研究目的
To achieve accurate and consistent calibration for VIRR reflective solar bands (RSBs) onboard FY-3A, FY-3B, and FY-3C satellites using a multisite calibration method, addressing the lack of onboard calibrator and limited in situ calibration frequency.
研究成果
The MST calibration method effectively corrects sensor drift, providing consistent VIRR reflectances across FY-3 satellites. It reduces TOA reflectance differences from 5-10% to less than 3%, shows good agreement with automatic calibration, and has lower temporal oscillations. The method is suitable for daily calibration updates and improves long-term data consistency for climate applications.
研究不足
The method relies on satellite and climatology datasets for atmospheric parameters, which may introduce uncertainties. Calibration accuracy is affected by aerosol and water vapor uncertainties, especially in shorter wavelengths. The approach is specific to VIRR sensors and may not be directly applicable to other instruments without modifications.
1:Experimental Design and Method Selection:
The study uses a multisite (MST) calibration method, an extension of a previous method, relying on radiative transfer modeling over multiple stable Earth sites without synchronous in situ measurements. It involves selecting calibration sites, identifying cloud-free conditions, calculating calibration references using the 6SV radiative transfer model, and determining calibration coefficients via least squares fitting.
2:Sample Selection and Data Sources:
Data from VIRR sensors on FY-3A, FY-3B, and FY-3C satellites are used. Calibration sites include 16 stable Earth sites (e.g., Libya 4, White Sands, Uyuni Salt Flats) and ocean sites. Atmospheric parameters are sourced from MODIS global products and NCEP datasets.
3:List of Experimental Equipment and Materials:
VIRR sensors, MODIS data, 6SV radiative transfer model, cloud-screening algorithms, and automatic calibration instruments (e.g., ATR) in Dunhuang for validation.
4:Experimental Procedures and Operational Workflow:
Cloud-free ROIs are identified using spatial homogeneity and temporal deviation tests. TOA reflectance is calculated from DN using operational coefficients. Calibration references are simulated with 6SV. Calibration coefficients are derived by solving least squares problems over accumulation periods (e.g., 30 days). Validation involves comparing with L1B products and automatic calibration results.
5:Data Analysis Methods:
Linear regression is applied to normalized calibration slopes to estimate radiometric changes. Statistical measures like RMSE, CV, and relative bias are used for accuracy assessment.
独家科研数据包,助您复现前沿成果,加速创新突破
获取完整内容