New York Times, the

In June 2008, a flood destroyed large parts of the University of Iowa campus, with damages reaching up to $750 million. The school has since built back most of its infrastructure, and has considered flood risk in the process. For example, some buildings near the river were rebuilt on higher ground and valuable equipment was moved away from the ground floor.

Artificial intelligence and machine learning can be vital tools in disaster risk management, enabling analyses of data on previously impossible scales. However, models generated by these technologies are only effective if they are a catalyst for action.

The destruction caused by Cyclone Idai in Mozambique highlights the need for investment in resilient infrastructure. Some of the projects that are part of the recovery plan include buildings that can withstand stronger winds, an urban park around a river, monitoring and data collection.

The cold wave in the U.S. and the heat wave and fires in Australia are consistent with the scientific consensus that more greenhouse gas emissions in the atmosphere bring a greater likelihood of abnormally high temperatures. With extreme heat possibly yielding severe and deadly consequences, communities must prepare for a wide range of climate change scenarios.

In Central Asia, a warming climate is shrinking the Tuyuksu glacier in Kazakhstan, potentially limiting the amount of water for millions of people and crops. While glaciers all over the world are retreating, Kazakhstan may experience a more imminent decline that could prove to be severe for its farmers, who are unprepared for water shortages.

Disaster-ravaged communities often face the dilemma of deciding whether to rebuild or retreat from hazard-prone areas. New Orleans exemplifies this debate, as much of the city has rebounded from Hurricane Katrina but some places remain affected. Moreover, climate change will possibly compromise many gains, strengthening the case for retreat.

While some have suggested that California should restrict housing growth to reduce fire risk, this solution could create a housing shortage. Instead, the U.S. can learn from the Dutch, who maximize smart planning and infrastructure to fight disasters. The community of Montecito, among others, is following this lead through defensible spaces and other plans.

With the help of A.I., a growing number of scientists say changes in the way they can analyze massive amounts of seismic data can help them better understand earthquakes, anticipate how they will behave, and provide quicker and more accurate early warnings. Scientists hope A.I. can help them one day establish quake predictions, but challenges remain.