Scholarlink

<p>Net Radiation (R<sub>net</sub>) combines the different radiative fluxes at the land–atmosphere interface and constitutes a key indicator for the study of surface energy balance. Its estimation is particularly challenging in desert regions, where in-situ measurements are scarce, making satellite observations indispensable. The Meteosat Second Generation (MSG) satellite provides multispectral images every 15 min, offering a comprehensive dataset for large-scale R<sub>net</sub> analysis. In this study, we propose a physical model to estimate global radiation, which integrates pixel-level cloud cover and the Linke turbidity factor to enhance accuracy under diverse atmospheric conditions. Consequently, a dedicated algorithm was developed to calculate cloud cover while accounting for atmospheric effects and sand transport. Pixels identified as clear-sky by this algorithm are subsequently used to generate clear-sky images, which serve to produce surface albedo maps using a parabolic model. Surface temperature is then estimated by combining these images with MSG infrared data at 10.8 µm and 12.0 µm and the Normalized Difference Vegetation Index (NDVI). The developed system, ShamSat, integrates all these steps and generates maps of global, atmospheric, surface, and net radiation, along with associated parameters (albedo, emissivity, NDVI, and surface temperature). Spatial interpolation using the inverse distance weighting (IDW) method ensures continuous coverage even in areas lacking direct data. Model validation was performed by comparing the calculated values with CERES observations (MODIS TERRA and AQUA) after resampling MODIS images to the same resolution. The Pearson correlation coefficients between the calculated and observed R<sub>net</sub> values are 0.95 for completely clear-sky days and 0.77 for cloudy days, demonstrating ShamSat’s reliability for estimating R<sub>net</sub> under varying atmospheric conditions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Contribution of Cloud Cover Images to the Quantification of the Different Components of Ground Radiation

  • 期刊论文

  • 发表日期：

    2026年3月2日

查看全文
  • M’hammed Boussir,
  • Nour EL Islam Bachari,
  • Abdelhak Razagui,
  • Aicha Belalia

摘要

Net Radiation (Rnet) combines the different radiative fluxes at the land–atmosphere interface and constitutes a key indicator for the study of surface energy balance. Its estimation is particularly challenging in desert regions, where in-situ measurements are scarce, making satellite observations indispensable. The Meteosat Second Generation (MSG) satellite provides multispectral images every 15 min, offering a comprehensive dataset for large-scale Rnet analysis. In this study, we propose a physical model to estimate global radiation, which integrates pixel-level cloud cover and the Linke turbidity factor to enhance accuracy under diverse atmospheric conditions. Consequently, a dedicated algorithm was developed to calculate cloud cover while accounting for atmospheric effects and sand transport. Pixels identified as clear-sky by this algorithm are subsequently used to generate clear-sky images, which serve to produce surface albedo maps using a parabolic model. Surface temperature is then estimated by combining these images with MSG infrared data at 10.8 µm and 12.0 µm and the Normalized Difference Vegetation Index (NDVI). The developed system, ShamSat, integrates all these steps and generates maps of global, atmospheric, surface, and net radiation, along with associated parameters (albedo, emissivity, NDVI, and surface temperature). Spatial interpolation using the inverse distance weighting (IDW) method ensures continuous coverage even in areas lacking direct data. Model validation was performed by comparing the calculated values with CERES observations (MODIS TERRA and AQUA) after resampling MODIS images to the same resolution. The Pearson correlation coefficients between the calculated and observed Rnet values are 0.95 for completely clear-sky days and 0.77 for cloudy days, demonstrating ShamSat’s reliability for estimating Rnet under varying atmospheric conditions.