Shortly after the COVID pandemic hit, inspiration struck Ryan Abernathey and Joe Hamman. The pair had been watching the “mass mobilization” of the epidemiology research community, Abernathey said.
“Joe and I were both thinking at the time, ‘We wanted to be part of something with that level of intensity and urgency around climate change.’”
Abernathey and Hamman met while working on open-source projects, including Pangeo and Xarray, both of which gave them a taste of where the field was heading. Abernathey, who is an associate professor of Earth and environmental sciences at Columbia University, said he saw his lab’s work on tools having a greater impact than the results of their research projects.
Hamman, who previously worked at the National Center for Atmospheric Research, also foresaw how data tools could begin to change the field. The business world was benefiting from a slew of new tools that worked well for their data types. “But none of those really exist for scientific data,” he said.
The pair eventually connected with Tony Liu, a partner at Costanoa Ventures who specializes in data infrastructure.
“We’ve seen the transformation of the business analytics world over the last few years,” Liu said, “where you have this ecosystem of cloud-native tooling that’s emerged that really reduces the complexity for someone less specialized to set up data infrastructure in their company. We believe that a similar pattern will emerge here.
“We’re seeing use of climate data increase, even among our portfolio — there’s several companies that are making use of climate data at scale. And also we expect a continued, massive investment into climate tech companies,” he added.
The scale of the data, and of the problem, is why Abernathey and Hamman’s new company, Earthmover, raised a $1.7 million pre-seed round from Liu and Costanoa, TechCrunch learned exclusively.
Currently, many teams working with climate or environmental data build their data stack from scratch, cobbling together a system from various open-source libraries and databases.
“About six years ago, we were starting to see datasets that we wanted to analyze, the size being measured in petabytes,” said Abernathey, who is CEO of Earthmover. “At that point, I had this realization: ‘Oh, crap. Our tools are really not up to the task of trying to understand this volume of data.’”
With Earthmover, Abernathey and Hamman, who is CTO, are hoping to build a “quasi-universal” data model that’s based on multidimensional arrays.
Multidimensional arrays are commonplace in climate and environmental data. They might be composed of stacks of satellite images, each representing a different point in time. Or they might be the results of climate models, with each point on Earth containing information for variables like temperature, humidity, wind speed, aerosols (as seen above) and more.
Earthmover’s first customers will likely be in the climate tech space, Abernathey said.
“There’s this big exchange happening now between the labs we’ve worked in and all these startups and even much bigger companies like McKinsey, for example, which is making a huge play in this climate risk analytics space.”
But Abernathey and Hamman think that their data model will be useful far beyond environmental science. “In the long term, we think there’s going to be a lot of other sectors that might want to use this type of data lake rather than a traditional tabular data lake in sectors like biotech, logistics, energy, materials, science and even finance,” Abernathey said.
Liu agreed, saying that as the data analytics world matures, sectors like those will be looking for other, more sophisticated tools.
“Over the past few years, there has been this focus on tabular data, tackling that use case first,” Liu said. “But I think this is very consistent with our excitement about more specialized end-to-end data platforms.”
Abernathey and Hamman said they would use the funding to start hiring a team to flesh out their designs.
“We think this is a fantastic opportunity for some engineers who want to jump ship from mainstream tech and work on the climate problem,” Abernathey said.
Comment