Downscaling Thermal Infrared Radiance for Subpixel Land Surface Temperature Retrieval

Area of Science:

• Earth Science
• Remote Sensing
• Environmental Science

Background:

• Satellite-derived land surface temperature (LST) often contains mixed signals.
• Subpixel LST retrieval is crucial for detailed environmental and ecological research.

Purpose of the Study:

• To develop and validate methods for downscaling coarse-resolution thermal infrared (TIR) radiance.
• To enable accurate subpixel land surface temperature retrieval.

Main Methods:

• Developed a scale-invariant physical model for TIR radiance downscaling.
• Developed a statistical method relating TIR radiance to land cover fraction.
• Applied methods to downscale simulated 990-m ASTER TIR data to 90-m resolution.

Main Results:

• Both downscaling methods successfully captured general LST patterns.
• Methods resolved considerable spatial details at finer resolutions.
• Quantitative assessments showed strong agreement between true and estimated LST values.

Conclusions:

• The developed methods are effective for downscaling TIR radiance.
• Accurate subpixel LST retrieval is achievable using these techniques.
• The methods support enhanced environmental and ecological monitoring.