Scholarlink

<p>Accurate retrieval of Land Surface Temperature (LST) and Land Surface Emissivity (LSE) from thermal infrared (TIR) remote sensing data is challenging due to atmospheric attenuation and the coupled nature of temperature and emissivity. This study presents a novel Simplified Atmospheric Correction Temperature Emissivity Separation (SACTES) algorithm for the simultaneous retrieval of LST and LSE, specifically designed for ISRO’s forthcoming Geo Imaging Satellite (GISAT)-1A mission equipped with six Thermal InfraRed (TIR) bands. The proposed SACTES framework integrates a machine learning-based atmospheric correction module with a Modified Temperature Emissivity Separation (MTES) method. The atmospheric correction component uses a Machine Learning (ML) model trained on MODTRAN (MODerate resolution atmospheric TRANsmission) Radiative Transfer (RT) simulations, enabling accurate estimation of surface-leaving and downwelling radiances without relying on external atmospheric profiles. These outputs are then fed into the MTES module, which employs a convergence check after the Maximum-Minimum Difference (MMD) module to iteratively derive LST and LSE. Extensive validation using over 745,294 simulated cases across multiple band combination scenarios was conducted. The optimized 3-band configuration achieved the best performance with a bias of 0.19&#xa0;K and RMSE of 0.53&#xa0;K for the SACTES algorithm due to less atmospheric perturbation compared to other band combinations. Compared to conventional methods like Split Window TES (SWTES) and Split Window-driven TES (SWDTES), SACTES demonstrated an improved performance under varying water vapour concentrations, sensor geometry, and land cover types. These results highlight the potential of the SACTES algorithm for application in the next-generation GISAT-1A satellite mission.</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

Simplified Atmospheric Correction Temperature Emissivity Separation Algorithm: Advancing LST and LSE Retrieval from the Thermal Infrared Sensor Onboard ISRO’s GISAT‑1A satellite

  • 期刊论文

  • 发表日期：

    2026年5月27日

查看全文
  • Ashwin Gujrati,
  • Mehul R. Pandya,
  • Disha B. Kardani,
  • Jalpesh A. Dave,
  • Dhiraj B. Shah,
  • Himanshu J. Trivedi

摘要

Accurate retrieval of Land Surface Temperature (LST) and Land Surface Emissivity (LSE) from thermal infrared (TIR) remote sensing data is challenging due to atmospheric attenuation and the coupled nature of temperature and emissivity. This study presents a novel Simplified Atmospheric Correction Temperature Emissivity Separation (SACTES) algorithm for the simultaneous retrieval of LST and LSE, specifically designed for ISRO’s forthcoming Geo Imaging Satellite (GISAT)-1A mission equipped with six Thermal InfraRed (TIR) bands. The proposed SACTES framework integrates a machine learning-based atmospheric correction module with a Modified Temperature Emissivity Separation (MTES) method. The atmospheric correction component uses a Machine Learning (ML) model trained on MODTRAN (MODerate resolution atmospheric TRANsmission) Radiative Transfer (RT) simulations, enabling accurate estimation of surface-leaving and downwelling radiances without relying on external atmospheric profiles. These outputs are then fed into the MTES module, which employs a convergence check after the Maximum-Minimum Difference (MMD) module to iteratively derive LST and LSE. Extensive validation using over 745,294 simulated cases across multiple band combination scenarios was conducted. The optimized 3-band configuration achieved the best performance with a bias of 0.19 K and RMSE of 0.53 K for the SACTES algorithm due to less atmospheric perturbation compared to other band combinations. Compared to conventional methods like Split Window TES (SWTES) and Split Window-driven TES (SWDTES), SACTES demonstrated an improved performance under varying water vapour concentrations, sensor geometry, and land cover types. These results highlight the potential of the SACTES algorithm for application in the next-generation GISAT-1A satellite mission.