Rising temperatures, wildfires, shifting wildlife populations, extreme weather events, and rising seas have become commonplace across the globe, courtesy of climate change. As the Planet continues to warm, the impact of climate change seems to progress from bad to worse. But thanks to artificial intelligence (AI), we now have a tool that can help us manage this menace.
In recent years, AI has become a powerful tool for technological progress. Despite the growth in its use cases to solve societal problems, we need to apply the technology towards fighting climate change.
Tackling climate change involves adaptation and mitigation, both of which are multifaceted issues. Adaptation requires planning for disaster management, while mitigation entails reducing greenhouse gas emissions. So, let’s look at how we can use AI to mitigate and adapt to climate change.
1. Mitigating Emissions from Electrical Systems
Electrical systems account for about 25% of greenhouse gas emissions annually. To reduce emissions from electrical systems, we must embrace low carbon electricity sources and minimize emissions from carbon-emitting power plants. These changes should be implemented across all sectors in every country.
AI and its derivative machine learning (ML) can significantly contribute to mitigating the effect of CO2 emissions from electrical systems. Specifically, the technologies can expedite the development of clean energy technologies, improve the optimization of electrical systems, and provide forecasts for energy demand. Additionally, they can help in system monitoring to facilitate efficacy.
2. Smart Transportation & Mobility
Globally, transport systems are responsible for 25% of energy-related CO2 emissions. However, when compared to electrical systems, this sector has made significant steps towards reducing emissions.
Different modes of transport emit CO2 in varying proportions, with road travel constituting more than two-thirds. To reduce emissions, we need to reduce transport activity, adopt alternative fuels, improve vehicle efficiency, and change our modes of transportation.
Often, urban planners design the transport infrastructure with limited information. However, AI has filled the gap by providing sensors that turn the raw data into meaningful insights. Now, roads can be modeled by using machine learning to assess the known traffic patterns for similar roads.
AI can also be used to model demand and in planning infrastructure. Machine learning can provide insights on mobility patterns, thereby discouraging sprawl. From traffic counts, it is easy to use AI to estimate origin-destination demand, which will be instrumental in transport policy decisions. Additionally, modal shifts in passenger transportation may involve analyzing travel demand data using AI to find a solution that will reduce carbon emissions.
3. Construction and Smart City Design
The energy consumed in buildings constitutes 25% of global energy-related gas emissions. However, state of the art technologies can reduce these pollutants by up to 90%. Machine learning provides vital tools that can help policymakers to reduce greenhouse gas (GHG) emissions.
AI can help in building management by selecting and implementing strategies that are customized to individual buildings through smart control systems. In urban planning, AI can help to collate and analyze data that will help policymakers. Moreover, it can help cities to transition into the use of low carbon materials.
When designing new buildings, we can use several technologies to reduce GHG emissions. Artificial intelligence can help by modeling data on the power consumption of different buildings using data from energy monitors. By transferring the knowledge acquired from modeling one building to another, AI can provide precise forecasts on the energy demands of similar structures.
Smart control systems in buildings can remarkably reduce the carbon footprint by integrating low-carbon sources into the power mix. Specifically, AI can allow electrical systems and devices to adapt to usage patterns.
In the realm of cities, artificial intelligence has become very useful in improving efficiency, resulting in smart cities. To reduce the carbon footprint, these jurisdiction should capture any relevant data that relates to activities that consume energy. Fortunately, most of them are already embracing the need to obtain the data. Notably, the city of Los Angeles requires all Mobility as a Service (MaaS) providers to use an open-source API.
4. Demand Forecasting in Manufacturing and Logistics
Industrial production, building materials, and logistics are the leading causes of greenhouse gas emissions. Luckily, the sector is at the forefront in gathering data that, when processed, can make a positive impact on the environment.
One of the challenges facing the industry is overproduction. With AI, manufacturers can perform demand forecasting to reduce the emissions emanating from climate-controlled warehousing. In production, AI, supported by data on processes, can improve the efficiency of industrial control mechanisms, such as HVAC (heating, ventilation, and air-conditioning) systems. Besides, the technology can help to reduce food waste through demand forecasting, optimization of delivery routes, and improvements in refrigeration systems.
5. Precision Agriculture and Forest Management
Land use by humans accounts for about 23% of global greenhouse gas emissions. We can reduce these emissions through better land management and agriculture. AI can play a vital role in facilitating the reduction, primarily through the implementation of precision agriculture.
The adoption of precision agriculture can minimize carbon release from the soil and enhance crop yield, thereby reducing deforestation. AI can also track the health of forests and predict the risk of wildfires. Moreover, satellite images enable us to estimate the quantity of carbon sequestered while also tracking greenhouse emissions from a piece of land.
One technology that is playing a vital role in the management of forests is SilviaTerra. Powered by Microsoft technology, the tool uses AI and satellite imaging to predict the species, sizes, and health of forest trees. This saves time for conservationists who’d have to contend with a tedious manual assessment of the forests.
6. Carbon Dioxide Removal
While it’s possible to cut emissions, we still face the challenge of dealing with the greenhouse gases already in the atmosphere. One of the most commonly used methods of eliminating CO2 is natural uptake by plants. However, with AI, it is possible to use technology to reduce CO2 from the environment.
With real-time maps of GHGs, we can quantify emissions from forestry and agriculture while monitoring the ones originating from different sectors. The data will facilitate target-setting and pinpointing the areas that require strict regulations to minimize emissions.
Oil and gas industries have made significant progress in the use of machine learning for subsurface imaging using raw seismograph traces. The models used in the industry, coupled with the data, can help in trapping co2 as opposed to releasing it. Moreover, AI, through the use of sensors, can help to maintain and monitor active sequestration sites.
7. Climate Prediction
Recent developments have created opportunities for the use of AI in climate predictions. Already, we have cheaper satellites that can create petabytes of climate observation data. Additionally, several climate modeling projects are providing petabytes of simulated climate data. Finally, AI is facilitating fast-to-run climate forecasts, especially using next-generation computing hardware.
The Antarctic and Arctic are warming at a faster rate than any other parts of the Earth. These areas are important to the world because their climates hold the future of the global sea levels. Unfortunately, the areas are icy and dark, making them very difficult to observe. However, advancements in satellite technology have provided useful data that can be used with AI to solve future problems.
Climate scientists have increasingly embraced ML techniques in climate forecasting by building data-driven models to solve problems that were initially challenging. As satellite databases continue to grow, we are likely to see more applications of machine learning techniques.
An example of a company that has used artificial intelligence in climate forecasting is IBM. The company started the Green Horizon Project to help cities become more efficient. The initiative uses AI to provide weather and pollution forecasts. Between 2012 and 2017, the project was a huge success after it helped Beijing City to decrease its average smog levels by 35%.