Reviewed by Danielle Ellis, B.Sc.Nov 18 2021
Global warming is considered to be responsible for aggravating extreme weather and climate events. The interaction between various forms of hazards activated by climate change will result in future cross-sectoral impacts influencing a range of natural and human systems.
Image Credit: Maxime Valcarce on Unsplash.
The study can enhance the awareness of such interactions and dynamics to help decision-makers in regulating current and future climate change threats, also as a result of an enhanced potential to forecast anticipated risks and measure their impacts.
Over the past few years, the scientific community has started testing new methodological methods, technologies and tools, among which the application of machine learning can help utilize the potential of huge and varied amounts of environmental tracking data available at present (big data).
What are the outcomes of the exponential increase in the application of machine learning techniques for the evaluation of climate-induced risks?
In the study titled, “Exploring machine learning potential for climate change risk assessment,” a team of researchers from the CMCC Foundation and Ca’ Foscari University of Venice performed an in-depth review of over 1,200 articles on the subject, reported in the past two decades. This stresses the ability and limitations of machine learning in this field.
Machine learning is a branch of artificial intelligence. By simulating the processes of the human brain, certain mathematical algorithms can understand the relationships between a set of input data in order to predict the required output. In our research, we identified that floods and landslides are the most analyzed events through machine learning models, probably because they are the most relevant and frequent around the world.
Federica Zennaro, Study Main Author and Researcher, CMCC Foundation and Ca’ Foscari University Venice
Furthermore, the study discloses that machine learning has two significant potentials that make it specifically interesting when employed in this field of study.
The first is that algorithms can learn from data: the more data, the better algorithms learn. As a result of its potential to examine and process big amounts of data, machine learning enables scientists to unravel complicated relationships underlying the working of socio-ecological systems.
This exploits the big data gathered from several sources, such as sensors for environmental analysis at high temporal frequency, social media, satellite data and images and drones.
The second is that they have the ability to integrate various kinds of data, thus allowing an evaluation of the risk extent while considering all its dimensions. These include not only the activating hazard (for instance, an increase in rainfall) but also the exposure and vulnerability of the socio-economic system at stake. These are critical factors in an evaluation of total impacts.
For example, consider a model that is trained with detailed data on flood events over the past 20 years, including their location and information on the affected context (urban or natural). This model can project, in a scenario characterized by future climate conditions, what the probability of an event happening at a certain point will be, and calculate its risk of causing harmful impacts to society and the environment.
Federica Zennaro, Study Main Author and Researcher, CMCC Foundation and Ca’ Foscari University Venice
Zennaro added, “Machine learning represents the future of risk assessment, but its great potential is not yet widely exploited. Our research shows that there are still few studies that use these models to develop long-term future risk scenarios (up to 2100). The vast majority of studies focus on the short term, probably influenced by the reduced availability of extended time series data capable of supporting adequate model training for long-term projections.”
According to study co-author Elisa Furlan who is a scientist at the CMCC Foundation and Ca’Foscari University Venice, the next step is to design machine learning models that are highly efficient at studying and untangling the complicated spatiotemporal interrelationships among various environmental, climatic and socioeconomic variables, thereby enhancing understanding of the behavior of complicated systems.
Under the perspective of a rising abundance of data and machine learning models’ complexity, researchers will have the possibility (and duty) to improve the understanding of climate-related risks, with the main aim of providing accurate and sound multi-risk scenarios able to drive robust adaptation planning and disaster risk reduction and management.
Elisa Furlan, Study Co-Author and CMCC Foundation and Ca’Foscari University Venice
Journal Reference:
Zennaro, F., et al. (2021) Exploring machine learning potential for climate change risk assessment. Earth-Science Reviews. doi.org/10.1016/j.earscirev.2021.103752.