Drought roots in natural climate variability over mid- to long-term time periods and often causes serious water and environmental consequences for human societies and ecosystems. Due to the random nature of contributing factors, occurrence and severity of droughts may be treated by stochastic techniques. Early indication of droughts can help to activate the corresponding mitigation strategies and measures in advance. Therefore, drought forecasting comes to the aid of decision makers for planning and management of water resource systems. In arid and semi-arid regions, drought negative effects on crops may be reduced by storing sufficient water from available sources. As a result, it is quite important to forecast the drought duration in the growing season in terms of irrigation scheduling. In this study, duration of agriculture droughts, quantified by the Palmer Drought Severity Index (PDSI), is forecasted through the application of linear stochastic models for Mashhad, Khorasan-Razavi province, northwest of Iran. Study of the temporal characteristics of droughts indicated that the region is affected by severe and, more or less, prolonged droughts from 1971 to 2004. Results showed that over 64% of the studied period, droughts prevailed that increased in duration and severity in recent years. A SARIMA (Seasonal Autoregressive Integrated Moving Average) model was developed to predict drought duration and was found to provide acceptable results up to 2 months ahead.

Keywords