China's FengYun-3 Satellites Revolutionize Global Diurnal Land Surface Temperature Monitoring

A recent study published in the ISPRS Journal of Photogrammetry and Remote Sensing demonstrates how China's FengYun-3 (FY-3) meteorological satellites have significantly advanced global land surface temperature (LST) monitoring across diurnal cycles. Led by Prof. ZHAO Tianjie from the Aerospace Information Research Institute (AIR) of the Chinese Academy of Sciences (CAS), the research leverages the unique capabilities of FY-3's MicroWave Radiation Imagers (MWRI) to overcome limitations of traditional satellite-based LST retrieval methods.

LST—a critical parameter for weather forecasting, agricultural planning, ecosystem monitoring, and disaster management—has historically relied on thermal infrared (TIR) sensors. However, TIR measurements are hindered by cloud cover (affecting approximately 60% of land areas) and limited to capturing surface "skin" temperature. The FY-3 satellite constellation addresses these challenges through its advanced MWRI sensors, providing an all-weather solution and deeper thermal insights. By leveraging the satellite's multi-pass observation capability, researchers have achieved more frequent and consistent LST measurements throughout the day. 

"Our study confirms that MWRI sensors on FY-3 satellites offer unprecedented accuracy in LST monitoring," stated Prof. ZHAO. "Microwaves penetrate clouds, enabling continuous observation even in overcast regions. Unlike TIR's superficial skin temperature measurements, microwave data provides an integrated effective temperature reflecting combined soil and vegetation thermal states.”

The team developed a novel algorithm that distinguishes between frozen and thawed soil states, significantly improving temperature measurements in high-latitude and mountainous regions where phase changes in soil moisture drastically alter thermal properties and radiative transfer mechanisms. The algorithm incorporates multiple microwave indices to minimize errors from vegetation cover, snow conditions, and atmospheric water vapor interference.

According to the study, the FY-3 MWRI-derived LST data demonstrates high validation accuracy, showing a strong correlation of over 0.87 with ground-based measurements and an error margin of just 4 K. It also aligns well with widely used datasets such as MODIS and ERA5. 

Notably, FY-3 LST exhibits a narrower diurnal range and delayed peak temperatures compared to MODIS—a reflection of deeper soil thermal inertia that benefits hydrological and climate modelling. The FY-3 B/C/D satellite constellation reduces the average error in reconstructing 24-hour temperature cycles to 5.2 K, much lower than MODIS's four-time daily sampling, offering superior temporal resolution for improving climate models, tracking heat weaves and droughts, understanding climate-driven changes in eco-systems.

"FY-3's microwave sensors fill a critical gap in Earth observation," emphasized Prof. ZHAO. "By capturing the thermal inertia of deeper soil layers, operating day-night through clouds, and enabling multiple daily observations, these data are transformative for hydrological models and climate adaptation strategies."

FY-3 Satellite Constellation MWRI Multi-Pass Land Surface Temperature (LST). (Image by AIR)

Looking ahead, the researchers plan to refine their algorithm and extend its application to other Chinese satellite systems, further solidifying China's role in climate monitoring. This progress promises enhanced accuracy in weather forecasting and climate modeling, delivering global benefits across industries and communities worldwide.

This research, funded by China’s National Key R&D Program, has made all datasets publicly available via the National Tibetan Plateau Data Center.