China’s new vision started in space has been created for 39 years

Chinese researchers reveal a powerful new tool for Earth observation: a 30-meter resolution Landsat composite data cube covered from 1985 to 2023.

Seamless dataset, in Remote Sensing Magazineproviding the first annual “Leaf” season satellite imagery for the entire China. By addressing the long-term problems of cloud coverage, sensor inconsistency and data gaps, the resource can transform land use research, vegetation monitoring and climate policy across the country.

Filling the gap in key data

For years, Chinese scientists have lacked countries that are equivalent to the analytical data (ARD) of the U.S. Geological Survey (ARD), forcing researchers themselves to struggle to process raw satellite data. This new dataset, led by Dr. Yaotong Cai and colleagues, ends the gap with ready-made composite materials tailored specifically to Chinese ecological and remote sensing needs.

“This dataset is a major breakthrough in environmental monitoring in China,” said Dr. CAI. “It not only simplifies satellite data processing, but also provides long-term resources on land use, climate change and biodiversity conservation.”

What makes this dataset different?

The data cube is constructed using Landsat 4, 5, 7, 8 and 9 images that use complex processing chains in Google Earth Engine. Key innovation is a “Medoid synthesis” technology that selects the most representative pixels for each location and year, avoiding outliers and maintaining spectral integrity. To fill in multi-cloud or missing areas, the team used segmented linear interpolation to generate proxy values that closely match the real conditions.

• Time span: 1985 – 2023 annual data
• Spatial resolution: 30 meters, nationwide
• Clouds and shadow masking: Evaluate quality using CFMASK
• Data gap filling: Linear interpolation with breakpoint detection
• Sensor Coordination: Adjust across multiple Landsat platforms

Tracking China’s ever-changing landscape

The preliminary application of the dataset shows its potential to analyze long-term land cover trends. In northwestern China, researchers use these composite materials to record urban growth, forest regeneration and agricultural transformation over nearly four decades. They also used machine learning models to map tree covers and aboveground biomass, thus capturing the effects of major afforestation activities.

Unlike older median-based composites, this data cube maintains consistent color and clarity even when switching between Landsat sensors. The result is a cleaner, easier to explain view of China’s terrain – for national-scale research and decision-making.

The way forward

Despite its advantages, the dataset is not without warnings. Some frequent cloud covers (such as southern China) are still challenging areas. And because the data covers only the leaf season (about June to October), events outside of the growing season may be missed. The authors also point out that spectral coordination methods derived from U.S. data may introduce small bias when applied in various regions of China.

Nevertheless, the team plans to continue improving the dataset by integrating Sentinel-2 images, enhancing cloud detection algorithms and eventually expanding to include leaf switches for the period. The open format and annual update plan make it a promising backbone of China’s environmental science and potentially surpass it.

Science, policy and the tools of the planet

With 39 years of continuous, coordinated satellite imagery, the new data cube provides researchers and decision makers with unprecedented insight into how China’s ecosystems change and how to cope with future stress. As climate change, urban expansion and conservation efforts continue to reshape the landscape, this dataset can help keep track.

“By providing open and accessible data, this resource will drive further research and innovation, thereby enabling more accurate and timely research to deepen our understanding of China’s dynamic environmental environment,” the authors wrote.

Diary and doi

Published on July 2, 2025 Remote Sensing Magazine

doi:10.34133/remotesensing.0698

Related