Radiometric Calibration

Radiometric calibration is the process of converting the raw digital numbers (DNs) output by a remote sensingRemote SensingRemote sensing is the science of collecting data about Earth's surface without direct physical contact, primarily usi... instrument into calibrated physical quantities such as spectral radiance (W/m²/sr/µm) or top-of-atmosphere reflectance. Every imaging sensor has a unique response function, and without calibration, pixel values are arbitrary numbers that cannot be compared between scenes, dates, or sensors. Calibration methodsPre-launch calibration characterizes sensor response in a laboratory using known light sources and targets. On-orbit calibration uses onboard reference lamps, solar diffusers, or observations of stable ground targets like desert playas and deep ocean to track sensor degradation over time. Cross-calibration transfers the radiometric scale from a well-characterized reference sensor to another using simultaneous observations of the same target. Why calibration mattersCalibrated data is the foundation for all quantitative remote sensing. Vegetation indices, atmospheric correctionAtmospheric CorrectionAtmospheric correction removes the effects of atmospheric scattering and absorption from satellite and aerial imagery... algorithms, and change detectionChange DetectionChange detection uses geospatial data and imagery to track and analyze alterations in landscapes, infrastructure, or ... analyses all depend on consistent radiometric values. Programs like Landsat and Sentinel maintain rigorous calibration protocols so that data spanning decades can be reliably compared for long-term environmental monitoringEnvironmental MonitoringEnvironmental Monitoring is the systematic collection and analysis of data about environmental conditions, including .... Radiometric calibration ensures that remote sensing data retains its scientific integrity across sensors and time. As the number of Earth observation satellites grows, standardized calibration practices become increasingly vital for interoperable, analysis-ready data products.