Modeling of environmental causes of hybrid by sowing date interaction in sunflower: a statistical approach

Abstract

The analysis of agronomic crop trials is usually complicated by the presence of hybrid by environment interaction, In this study the effect of selected environmental variables on hybrid by sowing date interaction in sunflower, using statistical approach, were investigated. For the statistical modeling of interaction six environmental variables (minimum, maximum and mean temperatures, total precipitation, sunshine hours and relative air humidity) were used as input information in factorial regression model (FR). The calculation of environmental variables for each sowing date, were made according to approximate sunflower growth stages (six leaves, budding, flowering, physiological maturity). Three hybrids Miro, Rimi and Pobednik were grown during three vegetation periods and sown at eight sowing dates. The effects of environmental variables on the hybrid by sowing date interaction for seed yield and yield components (oil yield, oil content, 1000 seed weight) were tested.