Abstract Geostationary satellites offer the possibility of producing high-frequency time series irradiance datasets, covering a large geographical area. Here, we derive half-hourly global irradiance (GHI) at 4 km spatial resolution from visible channel radiance of INSAT-3D satellite using a semi-empirical model and validate it against 19 ground stations over 2015–16, across India. Our model uses the Advanced Solis clear-sky model with monthly climatological values of modern-era retrospective analysis for research and applications-2 (MERRA-2) aerosol optical depth (AOD) and precipitable water (PW) as inputs. The relative root mean squared error (rRMSE) and relative mean bias error (rMBE) are 24.02% and −0.05%, respectively. Over the validation period, INSAT-3D derived GHI outperformed SARAH-E model and MERRA-2 reanalysis GHI.

中文翻译：

---

验证来自 INSAT-3D 的全球辐照度在印度上空

摘要 地球静止卫星提供了产生覆盖大地理区域的高频时间序列辐照度数据集的可能性。在这里，我们使用半经验模型从 INSAT-3D 卫星的可见信道辐射中以 4 公里空间分辨率推导出半小时全球辐照度 (GHI)，并在 2015-16 年间针对印度各地的 19 个地面站对其进行验证。我们的模型使用 Advanced Solis 晴空模型，将现代时代回顾性分析的月气候值用于研究和应用-2 (MERRA-2) 气溶胶光学深度 (AOD) 和可降水量 (PW) 作为输入。相对均方根误差 (rRMSE) 和相对平均偏差误差 (rMBE) 分别为 24.02% 和 -0.05%。在验证期间，INSAT-3D 衍生的 GHI 表现优于 SARAH-E 模型和 MERRA-2 再分析 GHI。