摘要
为探讨不同大豆品种(系)的耐荫特性,本研究筛选耐荫性鉴定指标,建立相对准确、稳定、高效的大豆耐荫性评价数学模型。以36个大豆品种(系)为试验材料,设遮光(玉米-大豆带状复合种植条件下形成的自然荫蔽条件)和自然光(清种)2个处理。将11个单项性状指标耐荫系数作为数据,运用主成分分析、隶属函数法、系统聚类和逐步回归分析方法对大豆耐荫性进行综合评价。结果表明:通过主成分分析,将11个单项性状指标耐荫系数转换为4个相互独立的综合指标,携带了74.722%的信息量;通过隶属函数法计算综合耐荫性评价值(D),并对其进行聚类分析,将36个大豆品种(系)划分为强耐荫型(2个)、中等耐荫型(9个)和弱耐荫型(25个)3大类;将11个单项性状指标耐荫系数作为自变量,D值作为因变量,采用逐步回归分析法建立最优回归方程:D=-0.585+0.662X2+0.152X3+0.084X5+0.211X8,R=0.962,调整的R^(2)=0.915,残差的标准差=0.036 16。筛选出底荚高、主茎节数、单株荚数和每荚粒数4个性状指标作为广西玉米-大豆带状复合种植条件下的大豆耐荫性鉴定指标。
This research aimed to explore the shade tolerance characteristics and identification index of soybean varieties(lines), and establish a relatively accurate, stable and efficient mathematical model for evaluating shade tolerance of soybean. There were two treatments(intercropping of soybean and maize;pure planting) with 36 soybean varieties(lines) as experimental materials. Eleven single character indexes were used as the data of shade tolerance coefficient, and the shade tolerance of soybean was comprehensively evaluated by using principal component analysis, membership function method, systematic clustering and stepwise regression analysis.The results showed as follows, the evaluation value of comprehensive shade tolerance was calculated by membership function method and analyzed by cluster analysis, 36 soybean varieties(lines) were divided into three categories: Strong shade tolerance(two varieties), medium shade tolerance(nine varieties) and weak shade tolerance(25 varieties/lines). The shade tolerance coefficient of eleven single trait indexes took as independent variable and D value as dependent variable, the optimal regression equation was established by stepwise regression analysis: D=-0.585+0.662x2+0.152x3+0.084x5+0.211x8(R=0.962,R^(2)=0.915,SD=0.036 16). The lowest of pod height, nodes number of main stem, pods number per plant and seeds number per pod were selected as the identification indexes of soybean shade tolerance under intercropping of corn and soybean strip compound planting in Guangxi.
作者
汤复跃
梁江
郭小红
韦清源
陈文杰
陈渊
TANG Fu-yue;LIANG Jiang;GOU Xiao-hong;WEI Qing-yuan;CHEN Wen-jie;CHEN Yuan(Cash Crops Research Institute,Guangxi Academy of Agricultural Sciences/Nanning Comprehensive Station of Nation Soybean Industry Technology,Nanning 530007,China)
出处
《大豆科学》
CAS
CSCD
北大核心
2022年第6期645-653,共9页
Soybean Science
基金
国家现代农业产业技术体系建设专项(CARS-04-CES30)
广西自然科学基金(2019GXNSFBA185007)
广西农业科学院稳定资助科研团队项目(2021YT055)
广西农业科学院基本业务费项目(桂农科2020YM118)。
关键词
大豆
耐荫性
主成分分析
隶属函数法
系统聚类分析
逐步回归
soybean
shade tolerance
principal components analysis
membership function method
system cluster analysis
stepwise regression analysis method