%0 Journal Article %T Drought monitoring with Standard Precipitation Index (SPI) and drought forecasting with Multi-layers perceptron (Case study: Tehran and Alborz Provinces) %J Iranian journal of Ecohydrology %I Faculty of New Sciences and Technologies, University of Tehran %Z 2423-6098 %A Jahangir, Mohammad Hossein %A Khoshmashraban, Meimaneh %A Yousefi, Hossein %D 2015 %\ 12/22/2015 %V 2 %N 4 %P 417-428 %! Drought monitoring with Standard Precipitation Index (SPI) and drought forecasting with Multi-layers perceptron (Case study: Tehran and Alborz Provinces) %K Standard Precipitation Index %K Multi-layers perceptron %K Drought Forecasting %K zoning %K Tehran Provin %R 10.22059/ije.2015.58068 %X Drought is the one of the repeating phenomenon in all areas with high rainfall and low rainfall climates and is known as a natural disaster. Iran is one of the countries those involved with this phenomenon in different parts of it especially in river basins. Tehran Province due to its importance in terms of social and political faced with growing population that it would contributes to the reduction of water sources in the province. Alborz province which in the past was considered one of the cities of Tehran, in this study were investigated. Losses incurred from drought in this area is socio-economic. In this study, we monitor and forecast drought, with rainfall data from 38 synoptic stations in Tehran and Alborz provinces. By Standard Precipitation Index (SPI) during 31 hydrological years between 1983-84 to 2013-14 at 3, 6, 9, 12 and 24 months average times. Study on SPI12 index showed that about half of the stations entire the study period were normal and about a third of the stations in this period had drought conditions. According to the numerical values SPI index 1996-97 and 1998-99 were selected as years those have been faced with drought. Severe and very severe periods of drought, the most severe drought level (lowest SPI) was calculated for some of the stations in periods of 3, 6, 9, 12 and 24 months. Also forecasted with Multi-layers perceptron neural network method and the results was very close to the observed data.  ce.     %U https://ije.ut.ac.ir/article_58068_71ac8a964a2c651b95329a5c0d366d94.pdf