A team led by researchers at the University of Tokyo have created a dataset of the whole atmosphere, enabling new research to be conducted on previously difficult-to-study regions. Using a new data-assimilation system called JAGUAR-DAS, which combines numerical modeling with observational data, the team created a nearly 20-yearlong set of data spanning multiple levels of the atmosphere from ground level up to the lower edges of space. Being able to study the interactions of these layers vertically and around the globe could improve climate modeling and seasonal weather forecasting. There is also potential for interdisciplinary research between atmospheric scientists and space scientists, to investigate the interplay between space and our atmosphere and how it affects us on Earth.
Complaining about the weather, and about weather forecasters when they get things wrong, is a popular pastime for many. But a meteorologist's job is not easy. Our atmosphere is multilayered, interconnected and complex, and global climate change is making it even harder to forecast both long-term and sudden, extreme weather events.
To help overcome these increasing challenges, researchers have created a dataset of the entire atmosphere. Ranging from September 2004 to December 2023, it spans multiple levels of the atmosphere from ground level up to the lower edge of space, about 110 kilometers above Earth's surface. The region between about 50 km to 110 km (though exact ranges vary) is particularly of interest, as it is so notoriously difficult to study that it had previously been dubbed the "ignorosphere." This region is too low for satellites and too high for weather balloons to observe, resulting in a shortage of data and consequently research. However, it is a fascinating area, characterized by vast global atmospheric tides and small-scale gravity waves which affect wind and temperature. It also plays an important role in the intensity of the impact of space weather events.
"The JAWARA (JAGUAR-DAS Whole neutral Atmosphere Reanalysis) dataset is a strong research tool which, for the first time, makes it possible to quantitatively understand atmospheric general circulation and the hierarchal structure of waves and vortices in the mesospheric layer (which is above the stratosphere and about 50-90 km above Earth's surface) and lower thermospheric layer (about 90-110 km above Earth's surface) of the atmosphere, including the ignorosphere," explained Professor Kaoru Sato from the University of Tokyo. "If we can better understand these layers, it would improve our ability to respond to climate change, extend the lead time of seasonal forecasts and advance our understanding of space weather phenomena."
The team developed its new JAGUAR-DAS high-speed data assimilation system as part of an international project led by Sato. The system integrates observational data into a numerical model which can then produce data on atmospheric conditions. The resulting dataset, named JAWARA, makes it possible to perform detailed analysis of the general circulation of the atmosphere and its hierarchical structure.
"Atmospheric general circulation models which range up to the lower edge of space have only been developed by a limited number of research institutions around the world, including our own," said Sato. "Recent studies indicate that extreme stratospheric phenomena can start at least in the upper mesosphere. Therefore, quantitative elucidation of phenomena in the mesosphere and lower thermosphere is extremely important for weather forecasting."
The dataset is now openly available, and the team intends to use it to study the large-scale circulation and the hierarchical structure in the atmosphere, as well as vertical and interhemispheric (i.e., between the Northern and Southern hemisphere) couplings. They also hope to work in collaboration with space scientists to study the interactions between the atmosphere and space, particularly the mesosphere (where the highest clouds form) and ionosphere (located within the thermosphere and about 60-300 km above Earth's surface, where many satellites are based).
JAWARA dataset site: https://jawara.nipr.ac.jp
