Preparing for Disasters: What Lies Ahead?

Although artificial intelligence (AI) and machine learning (ML) are often used for entertainment purposes such as generating text and creating art, they also have the potential to address significant challenges. The rise of unpredictable and devastating natural disasters has made life increasingly uncertain in certain regions. By incorporating AI and ML models, the field of disaster management can undergo a revolutionary transformation, enabling us to better prepare for, respond to, and minimize the impact of catastrophic events. These new tools can be instrumental in safeguarding communities and saving lives.

In recent years, it has been difficult to keep up with the changing landscape of natural disasters. According to WMO’s 2020 State of Climate Services report, one in three people worldwide is not sufficiently covered by an early warning system. Climate change, urbanization and population growth have increased the vulnerability of communities, which is why traditional disaster planning methods are insufficient in many cases.

While concerns about climate change continue to confound, the world’s organizations and institutions are paying attention and reaching out to technology to help, where AI and ML have emerged as strong allies.

“We are all exposed to natural hazards, and this will get worse in the future,” Jürg Luterbacher, director of science and innovation at the World Meteorological Organization (WMO), said at a recent AI for Good webinar. “We have to act on them.”

WMO, the International Telecommunication Union (ITU) and the United Nations Environment Program (UNEP) are exploring the potential of artificial intelligence to enhance disaster mitigation and have established an Expert Focus Group on Artificial Intelligence in Natural Disaster Management to support global efforts to integrate artificial intelligence into disaster management systems.

If only we could predict the arrival of disaster. Recently, technology has helped model natural disasters, leading to successful tools for responding to weather events. An example is cyclone Phailin in eastern India in 2013, when accurate numerical model control prevented the tragedy that followed a similar storm 15 years earlier.

AI and ML will further strengthen this.

Anticipatory information about proactive measures

AI and ML models are particularly adept at processing massive amounts of data and identifying patterns that might elude human analysis. This feature is a game changer in disaster planning. By analyzing historical data, weather patterns, seismic activity and many other factors, these models can predict the likelihood and intensity of various disasters.

For example, Microsoft has partnered with SEEDS (Sustainable Environment and Ecological Development Society) and Gramener, a design-driven data science company, to work on a project to develop an advanced machine learning-based model to plan and respond more effectively to natural situations. disasters such as floods, heat waves, earthquakes, hurricanes and more. The model provides users with risk points down to every house in the affected area, allowing first responders to warn at-risk houses earlier, help them protect their property and evacuate.

Sundeep Reddy Mallu, Head of ESG and Analytics at Gramener, explains how AI can help in disaster risk monitoring. “In addition to destruction, natural disasters leave tons of data in their wake. AI and ML models can mine historical data and identify different patterns that can help future decision-making processes,” he says.

Artificial intelligence models are good at performing risk assessment of large areas by combining geospatial intelligence and on-site information.

“This kind of risk assessment helps early warning systems that can minimize the damage caused by natural disasters. Less damage means faster and cheaper recovery after a disaster,” he adds.

Another example is the Stanford Earthquake Detecting System (STEDS), which is an ML model that detects earthquakes that are often missed by traditional methods.

Predictive models can provide early warnings of a tsunami, giving coastal communities precious hours to evacuate. Similarly, AI-based forest fire prediction systems can analyze terrain, weather conditions and vegetation to predict areas at the highest risk of ignition, enabling the strategic deployment of firefighting resources.

Enhancing emergency response

The power of AI and ML is such that they can improve emergency response through real-time insights and optimized resource allocation. Drones equipped with artificial intelligence can be used, for example, to map disaster areas, creating detailed maps that help responders identify hazards and locate survivors. ML algorithms can also process incoming data from various sources, such as social media and 911 calls, to provide a comprehensive picture of a chaotic event.

In addition, AI-powered chatbots and virtual assistants can help send critical information to affected populations, guiding evacuation routes, shelters, and medical resources. While this ensures that correct and up-to-date information reaches people, it also reduces the burden on people to respond.

Google’s flood forecasting system, which has been deployed by India and Bangladesh to predict the severity and location of floods, uses a combination of computational hydrology and ML. The system simulates how water flows across the land, while taking into account factors such as terrain and historical flood data. The system then generates maps and alerts that give local communities and authorities time to prepare and respond.

According to Google, their flood warnings have reached millions of people and are providing them with potentially life-saving information. In India, where more than 20 percent of global flood-related deaths occur, this system has helped improve the country’s disaster preparedness.

Mitigation and resilience

While preparing for disasters is critical, efforts to mitigate their effects and improve community resilience are equally important. AI and ML models also affect these aspects. By simulating various disaster scenarios, urban planners can optimize infrastructure plans to withstand natural forces. This might include flood-resistant buildings, earthquake-resistant bridges, or wildfire-resistant landscaping strategies.

ML algorithms can also analyze historical data to identify areas prone to repeated damage, guiding decisions about land use and zoning regulations. Additionally, these technologies can help create more accurate risk assessment models for insurance companies and financial institutions, leading to better pricing and comprehensive options for risk entities.

In this regard, IBM’s PAIRS (Physical Analytics Integrated Data Repository and Services) geoscope uses artificial intelligence to analyze satellite images and assess disaster damage, helping to create successful disaster response and recovery planning.

This system uses artificial intelligence to analyze large and complex geospatial-temporal data sets and ML algorithms to analyze satellite images to identify changes caused by natural disasters such as floods, forest fires or earthquakes. For example, it can detect changes in vegetation, waterways and infrastructure and give a clear picture of the extent of the damage.

A collaborative future

AI and ML can do wonders for disaster planning, but we need to make sure we use them collaboratively by bringing together experts in climate science, engineering, data science and social science. Governments, NGOs, technology companies and local communities must form partnerships that prioritize the safety and well-being of all citizens.

United Nations (UN) Artificial Intelligence Advisor Neil Sahota, who published the bestseller ‘Own the AI Revolution’, says: “Artificial intelligence has become an important ally in disaster prevention, acting as a vigilant watchdog constantly monitoring a wide range of elements, from subtle changes in the weather to changes in geological formations , to anticipate potential disasters.”

Even as natural disasters become more frequent and severe, AI and ML models may prove to be the best option. We can harness the power of data-driven insights, predictive capabilities, and real-time decision support to save lives in the face of nature’s fury.

By Navanwita Sachdev, The Tech Panda