Simulating agricultural drought periods based on daily rainfall and crop water consumption

Even though drought is difficult to define precisely, rainfall is the most widely used indicator of drought. This paper presents a methodology on modeling of the agricultural drought duration. For this purpose, study area was divided into four hydrologic homogeneous sections as W, CN, CS and E. To constitute the monthly time series of each section, the number of days in each month in which daily rainfall was less than the water consumption of the critical crop was assumed to be drought periods. Then, constituted monthly time series of drought durations of each hydrologic homogeneous section was simulated using ARIMA model. No linear trend was observed for the time series except that of the CS section. In general, the predicted data from the selected best models for the time series of each section represents the actual data of that section.