夏玉米主要生育阶段倒伏类型特征曲线的构建与验证

Construction and verification of characteristic curve of lodging type in key growth stages of summer maize

在夏玉米各主要生育阶段不同类型倒伏造成的产量损失差异很大,因此明确不同阶段倒伏类型发生规律是开展倒伏灾害研究的基础。本研究将夏玉米倒伏类型划分为“根倒伏”和“茎倒折”2类,利用2003—2019年鹤壁市品种区域试验中12个年份的典型灾情数据,初步构建了夏玉米主要生育阶段倒伏类型特征曲线,并根据2013年南阳市倒伏调查结果,分析抽雄期前后倒伏类型变化特征,对初步构建的倒伏类型特征曲线进行“抽雄前”和“抽雄后”的分段修订,建立了倒伏类型分段线性模型。为便于实际应用,利用分段线性模型模拟结果构建新数据序列,对其进行三次多项式拟合,建立主要生育阶段的综合曲线模型。结果表明,倒伏一般发生在播种后45~97 d,其中近一半的典型年份倒伏集中发生在播种后45~55 d。夏玉米平均抽雄期为播种后53 d,如果倒伏发生在抽雄前,以根倒伏为主,灾年根倒伏发生比例平均为85.1%,且随生育期推迟显著上升;如倒伏发生在抽雄后,根倒伏发生比例随生育期推迟显著下降,茎倒伏比例显著上升,而成熟期前后基本为茎倒折。利用历史典型倒伏案例和灾情图片信息,对所构建的分段线性模型和综合曲线模型进行验证,分段线性模型和综合曲线模型平均模拟误差分别为11.9%和11.1%。

The yield loss of maize caused by different types of lodging is very different.Therefore,it is the basis for carrying out studies on lodging disasters to clarify the types of lodging.In the study,lodging types were divided into two types:“root lodging”and“stem lodging”.The characteristic curve of lodging types during the whole growth period was initially constructed,using the data of 12 typical disaster years from the maize variety regional trials in Hebi city during 2003 to 2019.In addition,based on the results of the lodging survey data in Nanyang city in 2013,the characteristics of lodging types before and after the booting period were analyzed.The initially constructed lodging type characteristic curve was revised in sections of“before booting”and“after booting”,and a piecewise linear model of lodging type was established.In order to facilitate practical application,a new data sequence was constructed using simulation results of the piecewise linear model,and a cubic polynomial fitting was performed on it to establish a comprehensive curve model in main developmental stages.The result showed,lodging generally occurred 45 to 97 days after sowing,and the occurrence of lodging on 45 to 55 days after sowing was more concentrated,which was close to half of the total disaster year samples.The average tasseling period of summer maize was 53 days after sowing.If lodging occurred before tasseling,“root lodging”was the dominant type.The average proportion of root lodging in disaster years was 85.1%,and it increased significantly with the increase in the number of growth days.After tasseling,the proportion of root lodging decreased significantly with the postponement of growth period,and the proportion of stem lodging increased significantly.Stem lodging happened around the mature period.Using historical typical lodging cases and disaster picture information to verify the constructed piecewise linear and comprehensive curve models,the average simulation errors of the piecewise linear model and the comprehensive curve model were 11.9%and 11.1%,respectively.