含油量对提高菜油产量有十分重要的意义, 传统的高含油油菜新材料筛选方法工作量大, 耗时长。为找出一种快速诊断高含油油菜育种材料的新方法, 本研究以3组不同含油量的9个甘蓝型油菜为材料, 测定其不同生育期的高光谱反射率与叶绿素含量, 分析其内在关联。研究发现, 油菜高光谱反射率在可见光波段呈现出“两峰三谷”特性, 幼苗期含油量越高则高光谱反射率越低。叶绿素a含量是导致总叶绿素含量差异的主要原因。在490 nm和670 nm处, 幼苗期光谱反射率和叶绿素含量负相关, 相关系数分别为-0898和-0783；叶绿素和含油量显著正相关, 相关系数达到0962；490、560、670 nm反射率和含油量之间呈负相关, 相关系数分别为-0949、-0962和-0937。该结果表明, 油菜幼苗期光谱反射率可用于预测其含油量。本研究结果可用于高含油油菜育种材料早期筛选, 加快其育种进程, 促进新品种选育。但样本量偏少, 在后续研究中将进一步扩大样本量, 提高预测准确度。

The oil content of rapeseed is important for improving oil yield And the traditional method for new oilseed rape materials is complex and time-consuming In order to find a new method for rapid diagnosis of oilseed rape breeding materials, 3 groups of 9 different oil content Brassica napus lines were uesed as raw materials, meanwhile, hyperspectral reflectance and chlorophyll were tested in different stage and the relationship between them was analyzed The results showed that the hyper spectral reflectance of rape displays two peaks and three valleys in the visible band Meanwhile, in rapeseed seedling stage, the higher oil content rapeseed hed a lower hyperspectral reflectanceThe chlorophyll a was the main reason which led to the difference of total chlorophyll content There was a negative correlation at 490 nm and 670 nm between the spectral reflectance and chlorophyll content, the correlation coefficient was -0898 and -0783 respectively There was a significant positive correlation between chlorophyll and oil content, which reached to 0962There was a negative correlation between reflectivity and oil content at 490 nm, 560 nm and 670 nm, and the correlation coefficient was -0949, -0962 and -0937 respectively At last, the oil content of rapeseed can be predicted by Hyper spectral method in seedling stage and provide reference for rape breeding with high oil content These results can be used to choose high oil content rapeseed materials in early stages, which can enhance the breeding process and promote new materities breeding While, the samples in this study is small, and samples will be enlarged to improve the prediction accuracy of oil content by Hyper spectral method