were published openly by Gizmodo and yet didn’t attract much criticism. This trend of powerful companies venturing into oil businesses even after knowing the effects of dangerous climate changes is depressing.
Last year, Amazon quietly launched the ‘Amazon Sustainability Data Initiative’.It helps researchers store many weather observations and forecasts, satellite images and metrics about oceans, air quality so that they can be used for modeling and analysis. This encourages organizations to use the data to make decisions which will encourage sustainable development. This year, Amazon has expanded its vision by announcing ‘Shipment Zero’ to make all Amazon shipments with 50% net zero by 2030, with a wider aim to make it 100% in the future. However, Shipment Zero only commits to net carbon reductions. Recently, Amazon ordered 20,000 diesel vans whose emissions will need to be offset with carbon credits. Offsets can entail forest management policies that displace indigenous communities, and they do nothing to reduce diesel pollution which disproportionately harms communities of color. Some in the industry expressed disappointment that Amazon’s order is for 20,000 diesel vans — not a single electric vehicle.
In April, Over 4,520 Amazon employees organized against Amazon’s continued profiting from climate devastation. They signed an open letter addressed to Jeff Bezos and Amazon board of directors asking for a company-wide action plan to address climate change and an end to the company’s reliance on dirty energy resources.
Recently, Microsoft doubled its internal carbon fee to $15 per metric ton on all carbon emissions. The funds from this higher fee will maintain Microsoft’s carbon neutrality and help meet their sustainability goals. On the other hand, Microsoft is also two years into a seven-year deal—rumored to be worth over a billion dollars—to help Chevron, one of the world’s largest oil companies, better extract and distribute oil. Microsoft Azure has also partnered with Equinor, a multinational energy company to provide data services in a deal worth hundreds of millions of dollars. Instead of gaining profit from these deals, Microsoft could have taken a stand by ending partnerships with these fossil fuel companies which accelerate oil and gas exploration and extraction.
With respect to smaller firms, often it is difficult for a climate-focused conservative startup to survive due to the dearth of finance. Many such organizations are small and relatively weak as they struggle to rise in a sector with little apathy and lack of steady financing. Also startups being non-famous, it is difficult for them to market their ideas and convince people to try their systems. They always need a commercial boost to find more takers.
Pitfalls of using Artificial Intelligence for climate preservation
Though AI has enormous potential to help us create a sustainable future, it is only part of a bigger set of tools and pathways needed to reach the goal. It also comes with its own limitations and side effects. An inability to control malicious AI can cause unexpected outcomes. Hackers can use AI to develop smart malware that interfere with early warnings, enable bad actors to control energy, transportation or other critical systems and could also get them access to sensitive data. This could result in unexpected outcomes at crucial output points for AI systems. AI bias, is another dangerous phenomena, that can give an irrational result to a working system. Bias in an AI system mainly occurs in the data or in the system’s algorithmic model which may produce incorrect results in its functions and security.More importantly, we should not rely on Artificial Intelligence alone to fight the effects of climate change. Our focus should be to work on the causes of climate change and try to minimize it, from an individual level. Even governments in every country must contribute, by initiating “climate policies” which will help its citizens in the long run. One vital task would be to implement quick responses in case of climate emergencies. Like the recent case of Odisha storms, the pinpoint accuracy by the Indian weather association helped to move millions of people to safe spaces, resulting in minimum casualties.