In order to detect and quantifying drought (its frequency, duration and intensity), Palmer Drought Severity Index and Palmer Moisture Anomaly Z-Index were used in this paper. These indices were calculated by using monthly values of meteorological elements from meteorological station at Rimski Šančevi, Novi Sad, Serbia; for period 1965/66 to 2009/10. Also, yields of winter wheat from stationary long-term field trial, containing 20 different fertilization treatments with increasing doses of nitrogen, phosphorus and potassium were analyzed for the same region and period of time. For determining the most suitable index for quantifying effects of drought on wheat yield, correlation coefficients between two examined drought indexes and detrended wheat yields in long-term trial and Vojvodina province were calculated. Larger number of significant correlations was obtained between the wheat yields and Palmer's Z-index. On the basis of calculated r coefficients, it can be concluded that humidity/ drought conditions in October, December, March and April had the most important impact on grain yield. However, correlation coefficients in X and IV were positive, i.e. lower moisture conditions (drought) reduces yields, while in XII and III negative r coefficients indicated increase in yields at low moisture conditions. By observing the r values at different fertilizing variants in long-term experiment, it could be concluded that drought conditions in X and IV influenced all treatments more or less equally. Treatments with higher nitrogen doses were more sensitive to moisture conditions in XII, while in moisture conditions in III more sensitive were treatments without, or with small doses of nitrogen applied.

drought, drought indices, winter wheat, yield, mineral nutrition, correlations

Presentation: oral