East China Normal University in Shanghai unveiled its first hyperspectral satellite on Sunday, alongside a plan to build a satellite constellation that is expected to support a wide range of sectors, including smart agriculture, ecological conservation and marine research.
The development of the hyperspectral satellite, named ECNU-KEDA, began in 2023 and was led by the university's Shanghai Key Laboratory of Multidimensional Information Processing.
The satellite features world-leading advances in wide spectral coverage, high spectral resolution and high spatial resolution, enabling it to efficiently and accurately identify the composition of objects on Earth.
Experts explained that conventional remote sensing satellites are like "digital cameras" suspended in space. Relying primarily on a few spectral bands, they can capture the shape and color of terrestrial objects, much like taking color photographs of Earth's surface to determine what is located where.
Hyperspectral satellites, however, can go much further. Every substance in the world — and even the same substance in different states — has unique spectral characteristics, much like a human fingerprint.
A hyperspectral satellite is more like a precision "composition analyzer" installed in space, as it can capture in fine detail the unique "spectral fingerprints" of different substances, thereby accurately determining their physical and chemical composition and health status.
Healthy rice, water-stressed rice and pest-infested rice all have different spectral characteristics. Therefore, hyperspectral satellites can identify and assess whether a field lacks fertilizer, whether a lake is polluted or if a forest is thriving, experts said.
Balancing spatial resolution, spectral resolution and spectral coverage has long been a bottleneck in the hyperspectral remote sensing sector. The new satellite has made significant breakthroughs in all three areas.
Compared with existing hyperspectral satellites in orbit, it can achieve meter-level spatial resolution in Earth observation while covering the visible and near-infrared spectral ranges with more than 200 spectral bands.
These advances give the satellite extremely high sensitivity to moisture content, organic matter and biochemical indicators of crops, enabling it to precisely identify minute ground features that were previously obscured or conflated because of lower resolution.
The satellite is expected to support a diverse range of industries in the coming years.
In agriculture, it can dynamically monitor crop growth and soil moisture, provide early warnings of pests and diseases, estimate yields and more, serving as an important technical tool for precision agriculture.
It can also provide large-scale, high-frequency and quantitative monitoring of ecosystem types, quality and functions, offering strong support for ecological conservation. In the maritime sector, it can monitor nearshore waters, red and green tides, coastal and island ecosystems, fishery resources and other areas.
He Peixiong of Heilongjiang Agriculture Company Limited, which has collaborated with the university for about two years, said he looked forward to the satellite's launch. He said it could help monitor crop growth and nutrient status over large areas, enable more precise pesticide application, improve disaster prevention and loss reduction, and optimize the overall management of high-standard farmland.
"We are managing more than 10 million mu (666,667 hectares) of cropland, which is extremely challenging for manual management. With the satellite, we will be able to better understand the conditions of the farmland and achieve more predictable production with better quality," said He at the unveiling of the satellite.
The satellite will also serve as the first satellite in the Liangtian satellite constellation, which was announced by the university on the same day.
A total of 200 low-Earth-orbit hyperspectral satellites are planned for the constellation over the next five years. Through a multi-orbital-plane network and coordination among multiple satellites, the constellation will form a global dynamic integrated observation network capable of all-weather monitoring with high spatiotemporal and spectral resolution.
Xiong Hongkai, head of the university's School of Information and Electronic Engineering, said future satellites in the constellation will be equipped with domestically produced onboard intelligent computing chips and deploy the team's proprietary deep-learning and hyperspectral large-scale models.
The move will enable real-time in-orbit processing, onboard intelligent recognition and direct delivery of information to users on the ground, further improving the efficiency of monitoring and alert delivery.
wangxin2@chinadaily.com.cn
