The Geoscience Data Exchange (GDEX) is NSF NCAR’s platform for sharing and stewarding Earth system science (ESS) data and model output. It unifies datasets into a reliable environment, enabling analytics at scale using high-performance computing resources. GDEX is built for AI-ready research, and designed to accelerate collaboration and discovery.
 

 

"I have really loved using GDEX!" - Watch Mya Sears' success story above.


For many ESS researchers, the "science" part of a project often seems secondary to the logistical hurdle of data acquisition. Between navigating external repositories and managing massive local downloads, the workflow can grind to a halt before real analysis even begins. 

Mya Sears

"Really, all you have to do is this..." Sears explains the process behind her simplified GDEX workflow.

In the premiere of the new video series, “GDEX Unlocked” (watch on YouTube or above), Mya Sears of the Earth Observing Laboratory (EOL) shares how transitioning to GDEX transformed her research from a week-long process of waiting into a streamlined, high-speed operation.

Before migrating to GDEX, Sears relied on pulling ERA5 data from external climate data stores—a process that was notoriously time-consuming.

“I was spending a couple of days just getting that dataset,” Sears explained. “Before my data was in GDEX, this workflow was really long. It involved a lot of notebooks and Python scripts that you don’t even see now because the data is already ready for use.”

By utilizing GDEX’s direct integration with the Casper cluster, Sears was able to subset the larger ERA5 dataset into her own specific Zarr store. The results were immediate: a process that previously took days of downloading and programmatic reading now runs almost instantly. 
 

Mya Sears

"It's a lot faster, and I have really loved using GDEX for ERA5 specifically." 

“Now running that code takes seconds—on the scale of less than a minute. It goes right to the science, which is really nice. There are not many hours spent in the preprocessing.”


The shift to GDEX didn't just save time; it simplified the technical overhead. Sears points to a workflow that now consists of just a few lines of code, with complex tasks like data alignment being handled easily via the interp function.

However, Sears is the first to admit that moving a workflow can be daunting. “Anytime your data takes a different format or is located in a different place, that is pretty scary and I get it,” she says. “There was a lot of lift for me to get my data here. A lot of really, really helpful people [in CISL] worked with me to get the data into GDEX.”
 

"CISL was really helpful in working with me every step of the way."


This GDEX success story testifies to the excellent results when data is freed from local lab servers and integrated into a high-performance environment. GDEX is currently building a migration roadmap—act now to secure your spot.

Watch Mya Sears’ full success story on YouTube, or press play on the embedded video above. Submit your dataset details today.