Jonathan Poterjoy and Kayo Ide, both from the Department of Atmospheric and Oceanic Science, will lend their expertise to the new $6.6 million initiative.

The University of Maryland joined a $6.6 million consortium to improve weather predictions and train the next generation of atmospheric scientists.

Recommended for funding by the National Oceanic and Atmospheric Administration (NOAA) through the federal Inflation Reduction Act, the new Consortium for Advanced Data Assimilation Research and Education (CADRE) will focus on improving data assimilation—the science of using observations to improve model predictions of natural systems, like Earth’s atmosphere, over time. The initiative will also bring students up to speed on a complex area of study that few people have mastered, creating a high demand for data assimilation specialists.

“The U.S. has some catching up to do in terms of data assimilation implementations,” said Jonathan Poterjoy, an assistant professor in the Department of Atmospheric and Oceanic Science (AOSC) who studies data assimilation and was named UMD’s principal investigator for this collaboration. “The U.S. has a massive shortage of students coming from grad schools to fill positions at places like NOAA and push the boundaries of what we can do with our current models.”

Although weather forecasts have vastly improved in the last several decades, the computer models used to create them need to be continuously upgraded to reflect new mathematical and technological developments. A recent example underscoring these shortcomings was the sudden onset of Hurricane Otis, which struck Mexico’s southern Pacific coast last fall and caused catastrophic damage.

“We had a very high-profile event this last hurricane season where a major hurricane made landfall right off the coast of Acapulco and there was very little lead time,” Poterjoy said. “The storm went from virtually nothing to a major hurricane in less than a day, and none of the models got it right. That’s something that shouldn’t happen.”

Extreme weather events are also becoming more common, creating an urgent need for more accurate forecasts. 

“The U.S. is experiencing nearly six times more major weather and climate disasters per year than it did 40 years ago, and the Biden-Harris Administration is committed to ensuring we have the most accurate data possible to mitigate the impact of these disasters and fight climate change,” said U.S. Secretary of Commerce Gina Raimondo.

“This investment, made possible thanks to President Biden’s Inflation Reduction Act, will upgrade and improve NOAA’s technology for numerical weather prediction capabilities to ensure accurate and timely information is available to the public and public safety officials in the face of extreme weather and climate events—making our communities more climate resilient.”

AOSC Associate Professor Kayo Ide, a data assimilation expert who teaches a course on the subject, also joined the UMD team participating in CADRE. Ide has appointments in AOSC, the Department of Mathematics and the Institute for Physical Science and Technology.

In addition to UMD, the CADRE collaboration includes five other universities: Colorado State University, Howard University, Pennsylvania State University, the University of Oklahoma and the University of Utah. Most of these institutions will focus on land surface or atmospheric applications, but Poterjoy and Ide will explore ways to improve data assimilation for two lesser-studied parts of global weather systems: the ocean and cryosphere.

“On the UMD side, we’re focusing primarily on marine applications, so that’s one thing that’s unique to us,” Poterjoy said. “We’re focusing on changes in ocean ice over relatively short timescales—days to weeks—because it’s becoming increasingly important to get a good handle on what sea ice looks like to forecast Arctic weather, which then has an impact on mid-latitude weather.”

Data assimilation can help paint a more accurate picture of what’s happening in a weather system and can lead to more accurate predictions of tropical cyclone intensity, rainfall, snow depth, thunderstorm wind speeds and more. It corrects a weather model in real time by taking new observations into account, and models such as the Global Forecast System—used by NOAA to produce weather forecasts—rely on these constant updates.

By identifying better numerical weather prediction systems and data assimilation methods, CADRE’s collaborators hope to more accurately predict the weather with the Unified Forecast System (UFS), a community-based and comprehensive Earth modeling system.

“The more precisely you can characterize what’s happening in the atmosphere right now, the more accurately you can predict in the future,” Poterjoy explained. “Any improvements you make in data assimilation can lead to better forecasts."

Experts from around the world will be tapped to solve this issue. CADRE will foster collaboration, student training and an exchange of expertise between NOAA, participating universities and the Joint Center for Satellite Data Assimilation.

Support from additional academic partners, including minority-serving institutions and international institutions such as the Met Office in the UK with its Met Office Academic Partnership and the new Transatlantic Data Science Academy, will further support improvements in weather and climate modeling. CADRE will also work closely with NOAA’s Earth Prediction Innovation Center to put new data assimilation science into practice within the UFS.

Poterjoy said he’s most excited to get UMD graduate students and postdocs involved in this collaboration, which he believes will strengthen their scientific expertise and career opportunities in the long-run.

“You’re going to end up with students graduating from our program with a better understanding of data assimilation as well as some of these outstanding issues with modeling,” Poterjoy said. “And if you’re graduating from our program with expertise in data assimilation, you’re going to have excellent job prospects.”

###

This article is adapted from text provided by NOAA.