Abstract
In order to determine the critical level and classify the soil Zn in Western Azerbaijan, Eastern Azerbaijan, Kurdistan and Kermanshah dryland areas, a study was conducted in a complete randomized block design having 4 treatments (0, 5, 10 and 15 kg.ha-1 of zinc sulfate) with three replications for four years (1998-2002).When the experiment ended, the crop and soil data uniformity test were performed for all experimental sites. The results of these experiments were interpreted by different methods including: Cate-Nelson graphical method, Cate-Nelson two and three classes ANOVA models, Mitshcherlich equation, plant response column order procedure and interaction chi-square methods. The results showed that the Zn critical levels using the mentioned methods were 0.75, 0.55, 0.65, 0.61, 0.80 and 0.66 mg.kg-1, respectively. Different Zn critical levels calculated by different soil testing interpretation methods were compared by using contingency table. The results showed that Cate-Nelson two classes ANOVA model with 0.55 mg.kg-1 Zn and 0.47 predictability value was a better model for determining the Zn critical level than all other models for Northwestern dryland region of Iran. Using different soil testing interpretation methods for determining the Zn critical levels it was concluded that all the values were to some extent similar; however, Cate-Nelson two classes ANOVA model seemed to be more suitable for this purpose.
Key words: Zn critical level, dryland wheat, Western Azerbaijan, Eastern Azerbaijan, Kurdistan, Kermanshah. |
