Reduce losses and save lives, use artificial intelligence to fight natural disasters

According to data released by the National Center for Environmental Information, as of July 2022, a total of nine climate disaster events have caused economic losses of more than US$1 billion. According to reports, Hurricane Ian has caused more than 100 deaths and up to $47 billion in insured losses. This may be the most damaging storm disaster in Florida's history.

Since June 2022, floods in Pakistan have also killed 1,678 people, destroyed a large number of villages and infrastructure, and put 3.4 million children at risk of water-borne diseases, drowning and malnutrition. In addition, Hurricane Fiona left 90,000 people without power in Puerto Rico.

As natural disasters become more common and dangerous around the world, the application of technologies such as artificial intelligence (AI) may have the potential to prevent and mitigate related damage.

According to Neil Sahota, IBM Master Inventor, Chief AI Advisor to the United Nations, and co-founder of the AI ​​for Good Global Summit, people have long regarded it as a purely sudden extreme event. But in fact, thousands of subtle, slow-moving clues already point to the likelihood and severity of natural disasters.

Sahota explained, “As humans, we are very sensitive to immediate threats that erupt quickly, but are not good at recognizing long-term threats that advance slowly. Fortunately, the emergence of AI has given us the ability to A powerful tool for predicting natural disasters and taking prevention and mitigation measures."

Sahota used wildfires as an example to introduce how AI can process large amounts of data in real time and find subtle connections between different variables. “We’ve historically tended to assess fire risk through climate conditions, the distribution of brush, the amount of other potential fuels, and the topography of the area,” Sahota said. “But with the advent of AI wildfire tools, we’re able to include more variables. Especially the ignition factor."

Sahota pointed out that based on research data from mining companies, lightning strikes may be the main source of wildfires. But how do we evaluate such extremely random events?

“Humans may be helpless when faced with such problems, but AI can more easily predict where thunderstorms are likely to occur, the likelihood of thunderstorms hitting the ground, and “hot spots” with a higher risk of fire. This allows us to examine more fire sources, such as static electricity, hot surfaces and even friction, to assess wildfire threats. "

Sahota believes that AI can completely determine the chain reactions or indirect effects of potential disasters, and even prevent the next natural disaster.

"Let us take coastal dipping as an example. In the past, we mostly used key indicators such as sea level, but something unusual has recently occurred in the southeastern United States - an area that has almost never been affected by Lyme disease has experienced an outbreak of this disease. "

While studying this issue, Sahota and colleagues found that the ticks that cause the disease began to migrate further inland from the coastline. "Subtle changes in the coastal environment may be a problem for the ticks. drastic changes, triggering large-scale migration. "

"Using this clue, coupled with artificial intelligence, scientists can better understand how a single event affects an entire ecosystem. Now, we are beginning to use AI technology to study marine life, ocean currents and even ocean temperatures to find subtle clues that can indicate flood damage. ”

From disaster prediction to rescue resource optimization to disaster root cause analysis, AI has played an active role in detecting and preparing for extreme weather and other disasters. A team from Lancaster University has established a disaster Mapping and damage detection system to help rescue teams prioritize different areas during their work. The platform uses crowdsourced labeled data to obtain information such as road blockages, flooded areas and damaged buildings through the hands of volunteers on the ground.

Sahota said, “With hybrid intelligence, that is, the combination of humans and artificial intelligence, the impact of Hurricane Ian has indeed been controlled to a certain extent. "

"The advantage of AI is that it can take thousands of pieces of climate data into consideration in real time, helping us better predict the entire transition from a tropical storm to a hurricane, then weakening back to a tropical storm, and then reintensifying into a hurricane. process. This will allow us to design disaster preparedness plans for major storm-affected states like the Carolinas and Florida. "

Sahota also proposed several other important application directions of AI. "Using AI technology, we can analyze the trajectory of hurricanes in advance and deploy medical resources, food and water before they hit. Using AI's efficient communication capabilities, we are also expected to speed up evacuation and reduce the number of casualties. "

"After identifying the areas that may be damaged by disasters, we can also consider how to deploy resources and how to restore basic services such as water, power and food supply more quickly. ”

IFS North America CTO Rick Veague pointed out that AI prediction will also become an important supporting force in post-disaster recovery efforts.

Veague explained, "The greatest value of AI is that it can digest multiple information sources at once, calculate the probability of various possible outcomes, and make recommendations based on different reasons - without human intervention in the entire process. Except In addition to prediction, AI can also perform prediction optimization based on currently observed data to ensure that the AI-based decision-making process produces better results."

Veague emphasized, "The consequences of natural disasters are often chaotic and confusing. During the epidemic, municipal services were paralyzed and transportation infrastructure was damaged. With geospatial information collected by satellites, aerial photography and street cameras, as well as meteorological/past historical data and mobile phone signals, AI can map the situation in the disaster area and find the areas most in need. Population centers that provide aid supplies and organize response strategies efficiently."

With its powerful amount of information, AI can accurately locate the demand for aid and determine the best ways to deliver aid supplies.

"Take Hurricane Ian as an example. When recovery efforts began, supplies of food, water, and fuel were very limited. Therefore, finding efficient delivery methods can not only shorten the response cycle, but also protect the best interests of the victims. , quickly build confidence. This ability to rely on a wide range of information sources to obtain a large amount of data and accurately draw conclusions in a few seconds is simply beyond the reach of humans."

Veague concluded, "Anytime As the frequency of natural disasters continues to increase over the past decade, I believe AI technology will play an indispensable role in helping humans effectively respond to disasters."

The above is the detailed content of Reduce losses and save lives, use artificial intelligence to fight natural disasters. For more information, please follow other related articles on the PHP Chinese website!