The city of Dehloran with an area of 6229 square kilometers forms about one third of the total area of the province of Ilam, Five kilometers northeast of Dehloran, an area known as the natural and national works of Dehloran. The above-mentioned complex consists of three phenomena of the Sirgro Springs Fountain, the Bitumen Springs, and the Bat Cave. The tectonic forces affected by the subduction of the plate of Saudi Arabia below the Zagros area caused the formation of fissures and faults in the region, including the Black Sea Aegean. Also, according to previous studies, the process of fractures and normal faults and the slipping faults that are perpendicular to the axis of the Siah anticline, follow the Balarod fault and north-south fault in the study area. In order to investigate the hydrochemical properties and the relationship between the Sergro Dehloran hysthrotermal pring and adjacent oilfields in both dry and seagrass during 1994-1994, four samples were taken from the main ions and three samples were taken from heavy metals. I, Br was analysed by a spectrophotometric method and other heavy metals and semi-metals (such as: (AS, Cr, CO, Ni, V, B, Br, I, Li, Sr, Si) were analysed by atomic absorption methods of graphite furnace type and ICP OES and ICP MS. The integration of the above data with the hydrochemical data of time series 85-96 it was concluded that the hydrochemical facies of the Sargro Dehloran hydrothermal spring is chlorine - Saudik type. Studies and investigation on hydrochemical and heavy metals in the spring showed that the source of this spring is the mixing of atmospheric waters and deep waters and the water from the oil brine fields from the Dehloran oil field with the brine of the Gachsaran Formation. The conceptual model of Sergro Dehloran spring indicates that deep faults in the area are affected by the fault zone of Balarood and the North-South fault. These faults cause the atmospheric waters to blend with the oil brines and deep brines. The relatively high temperature of this spring is the result of deep water circulation along the joints, fractures and deep faults, and the geothermal gradient of the region. Interstitial water is mixed with the dissolution of the Gachsaran Formation along deep faults with oil brine and deep water, and ultimately increases the concentration of ions and the temperature of the spring.

 Prolonged droughts, climatic fluctuations, rapid population growth and increasing need for water resources have been caused a shortage of water sources in Iran. So in order to manage water resources should be close monitoring of these resources. Groundwater modeling and simulation techniques are as a way to monitor, control and management of groundwater resources. Precipitation is of the most important climatic parameters that play a crucial role in water balance and determine the type of climate in drought. The accurate determination of rainfall in mathematical models is one of the main criteria in the accuracy of mathematical models and this parameter has high uncertainty in the prediction models. To solve the problem using stochastic models with mathematical models to predict is very effective so the main objective of this study is combination of mathematical and stochastic models to better manage water resources in Maydavood_Dallan plain. Groundwater modeling of Maydavood_Dallan aquifer located in the northeast Ramhormoz city developed by Modflow code in the GMS software. Based on available data, Maydavood_Dallan aquifer model was began from September 2013 to September 2014 and calibrated model validated for 6 months from September 2014 to March 2015. In this study used the seasonal multiplicative model (Box Jenkins models) to predict 33 years rainfall data of Meydavood, Jokanak and Roodzard (Mashin) rain gauge stations. Based on the results, ARIMA (1,0,0) (0,1,1)12 model for Maydavood station, ARIMA (0,1,1) (0,1,2)12 for Mashin station and ARIMA (0,0,0) (0,1,2)12 Jokanak station were the best models that selected by stocastical parameters. The stochastic models verified by the mathematical model and the ARIMA (1,0,0) (0,1,1)12 model with the lowest RMSE selected as the final model to predict. The results shown, the combination of mathematical models and stochastic models cased improve the mathematic model results, therefor ARIMA (1,0,0) (0,1,1)12 used for 5 years pridication of water balance of Maydavood aquifer.