The SPI index was developed based on the fact that when drought occurs when reduced precipitation causes water shortage relative to the required water demand. However, the SPI index works under the assumption of stationarity or normality in which the probabilistic properties of the time series data do not change over time. It has limitations in reflecting information changing over time sufficiently, such as climate change.
In this study, we used SPIt, a new frequency analysis method considering non-stationarity, which is different from the existing SPI index calculation. We first used daily precipitation data to construct the time series data of 7-day precipitation. For the duration of drought, both SPI and SPIt indices for 3 months were calculated on a monthly and weekly basis, and the reproducibility of each index was assessed for the areas experiencing restricted water supply. In addition, both indices were calculated for all national meteorological stations in South Korea, which was used to create a spatial distribution map and confirm their spatial reproducibility.