From a morphological point of view, we attempted to estimate rice (Oryza sativa L.) yields by multiplication of four yield components considering seasonal changes in meteorological conditions. We focused on the rice cultivar Kirara397, for which large amounts of cultivation test data were available. The multiple linear regression method was applied for modeling between each yield component of Kirara397 and meteorological data. The compensation effect between yield components of rice was also considered. For other varieties (Nanatsuboshi as a panicle weight-type cultivar and Fukkurinko as a panicle number-type cultivar, relative to Kirara397) with less cultivation test data available, we formed the following two assumptions: 1) the type of weather elements affecting the yield components would be the same, regardless of the rice variety; and 2) the ratios of each yield component among rice varieties would be constant under the same weather conditions. Therefore, yield components for these varieties were estimated using correction ratios determined previously with yield component data. The determination coefficients for brown rice yield of Kirara397, Nanatsuboshi, and Fukkurinko were 0.440, 0.698, and 0.669, respectively. This method was thought to be a convenient and useful method for evaluation of yield and four yield components with only two meteorological data (daily air temperature and solar radiation) and dates of rice development. However, both our assumptions for rice cultivars with a small number of cultivation data were not fully adequate for estimating the yield and four yield components quantitatively, while it could be useful for determining relative quantities, particularly for panicle number, thousand kernel weight, and brown rice yield.

中文翻译：

---

按发育阶段划分的气象要素估算水稻产量构成

从形态学的角度来看，考虑到气象条件的季节性变化，我们试图通过将四个产量分量相乘来估计水稻 (Oryza sativa L.) 的产量。我们专注于水稻品种 Kirara397，该品种拥有大量栽培试验数据。Kirara397各产量分量与气象数据之间采用多元线性回归建模。还考虑了水稻产量成分之间的补偿效应。对于其他栽培试验数据较少的品种（七星为穗重型品种，福林科为穗数型品种，相对于Kirara397），我们形成以下两个假设：1）影响产量的天气因素类型无论水稻品种如何，成分都是相同的；2）在相同的天气条件下，水稻品种中各产量构成的比例保持不变。因此，这些品种的产量成分是使用先前用产量成分数据确定的校正比率来估计的。Kirara397、七星、福临子的糙米产量决定系数分别为0.440、0.698、0.669。该方法被认为是一种方便且有用的方法，可用于仅用两个气象数据（每日气温和太阳辐射）和水稻发育日期来评估产量和四个产量组成部分。然而，我们对少量栽培数据的水稻品种的假设不足以定量估计产量和四个产量组成部分，但它可能有助于确定相对数量，