摘要
本文研究了播期、密度和氮肥对小麦白粉病及产量的定量效应,组建了病情指数及产量回归数学模型,并对其进行了检验和解析,结果表明:两回归模型均在0.01水平上显著,且拟合较好;三个因子对白粉病病指的影响为:播期>氮肥>密度,对产量的影响为:播期>密度>氮肥;播期与氮肥互作对病指的影响达到0.05显著水平,播期与密度互作对病指及产量的影响分别在0.1和0.05水平上显著;早播稀植低氮措施组合有利于控制小麦白粉病,减少产量损失。最后对回归模型进行了仿真优化,提出了控病增产的栽培措施决策方案。
The quantitative effects of sowing time (x_1), density (x_2) and nitrogenous fertilizer (x_3)on wheat powdery mildew and on wheat yidld were studied by applying 3 factors-5 levels generaland rotational regression design. According to the data obtained, mathematical models of diseaseindex (I) and wheat yield (Y) were built respectively through computation as follows: I=67. 98440+4.00769x_1-0.24826x_2+1.57484x_3-1.15000x_1x_2 -1. 6000x_1x_3+0. 80000x_2x_3-2. 69811x_1~2-0. 29431x_2~2-1.74366x_3~2 Y=199. 82800-36.37790x_1+0.06829x_2-0. 03585x_3+15.37500x_1x_2 +1. 67500x_1x_3+2. 82500x_2x_3+0. 96448x_1~2-7. 30742x_2~2-4. 19661x_3~2 (-1. 682≤x_1≤1. 682) Based on the results of F-test, the two regressive equations are all statistically significant atthe level of 0. 01 with good simulation result. The analysis of the models indicate that theimportance order of the effect of each factor on the disease index is x_1>x_3>x_2, and theimportance order on the yield is x_1>x_2>x_3. The interactive effects between sowing time andnitrogenous fertilizer (x_1x_3) on the disease index is significant at the level of 0.05; the interactiveeffects between sowing time and density (x_1x_2) on the disease index and on the yield aresignificant at the level of 0. 1 and 0. 05, respectively; the combinations of early sowing, lowdensity and low nitrogen are beneficial to the control of wheat powdery mildew and to thedecrease of the yield loss caused by the disease. Based on the simulation and optimization of themodels, a cultural scheme for controlling the disease and increasing wheat output wasproposed.
出处
《安徽农业大学学报》
CAS
CSCD
1993年第1期18-24,共7页
Journal of Anhui Agricultural University
基金
安徽农学院青年科学基金
关键词
小麦
白粉病
播期
蜜度
施肥
氮肥
Wheat powdery mildew
Time of sowing
Density
Nitrogenous fertilizer
Quantitative effect