Application of the self-calibrated palmer drought severity index and standardized precipitation index for estimation of drought impact on maize grain yield in Pannonian part of Croatia

Recent global warming and more frequent droughts are causing significant damage to maize production. A reliable estimate of drought intensity and duration is essential for testing maize hybrids to drought tolerance. For this purpose, the self-calibrating 10-day palmer drought severity index (scPDSI) and standardized precipitation index (SPI) for 1, 2, 3, 6, 9, 18, 27, and 36 10-day scales were used to estimate the effects of drought on grain yield of 32 maize hybrids evaluated in 2017 and 2018 at eight experimental locations in the Pannonian part of Croatia. Time series of observed 10-day mean air temperature, relative humidity, and precipitation totals for a set of “reference” weather stations of the croatian meteorological and hydrological service (DHMZ) for the period 1981–2018 were used to calculate the scPDSI and SPI indices. According to the 10-day scPDSI and SPI for different time scales, 2018 proved to be a “normal year,” while 2017 experienced a “mild to moderate drought,” which resulted in a 13% reduction in maize grain yield at eight experimental locations compared to 2018. The correlation between grain yield and drought indices for summer months was the highest for the 10-day scPDSI. To some extent, correlations between summer months’ SPI for the 3 10-day time scale and maize grain yield were comparable to the corresponding correlations for the 10-day scPDSI. However, for other SPI time scales considered, the corresponding correlations were weaker and less informative. The dependence of grain yield on scPDSI values was not the same for all hybrids, indicating their different tolerance to drought. The reduction in grain yield due to drought was primarily caused by insufficient grain filling (lower 1000-grain weight) and, to some extent, by a reduction in the number of grains. In this study, application of 10-day scPDSI data proved to be more relevant in detecting effects of drought on agronomic traits than application of SPI data for the most time scales.