PREDICTION OF METEOROLOGICAL DROUGHT USING MACHINE LEARNING: A CASE STUDY IN PHOENIX, AZ

Drought is a complex phenomenon that impacts humans and the environment significantly. Predicting drought is one of the most challenging tasks for hydrologists. In this study we use machine learning algorithms and climate data to predict meteorological drought using two popular drought indices: Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) in the greater Phoenix area, Arizona. Random forest, a supervised learning algorithm was used for both regression and classification tasks.