This work aimed to quantify the sunflower crop under different soil and climate conditions of the Southeastern region of the State of Paraná, and to hierarchise the main factors related to the variability of crop yield. Twenty-six cropping conditions established during 2007 and 2008 cropping seasons were evaluated. Data regarding yield components and nutritional status were collected in the field, while crop water status was estimated with the use of a simulation model. The data was analyzed though the Principal Component and Cluster Analysis. Two open-pollinated varieties (BRS 122 and CATISSOL) and one hybrid (Helio 250) were used. Grain yields were grouped into three classes, with the following average values: high yields: 1995 kg ha-1; average yields: 1359 kg ha-1; low yields: 961 kg ha-1. Oil yields were grouped into three classes, with the following average values: high yields: 927 kg ha-1; average yields: 543 kg ha-1; low yields: 423 kg ha-1. The variability in oil and grain yields was correlated with the number of grains per head. In 2008, variability in grains per head was correlated with water deficit in R1-R4 and R4-R6 and to the N, Ca, B and P leaf content. A competitive inhibition between Mg and K was observed. However, this was not correlated with the variability in the number of grains per head. Despite the strong water deficit occurred during grain filling, no correlation was found between water deficit in R6-R9 and grain and oil yields.
agronomic diagnosis; multivariate analysis; biodiesel; oil seed crops; Helianthus annuus
