AI has already made a significant contribution to the fight against global warming. Quantifying its importance and describing its impacts, however, are still unresolved issues.
This article offers an overview of initiatives and projects that rely on AI to understand and address climate change, highlights existing research that suggests that AI may have a positive impact on the issue, and outlines a number of challenges that must be overcome to ensure that such use of AI is both ethically and practically sound.
Problem with climate change
The effects of climate change on the environment, society, and economy will be significant. Numerous of its environmental effects, including as extended droughts and more severe storms, are already being felt.
According to a 2022 BCG Climate AI Survey report, 87% of business and public sector CEOs with climate decision-making authority believe AI is a critical tool in the fight against climate change. According to the same survey, the most important business benefit of climate-related advanced analytics and AI was mitigation (decrease), which was valued 57% higher than mitigation (measured emissions). As for the remaining percentages, they are divided into:
In terms of adjusting to climate change, 44% (hazard forecasting)
42% in terms of climate change adaptation (vulnerability and exposure management). 37 percent in reducing climate change (emissions removal)
in basics, 28% (facilitating climate research, climate finance, and education)
existing remedy
We need to examine metropolitan places from a variety of perspectives in order to fully understand how they are impacted by climate change. Some of them include the surface temperatures of the soil and water, weather conditions, the quantity of plants, and the amount of ice on the ground. Numerous climate models may predict the weather in a region using these characteristics. Two of the most significant models that are frequently used in this discipline are the Earth System Models (ESMs) and the Global Climate Models (GCMs).
In addition to simulating the carbon cycle and other crucial chemical and biological cycles that are crucial for predicting how much greenhouse gas will be in the atmosphere in the future, ESMs have all the same components as GCMs. ESM models may also anticipate huge areas because they simulate environmental indicators in expansive computational domains.
Higher-resolution ESM simulations were performed by Canadian, Swedish, and Finnish researchers. They uncovered flow-blockage effects, which were missing from the coarse instances but were very important to the behaviour of the global climate.
Researchers from the US have also discussed this issue and noted that when applied to a smaller area, global models (or large-grid models) yield results that are only minimally correct.
Academics from the University of Washington Seattle and other colleges believe that the best method for making forecasts is to simulate a variety of scenarios based on the features of a certain location.
In order to better understand the causes and effects of human activities on the environment, research was undertaken in Ireland on a variety of data types that smart cities should collect.
Researchers from the UK proposed that we could use the smart-city indicators published by the ISO (International Organization for Standardization) (ISO 37122:2019 Sustainable cities and communities, Indicators for smart cities) to decide what data needs to be gathered to monitor the sustainable growth of cities.
By assessing the surface heat stress of cities using satellite sensors, Italian researchers showed that it is possible to estimate and track changes in the urban heat island effect.
To monitor water quality and levels, soil and water surface temperatures, biomass, carbon, and air quality, German researchers have suggested employing remote sensing images.
We may use coarse data in conjunction with high-fidelity turbulent flow simulations, as researchers from Spain, Sweden, and the Netherlands have shown. With the help of remote sensing data and AI models, this work increases confidence in the prospect of high-fidelity flow and climate prediction in the future.
