Monthly Meteorological Drought Prediction in Iran Using Large-Scale Climatic Indices and Nonlinear Autoregressive Exogenous

Drought, a major global environmental and economic challenge, significantly impacts water resources, agriculture, food production, and socio-economic conditions. Iran, with its predominantly arid and semi-arid climate, faces high risks of severe droughts. This study utilizes large-scale ocean-atmosphere climate indices, including the El Niño/Southern Oscillation (ENSO), Indian Ocean Dipole (DMI), North Atlantic Oscillation (NAO), Pacific Decadal Oscillation (PDO), Atlantic Multidecadal Oscillation (AMO), Arctic Oscillation (AO), and Southern Annular Mode (SAM), over the period from January ۱۹۵۷ to December ۲۰۲۰, which are expected to remotely influence Iran’s climate. Employing the nonlinear autoregressive exogenous (NARX) technique, this research aims to model and forecast monthly drought conditions across ۱۸ regions in Iran. The study combines information from these climate indices to enhance drought forecasting capabilities, using them as predictor variables in a machine learning model to simultaneously simulate and predict monthly Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) drought metrics. Results demonstrate the NARX model's effectiveness in predicting monthly SPI with mean squared errors (MSE) ranging from ۰.۲۶ to ۰.۰۲, and SPEI with MSE ranging from ۰.۷۷ to ۰.۰۱ across the studied regions. This approach, utilizing a comprehensive suite of indices with the NARX model, offers new insights into the influence and relative importance of ocean-atmosphere oscillations on Iran's drought patterns, potentially improving drought forecasting and management strategies in the region.

کلیدواژه ها:

IRAN DROUGHT PREDICTION ، NARX NEURAL NETWORK ، SPI ، SPEI ، OCEAN-ATMOSPHERE CIRCULATIONS