---

Drought is one of the most destructive natural disasters in human societies that cause irreparable impacts on agriculture, environment, society and economics. So, awareness of occurrence of droughts can be effective in reducing losses. In this study, in order to modeling and forecasting drought severity in a 37 year time period (1971-2007) in 21 meteorological stations, located in the cold semi-arid region of north-west Iran, artificial neural networks was used. The input data was annual rainfall data and annual drought precipitation index for all stations that 80% of the data (1971-2000) used for training the network and other 20% (2001-2007) used for testing it and in the next step drought severity predicted for the years 2008 to 2012 by the trained algorithm without using actual and existed data in this period. The appropriate structure for the network, based on Multi Layer Perceptron with three hidden layer, Back Propagation algorithm, Sigmoid transfer function and 10 neurons in middle layer. The results show that the artificial neural networks are well able to predict the non-linear relationship between rainfall and drought as it can simulate drought precipitation index values largely consistent with the real values with more than 97% regression and less than 5% error. So, drought can be predicted by this method in future and also it is useful in water resources management, drought management and climate change. 

---

In this study is predicted the groundwater level of Sharif Abad catchment using some artificial intelligence models. For this purpose used of monthly groundwater levels for modeling in the three observed wells located in the Sharif Abad watershed of Qom. To compare the results of the hybrid model of wavelet analysis-neural network (WNN), genetic programming (GP) multiple linear regression (MLR) and artificial neural network (ANN), two criteria of root mean squared error (RMSE) and nash-sutcliffe coefficient of efficiency (E) is used. The results of the study indicated that the WNN models provide more accurate monthly groundwater level predicted in compared to the ANN, GP and MLR models so the nash-sutcliffe coefficient in WANN model for piezometers 1, 2 and 3 are 0.98, 0.98 and 0.95, respectively.

.

---

 
Suspended particles management is one of the important issues in controlling the air pollution of cities. These particles cause and develop heart and respiratory diseases in people. Mashhad is considered as one of the main and populous cities of Iran. Because of its climatic conditions and its tourism, the city is at the highest risk of this type of pollution. We attempted to use the multi-layer perceptron (MLP) artificial neural network and a Markov chain model to predict PM₁₀ concentrations the city. We applied hourly data of CO, SO₂, PM_2.5 and temperature in late March and April 2015. Out of 1488 data series, 1300 data were used for network training and 188 data were used for validation. The results indicated the optimal performance of these methods in predicting of the amount of pollutants and also the probability of occurrence of hours with different quality of contamination. The best MLP artificial neural network model predicted the amount of pollutant particles with a coefficient of determination (R²) 0.88, index of agreement of  0.91 and a mean square error of  2.26. Also, the Markov model with average absolute error predicted about 0.1 percent of the probability of transferring the condition and the continuation of different states of air pollution caused by suspended particles.
 

---

---

  nk href="moz-extension://8b922523-7922-435a-ac74-8ddb59e9beaf/skin/s3gt_tooltip_mini.css" rel="stylesheet" type="text/css" >

---

Human thermal comfort and discomfort of many experimental and theoretical indices are calculated using the input data the indicator of climatic elements are such as wind speed, temperature, humidity, solar radiation, etc. The daily data of temperature، wind speed، relative humidity، and cloudiness between the years 1382-1392 were used. In the First step، Tmrt parameter was calculated in the Ray Man software environment. Then UTCI and PMV index values were calculated using Bioklima software. The results showed that the most severe cold temperature stress on PMV index is in the winter and late autumn and UTCI index in January and February are the coldest stress. The power of neural networks, prediction of future performance network (generalized orientation) it simply is not possible and the new model presented in this paper have been restricted Boltzmann machine-based neural networks or neural networks is used deep belief. Using this structure, metrics Mean Squared Error (MSE) and mean absolute percentage error (MAPE) benchmark ate for seven indexes derived from data gathered by three factors related to the occurrence of weather conditions and other indicators of thermal comfort of human the system was evaluated. Assessment by dividing the data into training and testing parts, and the ratios have been of two-thirds, fifty percent and one-third And two benchmark MSE and MAPE were calculated. The proposed system performance in forecasting the human thermal comfort is desirable.

