Steps to mitigate climate change are written in the stars and for Los Angeles and Bangalore-based startup Pixxel, the stars are exactly where it's going to deliver the insights necessary to take those steps.
Not only is Pixxel intercontinental, but they are space-bound as the company recently announced a $36 million Series B funding round to help advance its mission: to build the world’s first and highest-resolution hyperspectral satellite constellation.
But before we go any further, what in the world is a high-res hyperspectral satellite constellation?
A satellite constellation is a network of units that together, work like a hive mind for the same purpose, under the same control. Such a swarm can provide global or near-global coverage and is used for a variety of purposes from weather analytics and data for agriculture to those used by SpaceX for internet access to remote locations.
Pixxel, however, adds another layer to this tech, by using hyperspectral imaging, a technique used for applications like food safety inspection and recycling that analyzes each pixel of an image.
By combining the technologies and launching the satellites into space, Pixxel will gather data integral to helping global climate and environment organizations, whether they need to closely monitor emissions, water pollution, and soil composition, or gas leaks, oil spills, and industrial runoff or even forest biodiversity, drought-stricken regions, and crop health.
The startup has worked with farming companies to one of the world's biggest mining companies Rio Tinto to use satellite imagery for critical mining — necessary for electric vehicles — and find resources and make decisions with sustainability in mind.
The level of detail these satellites bring is not visible to ordinary satellites, nevertheless the human eye, or normal cameras.
According to the startup, its tech can gather these images and draw data in “unprecedented detail and at faster speeds,” than today’s “multispectral satellites.”
The difference between multispectral satellites and hyperspectral is that the latter can detect thousands of different bands within the light spectrum, whereas the former can only capture data along five to ten bands of the spectrum. To put that into perspective, the human eye can only see seven bands of the spectrum, which can be thought of as all of the colors in the rainbow, equating to only 0.0035% of the full light spectrum.
A hyperspectral satellite is like a super eye that can take images down to the mineral, for example, all from a low orbit in space. According to Pixxel, their satellites provide ten times the information of multispectrals and multiplies the resolution by fifty-fold.
With help from the new funding round — which saw participation from Google, Radical Ventures, Lightspeed, Blume Ventures, growX, Sparta, and Athera — the startup plans to launch 6 satellites in 2024 and another 18 by 2025, all watching from different angles to monitor planetary health. Together, they’ll make up its next-gen Firefly constellation, which, unlike previous satellites which have a lifespan of two years, these satellites live for seven.
In total, the round brought the company’s lifetime funding to $71 million.
“At Pixxel, we believe that the future of our planet lies in our ability to monitor and protect its health precisely. With this round of funding, we are even closer now to realizing our mission of building a health monitor for the planet, and empowering people around the world to make informed decisions about our collective well-being,” Awais Ahmed, co-founder and CEO of Pixxel, said in a statement.
On top of launching the Fireflies, Pixxel is focusing on developing its "Aurora" analytics platform, an AI-powered software hyperspectral analysis accessible to everyone with a click of a button, giving customers access to data about their crops or notification where there is a gas leak.
“It’s important for us to not just dump data down to our customers and have them figure it out themselves,” Ahmed said via TechCrunch, noting Aurora’s ability to break down the models. “There’s very few people in the world with the skill set to actually analyze hyperspectral data so we realized to actually open it up to a lot more customers than would be possible without it, we will build and put the Aurora platform out.”
Next, some of the fresh funding could even go to Pixxel’s bigger, better, and higher-res future generation of satellites: the Honeybees.
The round comes just after the startup signed a five-year contract with the National Reconnaissance Office (NRO), a member of the United States Intelligence Community and an agency of the United States Department of Defense, to enhance the office’s geospatial intelligence.
This is Pixxel’s first government customer, and as the startup puts it, the agreement aligns with growing public sector and federal interest in climate monitoring tools, pointing to a 2022 initiative by the Department of Energy to use sensing tools like satellites and aircraft to better estimate the volume of methane emissions. The $32 million program funded companies like Sunnyville, California-based Kairos Aerospace, Petaluma, California-based Sonoma Technology, and Richland, Washington’s Pacific Northwest National Laboratory all to monitor and detect super-emitters from space.
In the future, Pixxel dreams that hyperspectral data be available and accessible to everyone, and with its big tech investors, that dream may be possible. Last year, Google launched its Earth Engine, a planetary-scale platform that combines satellite imagery, geospatial datasets, and Earth analyses to detect and quantify changes on the globe’s surface.
The engine is available for commercial use but remains free for academic and research use, and has real-world applications across climatic research, from landuse change to forest and water coverage, as one case study that studied over a decade of global tree cover extent, loss, and gain shows.
The leader of the team, Matt Hansen said the engine enabled research that would’ve taken 15 years to be done in a matter of days, and Ahmed hopes this kind of data is available to the everyday person to better understand the world around them.