Quantifying the risk of heavy rainfall and its damage to agriculture in Iran

This study has been performed with two goals. Firstly, zoning the country based on risk of heavy rain damage to agricultural lands; secondly; developing a model to determine the probability of heavy rainfall damage risk (Prsk) over various regions of Iran considering their land use (e.g. rainfed or irrigated farming regions). Daily precipitation data of 405 meteorological stations across Iran with acceptable and continuous records were collected, and then the Probability Density Function (PDF) for these data was worked out using maximum likelihood method. The occurrence probability of a daily rainfall event greater than or equal to 50mm (Prsk) was also calculated based on these PDFs. The statistical analysis showed that, there exist a linear relationship (significant at 1%) between Prsk, as the dependent variable, and annual precipitation R, elevation Z and latitude Y as the predictors. The obtained regression equation; Prsk = a + b Z +c R +d Y was used for generating the regional risk maps. Besides, the digital precipitation model of Iran based on climatic data of 1100 stations for a 40-year period as well as the digital elevation model of Iran with a resolution of 1x1 Km2 were prepared for calculating Prsk. Then, by using GIS tools, an error layer was prepared to correct the calculated values of Prsk. Finally these risk values were classified as :andlt; 20= low, 20 andndash; 35= medium, 35-50= high and andgt; 50= very high risk and corresponding maps were produced. Ultimately, for practical applications, a new agro-climatic risk software namely, AgroClimRisk was introduced by which, the Digital Land-use Model and digital boundaries of administrative regions of the country were incorporated in developed Prsk model. This package is capable to calculate the maximum, minimum, mean and coefficient of variation values of Prsk and quantification of its risk of damage to farmlands over each part of the country using geo-statistical methods.andnbsp;