Rice yield is an important factor for farmers when reviewing crop management planning and cultivation operations. In previous studies, a non-destructive, labor-saving method has been proposed to determine yield using spectral information from remote sensing. This method assumes that molecule number (e.g., chlorophyll content in leaves) and biophysical information (e.g., leaf area) cause spectral variations. However, spectral information cannot reflect the influence of physiological factors because of the limitations of this method. In this study, we investigated solar radiation use efficiency (RUE), which reportedly reflects physiological changes in crops. Time-series data of leaf area index were collected over five years in rice paddy fields of farmers to calculate and investigate the variation in RUE. The results showed that RUE varied among the survey points but did not differ significantly by year. The mean absolute error of yield estimation ranged from 68.2 g m⁻² to 102.2 g m⁻² depending on the yield estimation method and from 36.8 g m⁻² to 454.1 g m⁻² depending on the RUE. Among the estimation methods, the highest accuracy was achieved using leaf area dynamics. These results provide insights into both the effectiveness of leaf area dynamics in yield estimation and the potential errors when using remote sensing.

在审查作物管理计划和种植操作时，水稻产量是农民的一个重要因素。在以前的研究中，已经提出了一种非破坏性、省力的方法来使用来自遥感的光谱信息来确定产量。该方法假设分子数（例如，叶子中的叶绿素含量）和生物物理信息（例如，叶面积）会导致光谱变化。然而，由于该方法的局限性，光谱信息不能反映生理因素的影响。在这项研究中，我们调查了太阳辐射利用效率 (RUE)，据报道它反映了作物的生理变化。收集农民稻田5年的叶面积指数时间序列数据，计算和调查RUE的变化。结果表明，RUE 在调查点之间存在差异，但逐年差异不显着。产量估计的平均绝对误差范围为 68.2 g m^-2到 102.2 g m ^-2取决于产量估算方法，从 36.8 g m ^-2到 454.1 g m ^-2取决于 RUE。在估计方法中，使用叶面积动力学实现了最高精度。这些结果提供了深入了解叶面积动态在产量估计中的有效性以及使用遥感时的潜在错误。