About the Author

Connecting Atmospheric Structures with the “Godzilla” Dust Storm

By Jillian Kunze

In an average year, 180 million metric tons of dust from the Sahara Desert travels through the atmosphere across the Atlantic Ocean. When this dust reaches North and South America, it can degrade air quality, deposit pathogens, and cause other ecological concerns. The largest such event in recent years was the “Godzilla” dust plume in June 2020, which had a huge impact on air quality in the southern U.S. and Caribbean Basin.

During a minisymposium presentation at the 2023 SIAM Conference on Applications of Dynamical Systems, which is taking place this week in Portland, Ore., Albert Jarvis of Virginia Tech described an effort to uncover the dominant atmospheric structures that caused this dust plume’s evolution. This NASA-funded project is in collaboration with Shane Ross, Ali Hossein Mardi, and Hosein Foroutan (all of Virgina Tech). “Every year, huge amounts of dust are kicked up from the Sahara,” Jarvis said. “What we’re going to do is attempt to find structures within the velocity field that tell us something useful.”

The Modern-Era Retrospective analysis for Research and Applications (MERRA-2) dataset provides data on the velocity and streamfunction in the atmosphere at the time and location of the dust storm. While the approach that Jarvis described requires some simplifications in order to be feasible for real-time decision-making applications—such as averaging the velocity fields from this data across different pressure surfaces—the approach is still robust. Jarvis also used data on the aerosol index (AI), which measures dust and other aerosols in the atmosphere, from the Ozone Mapping and Profiler Suite (OMPS). Applying a threshold to the OMPS data focuses in on the plume. 

Figure 1. Aerosol index (AI) data from the Ozone Mapping and Profiling Suite (OMPS), in brown, mapped along with the finite-time Lyapunov exponent (FTLE) structures, in blue, from June 15-18, 2020. A vortex in the FTLE field meets a ridge structure, causing the dust storm to billow across the Atlantic. Figure courtesy of Albert Jarvis.

Jarvis calculated finite-time Lyapunov exponent (FTLE) fields based on the MERRA-2 data to reveal the major atmospheric structures that were present at the time of the “Godzilla” dust plume’s evolution. Higher values in the FTLE field indicate locations from which particles will travel farther distance, which one can essentially compute by determining the difference between initial and final particle positions. This tool works well for real-time applications. 

Given this information, Jarvis presented a coherent structure analysis of the “Godzilla” dust storm. He particularly wanted to find the regions with high values in the FTLE field, as these regions majorly influence other nearby trajectories and could thereby cause significant changes to the plume’s shape. “How do we get the regions?” Jarvis said. “The simplest approach we can take is a thresholding method.”

Jarvis mapped the OMPS AI data and FTLE field together, starting at the time when the vortex-like structure first appeared. The evolution of the FTLE field over time reveals that a vortex to the northwest of the Sahara meets a ridge structure near the desert, leading to a billowing effect. The vortex and ridge initially appear to be attracted to each other and subsequently collide (see Figure 1). The vortex then unravels, unleashing its energy to the dust plume that then continues across the Atlantic and develops a tail (see Figure 2). The FTLE field includes a ridge that hugs the plume’s boundary at first, and then seems to move towards the middle as the plume splits. Future work will look more into the effect of this kind of structure.

Figure 2. Aerosol index (AI) data from the Ozone Mapping and Profiling Suite (OMPS), in brown, mapped along with the finite-time Lyapunov exponent (FTLE) structures, in blue, from June 19-22, 2020. The plume travels across the Atlantic and eventually splits. Figure courtesy of Albert Jarvis.

“FTLE did a pretty good job,” Jarvis said. “With more powerful computational capacities, this could potentially be used for real-time decision-making.” Moving forward, he aims to identify important intersection points based on the data, perform similar computations at different pressure surfaces, and proceed with more detailed analysis to identify elliptic and parabolic structures. Hopefully, these efforts could help understand similar events and mitigate the effects of future dust storms.

  Jillian Kunze is the associate editor of SIAM News