---

Rainfall prediction plays an important role in flood management and flood alert. With rainfall information, it is possible to predict the occurrence of floods in a given area and take the necessary measures. Due to the fact that the three months of January, February and March are most floods and most precipitation is occurring this quarter, this study aimed to investigate the factors affecting precipitation and modeling of this quarter. For precipitation modeling, the monthly rainfall data of the Hamadid and Baranzadeh station in the statistical period (1984-2014) for 30 years as a dependent variable and climatic indexes, large-scale climatic signals including sea surface temperatures and 1000 millimeter temperatures Altitude of 500 milligrams, 200 milligrams of omega and climatic elements have been used as independent variables. Due to the nonlinear behavior of rainfall, artificial neural networks were used for modeling. Factor analysis was used to determine the best architecture for entering the neural network. For prediction of precipitation, the data that showed the most relationship with precipitation was used in four patterns, in January the fourth pattern with entropy error was 045/0, the number of input layers was 91, the best makeup was 15-1, and the correlation coefficient was 94% Was. In February, the third pattern with a correlation coefficient of 97%, entropy error, was 0.36. Percentage, number of input units was 8 units, and the best type of latency layout was 10-1. The precipitation of March with all patterns was high predictive coefficient. The first pattern with entropy error was 0.038, the number of input units was 67, the hidden layer arrangement was 17-1, the correlation coefficient was 98%.

---

The structure of urban resilience consists of four dimensions: social, economic, institutional and physical. In the desired situation, a resilient city has a strong local society with a dynamic and sustainable economy that is governed institutionally and institutionally in a participatory manner. As a result of these super-physical factors, the city is formed in a cohesive manner and no withering is observed at any point. The purpose of this study is to assess the resilience and stability of Semnan in the face of natural hazards (earthquake). The research method in this study is descriptive-analytical and its statistics and information have been obtained using a questionnaire. The statistical population of this study was citizens living in Semnan city. Using Cochranchr('39')s formula, a sample of 384 people was selected from them by random sampling method. To assess the validity (validity), using face validity, the opinions of related people were examined and its reliability was assessed using Cronbachchr('39')s test in the SPSS software environment equal to / 863. It was found to indicate high coordination and reliability of the data. In data analysis, descriptive and inferential statistical tests were used in Spss software. percentage; At the level of inferential statistics, Pearson correlation coefficient and sample titech test as well as neural network model were used to examine the relationship between variables. Findings showed that the socio-cultural dimension with an average rank of 2.59 and the physical dimension with an average rank of 3.05 and the economic dimension with an average rank of 2.17 and finally the institutional-organizational dimension with an average rank of 2.56 show the current situation of resilience in Semnan.

---

The purpose of this study was to estimate the amount of sediment of Vanai River in Borujerd. In this research, the characteristics of the sub-basins of this river have been extracted first. These specifications include the physical characteristics of the sub-basins, including the area, the environment and length of the waterways, and the characteristics of the river flow, and its sediment content. In the following, multivariate linear regression, multilevel prefabricated neural network (MLP) and radial function-based neural network (RBF) models are used to model sediment estimation. After estimating the model, the mean square error index (RMSE) was used to compare the models and select the best model. Evidence has shown that initially the MLP's neural network model had the best estimate with the lowest error rate (90.44) and then the RBF model (151.44) among the three models. The linear regression model has the highest error rate because only linear relationships between variables are considered.

---

This research study aims to investigate the effect of climatic variables, specifically precipitation, temperature, and humidity, on changes in vegetation indices of orange orchards in Hassan Abad, Darab County, using satellite data. Consequently, observational data, including orange tree phenology data and meteorological data from the agricultural weather station, were collected over a period of more than 10 years (2006 to 2016). MODIS images from 2006 to 2016 were referenced based on territorial data and 1:25000 maps from the Iran National Cartographic Center. These images were used to calculate remote sensing vegetation indices, namely the normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI). The results demonstrated that the variables of maximum humidity, minimum temperature, and precipitation have a significant positive effect on the NDVI variable. Additionally, the variables of maximum temperature and minimum humidity have a significant negative effect on both the NDVI and EVI. To determine the significance of each independent variable in predicting the dependent variables, the artificial neural network method was employed. The findings showed that the climatic elements of precipitation, minimum temperature, maximum temperature, minimum humidity, and maximum humidity had the greatest effect on EVI, with values of 0.39, 0.3, 0.13, 0.1, and 0.06 respectively. Moreover, the effect of these variables on the NDVI index is equal to their coefficients, which are 0.2, 0.28, 0.22, 0.11, and 0.17 respectively. Finally, the ARMAX regression method was used to improve the explanatory power of the model. The results indicated that this method enhanced the explanatory power of the model and reduced the forecasting error.