Accurate estimation of ET is vital for water resource management. In recent decades, researchers have focused on utilizing satellite imagery for this purpose. The use of RS data has enabled the development of new models that provide detailed spatial assessments. GeeSEBAL, an automated ET estimation tool, employs the SEBAL algorithm via GEE. The current version of GeeSEBAL utilizes Landsat images and ERA5 global reanalysis data to produce time series estimates. Landsat 8 images were processed into a 16-day time series spanning 2013–2022, specifically during the wheat growing season. To validate the GeeSEBAL model for 2013–2014, results were compared against lysimeter data. Subsequently, ET was calculated for the years 2015–2022. The evaluation of GeeSEBAL against lysimetric data, by metrics such as R², RMSE, MAE, NSE, and NRMSE, yielded values of 0.94, 0.98, 0.07, 0.86, and 0.62, respectively. Those findings underscore the importance of GeeSEBAL for estimating wheat ET in regions with limited data availability.

准确估算ET对于水资源管理至关重要。近几十年来，研究人员一直致力于利用卫星图像来实现这一目的。遥感数据的使用促进了提供详细空间评估的新模型的开发。 GeeSEBAL 是一种自动 ET 估算工具，通过 GEE 采用 SEBAL 算法。当前版本的 GeeSEBAL 利用 Landsat 图像和 ERA5 全球再分析数据来生成时间序列估计。 Landsat 8 图像被处理为 2013 年至 2022 年期间的 16 天时间序列，特别是在小麦生长季节。为了验证 2013-2014 年的 GeeSEBAL 模型，将结果与蒸渗计数据进行了比较。随后，计算了 2015-2022 年的 ET。通过R ² 、RMSE、MAE、NSE 和 NRMSE 等指标对 GeeSEBAL 对照测渗数据进行评估，得出的值分别为 0.94、0.98、0.07、0.86 和 0.62。这些发现强调了 GeeSEBAL 在数据可用性有限的地区估算小麦蒸散量的重要性。