Еstimation of the effect of sowing time on the productivity of winter wheat by the method of multivariate analysis of variance

The choice of sowing dates for winter wheat in the conditions of aridization of the climate of the Orel region is very important. This is due to an increase in the duration of the warm autumn period by 2-3 weeks with an average daily temperature of 10-12 0C and often with a lack of moisture in the soil, in comparison with the previously recommended sowing dates for winter wheat. Sowing winter wheat at a later date increases the duration of the field preparation period, which leads to an increase in yield and has a positive effect on improving the quality of grain. Sowing at a later date of wheat of the Moskovskaya 39 variety provided an increase in nature, the accumulation of protein and gluten in the grain by 8-24 g / l, 0.7-1.3%, respectively, and an increase in yield by 0.14-0.33 t/ha, in comparison with the recommended period. From the analysis of the composition of homogeneous subgroups, it follows that the variants of shifting the sowing dates by 10, 20 and 40 days in terms of gluten content form independent subgroups, while the gluten content in the experiment with a shift of 10 days is higher - 25.0%, respectively, against 23.7% in the control, and in the experiment with a shift of 30 and 40 days - less than the control (22.1% and 19.1%, respectively). At the same time, the difference in the gluten content when the sowing period is shifted by 20 days from the control is not statistically significant. It was also revealed that the variants of shifting the sowing time by 20 days in nature form the only independent subgroup, and the nature in this experiment is higher than the control - 805 g/l versus 780 g/l in the control, and when the sowing time is shifted by 10 days, the nature does not statistically differ from the control. Another subgroup combines experiments with a shift in the sowing time by 30 and 40 days, but in it the nature is only slightly higher than the control. The option of shifting the sowing time by 10 days after sedimentation forms a common subgroup 3 with the control, and the indicator of 47 cm cubic meters in this experiment does not statistically differ from the control level of 45 cm cubic meters. Subgroup 2 combines experiments with a shift in the sowing time by 20 and 30 days, in which the sedimentation is slightly lower than the control - 39 cm cu. and 37 cm cu. accordingly. Subgroup 1 with a 40-day shift in sowing dates is characterized by a minimum sedimentation value of 31 cm cu. According to the results of the correlation analysis, the yield is positively and strongly correlated with the quality indicators of wheat grain. According to the results of the performed statistical studies, the procedure of a multidimensional generalized linear model is recommended for modeling the influence of agricultural conditions on the yield and quality indicators of wheat grain. Multiple comparison of averages, based on the methods of variance analysis, increases the reliability of statistical conclusions, allows us to identify homogeneous variants of experiments in agricultural production. It is statistically justified that the maximum yield and quality of wheat grain is provided by the sowing period 10 days later than previously recommended for the region.