Using normalized difference vegetation index to estimate sesame drydown and seed yield

Sesame (<i>Sesamum indicum</i> L.) as an ancient crop has received wide attention in subtropical and temperate regions of the world. Yet there is limited data linking maturity with seed yield, which hampers sesame improvement. We used a multi-channel spectral sensor to measure canopy vegetation indices during the period of drydown to facilitate a rapid yield estimation for sesame genotypes under field conditions. We hypothesized that (a) a high normalized difference vegetation index (NDVI) at initial drydown would indicate a high yield; and (b) a rapid drydown rate would lead to a reduced seed yield. The results for 60 sesame genotypes confirmed the first hypothesis. Analysis of the NDVI data for the entire drydown period indicated that sesame genotypes with a higher initial NDVI showed an apparent trend of drying down faster and vice versa. This contradicts the second hypothesis and likely resulted from treating the full drydown as a linear process, which masked variations in NDVI in a short timeframe during initial drydown that could significantly impact the seed yield. However, the variations in NDVI during the initial nine days of drydown had a significant relationship with the measured seed yields, which was compatible with the prediction of the second hypothesis. To capture detailed changes in canopy features during sesame drydown, the value of making more frequent measurements of the vegetation indices from a wider time window was considered. Our study demonstrated that vegetation indices derived from a ground-based sensing tool were useful for characterizing the drydown process of sesame.

芝麻（Sesamum indicum L.）作为一种古老作物，已在全球亚热带与温带地区受到广泛关注。然而目前鲜有将生育期与种子产量关联的研究数据，这制约了芝麻的遗传改良。本研究采用多通道光谱传感器，在芝麻枯熟脱水阶段测量冠层植被指数，以快速估算田间条件下不同芝麻基因型的产量。我们提出两项假说：其一，枯熟脱水初期的归一化差异植被指数（normalized difference vegetation index, NDVI）较高，预示着种子产量更高；其二，枯熟脱水速率越快，种子产量越低。针对60份芝麻基因型的试验结果验证了第一项假说。对整个枯熟脱水周期的NDVI数据进行分析后发现，初始NDVI较高的芝麻基因型，其枯熟脱水速率看似更快，反之亦然——这一结果与第二项假说相悖，其原因可能在于我们将完整的枯熟脱水过程视作线性过程，从而掩盖了枯熟初期短时间内NDVI的波动，而这类波动本可对种子产量产生显著影响。不过，枯熟脱水初始9天内的NDVI波动与实测种子产量存在显著关联，这一结果契合第二项假说的预测。为精准捕捉芝麻枯熟脱水过程中冠层特征的细微变化，本研究探讨了在更宽的时间窗口内更频繁地测量植被指数的价值。本研究证实，基于地面传感工具获取的植被指数，可有效表征芝麻的枯熟脱水过程。