A new machine learning tool can reduce errors in national flood prediction programing, resulting in more accurate predictions of where floods will occur.  In a new study published in AGU Advances, scientists found that when the AI was used in combination with the National Water Model, developed by the U.S. National Oceanic and Atmospheric Administration, the resulting hybrid model was four to six times more accurate.

University of Oklahoma researchers have received $1.2 million from the National Science Foundation to develop an AI-powered system that predicts which trees are likely to fall during storms. The TREE-CARE project combines satellite imagery, computer modeling and community input to help cities proactively manage tree risks and reduce damage from extreme weather events.

According to a new study, most climate models exploring wildfires’ impacts on air quality overlook the effect of heat from a fire in one location on altering weather patterns – and in turn air quality – in locations farther afield. The authors of this study report that wildfires in the western United States are worsening air quality in the West but, paradoxically, may be improving it in the East. Over recent decades, wildfires in the western United States have become more frequent and intense, releasing vast amounts of smoke, which can greatly degrade air quality both locally and in regions far downwind. Often assumed to worsen when wildfire smoke drifts eastward from the West, fine particulate pollution (PM2.5) from wildfires has long been a major health concern in the eastern United States, potentially contributing to thousands of premature deaths each year. While this conventional wisdom has long informed air quality and fire management policies nationwide, the intense heat released by fires, which is capable of altering air temperatures, destabilizing the atmosphere, and driving strong convection, isn’t fully understood. Because wildfire heat can alter large-scale atmospheric circulation, its role in shaping distant air quality has been underappreciated. Most climate models consider only smoke emissions and not fire heat, which may explain why they tend to overestimate wildfire-driven pollution in the East.

 

Using both observational data and climate model simulations that incorporate daily heat measurements, Qihan Ma and colleagues found that, during extreme wildfires in the West, PM2.5 actually declines in the East, not only during extreme events but also throughout the fire season. This occurs because the intense heat released by large fires drives strong convection, which changes weather patterns, suppresses eastward smoke transport, and promotes rainfall that helps cleanse the air. However, while fire heat can improve air quality in the East, wildfires still worsen pollution in the West. According to Ma et al., ignoring fire heat in current climate models leads to inflated nationwide health and economic impact estimates – by about 1,200 premature deaths and $3.3 billion in damages – highlighting the need for policies that recognize its role in shaping regional air quality and environmental equity. “The differential impacts of wildfires necessitate the more appropriate allocation of resources and policy intensity for fire management and regional pollution control,” write the authors. “With greening in a warming world, the accumulation of fuels will lead to increased fire heat release when wildfires occur. It is imperative to consider fire heat in the Earth system when conducting comprehensive climate simulations and wildfire risk assessments.” In a related Perspective, Yun Qian discusses the study and the implications of its findings in greater detail.

Lightning will ignite more wildfires in the western United States in the coming decades as temperatures increase due to climate change, according to a new study published in AGU's journal Earth's Future that combined predictions for fire-friendly and lightning-friendly weather conditions.