New Technique Improves Understanding of Changes in the Frequency of Heavy Precipitation
Extreme weather events such as floods and droughts, are among the costliest natural disasters our society faces. Yet earth system model simulations of extreme precipitation events are not directly comparable to what we measure in our rain gauges. This is primarily due to relatively coarse spatial resolution, which precludes adequate representation of highly influential, small-scale features such as moisture convection and topography. Overcoming these drawbacks, a new algorithm developed by a research team from the Massachusetts Institute of Technology (MIT) produces more precise projections by pinpointing telltale large-scale atmospheric patterns associated with the occurrence of these smaller-scale events.
More precise projections of extreme precipitation can strengthen assessments of impacts and adaptation and thereby improve public safety and better target infrastructure investment. In addition, the algorithm that enables these projections can serve to diagnose earth system model deficiencies and identify model subcomponents where extreme-event processes can be better represented.
Extreme precipitation events pose a threat to public safety, natural and managed resources, and infrastructure. Identifying how such high-impact, low-probability events will change can aid communities in being better prepared for these natural disasters. However, any projected change in extreme precipitation events based on earth system model-simulated precipitation, especially on the local scale, lacks informative details, mainly due to the coarse spatial resolution, which precludes adequate representation of highly influential, small-scale features such as moisture convection and topography. To address this challenge, a team of researchers at the MIT Joint Program on the Science and Policy of Global Change and allied MIT departments has developed an algorithm that detects the occurrence of heavy precipitation events based on well-resolved, large-scale atmospheric circulation conditions associated with those events—rather than relying on the models’ representation of precipitation. The algorithm significantly reduces the uncertainty of extreme storm predictions in comparison with model-simulated precipitation. In multiple tests over different U.S. regions during different seasons, the algorithm provides more reliable estimates of late 20th-century heavy precipitation frequency than model-simulated precipitation. Applying the algorithm to project extreme precipitation events under a future business-as-usual scenario, the researchers found that California would likely undergo an increase in annual extreme precipitation events.