Could a $99 Billion Hyperspectral Market Depend on a Chip?

A recent #spacenews analysis argued that the hyperspectral satellite industry is entering a new phase: moving from demonstrating what the technology can do to discovering what customers will actually pay for.

One forecast cited in the article suggests that the market could grow from $3.14 billion to $99 billion over the next decade.

It is an ambitious projection. But realizing it will require the industry to solve a fundamental problem:

Hyperspectral satellites can generate data much faster than they can economically transmit it.

Satellites can see more than they can send

A conventional optical satellite records a small number of color bands. A hyperspectral satellite may capture hundreds of narrow bands, producing a spectral signature for every pixel.

That makes it possible to identify materials, vegetation conditions, pollutants, and gas emissions that conventional imagery may miss.

But every additional band adds another layer of data.

As spatial resolution, revisit frequency, and spectral coverage improve, the amount of information generated in orbit can quickly outpace downlink capacity. The bottleneck is therefore no longer only about building better sensors.

It is also about deciding what information is worth transmitting.

A metasurface thinner than a human hair

Researchers at the Xi’an Institute of Optics and Precision Mechanics in China recently demonstrated continuous-wave upconversion imaging of a weak 4.33-micrometer mid-infrared signal using a thin-film lithium-niobate metasurface.

The principle is relatively simple: convert difficult-to-detect mid-infrared light into a wavelength that can be captured by more mature and potentially lower-cost detectors.

The metasurface is only about one-hundredth the diameter of a human hair in thickness, with tens of thousands of nanostructures integrated into an area roughly the size of a sesame seed.

The result remains a laboratory demonstration rather than a space-qualified device. But it points toward smaller, more integrated, and potentially less cooling-intensive mid-infrared detection systems. (opt.cas.cn)

That could matter for applications such as environmental monitoring and gas detection, where parts of the mid-infrared spectrum contain valuable molecular signatures.

The next step is to compute before transmitting

Detection is only the first layer.

Metasurfaces can be engineered not only to convert wavelengths but also to filter, separate, and manipulate light. Other research has demonstrated their potential for analog optical processing and operations such as edge detection. (advanced.onlinelibrary.wiley.com)

In parallel, photonic computing companies are commercializing chips that use light to accelerate specific computing workloads.

Lightelligence, a Chinese photonic computing company, began trading in Hong Kong on April 28, 2026. Its optical computing business remains small but reported strong growth in 2025—an indication that the technology is beginning to move from research toward commercialization. (www1.hkexnews.hk)

These are still separate technology paths. But together, they suggest a compelling future architecture:

• Metasurfaces select wavelengths and extract optical features.

• Onboard photonic processors identify useful patterns.

• Only high-value information is transmitted to Earth.

Instead of capturing everything and filtering it later, future satellites may increasingly sense, process, and prioritize data in orbit.

Where could the value accumulate?

If the hyperspectral market grows as predicted, value could be distributed across several layers:

• Satellite operators monetize data and analytics.

• Payload manufacturers sell advanced sensing systems.

• Computing providers enable onboard processing.

• Photonic component suppliers reduce the cost of the entire data chain.

The most strategically valuable layer may not be the satellite itself. It could be the combination of integrated spectral sensing and in-orbit processing.

However, the engineering gap remains significant.

A laboratory metasurface must still survive launch vibration, extreme thermal cycles, and radiation. It must also be manufactured consistently and integrated with complete optical, electronic, and computing systems.

That transition may take years.

Do you believe photonic computing will become a standard part of future Earth-observation satellites?

LU FRONTIER STRATEGIES

© 2026 LU Frontier Strategies. All rights reserved.