Comparability Analyses of Three Meteorological Drought Indices in Turkey

Abstract

The following chapter investigates the role of precipitation in the evaluation of meteorological drought in a mountainous region. For this, Mount Uludag in Turkey was taken as the case of study. Three meteorological stations with quite long precipitation records were used. Monthly precipitation time series between January 1980 and October 2018 at the Keles and Osmangazi stations in the northern and southern hillsides, together with the Uludag station near the summit were used in the analysis. Afterward, the patterns in the data run, frequency changes, and temporal events related to the time series were evaluated using precipitation anomaly, z-index, autocorrelation, mutual information, and power spectrum. It was concluded that there is a strong seasonality in the data at every 6 and 12 months, whereas the temporal persistence is quite low and decays after the second time lag. In the next stage, three drought indices, namely the Standardized Precipitation Index (SPI), Deciles Index (DI), and percent of normal (PN) were calculated at monthly, seasonal, and annual scales for each station. Finally, a model based on the spatial, temporal, and spatiotemporal properties of the precipitation time series was developed using the multivariate adaptive regression splines (MARS) model. It was concluded that the spatial scenario is the best predictive model in the assessment of precipitation and drought, and the SPI is the best one-parameter meteorological drought index for use in drought studies. © 2024 Taylor & Francis Group, LLC.